You are on page 1of 202

Cochrane Database of Systematic Reviews

Antibiotics for community-acquired pneumonia in children
(Review)

Lodha R, Kabra SK, Pandey RM

Lodha R, Kabra SK, Pandey RM.
Antibiotics for community-acquired pneumonia in children.
Cochrane Database of Systematic Reviews 2013, Issue 6. Art. No.: CD004874.
DOI: 10.1002/14651858.CD004874.pub4.

www.cochranelibrary.com

Antibiotics for community-acquired pneumonia in children (Review)
Copyright © 2013 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

TABLE OF CONTENTS

HEADER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
PLAIN LANGUAGE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Figure 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Figure 2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
AUTHORS’ CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
CHARACTERISTICS OF STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
DATA AND ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
Analysis 1.1. Comparison 1 Azithromycin versus erythromycin, Outcome 1 Mean age (months). . . . . . . . 66
Analysis 1.2. Comparison 1 Azithromycin versus erythromycin, Outcome 2 Male sex. . . . . . . . . . . . 67
Analysis 1.3. Comparison 1 Azithromycin versus erythromycin, Outcome 3 Wheezing present. . . . . . . . . 67
Analysis 1.4. Comparison 1 Azithromycin versus erythromycin, Outcome 4 Cure rate. . . . . . . . . . . . 68
Analysis 1.5. Comparison 1 Azithromycin versus erythromycin, Outcome 5 Failure rate. . . . . . . . . . . 69
Analysis 1.6. Comparison 1 Azithromycin versus erythromycin, Outcome 6 Side effects. . . . . . . . . . . 69
Analysis 1.7. Comparison 1 Azithromycin versus erythromycin, Outcome 7 Organisms identified by serology or
nasopharyngeal cultures. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
Analysis 1.8. Comparison 1 Azithromycin versus erythromycin, Outcome 8 Cure rate in radiographically confirmed
pneumonia. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
Analysis 1.9. Comparison 1 Azithromycin versus erythromycin, Outcome 9 Failure rate in radiographically confirmed
pneumonia. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
Analysis 2.1. Comparison 2 Clarithromycin versus erythromycin, Outcome 1 Age below 5 years. . . . . . . . 72
Analysis 2.2. Comparison 2 Clarithromycin versus erythromycin, Outcome 2 Cure rates. . . . . . . . . . . 73
Analysis 2.3. Comparison 2 Clarithromycin versus erythromycin, Outcome 3 Clinical success rate. . . . . . . 74
Analysis 2.4. Comparison 2 Clarithromycin versus erythromycin, Outcome 4 Failure rate. . . . . . . . . . 74
Analysis 2.5. Comparison 2 Clarithromycin versus erythromycin, Outcome 5 Relapse rate. . . . . . . . . . 75
Analysis 2.6. Comparison 2 Clarithromycin versus erythromycin, Outcome 6 Radiologic resolution. . . . . . . 75
Analysis 2.7. Comparison 2 Clarithromycin versus erythromycin, Outcome 7 Radiologic success. . . . . . . . 76
Analysis 2.8. Comparison 2 Clarithromycin versus erythromycin, Outcome 8 Radiologic failure. . . . . . . . 76
Analysis 2.9. Comparison 2 Clarithromycin versus erythromycin, Outcome 9 Adverse events. . . . . . . . . 77
Analysis 2.10. Comparison 2 Clarithromycin versus erythromycin, Outcome 10 Bacteriologic response. . . . . . 77
Analysis 3.1. Comparison 3 Azithromycin versus co-amoxyclavulanic acid, Outcome 1 Cure rate. . . . . . . . 78
Analysis 3.2. Comparison 3 Azithromycin versus co-amoxyclavulanic acid, Outcome 2 Failure rate. . . . . . . 78
Analysis 3.3. Comparison 3 Azithromycin versus co-amoxyclavulanic acid, Outcome 3 Improved. . . . . . . . 79
Analysis 3.4. Comparison 3 Azithromycin versus co-amoxyclavulanic acid, Outcome 4 Side effects. . . . . . . 80
Analysis 3.5. Comparison 3 Azithromycin versus co-amoxyclavulanic acid, Outcome 5 Organisms isolated. . . . . 80
Analysis 3.6. Comparison 3 Azithromycin versus co-amoxyclavulanic acid, Outcome 6 Mycoplasma serology positive. 81
Analysis 3.7. Comparison 3 Azithromycin versus co-amoxyclavulanic acid, Outcome 7 Failure rates in radiographically
confirmed pneumonia. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
Analysis 4.1. Comparison 4 Azithromycin versus amoxycillin, Outcome 1 Age in months. . . . . . . . . . . 82
Analysis 4.2. Comparison 4 Azithromycin versus amoxycillin, Outcome 2 Duration of illness. . . . . . . . . 83
Analysis 4.3. Comparison 4 Azithromycin versus amoxycillin, Outcome 3 Wheezing present. . . . . . . . . 83
Analysis 4.4. Comparison 4 Azithromycin versus amoxycillin, Outcome 4 Cure rate clinical. . . . . . . . . . 84
Analysis 4.5. Comparison 4 Azithromycin versus amoxycillin, Outcome 5 Cure rate radiological. . . . . . . . 84
Analysis 4.6. Comparison 4 Azithromycin versus amoxycillin, Outcome 6 Fever day 7. . . . . . . . . . . . 85
Antibiotics for community-acquired pneumonia in children (Review) i
Copyright © 2013 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Analysis 5.1. Comparison 5 Amoxycillin versus procaine penicillin, Outcome 1 Median age. . . . . . . . . . 85
Analysis 5.2. Comparison 5 Amoxycillin versus procaine penicillin, Outcome 2 Failure rate. . . . . . . . . . 86
Analysis 6.1. Comparison 6 Co-amoxyclavulanic acid versus amoxycillin, Outcome 1 Poor or no response. . . . . 86
Analysis 6.2. Comparison 6 Co-amoxyclavulanic acid versus amoxycillin, Outcome 2 Cure rate. . . . . . . . 87
Analysis 6.3. Comparison 6 Co-amoxyclavulanic acid versus amoxycillin, Outcome 3 Complications. . . . . . . 87
Analysis 6.4. Comparison 6 Co-amoxyclavulanic acid versus amoxycillin, Outcome 4 Age (months). . . . . . . 88
Analysis 6.5. Comparison 6 Co-amoxyclavulanic acid versus amoxycillin, Outcome 5 Weight. . . . . . . . . 89
Analysis 6.6. Comparison 6 Co-amoxyclavulanic acid versus amoxycillin, Outcome 6 Male sex. . . . . . . . . 89
Analysis 6.7. Comparison 6 Co-amoxyclavulanic acid versus amoxycillin, Outcome 7 Wheeze present. . . . . . 90
Analysis 6.8. Comparison 6 Co-amoxyclavulanic acid versus amoxycillin, Outcome 8 Side effects. . . . . . . . 90
Analysis 7.1. Comparison 7 Co-trimoxazole versus amoxycillin, Outcome 1 Age less than 1 year. . . . . . . . 91
Analysis 7.2. Comparison 7 Co-trimoxazole versus amoxycillin, Outcome 2 Male sex. . . . . . . . . . . . 92
Analysis 7.3. Comparison 7 Co-trimoxazole versus amoxycillin, Outcome 3 Mean Z score for weight. . . . . . . 92
Analysis 7.4. Comparison 7 Co-trimoxazole versus amoxycillin, Outcome 4 Non-severe pneumonia. . . . . . . 93
Analysis 7.5. Comparison 7 Co-trimoxazole versus amoxycillin, Outcome 5 Received antibiotics in previous week. . 94
Analysis 7.6. Comparison 7 Co-trimoxazole versus amoxycillin, Outcome 6 Severe pneumonia. . . . . . . . . 94
Analysis 7.7. Comparison 7 Co-trimoxazole versus amoxycillin, Outcome 7 Failure rate in non-severe pneumonia. . 95
Analysis 7.8. Comparison 7 Co-trimoxazole versus amoxycillin, Outcome 8 Failure rate severe pneumonia clinical
diagnosis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
Analysis 7.9. Comparison 7 Co-trimoxazole versus amoxycillin, Outcome 9 Failure rate radiological positive pneumonia. 96
Analysis 7.10. Comparison 7 Co-trimoxazole versus amoxycillin, Outcome 10 Failure rate radiological negative
pneumonia. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
Analysis 7.11. Comparison 7 Co-trimoxazole versus amoxycillin, Outcome 11 Death rate. . . . . . . . . . 98
Analysis 7.12. Comparison 7 Co-trimoxazole versus amoxycillin, Outcome 12 Lost to follow-up. . . . . . . . 98
Analysis 7.13. Comparison 7 Co-trimoxazole versus amoxycillin, Outcome 13 Wheeze positive. . . . . . . . 99
Analysis 7.14. Comparison 7 Co-trimoxazole versus amoxycillin, Outcome 14 Cure rate. . . . . . . . . . . 100
Analysis 7.15. Comparison 7 Co-trimoxazole versus amoxycillin, Outcome 15 Change of antibiotics. . . . . . . 100
Analysis 7.16. Comparison 7 Co-trimoxazole versus amoxycillin, Outcome 16 Failure rates after excluding study by Awasthi
2008. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
Analysis 8.1. Comparison 8 Co-trimoxazole versus procaine penicillin, Outcome 1 Age less than 1 year. . . . . . 101
Analysis 8.2. Comparison 8 Co-trimoxazole versus procaine penicillin, Outcome 2 Age 1 to 5 years. . . . . . . 102
Analysis 8.3. Comparison 8 Co-trimoxazole versus procaine penicillin, Outcome 3 Age 5 to 12 years. . . . . . . 103
Analysis 8.4. Comparison 8 Co-trimoxazole versus procaine penicillin, Outcome 4 Duration of illness in days. . . 103
Analysis 8.5. Comparison 8 Co-trimoxazole versus procaine penicillin, Outcome 5 Male sex. . . . . . . . . . 104
Analysis 8.6. Comparison 8 Co-trimoxazole versus procaine penicillin, Outcome 6 Cure rate. . . . . . . . . 105
Analysis 8.7. Comparison 8 Co-trimoxazole versus procaine penicillin, Outcome 7 Hospitalisation rate. . . . . . 105
Analysis 8.8. Comparison 8 Co-trimoxazole versus procaine penicillin, Outcome 8 Well at end of follow-up. . . . 106
Analysis 8.9. Comparison 8 Co-trimoxazole versus procaine penicillin, Outcome 9 Death. . . . . . . . . . 106
Analysis 8.10. Comparison 8 Co-trimoxazole versus procaine penicillin, Outcome 10 Treatment failure. . . . . . 107
Analysis 9.1. Comparison 9 Co-trimoxazole versus procaine penicillin and ampicillin, Outcome 1 Mean age in months. 107
Analysis 9.2. Comparison 9 Co-trimoxazole versus procaine penicillin and ampicillin, Outcome 2 Age less than 1 year. 108
Analysis 9.3. Comparison 9 Co-trimoxazole versus procaine penicillin and ampicillin, Outcome 3 Male sex. . . . 108
Analysis 9.4. Comparison 9 Co-trimoxazole versus procaine penicillin and ampicillin, Outcome 4 Cure rate. . . . 109
Analysis 9.5. Comparison 9 Co-trimoxazole versus procaine penicillin and ampicillin, Outcome 5 Hospitalisation rate. 110
Analysis 9.6. Comparison 9 Co-trimoxazole versus procaine penicillin and ampicillin, Outcome 6 Death rate. . . . 110
Analysis 10.1. Comparison 10 Cefpodoxime versus co-amoxyclavulanic acid, Outcome 1 Cure rate (response rate) at end
of treatment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
Analysis 10.2. Comparison 10 Cefpodoxime versus co-amoxyclavulanic acid, Outcome 2 Mean age (months). . . . 112
Analysis 10.3. Comparison 10 Cefpodoxime versus co-amoxyclavulanic acid, Outcome 3 Adverse effects. . . . . 112
Analysis 10.4. Comparison 10 Cefpodoxime versus co-amoxyclavulanic acid, Outcome 4 Age in years. . . . . . 113
Analysis 10.5. Comparison 10 Cefpodoxime versus co-amoxyclavulanic acid, Outcome 5 Follow-up. . . . . . . 113
Analysis 11.1. Comparison 11 Chloramphenicol versus penicillin plus gentamicin, Outcome 1 Adverse events. . . 114
Analysis 11.2. Comparison 11 Chloramphenicol versus penicillin plus gentamicin, Outcome 2 Death. . . . . . 114
Antibiotics for community-acquired pneumonia in children (Review) ii
Copyright © 2013 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Analysis 11.3. Comparison 11 Chloramphenicol versus penicillin plus gentamicin, Outcome 3 Change of antibiotics. 115
Analysis 11.4. Comparison 11 Chloramphenicol versus penicillin plus gentamicin, Outcome 4 Readmission before 30
days. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
Analysis 11.5. Comparison 11 Chloramphenicol versus penicillin plus gentamicin, Outcome 5 Absconded. . . . . 116
Analysis 11.6. Comparison 11 Chloramphenicol versus penicillin plus gentamicin, Outcome 6 Age (months). . . . 117
Analysis 11.7. Comparison 11 Chloramphenicol versus penicillin plus gentamicin, Outcome 7 Culture positive. . . 117
Analysis 11.8. Comparison 11 Chloramphenicol versus penicillin plus gentamicin, Outcome 8 Male sex. . . . . 118
Analysis 11.9. Comparison 11 Chloramphenicol versus penicillin plus gentamicin, Outcome 9 Received antibiotics in
previous 1 week. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
Analysis 11.10. Comparison 11 Chloramphenicol versus penicillin plus gentamicin, Outcome 10 Lost to follow-up. 119
Analysis 12.1. Comparison 12 Chloramphenicol with ampicillin and gentamicin, Outcome 1 Mean age. . . . . 119
Analysis 12.2. Comparison 12 Chloramphenicol with ampicillin and gentamicin, Outcome 2 Male sex. . . . . . 120
Analysis 12.3. Comparison 12 Chloramphenicol with ampicillin and gentamicin, Outcome 3 Number received antibiotics
in past 7 days. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
Analysis 12.4. Comparison 12 Chloramphenicol with ampicillin and gentamicin, Outcome 4 Failure rates on day 5. 121
Analysis 12.5. Comparison 12 Chloramphenicol with ampicillin and gentamicin, Outcome 5 Failure rates on day 10. 121
Analysis 12.6. Comparison 12 Chloramphenicol with ampicillin and gentamicin, Outcome 6 Failure rates on day 21. 122
Analysis 12.7. Comparison 12 Chloramphenicol with ampicillin and gentamicin, Outcome 7 Need for change in antibiotics
(day 5). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
Analysis 12.8. Comparison 12 Chloramphenicol with ampicillin and gentamicin, Outcome 8 Need for change in antibiotics
(day 10). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
Analysis 12.9. Comparison 12 Chloramphenicol with ampicillin and gentamicin, Outcome 9 Need for change in antibiotics
(day 21). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
Analysis 12.10. Comparison 12 Chloramphenicol with ampicillin and gentamicin, Outcome 10 Death rates. . . . 125
Analysis 13.1. Comparison 13 Chloramphenicol plus penicillin versus ceftriaxone, Outcome 1 Cure rates. . . . . 125
Analysis 14.1. Comparison 14 Chloramphenicol versus chloramphenicol plus penicillin, Outcome 1 Need for change of
antibiotics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
Analysis 14.2. Comparison 14 Chloramphenicol versus chloramphenicol plus penicillin, Outcome 2 Death rates. . 126
Analysis 14.3. Comparison 14 Chloramphenicol versus chloramphenicol plus penicillin, Outcome 3 Lost to follow-up. 127
Analysis 15.1. Comparison 15 Ampicillin alone versus penicillin with chloramphenicol, Outcome 1 Cure rates. . . 127
Analysis 15.2. Comparison 15 Ampicillin alone versus penicillin with chloramphenicol, Outcome 2 Age (months). . 128
Analysis 15.3. Comparison 15 Ampicillin alone versus penicillin with chloramphenicol, Outcome 3 Male sex. . . . 128
Analysis 15.4. Comparison 15 Ampicillin alone versus penicillin with chloramphenicol, Outcome 4 Duration of hospital
stay. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
Analysis 15.5. Comparison 15 Ampicillin alone versus penicillin with chloramphenicol, Outcome 5 Grade 2 to 4
malnutrition. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
Analysis 16.1. Comparison 16 Benzathine penicillin versus procaine penicillin, Outcome 1 Cure rate. . . . . . 130
Analysis 16.2. Comparison 16 Benzathine penicillin versus procaine penicillin, Outcome 2 Failure rate. . . . . . 130
Analysis 16.3. Comparison 16 Benzathine penicillin versus procaine penicillin, Outcome 3 Male sex. . . . . . . 131
Analysis 16.4. Comparison 16 Benzathine penicillin versus procaine penicillin, Outcome 4 Age between 2 to 6 years. 132
Analysis 16.5. Comparison 16 Benzathine penicillin versus procaine penicillin, Outcome 5 Age between 7 to 12 years. 133
Analysis 16.6. Comparison 16 Benzathine penicillin versus procaine penicillin, Outcome 6 Lost to follow-up. . . . 133
Analysis 16.7. Comparison 16 Benzathine penicillin versus procaine penicillin, Outcome 7 Failure rates in radiographically
confirmed pneumonia. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
Analysis 17.1. Comparison 17 Amoxycillin versus penicillin, Outcome 1 Nasopharyngeal aspirates for S. pneumoniae. 135
Analysis 17.2. Comparison 17 Amoxycillin versus penicillin, Outcome 2 Age less than 1 year. . . . . . . . . 135
Analysis 17.3. Comparison 17 Amoxycillin versus penicillin, Outcome 3 Male sex. . . . . . . . . . . . . 136
Analysis 17.4. Comparison 17 Amoxycillin versus penicillin, Outcome 4 Weight below 2 Z score (indicating severe
malnutrition). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
Analysis 17.5. Comparison 17 Amoxycillin versus penicillin, Outcome 5 Breast fed. . . . . . . . . . . . . 137
Analysis 17.6. Comparison 17 Amoxycillin versus penicillin, Outcome 6 Received antibiotics in last 7 days. . . . 138
Analysis 17.7. Comparison 17 Amoxycillin versus penicillin, Outcome 7 Failure rate at 48 hours. . . . . . . . 138
Analysis 17.8. Comparison 17 Amoxycillin versus penicillin, Outcome 8 Failure rate on day 5. . . . . . . . . 139
Antibiotics for community-acquired pneumonia in children (Review) iii
Copyright © 2013 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Comparison 19 Amoxycillin with cefuroxime. 158 Analysis 24. . . . . . influenzae. . . Outcome 4 Failure rates. . . . . . 150 Analysis 20. . .3. . . . . . 158 Analysis 24. Outcome 12 Respiratory syncytial virus (RSV) in nasopharyngeal swabs. . . . . . . . Comparison 23 Cefuroxime with clarithromycin. . Outcome 1 Mean age. . 155 Analysis 22. . . . . . . Comparison 19 Amoxycillin with cefuroxime. . . . . . . .4. 140 Analysis 17. . . 151 Analysis 20. . . . . . .1. . . . . . .14. . . . . . Comparison 17 Amoxycillin versus penicillin. .13. . 162 Analysis 24. . . . . . Comparison 24 Co-trimoxazole versus chloramphenicol. . . . .7. . Outcome 9 Need for change in antibiotics. . . . Comparison 24 Co-trimoxazole versus chloramphenicol. . .2. . Comparison 24 Co-trimoxazole versus chloramphenicol. Outcome 2 Male sex. . . 155 Analysis 23. . . . . . . . . . . . . . . . . . . 156 Analysis 23. .9. . Outcome 2 Male sex. . . 157 Analysis 23. . . . . Outcome 1 Number of children less than 1 year age. 161 Analysis 24. . . . Comparison 24 Co-trimoxazole versus chloramphenicol. Comparison 24 Co-trimoxazole versus chloramphenicol. . 142 Analysis 17. . 150 Analysis 20.4. . . .2. . . . Outcome 6 Relapse rates. . . . . 151 Analysis 21.8. . . 142 Analysis 18. 145 Analysis 18. 149 Analysis 19.2. . . Outcome 1 Mean age. . . . Comparison 18 Amoxycillin with IV ampicillin. . . . . Outcome 15 Failure rate on day 5 in radiographically confirmed pneumonia. . . .4. . . . . Comparison 24 Co-trimoxazole versus chloramphenicol. Comparison 22 Levofloxacin with comparator (co-amoxyclavulanic acid/ceftriaxone). Comparison 23 Cefuroxime with clarithromycin. . . . Outcome 14 Blood culture positive for S. . . . Outcome 2 Male sex. . .6. Outcome 6 Failure rate. . Outcome 7 Excluded. . . . . . .2. . . . . . . 145 Analysis 18. . . . . . . .1. Outcome 2 Male sex. . . . Outcome 3 Cure rates. 139 Analysis 17. . . . . Comparison 20 Amoxycillin with clarithromycin. . . . .3. . . Outcome 5 Cure rate. Published by John Wiley & Sons. 144 Analysis 18. 143 Analysis 18. . . . . Comparison 20 Amoxycillin with clarithromycin. .3. Comparison 22 Levofloxacin with comparator (co-amoxyclavulanic acid/ceftriaxone). . Comparison 18 Amoxycillin with IV ampicillin.9. Outcome 1 Age below one year. 152 Analysis 21. . . . . 148 Analysis 19. . . . . 147 Analysis 18. . . . . 157 Analysis 24. . . . Comparison 24 Co-trimoxazole versus chloramphenicol. . . . .1.9. . . . . . pneumoniae. . . . Outcome 10 Death rate. . . . .6. . .5. . .1. . . . . . Analysis 17. . . . . . . . . 161 Analysis 24. . . . . . . . .1. . . . . . . . .1. . . 146 Analysis 18. .10. . . . Comparison 17 Amoxycillin versus penicillin. . . . . . . . Outcome 13 Mean age. . . . . . Outcome 10 Death rates. . . Comparison 18 Amoxycillin with IV ampicillin. Outcome 4 Wheezing positive. . . . . . . . Outcome 3 Weight for age. .4. Ltd. Comparison 24 Co-trimoxazole versus chloramphenicol. Comparison 17 Amoxycillin versus penicillin. . . Comparison 24 Co-trimoxazole versus chloramphenicol. .12. . Outcome 11 Nasopharyngeal H. . Outcome 1 Age in months. . . . . . . . . . 154 Analysis 22. . . Comparison 19 Amoxycillin with cefuroxime. . . . . .8. . . Comparison 18 Amoxycillin with IV ampicillin. Outcome 3 Failure rates. . . . . . . . Comparison 18 Amoxycillin with IV ampicillin. 141 Analysis 17. . . . 159 Analysis 24. . . 147 Analysis 19. . Comparison 17 Amoxycillin versus penicillin. . . . . .1. . . . . Outcome 4 Failure rates. . . 160 Analysis 24. . . .4. . Comparison 17 Amoxycillin versus penicillin. . . . Outcome 9 Protocol violation. . . . . Comparison 17 Amoxycillin versus penicillin. . . . .10. . . . Outcome 8 Lost to follow-up. . . . . 149 Analysis 20. . . . . . . Outcome 3 Wheezing in infants. Comparison 20 Amoxycillin with clarithromycin. Comparison 21 Penicillin and gentamycin with co-amoxyclavulanic acid. . . . . . . . . . . Comparison 18 Amoxycillin with IV ampicillin. . Comparison 22 Levofloxacin with comparator (co-amoxyclavulanic acid/ceftriaxone). Outcome 3 Cure rates. . Outcome 3 Numbers received antibiotics in past 1 week. . . . . 156 Analysis 23. Outcome 8 Relapse rate. . . . . . . Comparison 19 Amoxycillin with cefuroxime. . . Outcome 4 Wheezing in age group one to five years. . . . Outcome 3 Cure rates. Comparison 18 Amoxycillin with IV ampicillin. . . . . .15. . . . . . . . Comparison 22 Levofloxacin with comparator (co-amoxyclavulanic acid/ceftriaxone). . . . . . . . 153 Analysis 22. . . . . . . . . . .5. . . . . . . . 152 Analysis 21.2. Comparison 20 Amoxycillin with clarithromycin. . Outcome 2 Male sex. Comparison 21 Penicillin and gentamycin with co-amoxyclavulanic acid. 154 Analysis 22. 148 Analysis 19. . . . . . Comparison 23 Cefuroxime with clarithromycin. 141 Analysis 17. . Comparison 23 Cefuroxime with clarithromycin. 140 Analysis 17. . Outcome 2 Male sex. . . . . . Comparison 18 Amoxycillin with IV ampicillin. .3. . Outcome 5 Failure rates. . . . . Outcome 7 Death rates. . . . . . . . . . . 144 Analysis 18. 146 Analysis 18. . . . . . . . . . . Outcome 4 Cure rates. . . . . . . . . . Outcome 2 Male sex. . . Comparison 17 Amoxycillin versus penicillin. . . .7.3. . . .11. . . . . . . . Outcome 1 Mean age in months. . Outcome 1 Mean age. .2. . 162 Antibiotics for community-acquired pneumonia in children (Review) iv Copyright © 2013 The Cochrane Collaboration. . 160 Analysis 24. Outcome 4 Failure rates. .3. . Comparison 18 Amoxycillin with IV ampicillin. . . . . . . . . . . . . . . . . . . . . . . . . . . .2. . . Comparison 24 Co-trimoxazole versus chloramphenicol. . Comparison 21 Penicillin and gentamycin with co-amoxyclavulanic acid. . . . 159 Analysis 24. . . . . . .4.3. Outcome 9 Failure rate on day 14. . . .

. . . . . . . . . . . . . . . .13. . . . Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia. . . . 177 Analysis 27. . . . . . . Outcome 2 Age below 12 months. . . .5. . . . . . . . . . Outcome 4 Received antibiotics before enrolment. Comparison 26 Oxacillin ceftriaxone versus co-amoxyclavulanic acid. Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia. . . . . . . . . . . . Outcome 15 Death rates. . . . . . . . . . . Outcome 3 Mean number of days before admission. . . . . Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia.2. . . . . . . . . . . . . . . . . . . . . .8. . . . . . . . . . 169 Analysis 27. . . . . . . . . Comparison 25 Ceftibuten versus cefuroxime. . . . . . . .12. . Outcome 8 Failure rate in children below 5 years of age. . . . Outcome 4 Cure rate. . .4. Outcome 4 Children with wheezing. . . . . . . .3. Outcome 16 Lost to follow-up. . . . . . . . . . . . . . 167 Analysis 26. . . . Outcome 11 Organisms isolated on blood culture or lung puncture. 176 Analysis 27. . . . . . . . . . . . . . . . . 173 Analysis 27. . . . . . . . . . Outcome 5 Failure rates. . . . . . . . . . . . . 180 Analysis 27. . . . . . . . . . . . . . . . . . . . Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia. . . . . . . Outcome 10 Failure rate in children receiving cotrimoxazole or injectable penicillin. . . . . . . .1. . . . . . . . . . . . . . . . . . . . Outcome 6 Mean time for improvement in tachypnoea. Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia. . .16. . . . . . . . . 181 Analysis 27. . . . . . . . . Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia.5. . . . . .6. . . . . . . . . . . . . . . . . .10. . . . . . . . . . Ltd. . . .6. . . . . . . . . 168 Analysis 26. Outcome 17 Cure rate. . . . . . . . . . Outcome 5 RSV positivity. .2. . . . . . . Outcome 1 Male sex. . . Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia. Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia. . . . . . . . Outcome 7 Failure rates on day 6. . . Outcome 3 Adverse reaction. .3. . . Comparison 25 Ceftibuten versus cefuroxime. . . . . . . . 165 Analysis 26. . . . . Comparison 26 Oxacillin ceftriaxone versus co-amoxyclavulanic acid. . . 181 Analysis 27. . . . . . . . 177 Analysis 27. . . .11. . . . . . . 167 Analysis 26. 168 Analysis 26. . . . .4. . Outcome 1 Male sex. . . Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia.11. . . . . Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia. Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia. .18. . . . .1. . . . . . Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia. . . . . . . .5. . . . . .7. . Analysis 24. . Comparison 26 Oxacillin ceftriaxone versus co-amoxyclavulanic acid. . .7. . . . . . . 171 Analysis 27. . . . . . . . . . Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia. Comparison 26 Oxacillin ceftriaxone versus co-amoxyclavulanic acid. . . . . 164 Analysis 25. . . . . . . . . . . . . . . . .9. 179 Analysis 27. . . . . . . . . . . . . . . . . . . . Outcome 9 Failure rates in children receiving oral amoxicillin or injectable antibiotics. . . . . . . . . . . . Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia. . Outcome 7 Mean length of stay. . . Outcome 6 Failure rates on day 3. . .3. . . . . . . . 172 Analysis 27. . . . Comparison 25 Ceftibuten versus cefuroxime. . . . . . . 171 Analysis 27. . . . . . . . . . . . . Outcome 2 Male sex. . . . . . . . . . . . . Published by John Wiley & Sons. . . . . . . . Outcome 13 Hospitalisation. . . . 175 Analysis 27. . . . . . . . . . . . Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia. . . . . . . . . . . . . Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia. . Outcome 18 Failure rates in radiographically confirmed-pneumonia. . . . . . . . . .15. 164 Analysis 25. . . . Outcome 5 Failure rate. . . . Comparison 25 Ceftibuten versus cefuroxime. . . . . . . . Outcome 12 Failure rate after removing one study. . . . . . . . . . . . . . . . . . . . . .4. . . . . . . . . . . . . . . . . . . . . . Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia. . . . . . . . . . . . . . . . . . . . . 166 Analysis 26. . . . . Comparison 26 Oxacillin ceftriaxone versus co-amoxyclavulanic acid. . . Outcome 14 Relapse rates. . Comparison 26 Oxacillin ceftriaxone versus co-amoxyclavulanic acid. . Outcome 3 Received antibiotics in the past week. . . . 163 Analysis 25. . . . . . . . . . . . . . 178 Analysis 27. . . . . .14. . . Comparison 24 Co-trimoxazole versus chloramphenicol. . . 174 Analysis 27. . . . . . . 165 Analysis 25. . . .2. . . . Comparison 25 Ceftibuten versus cefuroxime. . . . . . 179 Analysis 27. . . Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia. . . . . . . . . 163 Analysis 25. . . . . . . . . . . . . . . . 169 Analysis 27. Outcome 2 Positive for microbial agent. 182 Antibiotics for community-acquired pneumonia in children (Review) v Copyright © 2013 The Cochrane Collaboration. . . . . . . . .17. . . . . . . . . . . . . 170 Analysis 27. . . . . . . . . . Outcome 11 Failure rate in children treated with oral or parenteral antibiotics on ambulatory basis. . . .

. .3. . . . . . . . . . . . . . . . 186 Analysis 30. . . . 194 INDEX TERMS . . . . . . . . . . . 187 Analysis 30. . . . . . . . . . . . . . . . . . . . . . . . . .1. . . Outcome 1 Age (mean/median). . . .1. . . . . . . . . . . . . . . . . . . . . . . . .3. . . Comparison 29 Amoxycillin versus cefpodoxime. . . . . . . . 195 Antibiotics for community-acquired pneumonia in children (Review) vi Copyright © 2013 The Cochrane Collaboration. . . . Outcome 2 Male sex. .2. . .19. . . . . . . . . . . . . . . . . . . . . . . Published by John Wiley & Sons. . . . . Comparison 30 Amoxycillin versus chloramphenicol. . . . 185 Analysis 29. . . . . . . . . . . . . . . . . . Ltd. . . . . . . . . 194 SOURCES OF SUPPORT . . . . . . . . . . Outcome 2 Male sex. . . . . . . . . . . Comparison 29 Amoxycillin versus cefpodoxime. . . Outcome 1 Age in months. . . . . . . . . . . . . . . . . . . . . . . . . . . . . Outcome 1 Children below 1 year of age. . . . . . . . . . . Outcome 3 Cure rate. . . . . . .2. . Comparison 30 Amoxycillin versus chloramphenicol. . . . . Comparison 30 Amoxycillin versus chloramphenicol. . . . . . . . . . . . 182 Analysis 28. . . . . . . . . . . . Comparison 30 Amoxycillin versus chloramphenicol. . . .2. 183 Analysis 28. . . . . . . . . . . . . . . . . . . 183 Analysis 28. . 194 DIFFERENCES BETWEEN PROTOCOL AND REVIEW . 187 Analysis 30. . . . 194 DECLARATIONS OF INTEREST . . . . . . . Outcome 4 Failure rates. . . . . . . . . . . . . . . . 186 Analysis 30. Outcome 19 Death rates after removing one study. . .4. . . . . . . . . . . . . . . . . . Comparison 28 Co-trimoxazole versus co-amoxyclavulanic acid. . Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia. . . . . . . .1. . . . . . Comparison 29 Amoxycillin versus cefpodoxime. Outcome 3 Response/cure rate. . . . . . . . . . . . . . . . . . . Comparison 28 Co-trimoxazole versus co-amoxyclavulanic acid. . . . . . . . . . . . . . Outcome 2 Male sex. . 188 APPENDICES . 193 HISTORY . . . . . . . . . . 188 ADDITIONAL TABLES . . . . . . . .3. . . . . . . . . . . . . . . . . Analysis 27. . . . . . Outcome 3 Failure rate. . . . . . 184 Analysis 29. . . . 189 WHAT’S NEW . . . . . Comparison 28 Co-trimoxazole versus co-amoxyclavulanic acid. . . . . . . . . 193 CONTRIBUTIONS OF AUTHORS . . . . . . . . . 185 Analysis 29. . . .

Ansari Nagar. New Delhi. comparing at least two antibiotics for CAP within hospital or ambulatory (outpatient) settings. All India Institute of Medical Sciences. MEDLINE (1966 to October week 4. Sushil K Kabra2 . Citation: Lodha R. Department of Pediatrics.: CD004874. EMBASE (1990 to November 2012). Ansari Nagar. All India Institute of Medical Sciences.1002/14651858. skkabra@rediffmail. India.com. Antibiotics for community-acquired pneumonia in children (Review) 1 Copyright © 2013 The Cochrane Collaboration. 2 Pediatric Pulmonology Division. Ansari Nagar.188 children. No. Kabra SK. suggesting the evidence for these comparisons was of high quality compared to other comparisons.pub4. Department of Pediatrics. classifying them as good quality studies. oral amoxycillin with injectable penicillin/ampicillin and chloramphenicol with ampicillin/penicillin and studies were of good quality. published in Issue 6. Published by John Wiley & Sons. Editorial group: Cochrane Acute Respiratory Infections Group. Review content assessed as up-to-date: 7 November 2012. Objectives To identify effective antibiotic drug therapies for community-acquired pneumonia (CAP) of varying severity in children by comparing various antibiotics. Copyright © 2013 The Cochrane Collaboration. Antibiotics for community-acquired pneumonia in children.CD004874. Search methods We searched CENTRAL 2012. Assessment of quality of study revealed that 5 out of 29 studies were double-blind and allocation concealment was adequate. All India Institute of Medical Sciences. Ltd. All India Institute of Medical Sciences (AIIMS). There was more than one study comparing co-trimoxazole with amoxycillin. Selection criteria Randomised controlled trials (RCTs) in children of either sex. Main results We included 29 trials. Issue 10. None compared antibiotics with placebo.com. Ansari Nagar.[Intervention Review] Antibiotics for community-acquired pneumonia in children Rakesh Lodha1 . Ravindra M Pandey3 1 Department of Pediatrics. India Contact address: Sushil K Kabra. Web of Science (2009 to November 2012) and LILACS (2009 to November 2012). DOI: 10. skkabra@hotmail. Ltd. Published by John Wiley & Sons. . Art. Cochrane Database of Systematic Reviews 2013. 2012). India. 3 Department of Biostatistics. Issue 6. India. ABSTRACT Background Pneumonia caused by bacterial pathogens is the leading cause of mortality in children in low-income countries. Data collection and analysis Two review authors independently extracted data from the full articles of selected studies. which enrolled 14. 110029. comparing multiple antibiotics. Publication status and date: New search for studies and content updated (conclusions changed). 2013. CINAHL (2009 to November 2012). Another 12 studies were unblinded but had adequate allocation concealment. Pandey RM. Pediatric Pulmonology Division. Early administration of antibiotics improves outcomes.

24).51) and cure rates (OR 1. co-amoxyclavulanic acid and cefpodoxime may be alternative second-line drugs. amoxycillin compared with co- trimoxazole had similar failure rates (odds ratio (OR) 1.89). amoxycillin is an alternative to co-trimoxazole. with an estimated incidence of 0. 95% CI 0. One study involved 1116 children.13.05 episodes per child-year in low-income and high-income (Liu 2012). a decreased level of oxygen). Wherever information on adverse events was available.18.e. Reports of adverse events were not available in many studies. Most cases of community-acquired pneumonia (CAP) in low-income countries are caused by bacteria. Ltd. Limitations of this review are that only five studies met all the quality assessment criteria and for most comparisons of the efficacy of antibiotics only one or two studies were available. respectively.56 to 1. The other alternative drugs for such patients are co- amoxyclavulanic acid and cefuroxime. PLAIN LANGUAGE SUMMARY Different antibiotics for community-acquired pneumonia in otherwise healthy children younger than 18 years of age in hospital and outpatient settings Pneumonia is the leading cause of mortality in children under five years of age.6 million deaths in children below five aged less than five years. For children hospitalised with severe and very severe CAP. Published by John Wiley & Sons. In children with severe pneumonia without hypoxaemia.188 children and comparing antibiotics for treatment of CAP in children.03. There is a need for more studies with radiographically confirmed pneumonia in larger patient populations and similar methodologies to compare newer antibiotics. oral antibiotics (amoxycillin/co-trimoxazole) compared with injectable peni- cillin had similar failure rates (OR 0. In very severe CAP.In ambulatory settings. Reducing mortality due to pneumonia may help in countries. 95% CI 0. Pakistan (10 million) and Bangladesh. 95% CI 0. it did not differ between two drugs compared except that gastrointestinal side effects were more commonly reported with erythromycin compared to azithromycin.38 to 3.82). Oral amoxycillin in children with severe pneumonia without hypoxia (i. We found that for outpatient treatment of pneumonia. China (21 million).07). Children with severe pneumonia without hypoxaemia can be treated with oral amoxycillin in an ambulatory setting. 1. 95% CI 0. these can be used as second-line therapies. may be effective. 95% CI 0. out of 7. amoxycillin is an alternative treatment to co-trimoxazole. death rates were higher in children receiving chloramphenicol compared to those receiving penicillin/ampicillin plus gentamicin (OR 1.76 to 2. . This systematic review identified 29 randomised controlled trials from many different countries enrolling 14. Recommendations in this review are applicable to countries with high case fatalities due to pneumonia in children without underlying morbidities and where point of care tests for identification of aetiological agents for pneumonia are not available. For very severe pneumonia.91 to 1. Most were single studies only.34) and relapse rates (OR 1. BACKGROUND in India (43 million). Three studies involved 3952 children.84.3%) deaths were due to pneumonia and 0. 95% confidence interval (CI) 0. for treatment of World Health Organization (WHO) defined non-severe CAP. Six studies involved 4331 children below 18 years of age.29 years of age. Indonesia and Nigeria (six million each) (Rudan Pneumonia is the leading single cause of mortality in children 2008). hospitalisation rates (OR 1.4 million (18. With limited data on other antibiotics. and who are feeding well. It is estimated that a total of around 156 reducing childhood and under five-year old mortality rates (Liu million new episodes occur each year and most of these occur Antibiotics for community-acquired pneumonia in children (Review) 2 Copyright © 2013 The Cochrane Collaboration.34 to 4. Until more studies are available.56 to 1. Authors’ conclusions For treatment of patients with CAP in ambulatory settings. penicillin/ampicillin plus gentamycin is superior to chloramphenicol.25. In 2010.28. a combination of penicillin or ampicillin and gentamycin is more effective than chloramphenicol alone.

co-trimoxazole. years of age. nasopharyngeal aspiration and immune assays of blood and azole (IBIS 1999. Since clinical or radiological findings cannot dif- under five years with pneumonia are Streptococcus pneumoniae (S.H. Some studies carried out nasopharyngeal aspirates and (identification of pneumonia on clinical symptoms/signs and ad- identification of virus and atypical organisms. it does not have the necessary sensitivity and specificity to identify aetio- Why it is important to do this review logical agents (i e. Administration of the most appropriate affect the outcome. It is not easy to identify aetiological agents in children In infants under three months of age. Straus most cases. The commonest bacterial pathogens isolated in children viral infection. common pathogens include with pneumonia. How the intervention might work low five years of age is caused by viral agents. The WHO has provided guidelines for early diagnosis and assessment of the severity of pneumonia on the Description of the condition basis of clinical features (WHOYISG 1999) and suggests adminis- Pneumonia is defined as an infection of the lung parenchyma (alve. It is difficult to identify the causative or. Ishiwada 1993. 1999). . ferentiate viral or bacterial pneumonia and due to the absence of pneumoniae) (30% to 50%) and Haemophilus influenzae (H. diagnosis of pneumonia is based on antibiotic as the first-line treatment may improve the outcome of clinical criteria. The yield from blood cultures is too low (5% to 15% 1998). either in hospital or in an ambulatory setting. pneumoniae) and Haemophilus influenzae (H. treatment regimens may be different in high- income and low-income countries. gram-negative bacilli and Staphylo- mortality due to pneumonia. studies conducted in the same urine tests. The causative organisms are different is relied upon in most instances. This is necessary in view of the in high-income countries and include more viral and atypical or- inability of most commonly available laboratory tests to identify ganisms (Gendrel 1997. tration of co-trimoxazole as a first-line drug. Ltd. The last review of all available randomised con- may lead to antibiotic prescription even for those cases caused by trolled trials (RCTs) on antibiotics used for pneumonia in children Antibiotics for community-acquired pneumonia in children (Review) 3 Copyright © 2013 The Cochrane Collaboration. it is important to know which works best for pneumo- ministration of antibiotics for all clinically diagnosed pneumonia nia in children. Therefore. Therefore. influenzae) (Berman 1990). However. Many antibiotics are prescribed to treat pneumonia. antibiotics for community-acquired pneumonia (CAP) include ganism in most cases of pneumonia. and 50% of deaths due to tibiotics in countries with high case fatalities due to pneumonia is pneumonia in this age group are attributed to these two organisms recommended by the World health Organization (WHO). rates with co-trimoxazole treatment when compared to treatment There are few studies that document the aetiology of pneumonia with amoxycillin in severe and radiologically confirmed pneumo- in children below five years of age from low-income countries. To meet the public health goal of reducing child S. Wubbel causative pathogens. Treatment of pneumonia includes administration pneumonia. Published by John Wiley & Sons. point of care tests for routine use. Numazaki 2004. in. with the commonest Pneumonia is the leading cause of mortality in children below five viral pathogen being respiratory syncytial virus (Maitreyi 2000). influenzae. or in ambulatory settings. (Shann 1995). Obtaining an X-ray film in all Empirical antibiotic administration is the mainstay of treatment suspected pneumonia cases may not be cost-effective as it does not of pneumonia in children. nia (Straus 1998). drugs. oral cephalosporins and macrolide fication of the aetiologic agents include blood culture. lung punc. A meta-analysis of all the trials on pneumonia Most studies carried out blood cultures for bacterial aetiology of based on the case-management approach proposed by the WHO pneumonia. A review of 14 stud. However. Studies using nasopharyngeal aspirates for identification of viral agents suggest that about 40% of pneumonia in children be. Ad. Lung puncture is an invasive procedure associated with time period showed good clinical efficacy of oral co-trimoxazole for significant morbidity and hence cannot be performed routinely in non-severe pneumonia (Awasthi 2008. of antibiotics. Selection Description of the intervention of first-line antibiotics for empirical treatment of pneumonia is Administration of appropriate antibiotics at an early stage of pneu- crucial for office practice as well as public health. empirical treatment with an- fluenzae) (10% to 30%) (Falade 2011). the common severities of pneumonia. it is important to reduce mortality due to pneumonia by appropriate intervention in the form of antibiotics. Timothy 1993). The reference standard for di- agnosis of pneumonia is X-ray film of the chest. Despite evidence of rising bacterial resistance to co-trimox- ture. empirical antibiotic administration coccus (WHOYISG 1999). Rasmussen 1997. one study reported a doubling of clinical failure for bacterial pathogens) to be relied upon (MacCracken 2000). The methods used for identi. To reduce the infant and under five-year child mor- tality rate. ministration of empirical antimicrobial agents) has found a reduc- ies involving 1096 lung aspirates taken from hospitalised children tion in overall mortality as well as pneumonia-related mortality prior to administration of antibiotics reported bacterial pathogens (Sazawal 2003). particularly when the causative agent is bacterial. amoxycillin. pneumoniae. Antibiotics are administered in hospital bacterial pathogens identified were Streptococcus pneumoniae (S. Therefore. In 27% of patients.2012). Various antibiotics have been used for varying in 62% of cases (Berman 1990). The commonly used oli) by microbial agents. bacterial or viral). monia improves the outcome of the illness.

Length of hospital stay: duration of total hospital stay (from day of admission to discharge) in days. METHODS 3. but quired pneumonia (CAP) of varying severity in children by com. convulsions. Bari 2011. paring various antibiotics. 2. with oral antibiotics. Nogeova 1997. Relapse rate: defined as children declared ’cured’. Types of studies 6. vaso-pressure agents. trolled Trials (CENTRAL) 2012. time. drawing. five new trials 2. There. EM- BASE (September 2009 to November 2012). Loss to follow-up or withdrawal from the study at any time after recruitment indicated failure in the analysis. Additional interventions used: any additional intervention in the form of mechanical ventilation. Additional information on the epi. Mortality rate. etc.thecochranelibrary. Antibiotics for community-acquired pneumonia in children (Review) 4 Copyright © 2013 The Cochrane Collaboration. developing recurrence of disease at follow-up in a defined period. We considered only those studies using the case definition of pneumonia (as given by the WHO) or radiologically confirmed pneumonia in this review. We Types of outcome measures used the Cochrane Highly Sensitive Search Strategy to identify randomised trials in MEDLINE: sensitivity. We included children under 18 years of age with CAP treated in a hospital or community setting. Randomised controlled trials (RCTs) comparing antibiotics for CAP in children. Hospitalisation rate (in outpatient studies only): defined as the need for hospitalisation in children who were getting treatment or in an ambulatory (outpatient) setting. Soofi is the presence of any of the following: development of chest in- 2012) have been published. Clinical cure.was published in 2010 (Kabra 2010). . we have updated this review and included new data and also completion of treatment. Published by John Wiley & Sons. respiratory rate above the age-specific cut-off point on fore. Secondary outcomes OBJECTIVES The clinically relevant outcome measures were as follows. Treatment failure rates. The definition of treatment failure (Ambroggio 2012. Web of Science (Appendix 4) and and involves clinical recovery by the end of treatment. 1. Since then. MEDLINE (September 2009 to October week 4. To search CENTRAL and MEDLINE we combined the follow- ing search strategy with the validated search strategy for iden- tifying child studies developed by Boluyt (Boluyt 2008).com (accessed 7 November 2012). We excluded studies describing pneumonia post-hospitalisation in immunocompromised patients Electronic searches (for example. Ribeiro 2011. intramuscular route or orally) with another are in Appendix 1. drowsiness or inability to drink at any demiology of pneumonia in children has been published. Ltd. brary. or oxygen saturation of less than 90% carried out a meta-analysis on the treatment of severe pneumonia (measured by pulse oximetry) after completion of the treatment. Issue 10. 2012). We checked the cross-references of all the studies Types of participants manually. antibiotic for the treatment of CAP. Ovid format (Lefebvre 2011). 4. LILACS (Appendix 5). CINAHL (Appendix 3). 5. part of The Cochrane Li- mune deficient state. steroids. Web of Science (2009 to November 2012) and We compared any intervention with antibiotics (administered by LILACS (2009 to November 2012). www.and precision-max- imising version (2008 revision). CINAHL (2009 to Types of interventions November 2012). Primary outcomes We adapted the search strategy to search EMBASE (Appendix 1. To identify effective antibiotic drug therapies for community-ac. For this update we searched the Cochrane Central Register of Con- derlying illnesses like congenital heart disease or those in an im. Search methods for identification of studies We retrieved studies through a search strategy which included cross-referencing. Details of the previous search intravenous route. The definition of clinical cure is symptomatic 2). following surgical procedures) or patients with un. Need for change in antibiotics: children required change in Criteria for considering studies for this review antibiotics from the primary regimen.

Allocation concealment: assessed as yes. 2 pneumon*. Two review authors (SKK. Incomplete outcome data: assessed as yes. carried out an assessment of heterogeneity using Review Manager 4 pleuropneumon*. we compared antibiotics A and C when 10 or/7-9 a comparison of antibiotics A and B was available and likewise a 11 6 and 10 separate comparison between antibiotics B and C. Blinding of participants. or gentamycin* or levofloxacin* or macrolide* or minocyclin* We did multiple analyses. personnel and outcome assessors: A person who was not involved in the review gave all relevant stud. no or unclear the trial authors. no or the design effect. need for change in antibiotics. We computed both the effect size 7 exp Anti-Bacterial Agents/ and summary measures with 95% confidence intervals (CIs) using 8 antibiotic*. Published by John Wiley & Sons. we also recorded or cefpodioxime* or cephradine* or cephalexin* or cefaclor* or additional data on potential confounders such as prior antibiotic cefetamet* or cephalosporin* or erythromycin* or gentamicin* therapy and nutritional status. assessed as yes.tw. using The Cochrane Collaboration’s ’Risk of bias’ tool (Higgins Unclear: either not described or incompletely described. baseline characteristics and the outcomes assessed were similar (Bucher 1997).tw.tw.MEDLINE (Ovid) where M is the average cluster size and ICC is the intracluster 1 exp Pneumonia/ correlation coefficient (Higgins 2011). We only did this type of comparison if the inclusion and exclusion criteria of these studies. no or unclear ies a serial number to mask the authors’ names and institutions.nm. We also searched bibliographies of selected articles to identify any additional trials not recovered by the electronic searches. Yes: when it was a double-blind study. we 3 bronchopneumon*. RL) independently reviewed Unclear: not clearly described.tw. Sequence generation: assessed as yes. augmentin* or benzylpenicillin* or b-lactam* or beta-lactam* or We collected data on the primary outcome (cure rate/failure rate) clarithromycin* or ceftriaxone* or cefuroxime* or cotrimoxazole* and secondary outcomes (relapse rate. reference lists and any other potential No: when it was an unblinded study. relevant studies based on their title and abstract. We assessed publication bias outcome. 1. . RL) independently selected potentially No: sequence not generated. Data extraction and management 3. No: described details where allocation concealment was not done. RevMan 2012. including attrition and exclusions from the analysis. unclear Antibiotics for community-acquired pneumonia in children (Review) 5 Copyright © 2013 The Cochrane Collaboration. RL) independently Yes: when the study described the method used to conceal the reviewed the results for inclusion. identifiers. For example. the complete texts of these studies electronically or by contacting 2. Assessment of risk of bias in included studies Data collection and analysis We assessed risk of bias in all included studies using The Cochrane Collaboration’s ’Risk of bias’ tool (Higgins 2011). Free of selective outcome reporting: assessed as yes. firstly on studies comparing the same or moxifloxacin* or penicillin* or quinolone* or roxithromycin* antibiotics. the location of the study. Before combining the studies for each of the outcome variables. We calculated the design effect by 1+(M-1) ICC. unclear about study quality through discussion. Two review authors (SKK. We used a random-effects model to combine the 9 (amoxycillin* or amoxycillin* or ampicillin* or azithromycin* or study results for all the outcome variables. We retrieved Unclear: when it was not described or incompletely described.tw. Two review authors (SKK. When available. We included data from cluster-RCTs after adjustment for 5.tw. (248104) available. (RevMan 2012) software. We resolved differences 4. the results for inclusion in the analysis. We performed a sensitivity analysis to 5 cap. check the importance of each study in order to see the effect of 6 or/1-5 inclusion and exclusion criteria. 2011). the dose and duration of the common intervention Searching other resources (antibiotic B). allocation sequence in sufficient detail. Unclear: when it was not described or incompletely described. need for additional ceftriaxone* or ceftrioxone* or cefditoren* or chloramphenicol* interventions and mortality). We also attempted to perform indirect comparisons or sulphamethoxazole* or sulfamethoxazole* or tetracyclin* or of various drugs when studies with direct comparisons were not trimethoprim*). We recorded data on a Yes: describe the completeness of outcome data for each main pre-structured data extraction form. Ltd. no or unclear Yes: when the study described the method used to generate the Selection of studies allocation sequence in sufficient detail. rate of hospitalisation and or co-trimoxazole* or co-amoxyclavulanic acid or cefotaxime* or complications.

. For each of the pharmaceutical companies to be unclear unless the study defined outcome variables. we carried out an assessment of heterogeneity the role of the pharmaceutical companies. the compli. We calculated the design effect by 1+(M-1) ICC. we prepared 2 x 2 tables. One was a cluster-RCT (Awasthi 2008). ORs and 95% CIs. (Higgins 2011). pendently extracted data by using a pre-designed data extraction form. We used a random-effects model for all the Unclear: details of all the enrolled participants incompletely de. Kogan 2003. failure rate. we conducted a sensitivity analysis. relapse rate. We performed a subgroup analysis of or pharmaceutical companies. scribed. We short- listed 49 trials as potential RCTs to be included and we attempted Dealing with missing data to collect the full-text articles. cations needed for change in antibiotics and mortality rate) we • Co-trimoxazole with amoxycillin: three studies (Awasthi used a 2 x 2 table for each study and performed Breslow’s test of 2008.Yes: results of study free of selective reporting. Published by John Wiley & Sons. • Azithromycin with erythromycin: four studies (Harris 1998. rate of hospitalisation. Assessment of reporting biases • Clarithromycin with erythromycin: one study (Block Before combining the study results. We included data from cluster-RCTs after adjustment for Description of studies the design effect. Sensitivity analysis Measures of treatment effect Most comparisons were for two to three trials. where M is the average cluster size and ICC is the intracluster correlation coefficient (Higgins 2011). involving 2347 children homogeneity to determine variation in study results. We excluded the papers for identifiers. Straus 1998). involving 457 children aged two months to 16 years. CATCHUP 2002. Results of the search Two review authors (SKK. We con- ment effect in the form of failure rates was calculated by making 2 ducted multiple analyses after excluding one study data at a time. Assessment of heterogeneity For each of the outcome variables. we carried out an assessment Included studies of heterogeneity with Breslow’s test of homogeneity in accordance We identified 29 studies for inclusion. Total numbers of events and Antibiotics for community-acquired pneumonia in children (Review) 6 Copyright © 2013 The Cochrane Collaboration. We recorded funding agencies In this review we included RCTs that compared two antibiotics in as governmental agencies. We expressed the results as ORs with 95% confidence intervals (CIs). Wubbel 1999). For each comparison. Ltd. We obtained the full text for 48 We contacted trial authors for missing data. with the following com- with the Cochrane Handbook for Systematic Reviews of Interventions parisons. comparisons. we checked for publication 1995). We considered studies supported by children with radiologically confirmed pneumonia. the extracted data matched completely. x 2 tables and calculating odds ratios (ORs) for each comparison. Roord 1996. For each of the outcome variables (cure or radiographically confirmed pneumonia treated in an rate. Details of all the Data synthesis participants enrolled in the study are included in the paper. RL) screened the article titles. 6. However. icant heterogeneity. Other sources of bias Among the other sources of potential bias considered was funding Subgroup analysis and investigation of heterogeneity agencies and their role in the study. RESULTS Unit of analysis issues All except one study were RCTs. We also considered with Breslow’s test of homogeneity using RevMan 2012 (see Data studies not mentioning the source of funding as unclear under this collection and analysis). heading. universities and research organisations children with pneumonia. If there was signif- The main outcome variables were failure rates or cure rates. RL) inde- two new studies in this update (Bari 2011. ambulatory setting. Two review authors (SKK. Soofi 2012). involving 357 children below 15 years of age with clinical bias using a funnel plot. We calculated No: details of all the enrolled participants not given in the paper. we could trials. Treat. aged two months to 59 months. A third person who was not involved in the review masked not retrieve any missing data from any of the studies.

involving 115 children aged five months • Three studies included children with severe infections or to four years. . Mouallem 1976. Ruhrmann 1982).effective sample size in one cluster-randomised controlled trial (Bansal 2006). aged 0.5 who received intravenous antibiotics were not available and data years. • Parenteral ampicillin followed by oral amoxycillin with • One study (Lee 2008) was excluded because the outcome home-based oral amoxycillin: one study (Hazir 2008) involving was not in the form of cure or failure rates. van Zyl 2002). Results were available (Mulholland 1995). two to 24 months admitted to hospital with severe pneumonia. children between two and 12 years of age in one study (Sidal 1994) and 105 children aged between three months to 14 years in the other similar study (Camargos 1997). Peltola 2001. • Two cluster-RCTs (Bari 2011. months to 12 years. Higuera 1996. • Four studies were carried out in adult participants (Bonvehi • Ampicillin with chloramphenicol plus penicillin: one study 2003. involving 958 children between two to 59 Risk of bias in included studies months with very severe pneumonia. • Oxacillin/ceftriaxone with co-amoxyclavulanic acid: one • Single-dose benzathine penicillin with procaine penicillin: study involving 104 children between age two months to five two studies (Camargos 1997. effect. involving 126 (Keeley 1990. one • Chloramphenicol in addition to penicillin with ceftriaxone: study (Hasali 2005) also did not report the outcome in the form one study (Cetinkaya 2004). treatment received either intravenous antibiotics in hospital or • Co-trimoxazole with chloramphenicol: one study oral medications at home or no treatment. involving 170 children aged six months to 18 years.5 to 16 years of age with CAP treated in hospital or in an • The full-text article could not be obtained for one study ambulatory setting. • Chloramphenicol with ampicillin and gentamicin: one study (Asghar 2008). Sidal 1994). involving 1905 children aged three months to children below five years of age. of these. Soofi 2012) compared oral • Penicillin with amoxycillin: two studies (Addo-Yobo 2004. involving 111 children aged under five years. (Sanchez 1998). • Co-amoxyclavulanic acid with cefuroxime or • Co-trimoxazole with procaine penicillin: two studies clarithromycin: one study (Aurangzeb 2003). The overall risk of bias is presented graphically and summarised • Penicillin and gentamicin with co-amoxyclavulanic acid (Figure 1. Ltd. (Lu 2006). Published by John Wiley & Sons. Sidal 1994). 2037 children between three months to 59 months of age with WHO-defined severe pneumonia. Fogarty 2002. Paupe 1992). • Levofloxacin and comparator (co-amoxyclavulanic acid or • One compared azithromycin with symptomatic treatment ceftriaxone): one study (Bradley 2007) involving 709 children for recurrent respiratory tract infection only (Esposito 2005). Vuori-Holopaine 2000). (Hasali 2005. Patients on conventional 59 months. involving 97 children aged between of cure or failure rates. Figure 2) Antibiotics for community-acquired pneumonia in children (Review) 7 Copyright © 2013 The Cochrane Collaboration. could not be obtained from the trial authors. 1998). involving 134 children aged below five years. • Amoxycillin with co-amoxyclavulanic acid: one study (Jibril • Three studies only compared the duration of antibiotic use 1989). involving 100 children aged two months to 12 years. Separate data on patients (Klein 1995). • Ceftibuten with cefuroxime axetil: one study involving 140 • Chloramphenicol with penicillin and gentamycin together: children between one to 12 years of age with CAP that was one study (Duke 2002). month to five years. • One studied only sequential antibiotic use (Al-Eiden 1999). involving 348 children aged three months to 11. amoxycillin or standard treatment for severe pneumonia in Atkinson 2007). 2012. Ambroggio involving 47 children aged one month to 14 years. involving 71 children with severe and very severe (Awasthi 2008) were calculated after adjusting for the design pneumonia between two months to 59 months of age. • Co-trimoxazole with single-dose procaine penicillin • One study did not provide separate data for children followed by oral ampicillin: one study (Campbell 1988). • Three studies were not RCTs (Agostoni 1988. • Azithromycin with amoxycillin: one study (Kogan 2003). Excluded studies • Amoxycillin with procaine penicillin: one study (Tsarouhas We excluded 20 trials. involving 1116 children aged one radiographically confirmed (Nogeova 1997). involving 176 years with very severe pneumonia (Ribeiro 2011). as oral treatment with amoxycillin in comparison with standard • Cefpodoxime with co-amoxyclavulanic acid: one study treatment (referral and antibiotics). sepsis (Haffejee 1984. involving 723 children aged three children between two to 72 months of age. (Deivanayagam 1996).

. Antibiotics for community-acquired pneumonia in children (Review) 8 Copyright © 2013 The Cochrane Collaboration. Ltd. Published by John Wiley & Sons. ’Risk of bias’ summary: review authors’ judgements about each risk of bias item for each included study. Figure 1.

Wubbel 1999) and sequence 1998. Jibril 1989. Awasthi 2008. Asghar 2008. Mulholland 1995. tion and exclusions from the analysis were unavailable. Camargos 1997. Roord 1996. Wubbel 1999). Sidal 1994). Klein 1995. Hazir 2008. Campbell 1988. CATCHUP 2002. Deivanayagam 1996. Sidal 1994. Duke 2002. Klein Keeley 1990. Straus Only five studies (CATCHUP 2002. Keeley 2008. Cetinkaya 2004. Tsarouhas 1998. Camargos 1997. CATCHUP 2002. it was unclear in nine studies (Aurangzeb 2003. Roord 1996. Cetinkaya 2004. Ribeiro 2011. Camargos 1997. Klein 1995. Bansal 1990. . Awasthi 2008. Straus 1998. Block 1995. 1996. Tsarouhas 1998). Sidal 1994. Keeley 1990. Eight studies were supported by the WHO. Deivanayagam 1996. Shann 1985. Camargos 1997. Jibril 1989. The tical companies (Block 1995. Blinding Nogeova 1997. Harris 1998. Antibiotics for community-acquired pneumonia in children (Review) 9 Copyright © 2013 The Cochrane Collaboration. Jibril 1989. Aurangzeb 2003. Ribeiro 2011. Hazir 2008. Ribeiro 2011. Cetinkaya 2004. 1995. Atkinson 2007. Straus 1998) were double-blinded. Bradley 2007. The rest of the Tsarouhas 1998). Other potential sources of bias Sidal 1994). Ribeiro 2011. Jibril 1989. Published by John Wiley & Sons. Bradley 2007. Atkinson 2007. Five studies were funded by pharmaceu- 1998. Shann 1985. Hazir 2008. Figure 2. Data were fully detailed in 20 studies (Addo-Yobo 2004. Mulholland 1995. Tsarouhas 1997. Cetinkaya 2004. Incomplete outcome data Bansal 2006. Asghar Deivanayagam 1996. Roord rest of the studies were unblinded. Asghar 2008. Mulholland 1995. CATCHUP 2002. Block 1995. Wubbel 1999). 2007. Kogan 2003. 2004. Campbell 1988. Details of sequence generation were described in 17 studies (Addo-Yobo 2004. Harris 1998. Harris 1998. Aurangzeb 2003. Straus 1998. Cetinkaya 1985). Roord 1996. ’Risk of bias’ graph: review authors’ judgements about each risk of bias item presented as percentages across all included studies. Nogeova Nogeova 1997. Mulholland 1995. Shann 1985. Klein 1995. Bansal 2006. Allocation Allocation concealment was adequate in 17 studies (Addo-Yobo Selective reporting 2004. Duke 2002. Campbell 1988. Ltd. Kogan 2003. Bradley 2007. Duke 2002. Shann 2006. Straus 1998. were not clear in 10 studies (Aurangzeb 2003. Block 1995. Bansal Selective reporting of data was unclear in 12 studies (Atkinson 2006. Nogeova 1997. Bradley 2007. The source of funding was not mentioned in 15 studies (Aurangzeb 2003. Wubbel 1999) and in the remaining studies details of attri- was not generated in two studies (Kogan 2003. studies were free from selective reporting. Wubbel 1999) and no con- cealment was done in three studies (Kogan 2003. Awasthi 2008. Harris 1998. Harris 1998. Atkinson 2007. Campbell 1988. Deivanayagam 1996.

Medical Research Council or universities (Addo-Yobo 2004. pneumoniae (in five and three children for ment was unclear. The study was unblinded and allocation concealment was also inade- Clarithromycin versus erythromycin (Analysis 2) quate. 95% CI 0. The trial was single-blinded and allocation conceal.85. pneu- the two groups: cure rate (OR 1. Roord 1996).3) were similar in the two groups. (C. 95% CI 0. 95% CI 0. Two studies (Harris 1998.43 to 1.04 to 0. Duke 2002. Wubbel 1999) compared these two Effects of interventions drugs in 283 children below five years of age.90) (Analysis 2.92. 2003).53) (Analysis 3. while 62 out of 161 (39%) in the ery. respectively) and others (in one and 2.04) (Analysis 2.18 to of the 129 children (10%) tested in the azithromycin group and 2. Organisms included M.16) (Analysis 2.7) or radiologic failure rates (OR 2008. respectively). pneumoniae in 28 children. group). 95% CI 0.84 to 3.61) (Analysis 3. Four studies (Harris 1998.3). pneumoniae were positive in 13 out and six out of 156 (3. One study (Harris 1998) was double-blinded and allocation concealment was ade- quate while the other study (Wubbel 1999) was unblinded with inadequate allocation concealment. Awasthi 2008. influenzae in one.7 to 18. rately for each of the two treatment groups (Harris 1998.6).2) and improvement rates (OR 0.5%) group. Amoxycillin versus procaine penicillin (Analysis 5) tion of pneumonia (diagnosed radiologically) was more frequent in One study involving 170 children aged six months to 18 years the clarithromycin group as compared to the erythromycin group was identified (Tsarouhas 1998).34.5) and ad- verse events (OR 1. relapse rate (OR 0.Analysis 2. influenzae in one patient each in the azithromycin 2003): 104 out of 318 (33%) children experienced wheezing in group. H.23) (Analysis 2. There were Azithromycin versus erythromycin (Analysis 1) fewer side effects reported in the azithromycin group (OR 0.45) (Analysis 2.4) .45 to 8. 95% CI 1. The cure rates (available for Studies comparing ambulatory setting treatment of one study) (OR 1.54 to 1. Mycoplasma 623 children. there allocation concealment was adequate.73. Kogan 2003.5) There were no significant side effects in either four out of the 66 children (6%) tested in the co-amoxyclavulanic group. The following outcomes were similar between the azithromycin and amoxycillin groups. All children recovered at the end of treatment in both the One study (Block 1995) compared erythromycin and clar. Published by John Wiley & Sons.71) (Analysis 3. the 129 children (16%) tested in the azithromycin group and nine thromycin group experienced wheezing. The distribution of different organisms was similar in the One study involving 47 children aged between one month and 14 two groups.02 to 1. 95% CI 0. . Jibril 1989. Atkinson 2007. respectively). S. 95% CI 0. Keeley 1990.61. respectively (OR 0. latory setting.52. groups. Wubbel 95% CI 0. 2.4). However. Three studies reported data on aetiologic organisms sepa. Kogan moniae and H. The study was unblinded but (OR 2.23) (Analysis three. clinical success rate (OR 1. Wubbel 1999) were unblinded and did not have in the two groups were available in one study only (Harris 1998). 95% CI 0. 95% CI 0. Mulholland 1995). Sidal 1994). Resolu. One study (CATCHUP 0. The identification rates were similar in the two groups. adequate allocation concealment.8) both of which were 2002) was supported by the WHO in addition to pharmaceutical established with radiological evidence. There were 24 organisms identified in the fourth study years with classical pneumonia compared these two drugs (Kogan (Wubbel 1999) in 59 participants tested.6 to 1.89) (Analysis 1.21. 95% CI 0. There were 19 organisms identified in the 47 children ithromycin. Investigations for mycoplasma were positive in 21 out of the azithromycin group. acid group.02.51. 234 children below 15 years of age with clinical or tested (10 in the azithromycin group and nine in the amoxycillin radiographically confirmed pneumonia were treated in an ambu. Information on the presence The organisms isolated in this study (Harris 1998) were S. Investigations for C. Keeley 1990. Ltd. Asghar 2008.8%). Kogan 2003.12 to 2. pneu- of wheezing was available in two studies (Harris 1998. pneumoniae) in 20. Information on clearance by Ethics Committees or In- stitutional Review Boards was available for all except four studies Azithromycin versus co-amoxyclavulanic acid (Analysis 3) (Aurangzeb 2003.1).1.17. Children treated with azithromycin were older than those treated with amoxycillin (OR 58. 95% CI 35. companies.80) (. failure rate (OR 0.06 to 1.2).42 to 3. The organisms isolated were 1999) compared erythromycin with azithromycin and enrolled S.59.95) (Analysis 3. 80. The age distribution in the Antibiotics for community-acquired pneumonia in children (Review) 10 Copyright © 2013 The Cochrane Collaboration. The separate data for isolation of organisms Roord 1996. failure non-severe pneumonia rates (available for both studies) (OR 1.07. Roord 1996.61). One study (Harris 1998) was double-blinded with pneumoniae (M.02 to 6. there were 234 organisms identified in the azithromycin group and 135 in the erythromycin group (Roord Azithromycin versus amoxycillin (Analysis 4) 1996).9). Hazir 95% CI 0. were no differences in the radiologic improvement rates (OR 3. The failure rates in the out of the 66 children (14%) tested in the co-amoxyclavulanic acid azithromycin and erythromycin groups were six out of 236 (2.15. 95% CI 0.55.08) (Analysis moniae (in four and three. pneumoniae) in 36 and Chlamydia pneumoniae adequate allocation concealment and three studies (Kogan 2003.

CATCHUP 2002) and it was not different in either treatment group (OR 1.53) children who had received antibiotics earlier and loss to follow- (. There was no with severe pneumonia.16) Failure rate in severe pneumonia available in one up. 95% CI 0. 95% group and 1077 in the amoxycillin group) between two months CI 0. (Analysis 11. death rates (OR 1.19.61) (Analysis 9.69. 95% CI 0. hospitalisation rates (OR 1.17 to 3. influenzae in 28 par- the design effect.18) (Analysis 11. A third study (Awasthi One multicentre study (Klein 1995) enrolled 348 children be- 2008) was open-label and cluster-randomisation was done (the tween three months and 11.56 to Studies comparing treatment of hospitalised children 1.2).89) (Analysis 7.25) (Analysis 5. The age dis- assessment of the primary outcome of treatment failure was done tribution in the two groups was comparable.94 to 1. failure rates in significant difference between the two groups in positive cultures.6) were similar for the two groups. Straus 1998) were double-blinded and allocation concealment was adequate.88 to 7. 95% CI 0.20. The study was unblinded and allocation con- Co-trimoxazole versus amoxycillin (Analysis 7) cealment was not clearly stated.71.60) (Analysis 10. There was no this study.57. influenzae in 79 children (52 in the Chloramphenicol versus penicillin plus gentamycin (Analysis co-trimoxazole group and 27 in the amoxycillin group) and S.3).15. The cure rate could be extracted in two studies (Awasthi 2008.51) (Analysis 7.2).36 to 3. 95% CI 0. In pooled data the failure rate in non.26 to 9. These organisms were H. CI 0. the distribution was similar in the month and five years compared chloramphenicol with penicillin two groups. M.96 to Antibiotics for community-acquired pneumonia in children (Review) 11 Copyright © 2013 The Cochrane Collaboration. 95% CI 0.7) There was only cally diagnosed bacterial pneumonia (Jibril 1989).4). S. Two studies (Keeley 1990. parainfluenzae in four (6.69) (.59 to 1. Allocation concealment was adequate. The cure rate was similar in the two groups (OR Co-amoxyclavulanic acid versus amoxycillin (Analysis 6) 1. one study (Straus 1998) also included 301 children seven (11.25) (Analysis 9.9%) and H. 95% CI Three multicentre studies (Awasthi 2008. Cure rate was better with co-amoxy.03.5 years of age. pneumoniae in 14 (23%). .1). unblinded and allocation concealment was also inadequate.58. total numbers of events and effective sample size (OR 0. Co-trimoxazole versus single-dose procaine penicillin clavulanic acid (OR 10.75.Analysis 8. Cefpodoxime versus co-amoxyclavulanic acid (Analysis 10) Two studies (CATCHUP 2002.76 to 2. 95% CI 2. presence of wheeze and mean weight in the two groups were comparable. 95% Straus 1998) involving 2346 children (1270 in the co-trimoxazole CI 0.85 to 38.18 to 2. This was an open-label RCT in children with the primary outcome in one study (Awasthi 2008).07) 3.25) (. Published by John Wiley & Sons. the end of 10 days of treatment was comparable in the two groups moxazole group.14) Loss to follow-up was comparable in the two with severe/very severe pneumonia groups (OR 0.8)).6) Rate of hospitalisation One study involved 100 children between two and 12 years of was available in only one study and was similar in the two groups age. Both studies were un- blinded and allocation concealment was adequate in one study (Keeley 1990). The response rate at on day four for the amoxycillin group and day six for the co-tri. 95% CI 0.52. The failure rates were similar in the Co-trimoxazole versus procaine penicillin (Analysis 8) two groups (OR 0.Analysis 7.80.Analysis 8.12).11) (Analysis 7.92 to 1. Sidal 1994) enrolled 723 children be- tween three months and 12 years of age. significant difference in the bacteriologic efficacy of either group severe pneumonia was similar in the two groups (OR 1.two groups was comparable. 95% CI 0.18.01 to 4.7). The diagnosis of pneumonia was based on clinical criteria.91 to 1. Age and sex distribution. The study was un- randomisation unit was Primary Health Centre) and in this study blinded and allocation concealment was inadequate.44. followed by oral ampicillin for five days (Analysis 9) One study was included that had enrolled 134 children below five years of age with severe pneumonia as defined by WHO criteria (Campbell 1988). Need for change in antibiotics (OR 0.54 to study was similar in the two groups (OR 1. (OR 2.4%).2) and adverse events (OR 1. we performed severe pneumonia that was carried out in Papua New Guinea ( analysis for failure rates in non-severe pneumonia after excluding Duke 2002).25.96. The organisms isolated from blood cultures were H.26. the two groups were similar (OR 1. CATCHUP 2002.72) (Analysis 9. The study was one death. 11) pneumoniae in 49 children (36 in the co-trimoxazole group and One multicentre study including 1116 children aged between one 13 in the amoxycillin group).8).5) and death rates (OR 0. All studies included children with non-severe ticipants (47.25 to 9. catarrhalis in pneumonia. 95% CI 0. Organisms were in this study (Awasthi 2008) were calculated after adjusting for isolated in 59 cases. 95% CI 0. Ltd. The cure rates (OR 1.57) (Analysis 7. 95% (100% versus 96. and 59 months of age have compared co-trimoxazole and amoxy- cillin. In view of the difference in the time of assessment for and gentamycin. There were only two deaths in both the groups. 95% CI 0. It was an open-label study on children suffering from clini. The results did not alter significantly. 0. 95% CI 0.21) (Analysis 6.5%).

49. Mean age. 95% CI 0.58 to 2. were observed in the other study (Atkinson 2007). For the purposes of this analysis we considered Chloramphenicol alone versus chloramphenicol plus them as failure on day five. Published by John Wiley & Sons.9) were (Analysis 14.51) (Analysis 15. 95% CI 0.80 to 1. rolment were comparable in the two groups.1) and duration of hospitalisation combination over chloramphenicol (OR 1. Failure rates on day 95% CI 0. sex distribution and presence of wheeze were com- distribution and proportion of children with severe malnutrition parable in the two groups. Age. The isolation rate for S. the study suffering from severe pneumonia (diagnosed on the basis of WHO enrolled 97 children between 2 and 24 months of age diagnosed criteria) (Addo-Yobo 2004) and 203 children with radiographi- with severe CAP with probable bacterial aetiology (Cetinkaya cally confirmed pneumonia (Atkinson 2007). Ltd. children (67 in children receiving chloramphenicol and 77 in the other group). 16.27 to 1. in one (Camargos 1997).04 to 2. However.66) (Analysis 11. breast penicillin (Analysis 14) feeding and the number of children who had received antibiotics One study (Shann 1985) from Papua New Guinea involved 748 in the last week were similar in both the groups. Two studies fulfilled the inclusion criteria.19) (Analysis 12.04 to 2.8) Need for change in antibiotics (OR 0.15. pneumoniae 1.36. this study enrolled 958 dren between three months and 14 years of age (Sidal 1994). Bacterial pathogens were five (OR 1.93) (Analysis sation until the temperature was < 38 degrees celsius for 24 hours 13. Death rates were higher in those receiving chloram.1).81 to 1.43.99 to 2. Amoxycillin versus penicillin (Analysis 17) Chloramphenicol plus penicillin versus ceftriaxone (Analysis Two multicentre non-blinded studies were identified.86. was six out of 90 blood cultures performed (four participants in 95% CI 1. .4). There were seven deaths in the group re- (Analysis 14.12 to 1. 95% CI 0. day 10 (OR identified in only one study.1) were similar in the two groups. Cure rates in the study (Atkinson 2007) measured outcome as time from randomi- two groups were similar (OR 1. The failure rates hospitalised children (age not clear) with severe pneumonia.30) (Analysis 17.3) and deaths rates (OR 0. 95% CI 0. The studies were 2004). Isolation rates or sensitivity of the organism and Benzathine penicillin versus procaine penicillin (Analysis 16) failure rates did not differ between the two groups. 95% CI 0. The second groups were comparable (details not available). Number of infants unblinded and allocation concealment was adequate. were comparable in the two groups.53) similar in both groups. those receiving chloramphenicol as compared to ampicillin and The clinical outcome did not differ in relation to the organism gentamycin. 95% CI 0. five days (OR 1.93) (Analysis 15.48 to 1. one which included 176 children between two and 12 years of age with chest X-ray films Chloramphenicol with ampicillin and gentamycin (Analysis showing lobar consolidation or infiltration (presumed streptococ- 12) cal infection) (Camargos 1997) and another study of 105 chil- One multicentre study was identified.7).97) and 14 days (OR 1.03.2) were comparable in the two groups. phenicol (OR 1. Ages in the two unblinded and allocation concealment was adequate. However. 95% CI 1.9 to 11.03 to 1.55) (Analysis 11.84 to 1.09) ceiving penicillin in one study (Addo-Yobo 2004) while no deaths (Analysis 14.5) and day 21 (OR 1. 95% CI 0.51. Allocation concealment was adequate. 95% CI 0. it provided data on need for change of antibiotics due to worsening of respiratory/ra- diological findings. Failure rates (OR 0.11) (Analysis 16. Age and sex in each group.29) (Analysis 17.77) (Analysis 12. these en- 13) rolled 1702 children between three months and 59 months of age. Bacterial pathogens were identified in 144 CI -1. 95% CI 0. The study was unblinded and allo. sex. 95% 2. Both children who were hospitalised with WHO-defined very severe studies were unblinded and allocation concealment was adequate pneumonia (Asghar 2008). The measured at 48 hours (OR 1.47 to 3.1). The cure rates (OR 0.4). proportion of boys ent in the two groups (OR 0.06) (Analysis 12.65.53.2). Failure rates were also similar between the groups (OR 3.73.01) (Analysis and number of children who had received antibiotics before en. The study was 2008). One double-blind study fulfilled the inclusion criteria. identified.1. 95% admission rates before 30 days favoured the penicillin-gentamycin CI 0.63 Antibiotics for community-acquired pneumonia in children (Review) 12 Copyright © 2013 The Cochrane Collaboration.1.11. loss to follow-up (OR 1. Allocation concealment was adequate. and oxygen requirement had ceased.6) were significantly higher in the benzathine group and two in the procaine penicillin group).17.04.13 to 0. 95% CI 0. 95% CI 1.4).48.46.61. severe malnutrition. 95% CI 1. Cure rates were not significantly differ- cation concealment was adequate.02 to were similar in the two groups (mean difference (MD) 0.31) (Analysis study was unblinded but allocation concealment was adequate.98) (Analysis 12.1). 17. re.15 to 1. Ampicillin alone versus penicillin with chloramphenicol Amoxycillin with intravenous (IV) ampicillin (Analysis 18) (Analysis 15) One non-blinded study involving 237 children between two and One trial involving 115 children between five months and four 59 months of age with severe pneumonia was identified (Hazir years of age was identified (Deivanayagam 1996).10).

32 95% CI 0.03.83) (Analysis tion concealment was adequate.to 1. were comparable. influenzae (13). aureus) (eight).7) were similar in the two groups. . aureus One non-blinded study. involving 709 children below 16 years (three). 26) One study involving 104 children aged between two months to five years with very severe pneumonia was included (Ribeiro 2011).51. 95% CI 0.10) were similar in the two groups.42.25.87 to 7.4) was significantly higher and failure rate (OR 6.46 to 2. failure (Analysis 19. Cefuroxime with clarithromycin (Analysis 23) The study was unblinded.40) (Analysis 24. Moraxella catarrhalis (M. including 25.3) and failure rates (OR 0.05 to reus (S.05. 95% CI 0. received ceftibuten (OR 3.83) (Analysis 24. group A beta haemolytic streptococcus (four).51) 1.42) (Analysis 22.5) was significantly lower in children receiv- (Analysis 21) ing cefuroxime. group A beta haemolytic streptococcus 14.94) (Analysis 25. The study was non-blinded and alloca.21. Age and sex distribution were similar in the two groups. days before admission in hospital.74) (Analysis quence generation and allocation concealment were not clear from 20. Sequence generation and moniae (three). influenzae (eight). Cure rates (OR 2.05. Co-trimoxazole versus chloramphenicol (Analysis 24) Amoxycillin with cefuroxime (Analysis 19) One double-blind study involving 111 malnourished children un- One randomised.47 to 2.05.04 to 5. controlled study involving 85 concealment and reporting of data were adequate. sex and number who received antibiotics before enrolment were comparable in the two groups.24 to 4. Cure rate (OR 0. pneumoniae (17).3) and failure rates (OR 0.6).83.02. of participants requiring a change in antibiotics (OR 1.33 data in the form of mean age and proportion of boys were similar to 2. The study was unblinded. Published by John Wiley & Sons. The sequence generation and Ceftibuten with cefuroxime axetil (Analysis 25) allocation concealment in the study is not clear.4) were similar in the two groups.5) were similar in the two groups of partici- in the two groups.40) (Analysis cefuroxime axetil (Nogeova 1997). 95% CI 0. Cure rates (OR 0. 95% CI 0.19) (Analysis 18. respiratory syncytial virus (onr) and Mycoplasma pneumoniae (M. 95% CI 0. H. 95% CI -1. H. catarrhalis (seven).18 to 23. Failure were S. S. der five years of age fulfilled the inclusion criteria for this review this included 83 children with non-severe and severe pneumonia (Mulholland 1995).78.39) (Analysis 21. Organisms were isolated in 83 participants (53 One study involving 71 children between two months and 59 in the ceftibuten group and 30 in the cefuroxime group). 95% CI 0.81. 95% CI 0. 95% CI 0.70) (Analysis 25. the paper.18 to 23. Mean time for improvement (MD -1. 95% CI 1. Cure in the two groups. Iden- months of age with very severe pneumonia fulfilled the inclusion tification of organisms was significantly higher in children who criteria (Bansal 2006).89) ( pants. children with wheez- data in the form of mean age and proportion of boys were similar ing and numbers excluded were similar in the two groups. (seven).98. 95% CI 0. Allocation concealment was unclear.5).40) (Analysis 24.30) (Analysis 24. The organisms identified in children receiving cefuroxime Levofloxacin with comparator group (Analysis 22) axetil were: S.05. Baseline data in One study involving 140 children between one and 12 years of age the form of mean age and proportion of boys were similar in the with radiographically confirmed CAP compared ceftibuten with two groups.4) were similar in the two groups. 85 children with non-severe and severe pneumonia were enrolled (Aurangzeb 2003). catarrhalis) (four). 95% CI 1. The mean age. pneumoniae (seven). respiratory syncytial virus (three) and M. non-blinded controlled study was identified. 95% CI 0.06.3). Age and sex children with non-severe and severe pneumonia was identified distribution.46 to Penicillin and gentamycin with co-amoxyclavulanic acid 31. allocation One randomised. Cure rates were similar in the two Oxacillin/ceftriaxone with co-amoxyclavulanic acid (Analysis groups (OR 1. Allocation concealment was unclear. Se- 20.21) (Analysis 18.86. Cure rates (OR 1. 95% CI 0. Baseline otics before enrolment and failure rates (OR 0.00 day. 95% CI 0.03 to (Analysis 23.49. The (Aurangzeb 2003). relapse rates (OR 0. compared oral levofloxacin with either ceftriaxone or co.46 to 4. non-blinded.4). 95% CI 0.6) and death rates (OR 0.04 to 5.3) and failure rates (OR 2. 2.63 to 7. allocation concealment is not clear from the study. non-blinded controlled study compared these two drugs.8) and death rates (OR 2. receipt of antibi- (Aurangzeb 2003). nutritional status. pneu- amoxyclavulanic acid (Bradley 2007).5). Allocation concealment was adequate. relapse rates (OR 1. 95% CI 0. 95% CI Amoxycillin with clarithromycin (Analysis 20) 0.92) (Analysis 26.33) (Analysis 18.06 to 17.45 to 2. Moraxella of age. One randomised.9). Organisms identified in children who received ceftibuten number of infants and sex distribution.4) were similar in the two groups. Ltd. 95% CI 0.89 Antibiotics for community-acquired pneumonia in children (Review) 13 Copyright © 2013 The Cochrane Collaboration. Staphylococcus au- rates in the two groups were similar (OR 0.95.06 to 15.51) rates (OR 1.59) rates (OR 1. Baseline characteristics.46 to Analysis 23. Baseline age and sex distribution. random sequence generation. number (Analysis 19.11 to 0.33) (Analysis 24. pneumoniae) (one).2).

Ltd.6) and total hospital stay (MD -3. (Analysis 15. The failure rate on day five was similar in the in the two groups (OR 0.72. Azithromycin versus amoxycillin (Analysis 4) 95% CI -5. Wubbel 1999). Azithromycin versus erythromycin (Analysis 1) Studies comparing treatment of hospitalised children Out of four studies (Harris 1998. 95% CI 0.4). In four studies (Aurangzeb 2003.61.02 to 6. Straus 1998) used 1998). . was identified (Tsarouhas 2004. Cure rates in the two groups were similar (OR 0.1.34.34) (Analysis 26. only assessed in one study which included 176 children between However.63) (Analysis 1. one study ( Out of two studies.51. ithromycin.1). 95% CI 0. Keeley 1990. clinical criteria to diagnose pneumonia. Bansal 2006. years of age.48. 95% CI 0.80) (Analysis 2. Tsarouhas 1998) enrolled children with radiographically con.9) and cure rates (Analysis 15) (OR 1. there were no differences in radiologic cure rates (OR two and 12 years of age with chest X-ray films showing lobar 3. radiographically confirmed pneumonia was thromycin group (OR 2.62.to -0. Duke 2002. (Camargos 1997).16) (Analysis 2.7 to 18.7) two groups (OR 2.75. Bradley 2007.51) (Analysis 15. Klein 1995.13 to 0.5). Cetinkaya pneumonia. Roord 1996) used clinical criteria or radiogra- phy for diagnosis of pneumonia.40 days. radiographs were performed for diagnosis of pneu- monia in only two studies (Kogan 2003. Kogan 2003. 95% CI 0.23 to 1. was iden- tified (Deivanayagam 1996). Resolution of pneumonia (diagnosed radiologically) was Benzathine penicillin versus procaine penicillin (Analysis 16) more frequent in the clarithromycin group compared to the ery. Failure rates were similar between the groups (OR 1. Ten studies (Addo-Yobo 2004. between five months and four years of age.08) (Analysis 3.69.62. children with radiographically confirmed Wubbel 1999) involving 88 children enrolled participants with pneumonia were enrolled in one study involving 203 children radiographically confirmed pneumonia.61) (Analysis 4.1) and duration of hospitalisation Clarithromycin versus erythromycin (Analysis 2) were similar in the two groups (MD 0. 95% CI Out of 29 studies.25) (Analysis 5. Antibiotics for community-acquired pneumonia in children (Review) 14 Copyright © 2013 The Cochrane Collaboration.18 to 2.2). Erythromycin versus co-amoxyclavulanic acid (Analysis 3) Amoxycillin versus penicillin (Analysis 17) Out of two studies (Harris 1998. The failure rates were similar in the two groups (OR 0.5 confirmed pneumonia. One multicentre study (Klein 1995) enrolled 348 children with The following comparisons were carried out in radiographically radiographically confirmed pneumonia aged three months to 11. Sidal 1994) the role of radiography in Cefpodoxime versus co-amoxyclavulanic acid (Analysis 10) the diagnosis of pneumonia was not clear from the description. Kogan 2003. 12 (Atkinson 2007. Shann 1985.39) (Analysis 17.46 to -1. 14 years with radiographically confirmed pneumonia compared azithromycin and amoxycillin (Kogan 2003).15 to 1. Ribeiro 2011. Jibril 1989. Deivanayagam 1996.45 to 5. One study involving 170 children with radiographically confirmed Awasthi 2008.60) (Analysis 10.36.7) or radiologic failure consolidation or infiltration (presumed streptococcal infection) rates (OR 0.1. 95% CI 0. Amoxycillin versus procaine penicillin (Analysis 5) firmed pneumonia. CATCHUP 2002. Asghar 2008.06 to 1.05 to 7. 95% CI 0. aged six months to 18 years. Three studies (Harris 95% CI 0.8). Published by John Wiley & Sons. Children treated with azithromycin were older than those treated with amoxycillin Antibiotics in radiographically confirmed pneumonia (OR 58. 0. Wubbel 1999). Roord 1996. 95% CI 0. Cure rates were not significantly different in the two groups (OR 2.04) (Analysis 2.70) (Analysis 16. Failure rates were similar (Atkinson 2007). Failure rates Ampicillin alone versus penicillin with chloramphenicol (OR 0. 95% CI 35.09) (Analysis 4.65 to 4. with severe/very severe pneumonia Wubbel 1999). Block 1995.7).55.85. pneumonia. Mulholland 1995. 95% CI -1.6). 1998.7) were significantly better One study involving 47 children aged between one month and in children receiving co-amoxyclavulanic acid.8) were not different One trial involving 115 children with radiographically confirmed in the two groups.17 to 3. Out of two studies. 95% CI 0.59 to 80.73 to 11.93) One study (Block 1995) compared erythromycin and clar. 95% CI 1. A total of 147 children were enrolled in these two studies. Campbell 1988. The cure rates (OR 0. 95% CI 0. Nogeova 1997. Hazir 2008.15).56) (Analysis 1. 234 children below 15 years of age with radiograph- ically confirmed pneumonia were treated on in an ambulatory setting. The study was unblinded and allo- cation concealment was adequate.1).59 to 9. Wubbel 1999.11) (Analysis 26. Camargos 1997.

3).8).11 to 0. Tsarouhas 1998) enrolled chil- included children with severe pneumonia and compared oral an- dren with radiographically confirmed pneumonia. 95% CI 0. The need for hospitalisation was similar CI 0.5). Failure rates were similar in the two groups (OR amoxyclavulanic acid (Bradley 2007). 95% CI 0. (OR 0.42) (Analysis separately.06.29) (Analysis 27.17 to 1. Cure rates (OR 1. Tsarouhas Antibiotics for community-acquired pneumonia in children (Review) 15 Copyright © 2013 The Cochrane Collaboration.4) and failure seven deaths in another study (but only in those receiving intra- rate (OR 6. 2174 received oral antibiotics months of age with very severe.46 to 2. Ceftibuten with cefuroxime axetil (Analysis 25) Hazir 2008) and was significantly higher in those who received One study (Nogeova 1997) involved 140 children between one injectable treatments (OR 0.19).81. Cure 27. Analysis of studies that treated both the groups in an ambulatory setting (after removing studies that gave both the treatments in hospital) showed that failure rates in the two groups were not Co-trimoxazole versus chloramphenicol (Analysis 24) different (OR 0.5) were sig.45 to 2.76 to 1.14). The failure rates were similar in the two cations.03. Sidal 1994. 95% CI 0. 95% CI 1. Failure rates were simi- lar in the two groups (OR 0. 95% CI in the two groups (OR 1. Sidal of participants requiring a change in antibiotics (OR 1. penicillin/ampicillin (Addo-Yobo 2004.86.18). 22. The sequence generation 0.9) or cotrimoxazole ment were comparable in the two groups.15.46 to 31. 95% CI 0.92.10) were analysed lar in the two groups (OR 1. 95% CI 0. of age. A total of 4331 children be- One study involving 71 children between two months and 59 low 18 years of age were enrolled. Atkinson 2007. study (one in the oral group and four in the intravenous ampi- cillin group) (Hazir 2008). Re-analysis after removing one study with seven deaths in only one group (Addo-Yobo 2004) suggests Oral treatment of severe pneumonia with parenteral no significant difference between the two groups (OR 0.38. Failure rates intramuscular antibiotics (penicillin or ampicillin). (cotrimoxazole or amoxycillin) and 2157 received intravenous or monia fulfilled the inclusion criteria (Bansal 2006).9).63 to 7.33.34 to 4. Sidal 1994. .21) (Analysis 27.25.8) and death rates (OR 2. 95% 1994.24) (Analysis Levofloxacin with comparator group (Analysis 22) 27.4). 95% CI 0. 95% CI 0. 95% CI 0.02. Ltd.84.6). Published by John Wiley & Sons. Hazir 2008. Separate data for children below five years of age were not available. venous penicillin (Addo-Yobo 2004)) and five deaths in the third nificantly better in the children receiving cefuroxime. radiographically confirmed pneu. Campbell 1988. There were no deaths in one study (Atkinson 2007) and rate (OR 0. Sidal 1994) with radiographically confirmed pneumonia under five years of age and were significantly better in children receiving oral antibiotics fulfilled the inclusion criteria for this review (Mulholland 1995).9).42.40) (Analysis 24. Hazir 2008. Failure rates did not and allocation concealment were not clear from the study.15).19 to 21. (OR 0.94) (Analysis 25.29) (Analysis 27.21. sex and number who received antibiotics before enrol.24 to 4.Penicillin and gentamycin with co-amoxyclavulanic acid 1998) children were treated in an ambulatory setting with injec- (Analysis 21) tions as well as oral medications. The show significant differences when children receiving amoxycillin mean age. relapse rates (OR 1. Cure rates were simi.33) (Analysis 24.33). proportion of children who had received antibiotics before enrol- ment were comparable in the two groups.03 to 3. 95% CI 0.10) (Analysis 27. failure Hospitalisation rate in children receiving treatment in an ambu- rates (OR 1. 95% CI 1.05 to 14. 95% treatment (Analysis 27) CI 0. (OR 5.05. We re-analysed data after removing studies that also en- One non-blinded study.82) (Analysis 27.70) (Analysis 25.83) (Analysis 24.77 to 1. Atkinson 2007. Four studies compared oral amoxycillin with intravenous groups (OR 1. Sidal 1994. involving 709 children below 16 years rolled children above five years of age (Atkinson 2007.20) (Analysis 27.47 to 2. A total of 373 timicrobial agents with initial intravenous or intramuscular medi- children were enrolled. groups (OR 0. Death rate was available in three studies (Addo-Yobo 2004.45.39) characteristics (age and sex distribution) in the two groups and (Analysis 21.34) (Analysis 27.7). Data on loss to follow-up were available in one study (Hazir 2008) and were similar in the two There were six studies (Addo-Yobo 2004.77 to 1. In Identification of aetiological agents four studies (Campbell 1988. number latory setting was available in three studies (Campbell 1988. 95% CI 0. Sidal 1994).28.32.03. 95% CI Relapse rates were available in two studies (Atkinson 2007.92.05. 95% CI 0. 0.09) (Analysis 27.31. 95% CI 0. Two studies compared oral cotrimoxazole with intramuscular penicillin (Campbell 1988.10) were similar in the two groups. 95% CI 0.40) (Analysis 24.56 to 1. 95% CI 0.03 to 0.30) (Analysis 24.13. The baseline in the two groups were similar (OR 0. Tsarouhas 1998). Tsarouhas 1998).91. 2008) and there was no significant difference in the two groups (OR 1.13).46 to 4.87) (Analysis and 12 years of age with radiographically confirmed CAP. Atkinson 2007. Tsarouhas 1998) that Only two studies (Atkinson 2007.10) (Analysis 27. 2. 3. Hazir 2008.16).41 to 4. Hazir 0. Tsarouhas 1998). Cure One double-blind study involving 111 malnourished children rates were available in two studies (Atkinson 2007. 95% CI 0. compared oral levofloxacin with either ceftriaxone or co.

erythromycin and gentamycin. co-trimoxazole with single-dose procaine penicillin followed Antibiotics for community-acquired pneumonia in children (Review) 16 Copyright © 2013 The Cochrane Collaboration. We utilised this process to compare co-trimoxa- Out of the bacterial pathogens identified. 95% CI 2. A limited number of randomised In the study by Bansal 2006. Out of a total of 47 isolates of S. 12/17 to gentamicin. Death rates were higher in children getting Nasopharyngeal aspirates were tested for RSV in four studies chloramphenicol as compared to those getting penicillin/ampi- (Atkinson 2007. 13/14 were sensitive to chloramphenicol. were positive in 158 out of 1534 (10%) ithromycin with erythromycin. 19/37 were sensitive to The aim of this review was to establish the most effective antibiotics chloramphenicol. be carried out. Most of the were sensitive to penicillin. In the comparison of amoxycillin with chloram- Information regarding the sensitivity pattern of bacterial iso. This information was only avail. Addo-Yobo 2004. influenzae. Out of a total of eight isolates of H. gentamycin. A single iso- nia was better with clarithromycin as compared to erythromycin late of H. Baseline data for age and sex were not comparable in negative bacilli M. clarithromycin. influenzae was resistant to co-trimoxazole. For children with severe pneumonia. Ltd. tamycin. phenicol (CATCHUP 2002. 30. influenzae. amoxycillin. treat- mycin while three organisms were resistant to erythromycin. nia (CAP) of different severity.57 to 7. 150 (25%) had H.26. In participants aged under The rest of the comparisons for treatment in ambulatory settings five years 141 out of 659 (21%) tested positive for mycoplasma. out of a total of 22 S. antibiotic comparisons were available in single studies only. all 20 isolates were sensitive to azithro- amoxyclavulanic acid. 95% CI 0. 69 (12%) had cefpodoxime (Analysis 29) and amoxycillin with chloramphenicol S. Mulholland 1995. 15/16 to ampicillin and DISCUSSION 6/6 to third-generation cephalosporins. (Analysis 30). involved azithromycin with erythromycin. one was sensitive and the other isolate was resistant to penicillin. Resolution of radiographically confirmed pneumo- susceptible to co-trimoxazole and chloramphenicol. pneumoniae. amoxycillin with had S. 12/ 0. For Indirect comparisons this review. to chloramphenicol. 17 to gentamycin.41 to 1.08) (Analysis the isolates.3) and failure rates were lower in the amoxycillin group (OR pneumoniae isolates. erythromycin and gen. albus) the first two comparisons and therefore no valid comparison could and Group A beta haemolytic streptococcus (Table 1). Straus 1998) sex lates was available in four studies (Asghar 2008. 29/45 to gentamycin. In children under five years. Cure rates were better in the amoxycillin group compared to the able for the antibiotics studied and sensitivity was not tested in all chloramphenicol group (OR 4. The meth- ods used in these studies for identification of bacteria were a blood culture. Wubbel 1999). co-amoxyclavulanic acid was better than ceptible to chloramphenicol. pneumoniae. Out of the 1734 participants tested for M.00) (Analysis 30.4). amoxy- cillin. Published by John Wiley & Sons. 1999) involving 1916 children and RSV was identified by posi- For severe/very severe pneumonia. distribution was not comparable although age distribution was. The one that was resistant Summary of main results was sensitive to ciprofloxacin. influenzae. failure rates were significantly higher (Block 1995. 236 (40%) participants zole with co-amoxyclavulanic acid (Analysis 28). demonstrate the aetiological organisms in 14 studies. all the three isolates of S. clar- Tests for Chlamydia spp. Mulholland 1995. results of a throat swab for bacterial isolation were We attempted to compare various antibiotics (A and C) when ignored. ampicillin or penicillin. there were positive test cillin. Kogan 2003. Nogeova in those receiving chloramphenicol as compared to ampicillin and 1997. sputum examination or urinary antigen detection. isolates were amoxycillin. 15/16 to ampicillin and for first-line empirical treatment community-acquired pneumo- 6/6 to third-generation cephalosporins.Out of 29 studies reviewed. pneumoniae ting suggest that the failure rate with co-trimoxazole was com- were susceptible to co-trimoxazole and nine of these were also sus- parable to amoxycillin. In the study by Asghar 2008. 15/16 to ampicillin and 12/12 to third-gener- ation cephalosporins. Studies comparing treatment of pneumonia in an ambulatory set- In the study by Mulholland 1995. In the study and side effects were fewer with azithromycin as compared to co- by Roord (Roord 1996). attempts were made to isolate or results for Chlamydia spp. penicillin/ampicillin plus gen- tive serology in one study (Nogeova 1997) involving 140 children. were available. amoxycillin with procaine peni- participants. out of two isolates of H. cefotaxime and chloramphenicol. aureus and 136 (23%) had other pathogens including the gram. 6/7 were sensitive to chloramphenicol. 385 (22%) tested positive. Bacterial pathogens could be identified in blood cultures comparisons of antibiotics A and B were available and B and C or serology/sputum in 591 (12%) out of 4882 participants tested. in 45 out of 658 (7%) participants. All three Salmonella spp. aureus.64. Bradley 2007. all 10 isolates of S. Mulholland 1995. Roord 1996). Bansal 2006. Wubbel cillin plus gentamycin. pneumoniae controlled trials (RCTs) fulfilled the inclusion criteria. Harris 1998. . However. catarrhalis and Staphylococcus albus (S. tamycin was associated with lower re-admission rates as compared RSV was identified in 407 children (20%). Identification of atypical organisms was attempted in six studies For very severe pneumonia. ment with oral antibiotics was similar to treatment with injectable Respiratory syncytial virus (RSV) was tested in five studies.

children with non-severe pneumonia treated with amoxycillin or placebo had similar failure rates. Respiratory symptoms and rapid respiratory and there were no statistically significant differences in these com. Giving them and chloramphenicol with co-trimoxazole and were comparable. Reports of in vitro resistance of The WHO recommends admission to hospital and treatment with common organisms of pneumonia to cotrimoxazole and relatively penicillin for severe pneumonia and chloramphenicol for very se- more sensitivity to amoxycillin have not resulted in more failure vere pneumonia (WHO 1999). be cost-effective and therefore empirical treatment of pneumonia Amoxycillin was comparable with macrolides (azithromycin and is justified. either ampicillin/penicillin or amoxycillin. Injectable peni- and H. oxacillin plus ceftriaxone combination. cefuroxime with clarithromycin cover over three to seven days with supportive care. amoxycillin with clarithromycin. Com- graphic location. In this review it clearly emerged rates in the co-trimoxazole group. This with injection-site problems and therefore have a limited role in review suggests that amoxycillin and co-trimoxazole are associ- non-severe pneumonia. ceftibuten. ity rates were higher in children receiving injectable antibiotics. ception of more radiological clearance with clarithromycin with- and identification of aetiological agents was limited to a few stud- out any clinical implications. the likely aetiological agents and their resistance clavulanic acid. procaine penicillin and benzathine penicillin. The results are based on The World Health Organization (WHO) recommends treatment single studies for each drug. pneumoniae co-amoxyclavulanic acid cannot be administered. oral cephalosporins (cefpodoxime. Overall completeness and applicability of Therefore. Another study from Pakistan observed that cefuroxime was better than ceftibuten. Macrolides may acquire resistance ies. Amoxycillin cable to the management of pneumonia in developing countries. Treatment of pneumonia depends on the age of the child. furoxime. It should be noted that medications reveals adequate allocation concealment but all were amoxycillin is more expensive than co-trimoxazole for five days of unblinded. Two recent studies (Agarwal 2004. severe pneumonia using the WHO case definition (Hazir 2006). who are this review included children with pneumonia diagnosed by the feeding well. penicillin and gen. levofloxacin with ceftriax. co-amoxyclavulanic acid may be used as an alter- ical antimicrobial treatment in countries with an infant mortality native to penicillin. There is no explanation for the increased death rates in treatment for a child weighing between 5 kg and 10 kg (in India those who received injectable antibiotics. except those with bacterial pneu- tamycin with co-amoxyclavulanic acid. Most of the studies included in this review were parisons of various macrolides shows similar efficacy. rates in children may be due to bacterial pneumonia. the difference in death rate becomes non-sig- reduced to some extent if the treatment duration of amoxycillin nificant. influenzae) to co-trimoxazole have been raised (Krishnan cillins (procaine penicillin or benzathine penicillin) are associated 2011) and amoxycillin has been suggested as an alternative. Attempts to isolate aetiological agents may not macrolides should not be used as a first line drug in pneumonia. In children with severe and very se- of non-severe pneumonia with co-trimoxazole as a first-line empir- vere pneumonia.by oral ampicillin and cefpodoxime with co-amoxyclavulanic acid (rapid respiration). the Alternative antibiotics for CAP include macrolides. Published by John Wiley & Sons. It can be concluded that there are insufficient data to show supe- Quality assessment of these trials comparing oral with injectable riority of amoxycillin to co-trimoxazole. None had chest X-rays. with the ex- from underdeveloped countries with age groups below five years. ated with similar failure rates. In view of the similar antimicrobial spectrum of all these was lowered to three days. monia. Ltd. suggesting that the specificity of the WHO criteria for diagnosis of true bacterial pneumonia is low. can be treated with oral amoxycillin. . The majority of such children. The cost of amoxycillin would be (Addo-Yobo 2004). viral infec- parisons. procaine penicillin and cefuroxime. The mortal- WHO clinical definition of pneumonia. amoxycillin with ce. Co-amoxyclavulanic However. may therefore be preferable over these drugs. as they were treated with US USD 0. co-amoxy- severity of illness. data comparing two different antibiotics may also be acid has been shown to give better results than amoxycillin and useful in guiding antibiotic therapy in industrialised countries.3). radiographs were normal in 82% of children diagnosed with non- ramphenicol plus ampicillin with penicillin. Concerns clavulanic acid in a single study and may be an alternative where about increasing resistance of common pathogens (S. tion associated wheeze. In a study from Pakistan chest Comparisons for severe and very severe pneumonia involved chlo.6 versus USD 0. ce- pattern. may not require antimicrobial agents and are likely to re- one or co-amoxyclavulanic acid. The results of this review may therefore be more appli- clarithromycin). co-trimoxazole or amoxycillin or any other antibiotics may not Co-amoxyclavulanic acid was better than oxacillin/ceftriaxone and alter their outcome. Therefore oping countries. asthma etc. Most studies comparing co-trimoxa- drugs (ampicillin/amoxycillin/cotrimoxazole/penicillin) and the zole and amoxycillin used clinical case definition of pneumonia Antibiotics for community-acquired pneumonia in children (Review) 17 Copyright © 2013 The Cochrane Collaboration. After excluding one MASCOT Group 2002) reported similar cure rates with amoxy- study that reported seven deaths in children receiving injections cillin given for three or five days. Cefpodoxime was comparable to co-amoxy- rate higher than 40 per 1000 live births (WHO 1991). The aetiological agents vary with age and possibly geo- furoxime). The burden of pneumonia is significant in infants from devel- very quickly if used indiscriminately (Inoue 2006). it is important to have well-designed clinical trials in evidence children with true pneumonia (radiologically confirmed/direct or indirect evidence of bacterial pneumonia). All clinical trials included in that children with severe pneumonia without hypoxia.

Bacterial pathogens could be isolated in only 12% of the study Duke 2002. pneumoniae and H. mycin. the sensitivity pattern of the bacterial pathogen. these studies reveal that the outcome of patients with oral amoxy. pneumoniae. In this review. atypical organisms showed equal cure rates between erythromycin cillin/ampicillin plus gentamycin. Out of 1734 or follow the instructions. co-amoxyclavulanic acid and cefurox. occurred in 45 out of 658 (7%). In one study Respiratory syncytial virus (RSV) could be isolated in 20% of (Hazir 2008) antibiotic use in the last week was associated with patients. S. the possibility of tration of antibiotics. in terms of both clinical practice and research. However. a resistant organism increases (Chenoweth 2000). Many identified atypical organisms (Block 1995. Camargos 1997. From this observation it can be inferred that either the di- native to penicillin/ampicillin. a better outcome could be expected with use mixed infection (bacterial agents with viruses) has been observed of injectable antibiotics for the treatment of children with severe in 10% to 40% of cases (Kabra 2003). In this review. in school-going children who received macrolide in addition to In the management of CAP. agnostic tests used for atypical organisms in these studies may not Cure and failure rates of CAP depend not only on the choice of indicate invasive infections. amoxyclavulanic acid were comparable to those receiving azithro- bination suggesting that co-amoxyclavulanic acid may be an alter. until the result of the culture is available. While tam monotherapy and beta-lactam and macrolide combination information on resistance patterns was not included in the studies therapy on the outcomes of children hospitalised with CAP. or that the study was not adequately antibiotics but also on the aetiology of the pneumonia. Two studies (Harris 1998. and drugs are relatively more expensive. In this review identifi. Therefore. (Addo-Yobo 2004. Nogeova 1997. The most treated with oral amoxycillin. Harris 1998. it is policy to pneumonia. Asghar 2008. Six studies included in this review cillin is the same or better than with standard treatment. empirical antibiotic therapy was similar in the two treatment groups in all the studies. Harris patients on standard treatment did not take injectable medications 1998. More recently two cluster-RCTs (Bari 2011. positive monia with or without hypoxia to identify children who may be tests for Chlamydia spp. it was evident macrolides. Wubbel 1999) com- change the WHO guidelines. There is a need to re-define severe pneu. it may be concluded that positivity for Mycoplasma in children under five years age was 21% children with severe pneumonia without hypoxia may be treated (141/659). Bansal 2006. Block 1995. Chlamydia. Normann 1998. the child will need to be treated with empirical antibiotics addition of macrolides to beta-lactamase antibiotics. hypoxaemia compared oral amoxycillin with standard care (refer. avoid administration of antibiotics. Wubbel 1999). for CAP should be effective against these two pathogens. At present. Even if bacterial pathogens are iso. influenzae constituted 65% of bution of patients who had received antibiotics in the recent past all the bacterial isolates. Tests for Chlamydia spp. Wubbel 1999).) in CAP (Chaudhary 1998. A recent retrospective cohort the patient. Straus 1998). The cure rates in children receiving co- amoxyclavulanic acid is better than an oxacillin-ceftriaxone com. Ribeiro 2011. Mulholland information on past antibiotic use was available in eight studies 1995. paring azithromycin with co-amoxyclavulanic acid in children un- talised children with severe and very severe pneumonia include der five years of age also showed equal cure and failure rates. In children under five years of age. Nogeova 1997. The study did not observe a difference in order to make a decision about the choice of antibiotics is not re-admission rates or a difference in length of stay in children be- feasible in most circumstances. Hazir 2008. Bradley 2007. In ceftriaxone. Soofi treat all children with pneumonia with antibiotics due to a lack of 2012) in children with WHO-defined severe pneumonia without point of care tests that can reliably rule out bacterial pneumonia. this is likely to be of major importance in results of this study suggest that mean hospital stay was 20% less the future. Pandey 2005). suggesting that a sizeable proportion of patients may have increased failure rates on univariate analysis. there is a need to and azithromycin. Results of isms (Chlamydia and Mycoplasma spp. Straus 1998. Based on the results of these studies and children tested forM. Another study showed that co. 385 (22%) tested positive. 1534 children (10%). Roord 1996. subgroup analysis was not available in these studies. levofloxacin. A co-trimoxazole and amoxycillin the number of patients who had child with viral pneumonia can be identified from rapid diagnostic received antibiotics in the recent past was higher in the amoxycillin tests such as nasopharyngeal aspirates (Maitreyi 2000) and can group (34% compared with 25. The evaluated in the review. These patients may not need antibiotics. these studies the incidence of atypical organisms in children un- ime. .6% in the co-trimoxazole group) Antibiotics for community-acquired pneumonia in children (Review) 18 Copyright © 2013 The Cochrane Collaboration. Published by John Wiley & Sons. respectively. However. effective antibiotics against atypical organisms are tetracycline and In children with severe or very severe pneumonia. However. Klein 1995. the studies that attempted to identify that chloramphenicol was inferior to the combination of peni. In a study comparing a viral aetiology of CAP. Alternative antibiotics for hospi. The distri- participants. Kogan 2003. low six years of age. Another important issue is the aetiological role of atypical organ- ral to healthcare services with injectable antibiotics). the study (Ambroggio 2012) compared the effectiveness of beta-lac- severity of disease and any antibiotic usage in the recent past. Exposure to antibiotics in the recent past may adversely affect the cation of bacterial pathogens was attempted in 14 studies (Asghar outcome of bacterial pneumonia as the chances of infection with 2008. Atkinson 2007. isolation of bacterial pathogens in beta-lactamase therapy. comparisons were based on single studies and these der five years of age was 15% and 11% for Mycoplasma spp.possible benefit of better bioavailability with parenteral adminis. Bradley 2007. Duke 2002. were positive in 158 out of the with oral amoxycillin. More studies are required to recommend the lated. the age of powered to detect small differences. The the observations in the present review. Therefore. Ltd. Bradley 2007. Kogan 2003.

of the studies. We countries as most studies were done in these countries. There was more than one study comparing monia who received cotrimoxazole was 56/203 (27. In the present review we also found that oral review. Malnutrition may affect the treatment outcome of pneumonia. ting oral or intravenous antibiotics for the treatment of pneumonia tion of CAP and the response to antibiotic therapy. Failure rates penicillin/ampicillin and chloramphenicol with ampicillin/peni- in those with non severe pneumonia in the same study were 12. Therefore. Another 12 studies were unblinded but had adequate allocation concealment. 3. No cillin and co-trimoxazole versus procaine penicillin) for pneumo- data regarding antibiotic resistance were reported in the majority nia are equally effective. co-amoxyclavulanic acid) gave otic therapy. it is highly probable and intravenous antibiotics (amoxycillin versus penicillin/ampi- that this issue can influence the correct evaluation of the data. these comparisons is of high quality compared to other compar- spectively (Straus 1998). the distribution of viral cases is not uniform. for primary and secondary outcomes in the studies that also en. failure rates in children with severe pneu. is a need to divide WHO-defined severe pneumonia into those dation may be applicable to all age groups. isons. In the present (Rojas-Reyes 2006). we feel that the recommen. Ltd. The study did not show any significant difference in cure rates. fering from severe pneumonia who have received antibiotics in the recent past may benefit from treatment with amoxycillin. How- ever.5%). In this review we included one study (Awasthi 2008) of which one There was only one study in malnourished children (Mulholland of the authors of the present review (Kabra) was a co-author. conclude that there is a need for RCTs to document the advan- tages of the addition of macrolide antibiotics to conventional beta- lactam antibiotics. Agreements and disagreements with other The aetiology of pneumonia depends on the age of the patient. suggesting the evidence for and 12. A review comparing oral and intravenous antibiotics in pneumo- garding the efficacy of antibiotics can be debatable. with other antibiotics (amoxycillin. There not change the outcome. nia suggested no difference in cure and failure rates in children get- vidual factors. probably because the methods for identifying any study that compared beta-lactam antibiotics with and without these pathogens are too complicated or too expensive. Studies comparing macrolides These data are needed to more accurately define the best antibi. we did not find is not established. We tried to see the effect on outcomes after clinically diagnosed pneumonia are not different. A recent retrospective cohort study (Ambroggio 2012) suggests sistance to commonly used antibiotics can have a great influence that the addition of macrolide to beta-lactam antibiotics in chil- on the response to therapy. the conclusions re.5% in those receiving co-trimoxazole or amoxycillin re. identification of aetiological agents is lacking.(Straus 1998). The results may be more applicable for developing similar failure rates suggesting no advantage of macrolides. such as malnutrition. In studies or reviews this review. selective reporting those received antibiotics and those did not receive antibiotics are of data was unclear in 12 studies and 13 studies were funded by not available. classifying them as good-quality AUTHORS’ CONCLUSIONS Antibiotics for community-acquired pneumonia in children (Review) 19 Copyright © 2013 The Cochrane Collaboration. is higher co-trimoxazole with amoxycillin. However. 1995) which compared co-trimoxazole and chloramphenicol. 1. co-amoxyclavulanic acid may parisons are often performed among groups of children for whom be an alternative to ceftriaxone or penicillin/ampicillin. failure rates or need for change in antibiotics. There are limitations in reviewing antibiotic usage in CAP. These results suggest that children suf. these results are based on a single study and care should be Potential biases in the review process taken when drawing any definite conclusions. Quality of the evidence Five out of 29 studies were double-blind and allocation conceal- ment was adequate. can deeply modify the evolu. removing studies that included children older than five years for 2. In the present review. Several indi. Published by John Wiley & Sons. Data were fully detailed in 20 studies. or both. oral amoxycillin with injectable than those who received amoxycillin (18/99) (18%). However. WHO or universities.8% cillin and studies were of good quality. Com. It is well known that in some cases the level of re. only one study addressed this problem. For very severe pneumonia. can be treated with oral antibiotics in an ambulatory setting. In this study separate data regarding failure rates in studies. macrolide for treatment of CAP. This means that if gentamycin combination. Outcomes of children with radiographically confirmed or rolled older children. the majority of enrolled participants were below five The important changes in this updated review in comparison to years of age and separate data according to age were not available the previous version (Kabra 2010) include the following. The role of atypical bacteria in the dren above six years of age may improve outcomes in the form determination of CAP in children living in low-income countries of reduced hospital stay. more studies with hypoxia and those without hypoxia to identify children who are required for children older than five years of age. . WHO-defined severe pneumonia without hypoxaemia can severe pneumonia and observed that the age of participants did be managed with oral antibiotics in an ambulatory setting.

ACKNOWLEDGEMENTS cillin on an ambulatory basis. Smyth A.Pakistan 2003. We acknowledge all the help and infrastructure provided by the For children below five years of age. Group. graphically confirmed pneumonia in place of clinically diagnosed empirical antibiotics may be used as follows. Aurangzeb B. co-trimoxazole in the treatment of non-severe pneumonia BMJ 2008. New Delhi. Jeena P. penicillin/ampicillin plus gentamycin is superior authors serve as faculty.364:1141–8. ceftriaxone. Chisaka N. input in improving the quality of the previous version of this tibiotic treatment for pneumonia. Kabra SK. We More randomised controlled trials are required for a review of these are thankful to Dr Shivani Randev for helping in getting full- antibiotics in order to make more accurate recommendations for text articles for oral treatment of severe pneumonia. levofloxacin and cefuroxime. Pillai RM. of community acquired pneumonia in children. We are very their prescription. Until more Search Co-ordinator of the Cochrane Acute Respiratory Infections studies are available these can be used as second-line therapies.62:1102-6. et al. Agarwal G. amoxyclavulanic acid. For the treatment pneumonia. REFERENCES References to studies included in this review intravenous benzylpenicillin for community acquired pneumonia in children (PIVOT trial): a multicentre Addo-Yobo 2004 {published data only} pragmatic randomised controlled equivalence trial. Weston Bansal 2006 {published data only} V. where all the very severe CAP. failure who can be managed with second-line antimicrobials early. review. Effectiveness of 3-day amoxycillin vs. Banajeh S. Journal of Asghar 2008 {published data only} College of Physicians and Surgeons . et al. JM. Marilyn Bamford. five years of age. Comparative efficacy of amoxycillin. Lozano 2007. of pneumonia of varying severity. . Chloramphenicol versus ampicillin plus Awasthi 2008 {published data only} gentamicin for community acquired very severe pneumonia Awasthi S. in children aged 2-59 months of age: a multi-centric open labelled trial. or cefpodoxime and co-amoxyclavulanic acid. Journal Tropical Pediatrics 2008. thankful to the referees Dr Roger Damoiseaux. Studies should include radio- tification of aetiological agents for pneumonia are not available.54:382–9. Hassan M.Implications for practice Implications for research In children presenting with community-acquired pneumonia There is a need to compare various antibiotics for the treatment without underlying illness. and where point of care tests for iden. Ltd. Penicillin and gentamicin Antibiotics for community-acquired pneumonia in children (Review) 20 Copyright © 2013 The Cochrane Collaboration. among children aged 2-59 months in low resource settings: Singhi S. for doing the EMBASE search and obtaining the full- More studies are required to assess the role of the addition of text articles of studies. Dr Nicola Principi. Hibberd P. Comparison of oral amoxicillin and Bansal A. Lakhanpaul M. Hameed A. Oral amoxycillin versus injectable Aurangzeb 2003 {published data only} penicillin for severe pneumonia in children aged 3-59 months: a randomized multicenter equivalency study.13:704–7. Vyas H. The results of such studies will ternative to co-trimoxazole. Katep- Bwalya M. We acknowledge the help provided by to chloramphenicol. Dr Rajni Bhatia and Dr Mark Jones for their sistance in different geographic regions for developing empiric an. Jayashree M.336:80–4. We acknowledge all referees for critically macrolide antibiotics to beta-lactam antibiotics in children above reviewing and suggesting improvements to the quality of review. Thorax Addo-Yobo E. Severe pneumonia in children below five years of age. 5-day multicentre randomised controlled trial (SPEAR study). hospitalised with severe and All India Institute of Medical Sciences. Atkinson 2007 {published data only} Atkinson M. azithromycin and co. that lead to failure of treatment. et al. Managing Editor and Sarah Thorning. There is need for surveillance studies to document antibiotic re. Asghar R. Hibberd P. There are no apparent differences help in the formation of guidelines to identify children at risk of between azithromycin and erythromycin. Singhi SC. Other alternatives may be co-amoxyclavu- Elizabeth Dooley. Studies should try to identify the aetiological agents of WHO-defined non-severe community-acquired pneumonia and their susceptibilities to various antibiotics and the risk factors (CAP) in children below five years of age amoxycillin is an al. et al. can be managed with oral amoxy. without hy- poxia and accepting oral feeds. Sithole J. There are limited data on other antibiotics: co-amoxyclavu- lanic acid and cefpodoxime may be alternative second-line drugs. cefuroxime and clarithromycin in the treatment Lancet 2004. We also acknowledge the help provided by Mr Bharat Bhusan Pandey and Mr Rajat Prakash in masking the study articles. Singh JV. Iqbal I. Trials lanic acid. Egas J. Published by John Wiley & Sons.

children. Ashraf YP. Lancet Pediatric Pulmonology 2003. Guimarlies MDC. Odumah DU. Published by John Wiley & Sons.14:959–65. acquired pneumonia in young Gambian children.2(8457):684–6. Sa´ ez-Llorens X. Lancet 2008. Safety and efficacy of azithromycin Shann 1985 {published data only} in the treatment of community acquired pneumonia in Shann F. Clinical efficacy of cotrimoxazole Giadom PNRB.17: chloramphenicol plus penicillin for severe pneumonia in 865–71. Pediatric Infectious Disease Journal 1995. et al. Ashok TP. Rubilar L. Martinez MA. Arguedas A.14(Suppl):19–22. 43:353–60. Lancet 1985. Effectiveness of ampicillin and Ribeiro CF.26: penicillin for out patient treatment of childhood pneumonia 868-78. Bradley JS. Benzathine Klein M. O’Neill KP. Forgie IM. Ltd. Kogan 2003 {published data only} Campbell 1988 {published data only} Kogan R. A randomized trial of versus amoxicillin twice daily for treatment of pneumonia: chloramphenicol vs trimethoprim sulphamethoxazole for a randomized controlled clinical trial in Pakistan. Cetinkaya 2004 {published data only} Nogeova 1997 {published data only} Cetinkaya F. Byass P. Rathnam SR.40:2765–8. Michael A.33:813–6. WHO Bulletin 1990. Pediatric Infectious Diseases Journal 1995. Multicenter trial penicillin for unilateral lobar or segmental infiltrates of cefpodoxime proxetil vs. Nisar YB. Gossens MM. Craft a randomised equivalency trial. Keeley 1990 {published data only} Melkote R. Pediatric children 2-12 years old. Mgone J. Galova K.2:1182–4. Indian Journal of Pediatrics 2006. Cizmarova E.371:49-56. Deivanayagam N. clavulanic acid (Augmentin) in the treatment of bacterial pneumonia in children. Blumer JL.68:185–92. therapy for community-acquired bronchopneumonia: 71:969–72. The International Study Group. Fioretto JR.14: comparative evaluation of amoxycillin and amoxycillin plus 471–7. Mycoplasma pneumoniae and Chlamydia pneumoniae in pediatric community-acquired pneumonia: comparative Jibril 1989 {published data only} efficacy and safety of clarithromycin vs. Kolokathis A. et al. Poka H. Comparative study of levofloxacin Keeley DJ. Antibiotics for community-acquired pneumonia in children (Review) 21 Copyright © 2013 The Cochrane Collaboration. Falade AG. et al. Wal T. Ferrari GF. Kutluk G. Roord JJ. Mulholland 1995 {published data only} CATCHUP 2002 {published data only} Mulholland EK. JC. Nkrumah FK. Antimicrobial Agents and Harris JS. Lloyd Evans I. Noel GJ. CATCHUP Study Group.6:133–4. Sufliarska S. Hedrick J. Poore P. Greenwood BM. cefuroxime-axetil in initial childhood pneumonia. Puppo H. Fox MP. of Public Health 2011. Barker J. Pediatric Infectious Disease Journal 2007. Cassell GH. 1988. Pediatric Infectious Disease Journal 1998. erythromycin Jibril HB. Ambulatory short-course high-dose oral Block 1995 {published data only} amoxicillin for treatment of severe pneumonia in children: Block S. Nedunchelian K. et al. Corrah PT. An open label ethylsuccinate. antibiotic regimens in the empirical treatment of severe Raskova J. amoxycillin-clavulanate in acute presumptively caused by Streptococcus pneumoniae in lower respiratory tract infections in childhood. Chemotherapy 1996. acquired pneumonia. Duke 2002 {published data only} Roord 1996 {published data only} Duke T.35:91–8. Archives the treatment of malnourished children with community of Diseases in Childhood 2002. Randomized in the treatment of children with community-acquired trial of sulphamethoxazole + trimethoprim versus procaine pneumonia. Wolf BH. Hammerschlag MR. MacLeod WB. Mala N. Ribeiro 2011 {published data only} Sheela D. Chloramphenicol alone versus children. Infectious Disease Journal 1995. Adegbola RA. Comparative randomized trial of N. Comparison of two Nogeova A. Lancet 2002. Pan American Journal Indian Pediatrics 1996.73:305–9. Campbell M.11(9):585–92. Prospective Chloramphenicol versus benzylpenicillin and gentamycin open randomized study comparing efficacies and safety for the treatment of severe pneumonia in children in Papua of 3 day course of azithromycin and 10 day course of New Guinea. Ferreira CS. Indian Journal of Pediatrics 2004. Infectious Deivanayagam 1996 {published data only} Diseases in Clinical Practice 1997. in Zimbabwe. randomized multicentric study in 140 children.359:474–80. Quevedo Campbell H. Kapuyanyika C.86:113–8.6:444–50. Hammerschlag MR. Hazir T. erythromycin in children with community acquired acute Harris 1998 {published data only} lower respiratory tract infections. Fox LM. Ceftibuten vs. Antibiotic treatment combination of penicillin and chloramphenicol in the schemes for very severe community-acquired pneumonia in treatment of pneumonia: randomized controlled trial. Ifere OAS. Camargos 1997 {published data only} Klein 1995 {published data only} Camargos PAM. children: a randomized clinical study. Omoshigho C. Kimpen JL. Trial of cotrimoxazole versus procaine azithromycin versus erythromycin and amoxycillin for penicillin with ampicillin in the treatment of community treatment of community acquired pneumonia in children. therapy vs amoxicillin/clavulanate in severe hypoxemic Hazir 2008 {published data only} pneumonia. Current Medical Research and Bradley 2007 {published data only} Opinion 1989. Dale F. . Cassell GH. Gogremis A. Paya E. Journal of Tropical Pediatrics 1997. Krizan L.

Kallio MJ. Hodinka RL. Trujillo M. Immunology treatment of severe lower respiratory tract infections in and Infection 2008. amoxicillin-clavulanate in AA. Weber K. Published by John Wiley & Sons.24:1854–70. Vuori-Holopainen E. Nomani NK. Craig The Pakistan Cotrimoxazole Study Group.16:305–11. Haffejee 1984 {published data only} Effectiveness of intramuscular penicillin versus oral Haffejee IE. Khan IU. Delacourt C. Ibrahim MI. Chen JM. Journal of Microbiology.4:265–71. Comparison of minocyclin versus pneumonia. Evans efficacy and safety of cefaclor vs. Antimicrobial WA. Clinical trial of pediatric lower respiratory tract infection: Bonvehi 2003 {published data only} results and comments with cefetamet pivoxil. Droghetti cotrimoxazole versus procaine penicillin G and benzathine R. Scott MG. Ahmed A. days of parenteral beta-lactam treatment for common Antibiotics for community-acquired pneumonia in children (Review) 22 Copyright © 2013 The Cochrane Collaboration. Bell LM. Agostoni 1988 {published data only} Comparison of oral cefuroxime axetil and oral amoxycillin/ Agostoni C. Khan A. A comparison of cefditoren pivoxil and amoxicillin/ resistance and clinical effectiveness of cotrimoxazole versus clavulanate in the treatment of community acquired amoxycillin for pneumonia among children in Pakistan: pneumonia: a multicenter prospective randomized randomised controlled trial.18:98–104. pneumonia. Shortridge D. Pediatric JB. Collins JJ.6: therapy in children hospitalized with community-acquired 77–81.24: 40:301–4. Journal of Antimicrobial Chemotherapy 1984. Tabb LP. Jenkins J. Journal of Tropical Pediatrics 1994. Oguz F. et al. comparative study of clarithromycin and Al-Eidan FA. Pediatric Emergency Care 1998. clavulanate in the treatment of community-acquired Galluzzo C. Hidalgo H.161:1097–103. Giguere G. Pharmacy World and Science 2005. Taylor JA. Huang LM. Lee PI. Chinese Journal of Infection and Chemotherapy 2006. Palo WA. Schwartz B. Mouallem 1976 {published data only} Bari 2011 {published data only} Mouallem R. Begliatti E. Yang YJ.Sidal 1994 {published data only} Esposito 2005 {published data only} Sidal M. Journal of International Medical et al. SE-TU drug-resistant Streptococcus pneumoniae. Fogarty CM. Shah SS. Chen HZ. Clinical Drug Study Group. Journal Lee 2008 {published data only} of International Medical Research 1988. Role of atypical bacteria and penicillin + procaine penicillin G in the treatment of azithromycin therapy for children with recurrent respiratory childhood pneumonia. Scaglione F. Ahmad Z. Journal of Pediatrics 2012.23:491–501. . Lee CY. clinical response and effect on natural killer activity. Community case management of severe pneumonia Research 1976. Wu MH. and cephalexin in children.41(1):54–61. Sulaiman SA. Sorin M. Ltd. et al. Hasali 2005 {published data only} McCoig C. Tsarouhas 1998 {published data only} Tsarouhas N.27:249–53. Al-Eiden 1999 {published data only} randomized. A therapeutic trial of cefotaxime versus amoxicillin in the early treatment of outpatient pediatric penicillin-gentamycin for severe infections in children. Kearney MP. Neyzi O. Sequential antimicrobial therapy. Riva E.14(Suppl B): 147–52. Bosis S. Notario G. Trial of Esposito S. acquired pneumonia. Shaw KN. Lancet Paupe J. Kundi Z. Scheinmann P. Journal of Antimicrobial Chemotherapy 1999. Comparison of clarithromycin and amoxicillin/clavulanic Peltola 2001 {published data only} acid for community-acquired pneumonia in an era of Peltola H.352:270–4. investigator-blinded parallel group study.378:1796–803. 2011. Qing M. Sarbeji M. Comparative effectiveness of empiric β. Carubelli C. Clinical Therapeutics 2002. An open. Journal of Antimicrobial Chemotherapy 1996. Successful shortening from seven to four Investigation 2003. A Ambroggio 2012 {published data only} prospective multi-center randomized parallel study on Ambroggio L. Cyganowski M. pneumonia between single versus combination therapy.14:338–41. erythromycin in the treatment of children with community- Troughton KE. Faelli N.44: Lu 2006 {published data only} 709–15. with oral amoxicillin in children aged 2-59 months in Paupe 1992 {published data only} Haripur district. Hasali acquired pneumonia in ambulatory children. References to studies excluded from this review Higuera 1996 {published data only} Higuera F. tract infections. Tremolati E. Etiology and treatment of community. Giovannini M.38:29–32. children. Qazi SA. Hom RC. 1992. Lehri IA. Pakistan: a cluster randomised trial. Sarbat G. Straus 1998 {published data only} Fogarty 2002 {published data only} Straus WL. Newschaffer CJ. Sadruddin S. et al. Lu Q. Muniz L. A clinical and economic study of community-acquired Infectious Disease Journal 1999. 438–44. McElnay JC. Hasali MA. amoxycillin in lower respiratory tract infections in children: 37:555–64. Khan A. children with acute bacterial infection of lower respiratory lactam monotherapy and β-lactam-macrolide combination tract. Pediatric Infectious Disease Journal 2005. Zhang LE. Unuvar A. Busman T. Wubbel 1999 {published data only} Wubbel L. pneumonia. Fraschini F. Lancet 1998. Comparative efficacy and safety of cephradine Bari A. Chemotherapy Bonvehi P. Feris J.

Bucher HC. .159: March 2011]. 10(Suppl 2):86–91. Prospective and comparative pneumoniae: implications for patients with community- study between cefuroxime. S. Zameer multicenter. Klassen TP. editors(s). pathogens in Japan during PROTEKT years 1-3 (1999- 2002). Niimi H.5:3–8. et al. Journal of Maitreyi RS. Etiology of acute lower respiratory tract Clinical Epidemiology 1997. SA. Additional references IBIS 1999 Agarwal 2004 Invasive Bacterial infection Surveillance (IBIS) Group and Agarwal G. Martinez F. Griffith LE. meta-analysis of randomized controlled trials. Banos V. Kansenshogaku Zasshi Bucher 1997 1993. Ghosh M. www. Journal of Infection and Chemotherapy 2006. pneumonia in developing countries--a review. Iniguez JL. Dhawan B. Lancet 1999.24:1840–53. diagnosed with non-severe pneumonia as defined by World Vuori-Holopaine 2000 {published data only} Health Organization: descriptive multicentre study in Vuori-Holopainen E. Singhi S.0 [updated and cefuroxime. 791–6. Reviews of Infectious J. LaFata JA.50:683–91. Moulin F. Three day versus five day treatment with of Streptococcus pneumoniae disease in India.65:717–21. Archives of Ishiwada N. Peltola H. Lefebvre C. childhood infections: a prospective and randomized study. double-blind study. Antimicrobial susceptibility of respiratory tract Diseases 1990. van Zyl 2002 {published data only} European Journal of Clinical Microbiology and Infectious van Zyl L.75: clavulanic acid in the treatment of community-acquired 1161–8. Ltd.1. Available from 878–84. INCLEN.70:33–6. Chest radiography in children aged 2-59 months Clinical Therapeutics 2002. Broor S. Blenk H. Epidemiology. Indian for the treatment of pneumonia in children. et al. Indian Journal of Pediatrics 2003. Gomez J. Infection 1982. Published by John Wiley & Sons. Nazima N.13(Suppl 60):454–67. et al. Fox MP. ceftriaxone and amoxicillin. Boluyt N. BMJ 2004. The Cochrane Collaboration. European Journal of Pediatrics 2000. The Kabra 2003 results of direct and indirect treatment comparisons in Kabra SK. Ravilly S. Thea DM. Nisar YB. Awasthi S. Palo WA. Berman 1990 Inoue 2006 Berman S. rural Pakistan: a cluster. prospective. Etiology and response to antibiotic therapy randomised controlled trial. Kaul A. le Roux JG. Ahmed S. acquired pneumonia. African Soofi 2012 {published data only} Journal of Medicine and Medical Sciences 2011. Kabra SK. Effectiveness of community case management of Gendrel 1997 severe pneumonia with oral amoxicillin in children aged Gendrel D. et al. Saint S.67:642–7. Mayo Clinic Proceedings 2000. Kaneko K. Flamm R. Prevalence Ruhrmann 1982 {published data only} of mycoplasma pneumonia and chlamydia pneumoniae Ruhrmann H. pediatric in patients with pneumonia.328: Clinical Epidemiology Network (INCLEN). Akizawa K. Kurosaki T.16:388–91. MacLeod WB. Chenoweth 2000 Sanchez 1998 {published data only} Chenoweth CE. Habib F. International multicentre randomised controlled trial. et. infection. Khan SF. Chaudhary R. BMJ 2006. Erythromycin versus amoxycillin in children with community acquired pneumonia. et al.11(2):132–8. SE-TU Pakistan. Prospective multicentre hospital surveillance Walter SD. Toba T. management of childhood acute community-acquired Revista Espanola de Quimioterapia 1998. pneumonia [Estudio prospectivo y comparativo entre Falade 2011 cefuroxima. Ruiz Fendrick AM. Raza M. Antibiotics for community-acquired pneumonia in children (Review) 23 Copyright © 2013 The Cochrane Collaboration.33:629–33. Guyatt GH. Green S.40(4): Soofi S. Narrow versus broad spectrum parenteral Higgins 2011 antimicrobials against common infections of childhood: Higgins JPT. Etiology of Pediatrics & Adolescent Medicine 2008. Kabra SK. Fujita S. 2-59 months in Matiari district. Epidemiology of acute respiratory tract infection Inoue M. of community-acquired pneumonia in French children. Antimicrobial resistance in Streptococcus Gomez J. Cefditoren pivoxil versus cefpodoxime Hazir 2006 proxetil for community-acquired pneumonia: results of a Hazir T.org 2011. Walter SD. Qazi SA. Offringa M. et al. Volk RS. Lynch JP 3rd. Journal of Pediatrics 1998. Gomez Vargas J. Kallio MJ. Lodha R. Lancet 2012. aetiology and tratamiento de la neumonia adqirida en en la comunidad]. Chaudhary R. Raymond J. Sanchez ME. Diseases 1997. Tjosvold L. 353:1216–21.12: Boluyt 2008 9–21. Igari in children of developing countries. randomized. Usefulness of systematic review search strategies in finding Ishiwada 1993 child health systematic reviews in MEDLINE.379:729–37. Ayede AI.162(2):111–6. Study Group. Cochrane Handbook for a prospective randomized comparison between penicillin Systematic Reviews of Interventions Version 5. ceftriaxona y amoxicilina –clavulanico en el Falade AG. Invasive amoxicillin for non-severe pneumonia in young children: a Bacterial Infection Surveillance (IBIS) Group. Kaku M. Qazi 293–308.cochrane-handbook. Munoz L. Chaudhary 1998 International Journal of Infectious Diseases 2001.

86:408–16.379(9832):2151–61. Technical basis of WHO pneumonia: a multicenter randomised blind trial. regional. The management of pneumonia in children in pneumonia in children. Making a difference. Etiological agents of lower respiratory tract serious infections in young infants in developing countries: infections in Japanese children.20. The Cochrane Collaboration.19:924–8. Chlamydia WHO 1999 pneumoniae in children with acute respiratory tract World Health Organization. Lodha R. Granados Rugeles C.18:67–71. Scott S. Effect of pneumonia case management on mortality an updated systematic analysis for 2010 with time trends in neonates.2. [DOI: 10. infants. resistance. Cochrane Database Rasmussen Z.1.CD004874. Johnson HL. et Sazawal S. Antibiotics for Diseases 1997. Journal of Clinical Virology from children in Pakistan for surveillance for antimicrobial 2000. Handbook for Systematic Reviews of Interventions Version Campbell H. Boschi-Pinto C.Krishnan 2011 The Nordic Cochrane Centre.87:23–7. Lancet 2003. and national causes of child mortality: Group. Pandey RM. children at first level health facility. Numazaki 2004 WHOYISG 1999 Numazaki K. Wettergren B. [DOI: 10. WHO/ARI/91. Chichester: Wiley-Blackwell. Kabra SK. Gnarpe J. RK. Pneumonia Case Management Trials al.18(Suppl):17–22. 10. Clinical Infectious Diseases 1995.360:835–41.1002/14651858. Chiba S. Rajendran P. Mastro MD. Gnarpe H. pneumonia.12: MASCOT Group 2002 824–30. World infection. MacCracken 2000 Shann 1995 MacCracken Jr GH. Mulholland K. Manheimer E. Khobragade KJ. Journal of the versus parenteral antibiotics for severe pneumonia in Indian Medical Association 2011. Anitha C. and preschool children: a meta-analysis since 2000. community-acquired pneumonia in children. Geneva 1999. Ghafoor Dar L. Lawn JE. Pandey RM. Krishnan P. Health Report 1999. Epidemiology and etiology of childhood 5. Chaudhry R. Glanville J.CD004874. Kabra SK. Cousens S. Bulletin of the World Health Organization 2008. Issue 3. Green S editor(s). Rapid detection of respiratory viruses by A. children. Diagnosis and management of Shann F. Shaukat NF.47:97–101. MASCOT Group. Broor S. Published by John Wiley & Sons. Oral antibiotics paediatric cases of acute respiratory infections. The WHO Young Infant Study Group. Umetsu M. et al.pub3] Review Manager (RevMan). World Health Organization 1991. Liu 2012 Sazawal 2003 Liu L.21 2000. [DOI: The Nordic Cochrane Centre. Normann E. Version 5.1002/14651858. Lefebvre 2011 Issue 2. Cochrane RevMan 2012 [Computer program] Database of Systematic Reviews 2010. In: Higgins JPT. The Cochrane Collaboration. Use of nasopharyngeal isolates of centrifugation enhanced cultures from children with acute Streptococcus pneumoniae and Haemophilus influenzae lower respiratory tract infections. Maitreyi 2000 Timothy 1993 Maitreyi RS. Pandey 2005 Pandey A. . results of multicentre study.1:119. Kabra SK. Seth P. Lancet recommendation on the management of pneumonia in 2002.CD004979. Lancet 2012. Copenhagen: ∗ Indicates the major publication for the study Antibiotics for community-acquired pneumonia in children (Review) 24 Copyright © 2013 The Cochrane Collaboration. Black RE.pub2] Lefebvre C. Qazi SA.1002/ strength co-trimoxazole for the treatment of childhood 14651858. International Journal of Tuberculosis and Lung Kabra SK. Cochrane Database of Systematic Reviews 2006. (Suppl):218–25. Pediatric Infectious Diseases Journal 1993. Acta Paediatrica Scandinavica 1998. Biloglav Z. Standard versus double of Systematic Reviews 2006. of community-based trials. Bari Z. Nomani N. randomized multicentric trial Kabra 2010 in Pakistan.0 [updated March 2011]. Clinical efficacy of three day versus five WHO 1991 days of oral amoxicillin for the treatment of childhood World Health Organization. Cochrane Rudan I. et al. Chapter 6: Searching Rudan 2008 for studies. Pediatric Infectious Diseases Journal 1999. Kapoor L. Bacterial etiology of Kuniya Y. In Vivo 2004. Normann 1998 Geneva. Ishaq Z.109:241–2. Lodha R.3:547–56. Acute lower References to other published versions of this review respiratory tract infection due to Chlamydia species in children under five years of age.pub2] pneumonia: a double blind. Antibiotics for community Rasmussen 1997 acquired pneumonia in children. Evaluation of coamoxiclav and other Rojas-Reyes 2006 antibiotics against S pneumoniae and H influenzae from Rojas-Reyes MX. Timothy D. 2011. et al. Indian Journal of Chest Kabra 2006 Diseases & Allied Sciences 2005. Yoshimura H. Ghosh M. Sambandan AP. Perin J. Tanaka T. Issue 3. Chavda 2012.16:41–7. Ltd. Pediatric Infectious Disease Journal developing countries. Global.

Ltd.000 IU/kg or PO amoxycillin 45 mg/kg/day Outcomes Failure rate at 48 hours. stratified by site and pre- pared in advance by the WHO Blinding of participants and personnel High risk Open-label study (performance bias) All outcomes Blinding of outcome assessment (detection High risk Open-label study bias) All outcomes Selective reporting (reporting bias) Low risk No selective reporting Incomplete outcome data (attrition bias) Low risk Incomplete data adequately addressed All outcomes Other bias Low risk Funded by World Heath Organization and Applied Research Child Health Project. Boston University Antibiotics for community-acquired pneumonia in children (Review) 25 Copyright © 2013 The Cochrane Collaboration. penicillin allergy Risk of bias Bias Authors’ judgement Support for judgement Random sequence generation (selection Low risk Random sequence generated by World bias) Health Organization (WHO) Allocation concealment (selection bias) Low risk Randomisation codes were sealed in opaque envelopes in accordance with allo- cation sequence. very severe disease. non-severe pneumonia.CHARACTERISTICS OF STUDIES Characteristics of included studies [ordered by study ID] Addo-Yobo 2004 Methods RCT comparing amoxycillin and penicillin Participants Children 3 to 59 months with severe pneumonia Interventions Daily IM penicillin 200. 5 days and 14 days and death rate Notes Exclusion criteria: asthma. audible wheeze. . persistent vomiting. clinical HIV. Published by John Wiley & Sons.

known renal failure or not passed urine during past 6 hours. tachypnoea (≥ 50 breaths/ minute in children aged 2 to 11 months and ≥ 40 breaths/minute in children aged 12 to 59 months) and abnormally sleepy or difficult to wake 2. Treatment failure by 5 days after admission. Voluntary withdrawal or absconding 5. defined as new development or persistence of at least 2 of the following: inability to drink. known heart disease. renal failure or newly diagnosed co-morbid conditions. and gentamicin 7. empyema. with oxygen saturations ≤ 90%. septic shock. Ltd. evidence of cerebral malaria Evidence of bacterial meningitis. or known asthma. Modification of antibiotic treatment 4. with a history of 3 or more attacks. After that first group received oral amoxycillin (45 mg/kg/day in 3 divided doses) and the other group received oral chloramphenicol 75 mg/kg/day to complete 10 days Outcomes Primary outcome 1.Asghar 2008 Methods Randomised. Development or diagnosis of any of the following: bacterial meningitis. . non-blinded. clinical jaundice. previous enrolment in the study Admission to hospital for more than 24 hours within past 7 days Documented evidence of injectable antibiotic treatment for more than 24 hours before enrolment. empyema or presence of pneumatoceles on chest radiograph Risk of bias Bias Authors’ judgement Support for judgement Random sequence generation (selection Low risk Sequence generated by World Health Or- bias) ganization by using variable size blocks Antibiotics for community-acquired pneumonia in children (Review) 26 Copyright © 2013 The Cochrane Collaboration.5 mg/kg/d as in a single daily dose) or chloramphenicol (75 mg/kg/d given in 3 doses) every 8 hours for minimum of 5 days. Serious adverse drug reaction 3. Death Secondary outcomes Treatment failure as defined above at 48 to 60 hours Treatment failure as defined above plus relapse (hypoxaemic pneumonia at 10 to 12 days and 21 to 30 days. multi-site efficacy study Participants Children between 2 to 59 months of age with very severe pneumonia Interventions Ampicillin plus gentamicin (ampicillin 200 mg/kg/d in 4 doses every 6 hours. Published by John Wiley & Sons. duration of present illness more than 10 days. residence of patient in an area where follow-up was not possible. history of serious adverse reaction to any of the study drugs. or ≤ 88% in the 2 high altitude sites in Mexico and Yemen) Death by 30 days after enrolment Bacterial pathogens isolated from blood or other sterile sites Antimicrobial susceptibility of the isolated pathogens Notes Exclusion criteria: wheezing. stridor.

Baltimore Atkinson 2007 Methods Multicentre randomised controlled trial Participants Children admitted with pneumonia in 8 hospitals. further courses of antibiotics). Boston University and Johns Hopkins Bloomberg School of Public Health. immunodeficiency. complications (empyema. shock requiring 20 ml/kg fluid resuscitation. fever > 37.Asghar 2008 (Continued) Allocation concealment (selection bias) Low risk Separate randomisation lists were prepared for each site examination. sealed envelopes. Respiratory symptoms or signs. Before opening each en- velope the doctor in charge signed and dated the opening flap of the envelope Blinding of participants and personnel High risk Open-label randomised controlled trial (performance bias) All outcomes Blinding of outcome assessment (detection High risk Open-label randomised controlled trial bias) All outcomes Selective reporting (reporting bias) Low risk No selective reporting Incomplete outcome data (attrition bias) Low risk Incomplete data adequately addressed All outcomes Other bias Low risk Funded by World Health Organization and Center of International Health and De- velopment. Ltd. At least 3 inclusion criteria for diag- nosis of pneumonia. and individual patient assignments were placed in opaque. . Published by John Wiley & Sons. 85% in air.5 °C. radiographically confirmed pneumonia (defined as confluent area of consolidation agreed subsequently by 2 independent radiologists) Interventions Oral amoxycillin (doses for 6 months to 12 years of age 8 mg/kg/dose 3 times a day above 12 years of age 500 mg 3 times a day) or IV benzyl penicillin (doses 25 mg/kg/ dose 4 times a day) Outcomes The primary outcome measure was time from randomisation until the temperature was 38 °C for 24 continuous hours and oxygen requirement had ceased (the latter only applicable to those children who required oxygen during the admission) Secondary outcomes included time in hospital. duration of oxygen requirement and time to resolution of illness Notes Exclusion criteria were wheeze. pleural effusion at presentation requiring Antibiotics for community-acquired pneumonia in children (Review) 27 Copyright © 2013 The Cochrane Collaboration. re-admission. oxygen saturations.

penicillin allergy and age 6 months Treatment with oral antibiotics in the 5 days prior to admission. Published by John Wiley & Sons. Ltd. The sequence was accessed via the Internet. including amoxycillin.Atkinson 2007 (Continued) drainage. therefore allowing concealment of allocation Blinding of participants and personnel High risk Open-label study (performance bias) All outcomes Blinding of outcome assessment (detection High risk Open-label study bias) All outcomes Selective reporting (reporting bias) Unclear risk While the authors mention the primary out- come as “the time from randomisation un- til the temperature was less than 38 degree celsius for 24 continuous hours and oxygen requirement had ceased”. was not an exclusion criterion Risk of bias Bias Authors’ judgement Support for judgement Random sequence generation (selection Low risk Block randomisation list generated bias) Allocation concealment (selection bias) Low risk A block randomisation sequence stratified by centre was produced using a random number generator. admitted in the hospital with community- acquired pneumonia Antibiotics for community-acquired pneumonia in children (Review) 28 Copyright © 2013 The Cochrane Collaboration. non-blinded controlled clinical trial Participants Children between 3 to 72 months of age. . they calculated the sample size based on the proportion meeting the primary outcome measure at any time. The authors have not reported on these pro- portions in the results Incomplete outcome data (attrition bias) Low risk None All outcomes Other bias Low risk Funded by the British Lung Foundation Aurangzeb 2003 Methods Randomised. chronic lung condition (excluding asthma).

. Worse was defined as development of severe pneumonia or very severe disease 3. Group 2 was given cefuroxime 75 mg/kg/d IV in 3 divided doses and Group 3 was given clarithromycin 15 mg/kg/d IV divided into 2 divided doses Outcomes 1. between 2 months to 59 months with WHO-defined non-severe pneumonia Antibiotics for community-acquired pneumonia in children (Review) 29 Copyright © 2013 The Cochrane Collaboration.Aurangzeb 2003 (Continued) Interventions The patients were randomly allotted to 1 of the 3 groups Group 1 was given amoxycillin 75 mg/kg/d IV in 3 divided doses. open-label trial Participants Children of either sex. Cure was defined as return of respiratory rate to age specific normal range Notes - Risk of bias Bias Authors’ judgement Support for judgement Random sequence generation (selection Unclear risk No information provided bias) Allocation concealment (selection bias) Unclear risk No information provided Blinding of participants and personnel High risk Open-label study (performance bias) All outcomes Blinding of outcome assessment (detection High risk Open-label study bias) All outcomes Selective reporting (reporting bias) Unclear risk There are discrepancies in the number of patients in different study arms Incomplete outcome data (attrition bias) Low risk None All outcomes Other bias Unclear risk Source of funding not mentioned Awasthi 2008 Methods Cluster-randomised. Improvement defined as slower respiratory rate (either back to normal for the age of the child). or more than 5 as compared to the previous day evaluation without retractions. Published by John Wiley & Sons. Ltd. The same defined as still breathing fast as before as or higher than that with no chest in drawing or danger signs 2.

defined as devel- opment of signs of WHO-defined pneumonia among the clinically cured in either arm Notes - Risk of bias Bias Authors’ judgement Support for judgement Random sequence generation (selection Low risk Patients were randomly allocated bias) Allocation concealment (selection bias) Low risk This was an open-label study and the unit of randomisation was primary health centre Blinding of participants and personnel Unclear risk Open-label randomised controlled trial (performance bias) All outcomes Blinding of outcome assessment (detection High risk Open-label randomised controlled trial bias) All outcomes Selective reporting (reporting bias) Low risk No selective reporting Incomplete outcome data (attrition bias) Low risk Incomplete data adequately addressed All outcomes Other bias Low risk Funded by Indian Council of Medical Research Antibiotics for community-acquired pneumonia in children (Review) 30 Copyright © 2013 The Cochrane Collaboration. (v) lost to follow-up on day 4 or day 6 in the amoxycillin and co-trimoxazole arms. . (iv) death within the follow-up period of 14 days. or (vi) withdrawal at any time (requirement of intention-to-treat (ITT) analysis) The secondary outcome measure was clinical relapse on day 13 to 15. respectively. (iii) documented axillary tempera- ture > 38. Published by John Wiley & Sons.3 °C on the day of outcome assessment. Ltd. Doses of amoxycillin were between 31 to 51 mg/kg/day and trimethoprim 7 to 11 mg/kg/day Outcomes Primary outcome measure was clinical failure defined as presence of at least 1 of the following: (i) development of signs of WHO-defined severe pneumonia or very severe disease (ii) respiratory rate above age specific cut-off. that is day 4 for amoxycillin and day 6 for co-trimoxazole arm.Awasthi 2008 (Continued) Interventions Eligible children were randomised to receive oral dispersible scored amoxycillin (125 mg per tablet) given thrice a day (tds) for 3 days or co-trimoxazole (20 mg trimethoprim per tablet) given twice a day (bd) for 5 days.

prior antibiotic therapy > 24 hours. Group B patients were given amoxycillin-clavulanate 30 mg/kg IV q12h for at least 3 days and were changed to oral amoxycillin-clavulanic acid when able to feed Outcomes Treatment failure was defined as any change.. development of serious intercurrent illness or complications such as refractory septic shock. modification or discontinuation of allocated antibiotic therapy because of deterioration in patient’s condition.Bansal 2006 Methods Open-label randomised controlled trial Participants Children aged 2 to 59 months with WHO-defined severe or very severe pneumonia with hypoxaemia (Sp02 < 90%) were included in the study Interventions Patients in Group A received crystalline penicillin (benzyl penicillin) . heart disease and allergy to any of the study drugs were excluded Risk of bias Bias Authors’ judgement Support for judgement Random sequence generation (selection Low risk Patients were randomly allocated bias) Allocation concealment (selection bias) Low risk Randomisation list was prepared before start- ing the study and random treatment assign- ment was placed in serially labelled sealed en- velopes. q6h and gentamycin 2. bacterial meningitis. persistence of danger signs such as inability to drink after 48 hours of treatment or relapse of the hypoxaemic pneumonia during the following 2 weeks Notes Patients with fever > 10 days. q8h for at least 3 days.000 IV/kg IV. Published by John Wiley & Sons. stridor.5 mg/kg. The assignment was opened when the patient had met all the inclusion and exclu- sion criteria and written consent was available Blinding of participants and personnel High risk Open-label randomised controlled trial (performance bias) All outcomes Blinding of outcome assessment (detection High risk Open-label randomised controlled trial bias) All outcomes Selective reporting (reporting bias) Low risk No selective reporting Incomplete outcome data (attrition bias) Low risk Incomplete data adequately addressed All outcomes Other bias Unclear risk Source of funding not mentioned Antibiotics for community-acquired pneumonia in children (Review) 31 Copyright © 2013 The Cochrane Collaboration. acute renal failure. oral amoxycillin 15 mg/kg 8-hourly was substituted for crystalline penicillin. . After that. IV. Ltd. meningitis etc.50.

non-inferiority. severe renal or hepatic diseases. resolution of signs and symptoms. Published by John Wiley & Sons. improved but non-resolu- tion of signs and symptoms. Ltd. open-label.Block 1995 Methods RCT comparing clarithromycin with erythromycin in children with pneumonia Participants Children between 3 to 16 years of age with radiographically confirmed pneumonia Interventions PO clarithromycin (15 mg/kg/day) for 10 days or erythromycin 40 mg/kg/day for 10 days Outcomes Cure rates. active-comparator. active tuber- culosis. Open-label study. multicentre study Participants Children between 0. . severe infections requiring intravenous antibiotics Risk of bias Bias Authors’ judgement Support for judgement Random sequence generation (selection Unclear risk Patients were randomly allocated bias) Allocation concealment (selection bias) Unclear risk Not mentioned clearly. and the presence of 2 or more Antibiotics for community-acquired pneumonia in children (Review) 32 Copyright © 2013 The Cochrane Collaboration.5 to 16 years old with a diagnosis of community-acquired pneumonia (CAP). failure or worsening Notes Exclusion: hypersensitivity to macrolides. A diagnosis of CAP was defined as radiographic evidence of pulmonary infiltrate consistent with acute infection requiring antibiotic therapy. levofloxacin:comparator). improvement. Study drugs were dispensed and compli- ance was monitored by third party Blinding of participants and personnel High risk Unblinded study (performance bias) All outcomes Blinding of outcome assessment (detection Low risk Investigator-blinded study bias) All outcomes Selective reporting (reporting bias) Low risk No selective reporting Incomplete outcome data (attrition bias) Low risk Incomplete data adequately addressed All outcomes Other bias Unclear risk Funded by Abbott Laboratories and role of funding agency not clear Bradley 2007 Methods This was a randomised (3:1.

with dose determined by calculating amoxycillin 22. (3) clinical relapse: resolution or improvement of signs and symptoms at PTV evaluation with reappearance or deterioration of signs and symptoms of infection at test of cure visit (TOCV). improved.Bradley 2007 (Continued) of the indications of pneumonia: fever (rectal or oral temperature 38 °C for children 2 years. The comparator administration was (a) clarithromycin 7. Other exclusion criteria included hospitalisation or residence in a long-term care facility for 14 or more days before the onset of symptoms.000/L or 5000/L). Levofloxacin and comparators were given either orally or by intravenous (IV) administration. cough. levofloxacin was administered (a) as 10 mg/kg/dose as oral suspension qd (up to 500 mg/d). clinical failure. (b) as 1 250 mg tablet qd (for children weighing 22. The patients were randomised in a 3:1 levofloxacin:comparator ratio within 14 strata in the study (1 for the 2 age groups within each country). or (b) ceftriaxone 25 mg/kg/dose IV q12 hours (up to 4 G/d). The comparator administration was (a) amoxycillin and clavulanic acid (7:1) oral suspension bd. rales on auscultation. or physical signs of pneumonia on examination (e. history or presence of arthropathy or periarticular disease or any other musculoskeletal signs or symptoms that in the opinion of the investigator may have confounded a future safety evaluation of musculoskeletal complaints qd: once a day bd: twice a day Antibiotics for community-acquired pneumonia in children (Review) 33 Copyright © 2013 The Cochrane Collaboration.5 to 27. or (b) ceftriaxone 25 mg/kg/dose IV q12 hours (up to 4 G/d) For Group II (5 to 16 years).5 mg/kg/dose as oral suspension (or as a 250 mg tablet) bd (up to 250 mg bd).3 °C for children 0. infection acquired in a hospital (48 hours after hospital admission and 7 days after hospital discharge). For Group I (6 months to 5 years). and (5) unable to evaluate: unable to determine response because patient was not evaluated after PTV Notes Exclusion criteria: received systemic antibiotics for more than 24 hours immediately before enrolment. levofloxacin was administered (a) 10 mg/kg/dose as oral suspension bd (up to 500 mg/d) or (b) 10 mg/kg/dose IV q12 hours (up to 500 mg/d). Published by John Wiley & Sons. (2) improved: continued incomplete resolution of signs and symptoms with no deterioration or relapse after post-therapy visit (PTV) and no requirement for additional antimicrobial therapy. response was carried forward to TOCV. required a systemic antibiotic other than the study drugs.5 mg/kg/dose (up to 875 mg/d). chest pain. (4) failure: patient was considered a clinical failure at PTV. or 38. with either erythromycin lactobionate 10 mg/kg/dose IV q6 hours (up to 4 G/24 hours) or clarithromycin 7. Ltd. egophony) Interventions Levofloxacin or a comparator antibiotic for 10 days.5 to 2 years). relapse at test of cure visit (TOCV) 10 to 17 days after the last dose of study drug: (1) cured: resolution of signs and symptoms associated with active infection along with an improvement or lack of progression of abnormal findings of chest roentgenogram.5 mg/kg/dose as oral suspension (or as a 250 mg tablet) bd (up to 250 mg bd) Outcomes Clinical response was categorised as cured.5 kg). signs and symptoms of a bacterial infection of the central nervous system. or had a suspected infection with micro-organisms known to be resistant to the study drugs. abnormal white blood cell count (15. or (c) 10 mg/kg/dose IV q24 hours (up to 500 mg/d). clarithromycin 250 mg oral tablet bd. shortness of breath.g. dullness to percussion. .5 kg) or 2 250 mg tablets qd (for children weighing 45.

000 U for those above 20 kg). 200.000 U for patients below 20 kg weight and 1. hospitalisation in previous 2 weeks Risk of bias Bias Authors’ judgement Support for judgement Random sequence generation (selection Low risk Patients assigned randomly bias) Antibiotics for community-acquired pneumonia in children (Review) 34 Copyright © 2013 The Cochrane Collaboration. lost to follow-up Notes Exclusion criteria: severe disease. details of role of funding agency not mentioned Camargos 1997 Methods RCT comparing benzathine penicillin and procaine penicillin Participants Children 2 years to 12 years with non-severe pneumonia Interventions Single dose of benzathine penicillin (600. Published by John Wiley & Sons. Ltd. procaine penicillin 300. allergic to penicillin. failure rate. sickle cell car- diomyopathy.000 IU/kg/day IM for 7 days Outcomes Cure rate. . post-measles pneumonia. immunodeficiency.Bradley 2007 (Continued) Risk of bias Bias Authors’ judgement Support for judgement Random sequence generation (selection Unclear risk Sequence generation not mentioned bias) Allocation concealment (selection bias) Unclear risk Not mentioned clearly in the paper Blinding of participants and personnel High risk Open-label randomised controlled trial (performance bias) All outcomes Blinding of outcome assessment (detection High risk Open-label randomised controlled trial bias) All outcomes Selective reporting (reporting bias) Unclear risk Information not clear Incomplete outcome data (attrition bias) Unclear risk Information not provided clearly All outcomes Other bias Unclear risk Funded by Johnson & Johnson Pharma- ceutical Research and Development. atelectasis.

hospitalisation rate and death rate Notes Exclusion criteria: very severe disease. refusal of consent.Camargos 1997 (Continued) Allocation concealment (selection bias) Low risk Randomisation done by trained staff mem- ber blinded to control or treatment using 4 identifying letters randomly selected for benzathine (W and Z) and procaine (X Y) enclosed in sealed envelope Blinding of participants and personnel Low risk Randomisation done by trained staff mem- (performance bias) ber blinded to control or treatment All outcomes Blinding of outcome assessment (detection Unclear risk Information not provided bias) All outcomes Selective reporting (reporting bias) Low risk No selective reporting Incomplete outcome data (attrition bias) Low risk Incomplete data adequately addressed All outcomes Other bias Unclear risk Source of funding not mentioned Campbell 1988 Methods RCT comparing co-trimoxazole for 5 days and procaine penicillin single dose with ampicillin for 5 days Participants Children 1 month to 4 years of age with non-severe pneumonia Interventions Daily co-trimoxazole PO for 5 days or single-dose procaine penicillin with daily PO ampicillin Outcomes Cure rate. . Ltd. Published by John Wiley & Sons. unable to take tablets Risk of bias Bias Authors’ judgement Support for judgement Random sequence generation (selection Unclear risk Sequence generation and randomisation bias) not clear Allocation concealment (selection bias) High risk Eligible children were allocated sequen- tially to 2 treatment groups by study physi- cian Antibiotics for community-acquired pneumonia in children (Review) 35 Copyright © 2013 The Cochrane Collaboration.

very severe disease. acute bronchial asthma. failure rate. antibiotics in past 48 hours Risk of bias Bias Authors’ judgement Support for judgement Random sequence generation (selection Low risk Random numbers generated using a com- bias) puter program Allocation concealment (selection bias) Low risk The drug assignment was concealed from patients parents and study personnel Blinding of participants and personnel Low risk Double-blind randomised controlled trial (performance bias) All outcomes Blinding of outcome assessment (detection Low risk Double-blind randomised controlled trial bias) All outcomes Selective reporting (reporting bias) Low risk No selective reporting Antibiotics for community-acquired pneumonia in children (Review) 36 Copyright © 2013 The Cochrane Collaboration. exclusion criteria: severe pneumonia. change of antibiotics Notes Blinded. . chronic illness. Published by John Wiley & Sons. Ltd. past history of 2 or more episodes of wheeze.Campbell 1988 (Continued) Blinding of participants and personnel High risk Open-label study (performance bias) All outcomes Blinding of outcome assessment (detection High risk Open-label study bias) All outcomes Selective reporting (reporting bias) Unclear risk Data not recorded clearly Incomplete outcome data (attrition bias) Unclear risk Data not recorded clearly All outcomes Other bias Unclear risk Source of funding not mentioned CATCHUP 2002 Methods RCT comparing amoxycillin and co-trimoxazole in non-severe pneumonia Participants Children 2 to 59 months with non-severe pneumonia Interventions PO amoxycillin 25 mg/kg/day for 5 days or co-trimoxazole 20/4 mg/kg/day for 5 days Outcomes Cure rate.

Ltd.CATCHUP 2002 (Continued) Incomplete outcome data (attrition bias) Low risk Incomplete data adequately addressed All outcomes Other bias Low risk Funded by World Health Organization Cetinkaya 2004 Methods RCT comparing chloramphenicol in combination with penicillin with ceftriaxone Participants Children aged 6 months to 16 years with clinical or graphically confirmed pneumonia Interventions IV chloramphenicol 15 mg/kg every 6 hours plus penicillin 25. . Published by John Wiley & Sons.000 IU/kg every 4 hours for 10 days and ceftriaxone 50 mg/kg every 12 hours Outcomes Clinical recovery Notes Blinded. children clinically diagnosed with bacterial pneumonia were enrolled Risk of bias Bias Authors’ judgement Support for judgement Random sequence generation (selection Unclear risk Generation of random list/numbers not bias) mentioned Allocation concealment (selection bias) Unclear risk Allocation concealment not mentioned Blinding of participants and personnel High risk Open-label study (performance bias) All outcomes Blinding of outcome assessment (detection High risk Open-label study bias) All outcomes Selective reporting (reporting bias) Low risk No selective reporting Incomplete outcome data (attrition bias) Low risk Incomplete data adequately addressed All outcomes Other bias Unclear risk Source of funding not mentioned Antibiotics for community-acquired pneumonia in children (Review) 37 Copyright © 2013 The Cochrane Collaboration.

Deivanayagam 1996

Methods RCT comparing ampicillin in combination with penicillin with chloramphenicol for
pneumonia diagnosed by clinical/radiological evidence

Participants Children 5 months to 4 years with pneumonia admitted to hospital

Interventions IM/IV ampicillin (100 mg/kg/day) for 48 hours than PO, IV penicillin (100,000 IU/
kg/day) plus chloramphenicol (100 mg/kg/day)

Outcomes Cure rate, failure rate

Notes Not blinded. Exclusion criteria: acute bronchiolitis, allergy to penicillin, antibiotics in
past 2 days, other drugs by treating physician receiving anti-tuberculosis drugs

Risk of bias

Bias Authors’ judgement Support for judgement

Random sequence generation (selection Low risk Computer-generated random listen
bias)

Allocation concealment (selection bias) Unclear risk Information not provided

Blinding of participants and personnel High risk Open-label study
(performance bias)
All outcomes

Blinding of outcome assessment (detection High risk Open-label study
bias)
All outcomes

Selective reporting (reporting bias) Unclear risk Data not completely described

Incomplete outcome data (attrition bias) Unclear risk No intention-to-treat analysis and details
All outcomes of excluded patients not clear

Other bias Unclear risk Source of funding not mentioned

Duke 2002

Methods RCT comparing chloramphenicol with combination of penicillin and gentamicin in
children with severe pneumonia

Participants Children aged 1 to 59 months age, with severe pneumonia

Interventions IM chloramphenicol (25 mg/kg 6-hourly for at least 5 days) versus penicillin (50 mg/kg
6-hourly) and gentamycin (7.5 mg/kg/d single dose) for at least 5 days

Antibiotics for community-acquired pneumonia in children (Review) 38
Copyright © 2013 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Duke 2002 (Continued)

Outcomes Adverse outcome (death, change in antibiotics, absconded, readmission within 30 days)
, rate of hospitalisation, duration of hospital stay

Notes Not blinded
Exclusion criteria: wheezing, bronchiolitis, meningitis, tuberculosis, CHD, renal failure,
jaundice, received study antibiotics for more than 48 hours in last 1 week

Risk of bias

Bias Authors’ judgement Support for judgement

Random sequence generation (selection Low risk Computer-generated
bias)

Allocation concealment (selection bias) Low risk Sealed envelopes

Blinding of participants and personnel High risk Open-label study
(performance bias)
All outcomes

Blinding of outcome assessment (detection High risk Open-label study
bias)
All outcomes

Selective reporting (reporting bias) Low risk No selective reporting

Incomplete outcome data (attrition bias) Low risk Incomplete data adequately addressed
All outcomes

Other bias Low risk Funded by the World Health Organization
and the Papua New Guinea Health Depart-
ment

Harris 1998

Methods RCT comparing azithromycin, co-amoxyclavulanic acid and erythromycin in pneumo-
nia

Participants Children aged 6 months to 16 years with clinical or radiological evidence of pneumonia

Interventions PO azithromycin (10 mg/kg/day 1 followed by 5 mg/kg/day for 4 days) or amoxycillin
clavulanic acid (40 mg/kg/day) for 10 days or erythromycin (40 mg/kg/day) for 10 days

Outcomes Cure rate (day 15 to 19), improvement rate, failure rate

Notes Exclusion criteria: known hypersensitivity, intolerance to drugs, pregnancy, lactation,
need for parental antibiotics, severe pneumonia, antibiotics in past 72 hours, chronic
steroid therapy, on carbamazepine, ergotamine, terfenadine, loratadine

Antibiotics for community-acquired pneumonia in children (Review) 39
Copyright © 2013 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Harris 1998 (Continued)

Risk of bias

Bias Authors’ judgement Support for judgement

Random sequence generation (selection Unclear risk Sequence generation not mentioned
bias)

Allocation concealment (selection bias) Low risk Double-blind

Blinding of participants and personnel Low risk Double-blind
(performance bias)
All outcomes

Blinding of outcome assessment (detection Low risk Double-blind
bias)
All outcomes

Selective reporting (reporting bias) Unclear risk No details

Incomplete outcome data (attrition bias) Unclear risk Intention-to-treat analysis not performed
All outcomes and no details of excluded patients

Other bias Unclear risk Funded by Pfizer Inc., New York

Hazir 2008

Methods Randomised, open-label equivalency trial

Participants Children aged 3 to 59 months with WHO-defined severe pneumonia

Interventions Oral amoxycillin syrup (80 to 90 mg/kg per day in 2 doses) and sent home (ambulatory
group), or to receive intravenous ampicillin (100 mg/kg per day in 4 doses) for 48 hours
as an inpatient (hospitalised group)

Outcomes Primary outcome (treatment failure up to or on day 6). Any of the following: clinical
deterioration; inability to take oral medication due to persistent vomiting; development
of a comorbid condition requiring an antibiotic; persistence of fever > 38 ºC with lower
chest in-drawing (LCI) from day 3 to day 6; either fever or lower chest in-drawing alone
at day 6; hospitalisation related to pneumonia; serious adverse event; left against medical
advice or lost to follow-up; voluntary withdrawal of consent; death
Secondary outcome (treatment failure between day 6 and day 14; relapse). Any of
the following: clinical deterioration; development of a comorbid condition requiring an
antibiotic; development of lower chest in-drawing or fast breathing non-responsive to 3
trials in children with wheeze

Notes Exclusion criteria: known asthma, those with a history of 3 or more episodes of wheezing
in 1 year, or those in whom lower chest in-drawing resolved after 3 doses of a bron-
chodilator over 30 minutes were excluded. Children showing signs of WHO-defined

Antibiotics for community-acquired pneumonia in children (Review) 40
Copyright © 2013 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Hazir 2008 (Continued)

very severe pneumonia (panel 1) were also excluded; such individuals were admitted to
hospital for treatment with intravenous antibiotics. Children who were known to have
anaphylactic reactions to penicillin or amoxycillin, those with persistent vomiting, those
who had been hospitalised within the previous 2 weeks, and those with other infectious
diseases that needed antibiotic treatment, were also excluded

Risk of bias

Bias Authors’ judgement Support for judgement

Random sequence generation (selection Low risk Randomisation list generated by uneven
bias) block size

Allocation concealment (selection bias) Low risk The randomisation scheme was generated
by a computer program in uneven blocks
of 4, 6 and 8 by an individual not involved
in study. Randomisation codes were sealed
in opaque envelopes in accordance with the
allocation sequence and stratified by site.
After being deemed eligible for enrolment,
participants were assigned the next enve-
lope in the sequence to determine treatment
assignment. The randomisation code was
held at the co-ordinating centre and was
broken at the time of data analysis

Blinding of participants and personnel High risk Open-label study
(performance bias)
All outcomes

Blinding of outcome assessment (detection Unclear risk Open-label study
bias)
All outcomes

Selective reporting (reporting bias) Low risk No selective reporting

Incomplete outcome data (attrition bias) Low risk Incomplete data adequately addressed
All outcomes

Other bias Low risk Funded by the World Health Organiza-
tion and Family Applied Research Project,
Boston University

Antibiotics for community-acquired pneumonia in children (Review) 41
Copyright © 2013 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

.Jibril 1989 Methods RCT comparing amoxycillin and co-amoxyclavulanic acid with amoxycillin alone in bacterial pneumonia (non-severe) Participants Children aged 2 years to 12 years age. hypersensitivity to penicillin/cephalosporin Risk of bias Bias Authors’ judgement Support for judgement Random sequence generation (selection Low risk Serial number selected on random basis bias) Allocation concealment (selection bias) Unclear risk Not mentioned Blinding of participants and personnel High risk Unblinded study (performance bias) All outcomes Blinding of outcome assessment (detection High risk Unblinded study bias) All outcomes Selective reporting (reporting bias) Unclear risk No selective reporting Incomplete outcome data (attrition bias) Unclear risk Incomplete data adequately addressed All outcomes Other bias Unclear risk Source of funding not mentioned Keeley 1990 Methods RCT comparing co-trimoxazole and procaine penicillin Participants Children aged 3 months to 12 years with non-severe pneumonia Interventions Co-trimoxazole per oral for 5 days.5 mg or 500 + 125 mg tds) with amoxycillin (250 mg or 500 mg tds) for 10 days Outcomes Poor or no response. Published by John Wiley & Sons. with non-severe pneumonia Interventions Amoxycillin and co-amoxyclavulanic acid (250 mg + 62. treatment failure. Procaine penicillin IM daily for 5 days Outcomes Cure rate. cure rate Notes Exclusion criteria: renal/hepatic impairment. hospitalisation. well at final follow-up and death rate Notes Exclusion criteria: children with chest in-drawing. Ltd. unable to feed and requiring imme- diate referral Antibiotics for community-acquired pneumonia in children (Review) 42 Copyright © 2013 The Cochrane Collaboration.

congenital anomalies Risk of bias Bias Authors’ judgement Support for judgement Random sequence generation (selection Unclear risk Information not provided bias) Antibiotics for community-acquired pneumonia in children (Review) 43 Copyright © 2013 The Cochrane Collaboration.5 years Interventions Cefpodoxime 5 to 12 mg/kg/day PO for 10 days or co-amoxyclavulanic acid 6 to 13 mg/kg/day for 10 days Outcomes Response rate Notes Exclusion criteria: nosocomial infection. Harare Klein 1995 Methods RCT comparing cefpodoxime and co-amoxyclavulanic acid in LRTI Participants Children aged 3 months to 11. allergy to beta- lactam antibiotics. suspected/confirmed TB. .Keeley 1990 (Continued) Risk of bias Bias Authors’ judgement Support for judgement Random sequence generation (selection Unclear risk Information not provided in the paper bias) Allocation concealment (selection bias) Low risk Used sealed envelopes Blinding of participants and personnel High risk Open-label (performance bias) All outcomes Blinding of outcome assessment (detection High risk Open-label bias) All outcomes Selective reporting (reporting bias) Unclear risk No selective reporting Incomplete outcome data (attrition bias) Unclear risk Incomplete data adequately addressed All outcomes Other bias Low risk Funded by University of Zimbabwe. Ltd. Published by John Wiley & Sons. antibiotics in past 48 hours.

fever on day 3 and day 7. Ltd. chest X-ray on day 14 Notes Exclusion criteria: chronic pathology.Klein 1995 (Continued) Allocation concealment (selection bias) Unclear risk No mention about details of allocation con- cealment Blinding of participants and personnel Unclear risk Information not provided (performance bias) All outcomes Blinding of outcome assessment (detection Unclear risk Information not provided bias) All outcomes Selective reporting (reporting bias) Unclear risk No intention-to-treat analysis. . preterm. Published by John Wiley & Sons. No details All outcomes of children excluded from the analysis Other bias Unclear risk Source of funding not mentioned Kogan 2003 Methods RCT comparing azithromycin and amoxycillin Participants Children aged 1 month to 14 years with non-severe pneumonia Interventions Azithromycin (10 mg/kg/day) PO for 3 days or amoxycillin PO 75 mg/kg/day for 7 days Outcomes Clinical and radiological cure rates. received antibiotics in past 5 days Risk of bias Bias Authors’ judgement Support for judgement Random sequence generation (selection Unclear risk Information on sequence generation not bias) mentioned Allocation concealment (selection bias) High risk Allocated by investigators Blinding of participants and personnel High risk Open-label study (performance bias) All outcomes Blinding of outcome assessment (detection High risk Open-label study bias) All outcomes Antibiotics for community-acquired pneumonia in children (Review) 44 Copyright © 2013 The Cochrane Collaboration. No details of children excluded from the analysis Incomplete outcome data (attrition bias) Unclear risk No intention-to-treat analysis.

failure rate and exclusion. Ltd. clinical or radiological signs of TB. severe pneumonia TMP/SMX: trimethoprim/ sulphamethoxazole (co-trimoxazole) Risk of bias Bias Authors’ judgement Support for judgement Random sequence generation (selection Low risk Random list generated in advance bias) Allocation concealment (selection bias) Low risk Double-blind study.Kogan 2003 (Continued) Selective reporting (reporting bias) Low risk No selective reporting Incomplete outcome data (attrition bias) Low risk Incomplete data adequately addressed All outcomes Other bias Unclear risk Source of funding not mentioned Mulholland 1995 Methods RCT comparing chloramphenicol and co-trimoxazole in malnourished children with clinical or radiological pneumonia Participants Children below 5 years of age with malnutrition and clinical or radiological evidence of pneumonia Interventions Enrolled subjects received either chloramphenicol with TMP/SMX placebo or TMP/ SMX with chloramphenicol placebo Outcomes Cure rate. . Randomisation codes were kept with senior nurse and pharmacist Blinding of participants and personnel Low risk Double-blind study (performance bias) All outcomes Blinding of outcome assessment (detection Low risk Double-blind study bias) All outcomes Selective reporting (reporting bias) Low risk No selective reporting Incomplete outcome data (attrition bias) Low risk Incomplete data adequately addressed All outcomes Antibiotics for community-acquired pneumonia in children (Review) 45 Copyright © 2013 The Cochrane Collaboration. relapse rate. death rate Notes Blinded Exclusion criteria: already receiving antibiotics. Published by John Wiley & Sons.

Ltd.Mulholland 1995 (Continued) Other bias Low risk Funded by the World Health Organization Nogeova 1997 Methods Randomised. controlled. Details not included in (performance bias) the text All outcomes Blinding of outcome assessment (detection High risk Open-label study. multicentre trial on children between 1 to 12 years of age with radiologically documented pneumonia to compare efficacy of ceftibuten with cefuroxime axetil. Sputum and blood were tested for aetiological agents Participants Children 1 to 12 years of age with radiographically confirmed pneumonia Interventions Ceftibuten or cefuroxime axetil in 2 divided doses Outcomes Cure or failure rates Notes Duration of treatment not clear Risk of bias Bias Authors’ judgement Support for judgement Random sequence generation (selection Unclear risk Details of randomisation including se- bias) quence generation not mentioned Allocation concealment (selection bias) Unclear risk Details of allocation concealment not men- tioned Blinding of participants and personnel High risk Open-label study. . Published by John Wiley & Sons. Details not included in bias) the text All outcomes Selective reporting (reporting bias) Low risk No selective reporting Incomplete outcome data (attrition bias) Low risk Incomplete data adequately addressed All outcomes Other bias Unclear risk No mention about ethical clearance and source of funding Antibiotics for community-acquired pneumonia in children (Review) 46 Copyright © 2013 The Cochrane Collaboration.

Children receiving oxacillin + ceftriaxone continued to get IV antibiotics for 10 days while those receiving co-amoxy- clavulanic acid were switched over to oral medications after improvement at 48 hours Outcomes Cure rate. 5 bacterial isolates from blood culture. time for response in tachypnoea. failure rate. Pseudomonas aeruginosa (1 patient) Risk of bias Bias Authors’ judgement Support for judgement Random sequence generation (selection Low risk Computer-generated randomisation list bias) Allocation concealment (selection bias) Low risk Kept in opaque envelope Blinding of participants and personnel High risk Open-label (performance bias) All outcomes Blinding of outcome assessment (detection Low risk Physicians were blinded bias) All outcomes Selective reporting (reporting bias) Low risk No selective reporting Incomplete outcome data (attrition bias) Low risk Incomplete data adequately addressed All outcomes Other bias Unclear risk Source of funding not mentioned Antibiotics for community-acquired pneumonia in children (Review) 47 Copyright © 2013 The Cochrane Collaboration. In oxacillin + ceftriaxone group: Enterobacter and S. hospitalised with very severe pneumonia Interventions Patients received either intravenous (IV) oxacillin 200 mg/kg/day every 6 hours for 10 days and ceftriaxone 100 mg/kg/day every 12 hours for 10 days or co-amoxyclavulanic acid 100 mg/kg/day every 8 hours (amoxycillin base). in co- amoxyclavulanic acid group: coagulase-negative staphylococci (2 patients). Published by John Wiley & Sons. . total hospital stay Notes Unblinded study. aureus (1 each). Source of funding not mentioned. Ltd.Ribeiro 2011 Methods Randomised controlled trial comparing oxacillin + ceftriaxone with co-amoxyclavulanic acid for treatment of very severe pneumonia that was radiologically diagnosed Participants Children between 2 months to 5 years of age.

Roord 1996 Methods RCT comparing azithromycin and erythromycin in non-severe pneumonia (acute LRTI) Participants Children aged 2 months to 16 years with non-severe pneumonia (acute LRTI) Interventions Azithromycin 10 mg/kg/day for 3 days or erythromycin 40 mg/kg/day for 10 days Outcomes Cure rate. known hypersensitivity to azithro- mycin or erythromycin.BV Shann 1985 Methods RCT comparing chloramphenicol and chloramphenicol in combination with penicillin in severe pneumonia Participants Children Interventions IM chloramphenicol daily until switched over to oral. immunodeficiency. . cystic fibrosis. improvement at day 10 and between days 25 to 30 Notes Exclusion criteria: not able to take oral medications. Ltd. bacteraemia. Published by John Wiley & Sons. No mention about allocation concealment Blinding of participants and personnel High risk Open-label study (performance bias) All outcomes Blinding of outcome assessment (detection High risk Open-label study bias) All outcomes Selective reporting (reporting bias) Low risk No selective reporting Incomplete outcome data (attrition bias) Low risk Incomplete data adequately addressed All outcomes Other bias Unclear risk Funded by Pfizer . Block randomisation. receiving alterna- tive treatment Risk of bias Bias Authors’ judgement Support for judgement Random sequence generation (selection Unclear risk Information not provided bias) Allocation concealment (selection bias) Unclear risk Open-label randomised controlled trial. need for oxygen. nosocomial pneumonia. leucocyte count less than 300. failure rate at day 10 to 14.000 per litre. or IM chloramphenicol with benzyl penicillin until switched over to oral Antibiotics for community-acquired pneumonia in children (Review) 48 Copyright © 2013 The Cochrane Collaboration.

wheezing Risk of bias Bias Authors’ judgement Support for judgement Antibiotics for community-acquired pneumonia in children (Review) 49 Copyright © 2013 The Cochrane Collaboration. Ltd. evident improvement at day 5 and day 10.000 IU/ kg/day) for 10 days Outcomes Cure rate at day 5 and day 10.Shann 1985 (Continued) Outcomes Discharge from hospital and good improvement of symptoms Notes Not blinded Risk of bias Bias Authors’ judgement Support for judgement Random sequence generation (selection Low risk Random table was prepared bias) Allocation concealment (selection bias) Low risk Sealed numbered envelopes used Blinding of participants and personnel High risk Open-label study (performance bias) All outcomes Blinding of outcome assessment (detection High risk Open-label study bias) All outcomes Selective reporting (reporting bias) Low risk No selective reporting Incomplete outcome data (attrition bias) Low risk Incomplete data adequately addressed All outcomes Other bias Unclear risk Source of funding not mentioned Sidal 1994 Methods RCT comparing co-trimoxazole and penicillin in non-severe pneumonia (including moderate pneumonia) Participants Children aged 3 months to 14 years with non-severe pneumonia (including moderate pneumonia) Interventions PO co-trimoxazole (40 mg/kg/day) for 10 days or IM procaine penicillin (50. Published by John Wiley & Sons. . inability to eat or drink. moderate to severe malnutrition. failure rate Notes Exclusion criteria: severe chest in-drawing. antibiotics in last 2 weeks.

Published by John Wiley & Sons. Details not clear Blinding of participants and personnel High risk Open-label study. Ltd. Exclusion criteria: very severe pneumonia.Sidal 1994 (Continued) Random sequence generation (selection Unclear risk Information not provided bias) Allocation concealment (selection bias) High risk No details of randomisation or allocation concealment Blinding of participants and personnel High risk Open-label study (performance bias) All outcomes Blinding of outcome assessment (detection High risk Open-label study bias) All outcomes Selective reporting (reporting bias) Low risk No selective reporting Incomplete outcome data (attrition bias) Low risk Incomplete data adequately addressed All outcomes Other bias Unclear risk Source of funding not mentioned Straus 1998 Methods RCT comparing co-trimoxazole and amoxycillin in non-severe pneumonia Participants Children aged 2 months to 59 months with non-severe pneumonia Interventions PO co-trimoxazole 20 mg/kg/day for 5 days or amoxycillin 45 mg/kg/day for 5 days Outcomes Failure rate. determined by clinical and radiological evidence Notes Blinded. hypoxaemia Risk of bias Bias Authors’ judgement Support for judgement Random sequence generation (selection Unclear risk Sequence generation not mentioned bias) Allocation concealment (selection bias) Unclear risk Drug allotment was concealed from partic- ipants. antibiotics in past 48 hours. Details not included (performance bias) All outcomes Antibiotics for community-acquired pneumonia in children (Review) 50 Copyright © 2013 The Cochrane Collaboration. . hospi- talisation in past 7 days.

Ltd. failure rate. Details not included bias) All outcomes Selective reporting (reporting bias) Low risk No selective reporting Incomplete outcome data (attrition bias) Low risk Incomplete data adequately addressed All outcomes Other bias Unclear risk Source of funding not mentioned Tsarouhas 1998 Methods RCT comparing procaine penicillin and amoxycillin for radiographically confirmed pneumonia Participants Children aged 6 months to 18 years with pneumonia Interventions PO amoxycillin (50 mg/kg/day) or procaine penicillin IM (50. Published by John Wiley & Sons. or penicillin. ill appearance. . cystic fibrosis. concurrent febrile illness Risk of bias Bias Authors’ judgement Support for judgement Random sequence generation (selection Unclear risk Sequence generation not mentioned bias) Allocation concealment (selection bias) Low risk Sealed envelope opened by Emergency De- partment nurse Blinding of participants and personnel High risk Open-label study (performance bias) All outcomes Blinding of outcome assessment (detection High risk Open-label study bias) All outcomes Selective reporting (reporting bias) Low risk No selective reporting Antibiotics for community-acquired pneumonia in children (Review) 51 Copyright © 2013 The Cochrane Collaboration.Straus 1998 (Continued) Blinding of outcome assessment (detection High risk Open-label study. temperature more than 38. antibiotics in past 1 week.000 IU/kg/day) Outcomes Hospitalisation rate.5 °C. allergy to amoxycillin. asthma. sickle cell disease. increased respiratory rate Notes Unblinded Exclusion criteria: chronic illness. wheezing.

Exclusion criteria: hypersensitivity to study drugs. Ltd. nosocomial pneumonia. . failure rates and improvement Notes Non-blinded. hospitalisation. and erythromycin 40 mg/kg/day for 10 days in children over 5 years Outcomes Clinically diagnosed cure rates. Antibiotics for community-acquired pneumonia in children (Review) 52 Copyright © 2013 The Cochrane Collaboration.Tsarouhas 1998 (Continued) Incomplete outcome data (attrition bias) Low risk Incomplete data adequately addressed All outcomes Other bias Unclear risk Source of funding not mentioned Wubbel 1999 Methods RCT comparing azithromycin and erythromycin in children over 5 years of age with pneumonia. Published by John Wiley & Sons. antibiotics in last 7 days Risk of bias Bias Authors’ judgement Support for judgement Random sequence generation (selection Unclear risk Details not mentioned bias) Allocation concealment (selection bias) Unclear risk Allocation concealment not clearly de- scribed Blinding of participants and personnel High risk Unblinded study (performance bias) All outcomes Blinding of outcome assessment (detection High risk Unblinded study bias) All outcomes Selective reporting (reporting bias) Low risk No selective reporting Incomplete outcome data (attrition bias) Low risk Incomplete data adequately addressed All outcomes Other bias Unclear risk Funded by Pfizer Inc. and comparing azithromycin with co-amoxyclavulanic acid in children un- der 5 years of age Participants Children aged between 6 months a 16 years with pneumonia Interventions PO azithromycin (10 mg/kg on day 1 followed by 5 mg/kg/day for next 4 days) or co- amoxyclavulanic acid 40 mg/kg/day for 10 days in children under 5 years of age.

743 patients hospitalised with community-acquired pneumonia.bd: twice a day CHD: congenital heart disease CPZ: carbamazepine IM: intramuscular IV: intravenous LRTI: lower respiratory tract infection PO: orally PTV: post-therapy visit q6h: every 6 hours q8h: every 8 hours q12h: every 12 hours RCT: randomised controlled trial Sp02 : oxygen saturation TB: tuberculosis tds: three times a day TOCV: test of cure visit WHO: World Health Organization Characteristics of excluded studies [ordered by study ID] Study Reason for exclusion Agostoni 1988 Compares minocycline and amoxycillin in 23 children between 3 to 11. The study was excluded because of its adult study population Esposito 2005 Compared azithromycin in addition to symptomatic treatment with symptomatic treatment alone in chil- dren with recurrent respiratory tract infections. Ltd. Not a RCT Al-Eiden 1999 Describes results of sequential antibiotic therapy (SAT) in 89 patients with severe lower respiratory tract infection. . did not compare 2 antibiotics Bonvehi 2003 Compared clarithromycin and co-amoxyclavulanic acid in adult patients with CAP due to penicillin- resistant and/or macrolide-resistant S. The sequential antibiotic use was the reason for exclusion Ambroggio 2012 A retrospective cohort study. 24% received beta-lactam and macrolide combination therapy on admission. The study had an adult population Antibiotics for community-acquired pneumonia in children (Review) 53 Copyright © 2013 The Cochrane Collaboration. The study did not compare 2 or more antibiotics for pneu- monia Fogarty 2002 Compared cefditoren with co-amoxyclavulanic acid in the management of community-acquired pneumonia in adult patients. Of these. Compared outcome in form of hospital stay and relapse rates between children who received beta-lactam monotherapy or children who received beta-lactam plus macrolide combination therapy Excluded because it was not randomised controlled trial Bari 2011 A cluster-randomised controlled trial on children below 5 years of age with severe pneumonia compared oral amoxicillin with standard treatment (referral and parenteral/oral antibiotics). Published by John Wiley & Sons.5 years with pneumonia. pneumoniae. The study compared 2 modalities of treatment (oral amoxicillin with standard treatment). analysed 20.

aetiology of pneumonia and side effects of antibiotics. No clear data on cure or failure rates Lu 2006 Full paper could not be obtained Mouallem 1976 Compared cephradine and cephalexin for the treatment of bacterial infections in 162 children between 4 months and 11 years of age. There were no separate data for pneumonia available CAP: community-acquired pneumonia RCT: randomised controlled trial Antibiotics for community-acquired pneumonia in children (Review) 54 Copyright © 2013 The Cochrane Collaboration. The study compared 2 modalities of treatment (oral amoxicillin with standard treatment). The study was in adult patients Lee 2008 A randomised controlled trial comparing ampicillin versus ampicillin + gentamycin in children with com- munity-acquired pneumonia. Measured outcomes were duration of clinical symptoms. Ltd. did not compare 2 antibiotics van Zyl 2002 Randomised controlled trial compared cefditoren with cefpodoxime in community-acquired pneumonia in adult patients. The study had an adult study population Vuori-Holopaine 2000 Compared procaine penicillin and cefuroxime in children between 3 months and 15 years of age with suspected sepsis. Compared ceftriaxone.(Continued) Haffejee 1984 A single-blind therapeutic trial using cefotaxime or a benzyl-penicillin-gentamycin combination in 68 hospitalised paediatric patients with 72 episodes of severe infection (septicaemia. No separate data were available for pneumonia Hasali 2005 A randomised comparative study of clarithromycin and erythromycin in the treatment of community- acquired pneumonia in children. cefuroxime and amoxycillin-clavulanic acid. Outcome variables were total hospital stay and time taken for improvement in clinical symptoms. There were no separate data for pneumonia Paupe 1992 Compares cefetamet (2 doses) with cefaclor. Published by John Wiley & Sons. death rates or relapse rates Sanchez 1998 Randomised controlled trial involving 409 patients admitted to internal medicine department. Compared erythromycin with amoxycillin in the treatment of 120 children with community-acquired pneumonia. The study does not provide cure rates. Study does not provide separate data for children Soofi 2012 A cluster-randomised controlled trial on children below 5 years of age with severe pneumonia compared oral amoxycillin with standard treatment (referral and parenteral/oral antibiotics). . neonatal meningitis and others). The doses of antibiotics were inconsistent Peltola 2001 Describes results of treatment with a short (4-day) duration of antibiotics Ruhrmann 1982 Randomised controlled study. failure rates. Outcome in form of cure or failure available only for children with mycoplasma or chlamydia pneumonia Higuera 1996 Compared oral cefuroxime axetil and oral co-amoxyclavulanic acid in the treatment of community-acquired pneumonia in adult patients. pneumonia.

1. 95% CI) 0.50.0 [0.56] confirmed pneumonia 9 Failure rate in radiographically 2 147 Odds Ratio (M-H. Random. Random.DATA AND ANALYSES Comparison 1.83 [0. 4.12.65.08] 3 Clinical success rate 1 234 Odds Ratio (M-H.61 [0. 8.55 [0. Random. Random. 95% CI) 1. Random. Random. 95% CI) 1.22 [0.73 [0. 95% CI) 1. .94] 5 Failure rate 3 392 Odds Ratio (M-H. 95% CI) 3. Random.92 [0.23.70. 3. 18.51 [1.62 [0.55] 2 Cure rates 1 234 Odds Ratio (M-H.93 [0. Random.75 [0.23 [0.30.02. 4.31.52 [0. 95% CI) 0. 95% CI) 0. 95% CI) 0. 95% CI) 2. Azithromycin versus erythromycin No. 95% CI) 0.56. of No.15. Random. 1. Random.60.45. 9.87] or nasopharyngeal cultures 8 Cure rate in radiographically 2 147 Odds Ratio (M-H.89] 6 Side effects 2 153 Odds Ratio (M-H.73] 7 Organisms identified by serology 3 368 Odds Ratio (M-H. Published by John Wiley & Sons. 95% CI) 0.17 [0.18] 3 Wheezing present 2 479 Odds Ratio (M-H. Random. 95% CI) 0. 1.58. of Outcome or subgroup title studies participants Statistical method Effect size 1 Age below 5 years 1 260 Odds Ratio (M-H.07 [0. Ltd. 2.06. Random.45] 6 Radiologic resolution 1 209 Odds Ratio (M-H. 95% CI) 0. 4. of No.67] Antibiotics for community-acquired pneumonia in children (Review) 55 Copyright © 2013 The Cochrane Collaboration.57] 2 Male sex 3 564 Odds Ratio (M-H. Random. 95% CI) 1. of Outcome or subgroup title studies participants Statistical method Effect size 1 Mean age (months) 3 369 Mean Difference (IV.18. 1. Clarithromycin versus erythromycin No.92 [0.23] 4 Failure rate 1 234 Odds Ratio (M-H. 2. 1.02. 2.16] 7 Radiologic success 1 209 Odds Ratio (M-H. 95% CI) 1. 95% CI) 1.34 [0. Random.54. 6. 6. Random.80] 9 Adverse events 1 260 Odds Ratio (M-H. Random.48 [-18. 1.72 [0.87] 4 Cure rate 3 363 Odds Ratio (M-H. 95% CI) 0.90] 10 Bacteriologic response 1 45 Odds Ratio (M-H.84.23] 5 Relapse rate 1 226 Odds Ratio (M-H. Random.18. Random. 1. 95% CI) 1. Random. 95% CI) -4.63] confirmed pneumonia Comparison 2.04] 8 Radiologic failure 1 209 Odds Ratio (M-H.

75 [0. 95% CI) -0.19 [0.10 [35.52.73. 0. 0.22] 7 Failure rates in radiographically 1 88 Odds Ratio (M-H. Random. Random.42.25] Antibiotics for community-acquired pneumonia in children (Review) 56 Copyright © 2013 The Cochrane Collaboration.96] 4 Cure rate clinical 1 47 Odds Ratio (M-H. 6.30 [-0. Published by John Wiley & Sons.37 [0. 95% CI) 0. 95% CI) 1.08] confirmed pneumonia Comparison 4.0] 5 Cure rate radiological 1 47 Odds Ratio (M-H. Azithromycin versus amoxycillin No. 95% CI) 0.0 [0.27 [0. 3. Random. Random.24.53] 3 Improved 1 188 Odds Ratio (M-H. 3. 6.09] 6 Fever day 7 1 47 Odds Ratio (M-H. Random. Random. 1.74] 6 Mycoplasma serology positive 1 192 Odds Ratio (M-H. Ltd. of No. Azithromycin versus co-amoxyclavulanic acid No.02 [0. 2. Random. Random.Comparison 3. 95% CI) 1. 95% CI) 0. 4. Random.02 [0.05. of No. 1.43. 1.61] 2 Duration of illness 1 47 Mean Difference (IV.85 [0. 95% CI) 1. 95% CI) 2. 1. Random.17. Random. 95% CI) 2.54. 7. of Outcome or subgroup title studies participants Statistical method Effect size 1 Age in months 1 76 Mean Difference (IV.04.95] 2 Failure rate 2 276 Odds Ratio (M-H.12] 2 Failure rate 1 154 Odds Ratio (M-H.30] 3 Wheezing present 1 47 Odds Ratio (M-H. 95% CI) 0. of No.10 [-1. Random.71] 4 Side effects 2 276 Odds Ratio (M-H.59. Random. Random. 95% CI) 58. .64.59.62 [0. Random.41.15 [0.0. of Outcome or subgroup title studies participants Statistical method Effect size 1 Median age 1 170 Mean Difference (IV. 95% CI) 1.21 [0.85 [0. 95% CI) 1.61] 5 Organisms isolated 1 188 Odds Ratio (M-H. Amoxycillin versus procaine penicillin No.61] Comparison 5.50. of Outcome or subgroup title studies participants Statistical method Effect size 1 Cure rate 1 188 Odds Ratio (M-H. 11. 95% CI) 0. 80. 95% CI) 0.

89] 15 Change of antibiotics 1 1459 Odds Ratio (M-H. 111.01.37] 5 Received antibiotics in previous 1 595 Odds Ratio (M-H. 95% CI) 1.14 [0. Random.57] 13 Wheeze positive 1 1471 Odds Ratio (M-H. . 1. 95% CI) 1. Random.08 [0.46.26] 6 Male sex 1 100 Odds Ratio (M-H.91.06 [-0.06. 1.19 [0. Random. 3.97] week 6 Severe pneumonia 1 595 Odds Ratio (M-H. Random. 4. 0. Random. Random.24. 111.59. 95% CI) 5.71 [0.92. 1.73. of Outcome or subgroup title studies participants Statistical method Effect size 1 Age less than 1 year 3 2347 Odds Ratio (M-H. 3.69] 16 Failure rates after excluding 2 1750 Odds Ratio (M-H.51] pneumonia 8 Failure rate severe pneumonia 1 302 Odds Ratio (M-H. Co-trimoxazole versus amoxycillin No.09. 1.96.19] 14 Cure rate 2 1732 Odds Ratio (M-H.11] clinical diagnosis 9 Failure rate radiological positive 1 153 Odds Ratio (M-H.Comparison 6. Co-amoxyclavulanic acid versus amoxycillin No.58 [0.27.78] pneumonia 10 Failure rate radiological 1 424 Odds Ratio (M-H.70 [0.24. 38. 1. 95% CI) 0. 95% CI) 2.76 [0.85.59. Random.21 [0. 95% CI) 10.67] 2 Cure rate 1 100 Odds Ratio (M-H. 95% CI) 0. 3.69] 5 Weight 1 100 Mean Difference (IV. Random. of No. 17. Published by John Wiley & Sons. 95% CI) 4.10 [-1. Random.22. 95% CI) 1.18 [0.24] 4 Age (months) 1 100 Mean Difference (IV.44 [2.96. 95% CI) 1.97 [0. 95% CI) 0. 95% CI) 1. 1. 95% CI) 1. Random.21] 3 Complications 1 100 Odds Ratio (M-H.08 [0.09] negative pneumonia 11 Death rate 2 2050 Odds Ratio (M-H.45] 7 Failure rate in non-severe 3 1787 Odds Ratio (M-H.67 [0. 95% CI) 1.06] 12 Lost to follow-up 3 2325 Odds Ratio (M-H.97 [0.72 [0. 1. Random. 95% CI) 1.03 [0. Random. 1.96 [0. Random.21 [0. 3. 95% CI) 0.31 [0.03 [0.53] study by Awasthi 2008 Antibiotics for community-acquired pneumonia in children (Review) 57 Copyright © 2013 The Cochrane Collaboration.80 [-8.18. Random.56. 95% CI) 1. 0.29] 2 Male sex 3 2318 Odds Ratio (M-H. 1. of Outcome or subgroup title studies participants Statistical method Effect size 1 Poor or no response 1 100 Odds Ratio (M-H.15] 4 Non-severe pneumonia 1 595 Odds Ratio (M-H. Random.83] 3 Mean Z score for weight 2 2066 Mean Difference (IV. 20. 95% CI) 2.74. of No.26 [0. 0. Random. 95% CI) -0.69.03] 7 Wheeze present 1 100 Odds Ratio (M-H. 95% CI) 5.92] 8 Side effects 1 100 Odds Ratio (M-H. Random.49. Random.24] Comparison 7. Random.57. Random. 1.95. 95% CI) 0.94. 95% CI) 0. Random. 0. Random. Ltd. 95% CI) 0. Random. 95% CI) 0. Random.

Comparison 8. Co-trimoxazole versus procaine penicillin

No. of No. of
Outcome or subgroup title studies participants Statistical method Effect size

1 Age less than 1 year 2 723 Odds Ratio (M-H, Random, 95% CI) 1.30 [0.96, 1.75]
2 Age 1 to 5 years 1 614 Odds Ratio (M-H, Random, 95% CI) 0.84 [0.61, 1.16]
3 Age 5 to 12 years 2 723 Odds Ratio (M-H, Random, 95% CI) 0.79 [0.45, 1.39]
4 Duration of illness in days 2 723 Mean Difference (IV, Fixed, 95% CI) -0.15 [-0.49, 0.20]
5 Male sex 1 614 Odds Ratio (M-H, Random, 95% CI) 0.93 [0.67, 1.27]
6 Cure rate 2 723 Odds Ratio (M-H, Random, 95% CI) 1.58 [0.26, 9.69]
7 Hospitalisation rate 1 614 Odds Ratio (M-H, Random, 95% CI) 2.52 [0.88, 7.25]
8 Well at end of follow-up 1 614 Odds Ratio (M-H, Random, 95% CI) 0.90 [0.51, 1.57]
9 Death 1 614 Odds Ratio (M-H, Random, 95% CI) 3.09 [0.13, 76.13]
10 Treatment failure 1 614 Odds Ratio (M-H, Random, 95% CI) 1.72 [0.41, 7.27]

Comparison 9. Co-trimoxazole versus procaine penicillin and ampicillin

No. of No. of
Outcome or subgroup title studies participants Statistical method Effect size

1 Mean age in months 1 134 Mean Difference (IV, Random, 95% CI) 0.0 [0.0, 0.0]
2 Age less than 1 year 1 134 Odds Ratio (M-H, Random, 95% CI) 0.80 [0.39, 1.64]
3 Male sex 1 134 Odds Ratio (M-H, Random, 95% CI) 1.29 [0.65, 2.58]
4 Cure rate 1 134 Odds Ratio (M-H, Random, 95% CI) 1.15 [0.36, 3.61]
5 Hospitalisation rate 1 134 Odds Ratio (M-H, Random, 95% CI) 1.57 [0.25, 9.72]
6 Death rate 1 134 Odds Ratio (M-H, Random, 95% CI) 0.20 [0.01, 4.25]

Comparison 10. Cefpodoxime versus co-amoxyclavulanic acid

No. of No. of
Outcome or subgroup title studies participants Statistical method Effect size

1 Cure rate (response rate) at end 1 278 Odds Ratio (M-H, Random, 95% CI) 0.69 [0.18, 2.60]
of treatment
2 Mean age (months) 1 348 Mean Difference (IV, Random, 95% CI) 0.0 [0.0, 0.0]
3 Adverse effects 1 278 Odds Ratio (M-H, Random, 95% CI) 0.46 [0.16, 1.35]
4 Age in years 1 348 Mean Difference (IV, Random, 95% CI) 0.0 [0.0, 0.0]
5 Follow-up 1 278 Odds Ratio (M-H, Random, 95% CI) 0.37 [0.11, 1.31]

Antibiotics for community-acquired pneumonia in children (Review) 58
Copyright © 2013 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Comparison 11. Chloramphenicol versus penicillin plus gentamicin

No. of No. of
Outcome or subgroup title studies participants Statistical method Effect size

1 Adverse events 1 1116 Odds Ratio (M-H, Random, 95% CI) 1.26 [0.96, 1.66]
2 Death 1 1116 Odds Ratio (M-H, Random, 95% CI) 1.25 [0.76, 2.07]
3 Change of antibiotics 1 1116 Odds Ratio (M-H, Random, 95% CI) 0.80 [0.54, 1.18]
4 Readmission before 30 days 1 1116 Odds Ratio (M-H, Random, 95% CI) 1.61 [1.02, 2.55]
5 Absconded 1 1116 Odds Ratio (M-H, Random, 95% CI) 1.31 [0.83, 2.09]
6 Age (months) 1 1116 Mean Difference (IV, Random, 95% CI) 0.0 [0.0, 0.0]
7 Culture positive 1 1116 Odds Ratio (M-H, Random, 95% CI) 0.85 [0.60, 1.21]
8 Male sex 1 1116 Odds Ratio (M-H, Random, 95% CI) 0.88 [0.69, 1.12]
9 Received antibiotics in previous 1 1116 Odds Ratio (M-H, Random, 95% CI) 0.96 [0.75, 1.22]
1 week
10 Lost to follow-up 1 1116 Odds Ratio (M-H, Random, 95% CI) 1.31 [0.83, 2.09]

Comparison 12. Chloramphenicol with ampicillin and gentamicin

No. of No. of
Outcome or subgroup title studies participants Statistical method Effect size

1 Mean age 1 958 Mean Difference (IV, Random, 95% CI) -0.10 [-1.12, 0.92]
2 Male sex 1 958 Odds Ratio (M-H, Random, 95% CI) 0.85 [0.66, 1.11]
3 Number received antibiotics in 1 950 Odds Ratio (M-H, Random, 95% CI) 0.87 [0.67, 1.14]
past 7 days
4 Failure rates on day 5 1 958 Odds Ratio (M-H, Random, 95% CI) 1.51 [1.04, 2.19]
5 Failure rates on day 10 1 958 Odds Ratio (M-H, Random, 95% CI) 1.46 [1.04, 2.06]
6 Failure rates on day 21 1 958 Odds Ratio (M-H, Random, 95% CI) 1.43 [1.03, 1.98]
7 Need for change in antibiotics 1 958 Odds Ratio (M-H, Random, 95% CI) 1.81 [1.10, 2.98]
(day 5)
8 Need for change in antibiotics 1 958 Odds Ratio (M-H, Random, 95% CI) 1.71 [1.10, 2.66]
(day 10)
9 Need for change in antibiotics 1 958 Odds Ratio (M-H, Random, 95% CI) 1.65 [1.09, 2.49]
(day 21)
10 Death rates 1 958 Odds Ratio (M-H, Random, 95% CI) 1.65 [0.99, 2.77]

Antibiotics for community-acquired pneumonia in children (Review) 59
Copyright © 2013 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Comparison 13. Chloramphenicol plus penicillin versus ceftriaxone

No. of No. of
Outcome or subgroup title studies participants Statistical method Effect size

1 Cure rates 1 97 Odds Ratio (M-H, Random, 95% CI) 1.36 [0.47, 3.93]

Comparison 14. Chloramphenicol versus chloramphenicol plus penicillin

No. of No. of
Outcome or subgroup title studies participants Statistical method Effect size

1 Need for change of antibiotics 1 748 Odds Ratio (M-H, Random, 95% CI) 0.49 [0.12, 1.97]
2 Death rates 1 748 Odds Ratio (M-H, Random, 95% CI) 0.73 [0.48, 1.09]
3 Lost to follow-up 1 748 Odds Ratio (M-H, Random, 95% CI) 1.11 [0.80, 1.53]

Comparison 15. Ampicillin alone versus penicillin with chloramphenicol

No. of No. of
Outcome or subgroup title studies participants Statistical method Effect size

1 Cure rates 1 101 Odds Ratio (M-H, Random, 95% CI) 0.48 [0.15, 1.51]
2 Age (months) 1 101 Mean Difference (IV, Random, 95% CI) -1.69 [-5.54, 2.16]
3 Male sex 1 101 Odds Ratio (M-H, Random, 95% CI) 0.88 [0.41, 1.93]
4 Duration of hospital stay 1 101 Mean Difference (IV, Random, 95% CI) -0.10 [-1.13, 0.93]
5 Grade 2 to 4 malnutrition 1 101 Odds Ratio (M-H, Random, 95% CI) 0.88 [0.41, 1.93]

Comparison 16. Benzathine penicillin versus procaine penicillin

No. of No. of
Outcome or subgroup title studies participants Statistical method Effect size

1 Cure rate 2 281 Odds Ratio (M-H, Random, 95% CI) 0.53 [0.27, 1.01]
2 Failure rate 2 281 Odds Ratio (M-H, Random, 95% CI) 3.17 [0.90, 11.11]
3 Male sex 2 281 Odds Ratio (M-H, Random, 95% CI) 1.09 [0.67, 1.76]
4 Age between 2 to 6 years 2 301 Odds Ratio (M-H, Random, 95% CI) 1.08 [0.47, 2.48]
5 Age between 7 to 12 years 2 301 Odds Ratio (M-H, Random, 95% CI) 0.52 [0.31, 0.88]
6 Lost to follow-up 1 176 Odds Ratio (M-H, Random, 95% CI) 1.80 [0.16, 20.25]
7 Failure rates in radiographically 1 176 Odds Ratio (M-H, Random, 95% CI) 1.61 [0.45, 5.70]
confirmed pneumonia

Antibiotics for community-acquired pneumonia in children (Review) 60
Copyright © 2013 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Comparison 17. Amoxycillin versus penicillin

No. of No. of
Outcome or subgroup title studies participants Statistical method Effect size

1 Nasopharyngeal aspirates for S. 1 1486 Odds Ratio (M-H, Random, 95% CI) 0.90 [0.72, 1.13]
pneumoniae
2 Age less than 1 year 1 1702 Odds Ratio (M-H, Random, 95% CI) 1.06 [0.87, 1.29]
3 Male sex 2 1905 Odds Ratio (M-H, Random, 95% CI) 1.04 [0.87, 1.25]
4 Weight below 2 Z score 1 1686 Odds Ratio (M-H, Random, 95% CI) 0.92 [0.70, 1.19]
(indicating severe malnutrition)
5 Breast fed 1 1702 Odds Ratio (M-H, Random, 95% CI) 1.12 [0.92, 1.37]
6 Received antibiotics in last 7 2 1905 Odds Ratio (M-H, Random, 95% CI) 0.97 [0.69, 1.38]
days
7 Failure rate at 48 hours 1 1702 Odds Ratio (M-H, Random, 95% CI) 1.03 [0.81, 1.31]
8 Failure rate on day 5 2 1905 Odds Ratio (M-H, Random, 95% CI) 1.15 [0.58, 2.30]
9 Failure rate on day 14 1 1702 Odds Ratio (M-H, Random, 95% CI) 1.04 [0.84, 1.29]
10 Death rates 2 1905 Odds Ratio (M-H, Random, 95% CI) 0.07 [0.00, 1.18]
11 Nasopharyngeal H. influenzae 1 1482 Odds Ratio (M-H, Fixed, 95% CI) 1.00 [0.78, 1.29]
12 Respiratory syncytial virus 2 1731 Odds Ratio (M-H, Random, 95% CI) 1.04 [0.83, 1.31]
(RSV) in nasopharyngeal swabs
13 Mean age 1 203 Mean Difference (IV, Random, 95% CI) 0.0 [0.0, 0.0]
14 Blood culture positive for S. 1 203 Odds Ratio (M-H, Random, 95% CI) 0.34 [0.03, 3.29]
pneumoniae
15 Failure rate on day 5 in 1 203 Odds Ratio (M-H, Random, 95% CI) 2.36 [0.59, 9.39]
radiographically confirmed
pneumonia

Comparison 18. Amoxycillin with IV ampicillin

No. of No. of
Outcome or subgroup title studies participants Statistical method Effect size

1 Age below one year 1 2037 Odds Ratio (M-H, Random, 95% CI) 0.94 [0.79, 1.13]
2 Male sex 1 2037 Odds Ratio (M-H, Random, 95% CI) 1.09 [0.91, 1.30]
3 Wheezing in infants 1 1311 Odds Ratio (M-H, Random, 95% CI) 1.03 [0.78, 1.37]
4 Wheezing in age group one to 1 726 Odds Ratio (M-H, Random, 95% CI) 0.77 [0.56, 1.04]
five years
5 Failure rates 1 2037 Odds Ratio (M-H, Random, 95% CI) 0.86 [0.63, 1.19]
6 Relapse rates 1 1873 Odds Ratio (M-H, Random, 95% CI) 0.78 [0.46, 1.33]
7 Death rates 1 2037 Odds Ratio (M-H, Random, 95% CI) 0.25 [0.03, 2.21]
8 Lost to follow-up 1 2037 Odds Ratio (M-H, Random, 95% CI) 0.45 [0.17, 1.20]
9 Protocol violation 1 2037 Odds Ratio (M-H, Random, 95% CI) 0.92 [0.43, 1.96]

Antibiotics for community-acquired pneumonia in children (Review) 61
Copyright © 2013 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

95% CI) 0.74] 2 Male sex 1 709 Odds Ratio (M-H. 95% CI) 1.64.45.35] 3 Cure rates 1 82 Odds Ratio (M-H.30.93 [0. 95% CI) 4. Penicillin and gentamycin with co-amoxyclavulanic acid No.01.06. Amoxycillin with clarithromycin No. 95% CI) 1.32] 3 Failure rates 1 71 Odds Ratio (M-H.04. 1.64.69. of Outcome or subgroup title studies participants Statistical method Effect size 1 Mean age 1 85 Mean Difference (IV.55 [0. 17. Random.42. Random.36] 3 Numbers received antibiotics in 1 709 Odds Ratio (M-H. Random. 10. Random. 95% CI) 0.55.35 [0. 14.06.05 [0.98] 2 Male sex 1 85 Odds Ratio (M-H. Random. 95% CI) 1. 2. 1. 15. of No. 1. Random. Random. 0. Levofloxacin with comparator (co-amoxyclavulanic acid/ceftriaxone) No. 3. 95% CI) 0.Comparison 19. Random.11 [0.05 [-0.43] year age 2 Male sex 1 63 Odds Ratio (M-H. of No. Published by John Wiley & Sons. Amoxycillin with cefuroxime No. Random. 95% CI) 0. 95% CI) 2. 95% CI) 0. 4. 95% CI) 0.40] 4 Failure rates 1 82 Odds Ratio (M-H.39] 2 Male sex 1 85 Odds Ratio (M-H. 23. 5. of Outcome or subgroup title studies participants Statistical method Effect size 1 Mean age in months 1 84 Mean Difference (IV. .05.74] Comparison 21.97 [0.42] Antibiotics for community-acquired pneumonia in children (Review) 62 Copyright © 2013 The Cochrane Collaboration.05 [0. Random.05 [0. 95% CI) 0.18. of No. Random.90] 3 Cure rates 1 84 Odds Ratio (M-H.35] past 1 week 4 Cure rates 1 539 Odds Ratio (M-H. of Outcome or subgroup title studies participants Statistical method Effect size 1 Mean age 1 709 Mean Difference (IV.20. 95% CI) 0.16 [-10.46. 4. Random. of Outcome or subgroup title studies participants Statistical method Effect size 1 Number of children less than 1 1 71 Odds Ratio (M-H. 0.49 [0.39] Comparison 22. Random. Random.47 [-1. Ltd. Random.54 [0. of No. 95% CI) 1.51] 4 Failure rates 1 84 Odds Ratio (M-H.86 [0.95 [0. 95% CI) -3.59] Comparison 20.

Random.Comparison 23.03 [-13. of Outcome or subgroup title studies participants Statistical method Effect size 1 Male sex 1 140 Odds Ratio (M-H. 4.11.64. Random.42. 95% CI) 6. Random.06 [0. of Outcome or subgroup title studies participants Statistical method Effect size 1 Age in months 1 111 Mean Difference (IV.46.70] Antibiotics for community-acquired pneumonia in children (Review) 63 Copyright © 2013 The Cochrane Collaboration.44] 2 Male sex 1 111 Odds Ratio (M-H. -0. 7.04.32 [0. 118. 95% CI) 0. 1.42 [0.11. 31.55 [1. Random. 2.83] 3 Adverse reaction 1 140 Odds Ratio (M-H. 95% CI) 1. 95% CI) 1. 3.89] 3 Weight for age 1 111 Mean Difference (IV. 0.21 [0. 95% CI) 3. Random. Random.89] 4 Failure rates 1 84 Odds Ratio (M-H.0 [0.83 [1.16. 95% CI) 1.94 [0.40] 6 Failure rate 1 111 Odds Ratio (M-H.24.30] culture or lung puncture Comparison 25. Co-trimoxazole versus chloramphenicol No.78. 95% CI) 0.90 [-0.51] Comparison 24.15] 5 Cure rate 1 111 Odds Ratio (M-H.30] 9 Need for change in antibiotics 1 111 Odds Ratio (M-H.79 [0.47. Random. 95% CI) 2.29] 4 Cure rate 1 140 Odds Ratio (M-H. 95% CI) 1. Fixed. Fixed.94] 5 Failure rate 1 140 Odds Ratio (M-H.41.25 [0.83] 11 Organisms isolated on blood 1 111 Odds Ratio (M-H. Random.63. 95% CI) -7. 5. 95% CI) 0. 4.33] 7 Excluded 1 111 Odds Ratio (M-H.11. 95% CI) 14. 4. of No. Random. Published by John Wiley & Sons. Fixed.67 [0.54] 2 Positive for microbial agent 1 140 Odds Ratio (M-H. 95% CI) 2. Ltd. of No. 95% CI) 1.90] 2 Male sex 1 84 Odds Ratio (M-H. Ceftibuten versus cefuroxime No.51 [0. of No. Random.03 [0.35.46.76] 3 Cure rates 1 82 Odds Ratio (M-H.81 [1. Random. 2. 3. 23.12] 8 Relapse rate 1 111 Odds Ratio (M-H. 95% CI) 1. Random. 1.0 [-3. 4. 95% CI) 0. of Outcome or subgroup title studies participants Statistical method Effect size 1 Mean age 1 83 Mean Difference (IV. Random.45. 11. Cefuroxime with clarithromycin No.05 [0.89 [0. 95% CI) 0.11] 4 Wheezing positive 1 111 Odds Ratio (M-H. 95% CI) 2.02 [0. Fixed. . Random. Random. 95% CI) 0.47.87. 7.18. 2. Random. 95% CI) 0.42.40] 10 Death rate 1 111 Odds Ratio (M-H.

52] week 4 Children with wheezing 2 3739 Odds Ratio (M-H. 95% CI) 1. Published by John Wiley & Sons. 95% CI) 1.92] 6 Mean time for improvement in 1 104 Mean Difference (IV. 95% CI) 0. 2.10] oral amoxicillin or injectable antibiotics 10 Failure rate in children 2 219 Odds Ratio (M-H. Random.28. 1. 95% CI) 1.77. 95% CI) 1. 95% CI) -0. 1. Random.94. 4. Random. 95% CI) -3. -0. Random.76. Random. 95% CI) 0.45 [0.03. 95% CI) 0. 1.54] 3 Mean number of days before 1 104 Mean Difference (IV. Oral versus parenteral antibiotics for treatment of severe pneumonia No. 95% CI) 0.09] years of age 9 Failure rates in children receiving 4 4112 Odds Ratio (M-H.24.78. Random. 95% CI) 1. 3. 1. 1. 2.03.69. Random. 1. Random.0 [-1. Random. 1. Random.06 [0. 3. Random. 95% CI) 1.70 [0.29] receiving cotrimoxazole or injectable penicillin 11 Failure rate in children 4 2426 Odds Ratio (M-H.83] study 13 Hospitalisation 3 458 Odds Ratio (M-H.09 [0. 1.56 [0.82] 15 Death rates 3 3942 Odds Ratio (M-H.98 [0.90 [-2.82.91 [0.33] Antibiotics for community-acquired pneumonia in children (Review) 64 Copyright © 2013 The Cochrane Collaboration. 95% CI) 0.31] 6 Failure rates on day 3 3 3942 Odds Ratio (M-H.21] 2 Age below 12 months 4 3961 Odds Ratio (M-H.34] Comparison 27. Random. 95% CI) 1.13 [0.04 [0.87] 16 Lost to follow-up 1 2037 Odds Ratio (M-H.40 [-5. Random. 1. Random. Ltd. of Outcome or subgroup title studies participants Statistical method Effect size 1 Male sex 5 4164 Odds Ratio (M-H.89. 95% CI) 5.28 [0.11 [0.86.95 [0.95 [0. Random.68] 5 RSV positivity 2 1634 Odds Ratio (M-H.84 [0.34. of No.15 [0.34] 14 Relapse rates 2 2076 Odds Ratio (M-H.17 [0. 0.48] admission 4 Received antibiotics before 1 104 Odds Ratio (M-H. Random. Random. Random.14 [0.76] enrolment 5 Failure rates 1 104 Odds Ratio (M-H. 2. 95% CI) 0. 95% CI) -1.24] 8 Failure rate in children below 5 3 3870 Odds Ratio (M-H. Oxacillin ceftriaxone versus co-amoxyclavulanic acid No.15] 7 Failure rates on day 6 6 4331 Odds Ratio (M-H. 1. Random.70. Random.32] treated with oral or parenteral antibiotics on ambulatory basis 12 Failure rate after removing one 2 2240 Odds Ratio (M-H. Random. of Outcome or subgroup title studies participants Statistical method Effect size 1 Median age (months) with IQR Other data No numeric data 2 Male sex 1 104 Odds Ratio (M-H.33.46.92 [0. .05 [1. 1. of No.11] tachypnoea 7 Mean length of stay 1 104 Mean Difference (IV.56. Random. 21.20] 17 Cure rate 2 334 Odds Ratio (M-H.50. 95% CI) 0.31 [0. 1. 95% CI) 1.32. Random.30] 3 Received antibiotics in the past 3 3942 Odds Ratio (M-H.38.Comparison 26. 95% CI) 0. 95% CI) 0.19. 95% CI) 0.44.17. -1. 95% CI) 0. 0.

68] 2 Male sex 2 1032 Odds Ratio (M-H.18 Failure rates in radiographically 2 373 Odds Ratio (M-H.90 [16. Random.88] 3 Failure rate 2 1232 Odds Ratio (M-H.29] confirmed-pneumonia 19 Death rates after removing one 2 2240 Odds Ratio (M-H.06] Comparison 29. 4.0] 2 Male sex 1 51 Odds Ratio (M-H.41. 2. 95% CI) 0.60 [-10. Random. of Outcome or subgroup title studies participants Statistical method Effect size 1 Children below 1 year of age 2 1232 Odds Ratio (M-H. 95% CI) 0.39. of Outcome or subgroup title studies participants Statistical method Effect size 1 Age in months 1 284 Mean Difference (IV. Random. 95% CI) 0. 44. Published by John Wiley & Sons. -2. 95% CI) 0.08] 4 Failure rates 2 1065 Odds Ratio (M-H. Random.00] Antibiotics for community-acquired pneumonia in children (Review) 65 Copyright © 2013 The Cochrane Collaboration.0. 3. Fixed. 0. 95% CI) -6. Amoxycillin versus cefpodoxime No. of No. 95% CI) 1. 848.07.0 [0. . 0. 53. Random.57.08.52. of No.71 [0. Co-trimoxazole versus co-amoxyclavulanic acid No.54 [0. 1. 95% CI) 2. of No.03. Ltd. 7.41.53] Comparison 30.64 [0. Random. Random. 95% CI) 4.21] study Comparison 28. 0. 95% CI) 1. of Outcome or subgroup title studies participants Statistical method Effect size 1 Age (mean/median) 2 1032 Mean Difference (IV. Random.26 [2. Random. Random.18.09] 3 Response/cure rate 1 238 Odds Ratio (M-H.34 [1. Random.55.33 [0.98 [3. 95% CI) 12. Amoxycillin versus chloramphenicol No.53] 3 Cure rate 1 796 Odds Ratio (M-H.25 [0. 95% CI) 117. 37] 2 Male sex 2 1232 Odds Ratio (M-H.20 [0.33. 95% CI) 0.

Ltd.48 [ -18. Outcome 1 Mean age (months). Comparison 1 Azithromycin versus erythromycin.62 (P = 0.57 ] Heterogeneity: Tau2 = 0.6 (53.36). df = 1 (P = 0. 33.5 % -8. Review: Antibiotics for community-acquired pneumonia in children Comparison: 1 Azithromycin versus erythromycin Outcome: 1 Mean age (months) Mean Mean Study or subgroup Azithromycin Erythromycin Difference Weight Difference N Mean(SD) N Mean(SD) IV. 9. Analysis 1.53) Test for subgroup differences: Not applicable -10 -5 0 5 10 Azithromycin Erythromycin Antibiotics for community-acquired pneumonia in children (Review) 66 Copyright © 2013 The Cochrane Collaboration.Random.72 ] Roord 1996 45 58.Random.2 (39.5 % 6.0.8) 26.2 (52.95% CI Harris 1998 156 0 (0) 69 0 (0) Not estimable Kogan 2003 33 62.83. . Chi2 = 0.0% Test for overall effect: Z = 0.99 ] Total (95% CI) 234 135 100.6) 73.2) 40 67.95% CI IV.1. Published by John Wiley & Sons.40 [ -20.6) 26 56. 7.0 % -4. I2 =0.8 (37.54.92.40 [ -24.79.

Outcome 2 Male sex.94. Ltd.33.67 [ 0. I2 =83% Test for overall effect: Z = 0.63.10 ] Kogan 2003 17/33 14/26 11. Outcome 3 Wheezing present.29) Test for subgroup differences: Not applicable 0.48.Random.91 [ 0. Analysis 1.76) Test for subgroup differences: Not applicable 0. I2 =0. 3.43.95% H.8 % 2. df = 1 (P = 0.01 0.1 0.44). Chi2 = 5.02).1 1 10 100 1000 Azithromycin Erythromycin Antibiotics for community-acquired pneumonia in children (Review) 67 Copyright © 2013 The Cochrane Collaboration.3. Comparison 1 Azithromycin versus erythromycin.6 % 0.18 ] Total events: 198 (Azithromycin).73 [ 0.83 [ 0.6 % 1. Review: Antibiotics for community-acquired pneumonia in children Comparison: 1 Azithromycin versus erythromycin Outcome: 3 Wheezing present Study or subgroup Azithromycin Erythromycin Odds Ratio Weight Odds Ratio M.2 0.03 ] Kogan 2003 22/33 11/26 43. 7.0 % 0.0% Test for overall effect: Z = 1.2.73 [ 0. Chi2 = 1. M- H.Random. Comparison 1 Azithromycin versus erythromycin. df = 2 (P = 0. M- H.Random.06 (P = 0.87 ] Total events: 104 (Azithromycin).0 % 1.82. 62 (Erythromycin) Heterogeneity: Tau2 = 0. 2.55 ] Roord 1996 28/45 22/40 16.8 % 0.23 [ 0.35 [ 0. Published by John Wiley & Sons.2 % 0.57. 1.30 (P = 0.31. Review: Antibiotics for community-acquired pneumonia in children Comparison: 1 Azithromycin versus erythromycin Outcome: 2 Male sex Study or subgroup Azithromycin Erythromycin Odds Ratio Weight Odds Ratio M.89 ] Total (95% CI) 318 161 100. 1.5 1 2 5 10 Azithromycin Erythromycin Analysis 1.0. 1.21 ] Total (95% CI) 363 201 100.Random.95% H.81.95% n/N n/N CI CI Harris 1998 153/285 83/135 71. 119 (Erythromycin) Heterogeneity: Tau2 = 0. 4.001 0.95% n/N n/N CI CI Harris 1998 82/285 51/135 56. .58.

1 1 10 50 Azithromycin Erythromycin Antibiotics for community-acquired pneumonia in children (Review) 68 Copyright © 2013 The Cochrane Collaboration.50. Review: Antibiotics for community-acquired pneumonia in children Comparison: 1 Azithromycin versus erythromycin Outcome: 4 Cure rate Study or subgroup Azithromycin Erythromycin Odds Ratio Weight Odds Ratio M.0 % 1.22 [ 0. 100 (Erythromycin) Heterogeneity: Tau2 = 0.66) Test for subgroup differences: Not applicable 0.0 % 1. Comparison 1 Azithromycin versus erythromycin. df = 2 (P = 0.02 0.94 ] Total events: 179 (Azithromycin).44 (P = 0.93 ] Roord 1996 31/45 27/40 41.14 [ 0.43.15).Random. .90 [ 0. 325.45.0 % 0.66 ] Total (95% CI) 230 133 100. Outcome 4 Cure rate. Chi2 = 3. 1. Analysis 1. Published by John Wiley & Sons.28.07 [ 0. Ltd.95% H. I2 =47% Test for overall effect: Z = 0.78 ] Kogan 2003 33/33 21/26 8.80. M- H. 2.90.95% n/N n/N CI CI Harris 1998 115/152 52/67 51.0 % 17. 2.Random.4.

Outcome 6 Side effects. 4.18.1 0.91 ] Wubbel 1999 1/39 1/49 24.10 (P = 0. 6 (Erythromycin) Heterogeneity: Tau2 = 0.Random. Comparison 1 Azithromycin versus erythromycin. 14 (Erythromycin) Heterogeneity: Tau2 = 1.08.5. Published by John Wiley & Sons. Ltd. .94.6 % 2.0 % 0.86 ] Total (95% CI) 236 156 100.95% H.Random.2 0.Random.01 0. Outcome 5 Failure rate.0 % 1. 1.02. 16. M- H. Chi2 = 2. df = 2 (P = 0. 2.95% H. df = 1 (P = 0.89 ] Total events: 6 (Azithromycin).39 [ 0.5 1 2 5 10 Azithromycin Erythromycin Antibiotics for community-acquired pneumonia in children (Review) 69 Copyright © 2013 The Cochrane Collaboration.20 [ 0.4 % 0.27 ] Roord 1996 1/45 4/40 37. I2 =75% Test for overall effect: Z = 0.06 [ 0.6 % 0.73 ] Total events: 17 (Azithromycin).0 % 0.05).18.95% n/N n/N CI CI Harris 1998 4/152 1/67 38.4 % 1.69.11.20. Review: Antibiotics for community-acquired pneumonia in children Comparison: 1 Azithromycin versus erythromycin Outcome: 5 Failure rate Study or subgroup Azithromycin Erythromycin Odds Ratio Weight Odds Ratio M. Comparison 1 Azithromycin versus erythromycin. 20.26 [ 0.6. M- H. Analysis 1.92) Test for subgroup differences: Not applicable 0. I2 =1% Test for overall effect: Z = 0. 1.78 [ 0.92 [ 0.03. 6.73 [ 0.95% n/N n/N CI CI Roord 1996 12/45 6/40 51.34 ] Total (95% CI) 84 69 100.36).13 ] Wubbel 1999 5/39 8/29 48.1 1 10 100 Azithromycin Erythromycin Analysis 1.02.Random. Chi2 = 3.45 (P = 0.05. Review: Antibiotics for community-acquired pneumonia in children Comparison: 1 Azithromycin versus erythromycin Outcome: 6 Side effects Study or subgroup Azithromycin Erythromycin Odds Ratio Weight Odds Ratio M.65) Test for subgroup differences: Not applicable 0.

95% n/N n/N CI CI Harris 1998 103/156 43/69 39. Analysis 1.2 0.0 % 0.07.03).51.61 (P = 0.69 ] Roord 1996 24/44 20/40 33.1 % 0.30. Outcome 7 Organisms identified by serology or nasopharyngeal cultures. 1. 2. 83 (Erythromycin) Heterogeneity: Tau2 = 0.Random. Chi2 = 7. Ltd. Published by John Wiley & Sons. I2 =71% Test for overall effect: Z = 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 1 Azithromycin versus erythromycin Outcome: 7 Organisms identified by serology or nasopharyngeal cultures Study or subgroup Azithromycin Erythromycin Odds Ratio Weight Odds Ratio M.01.18 [ 0. 2. M- H.46.12 ] Kogan 2003 14/33 20/26 27. df = 2 (P = 0.5 % 1.Random.7.20 [ 0. .95% H.1 0.5 1 2 5 10 Azithromycin Erythromycin Antibiotics for community-acquired pneumonia in children (Review) 70 Copyright © 2013 The Cochrane Collaboration.65.83 ] Total (95% CI) 233 135 100. 0. Comparison 1 Azithromycin versus erythromycin.54) Test for subgroup differences: Not applicable 0.75 [ 0.22 [ 0.87 ] Total events: 141 (Azithromycin).4 % 1.

Random.82).9 % 1. .05.65.72 [ 0.56 ] Total events: 21 (Azithromycin). Published by John Wiley & Sons.95% n/N n/N CI CI Kogan 2003 20/33 12/26 87.09 (P = 0. 5.01 0.0 % 1. df = 1 (P = 0. I2 =0.79 [ 0.8. Outcome 8 Cure rate in radiographically confirmed pneumonia.0. 13 (Erythromycin) Heterogeneity: Tau2 = 0.08.1 1 10 100 Azithromycin Erythromycin Antibiotics for community-acquired pneumonia in children (Review) 71 Copyright © 2013 The Cochrane Collaboration. Chi2 = 0. Comparison 1 Azithromycin versus erythromycin. Analysis 1.28) Test for subgroup differences: Not applicable 0. Ltd. M- H.86 ] Total (95% CI) 72 75 100. 20.08 ] Wubbel 1999 1/39 1/49 12.26 [ 0. 4.95% H.1 % 1.0% Test for overall effect: Z = 1. Review: Antibiotics for community-acquired pneumonia in children Comparison: 1 Azithromycin versus erythromycin Outcome: 8 Cure rate in radiographically confirmed pneumonia Study or subgroup Azithromycin Erythromycin Odds Ratio Weight Odds Ratio M.63.Random.

Review: Antibiotics for community-acquired pneumonia in children Comparison: 2 Clarithromycin versus erythromycin Outcome: 1 Age below 5 years Study or subgroup Clarithromycin Erythromycin Odds Ratio Weight Odds Ratio M.55 ] Total events: 45 (Clarithromycin). I2 =0.29.98 (P = 0.27 (P = 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 1 Azithromycin versus erythromycin Outcome: 9 Failure rate in radiographically confirmed pneumonia Study or subgroup Azithromycin Erythromycin Odds Ratio Weight Odds Ratio M.01 0.86 ] Total (95% CI) 72 75 100. 1.0 % 0.26 [ 0.58 ] Wubbel 1999 1/39 1/49 12.56 [ 0.93 [ 0.0 % 0.23.Random.95% n/N n/N CI CI Block 1995 45/133 45/127 100. 45 (Erythromycin) Heterogeneity: not applicable Test for overall effect: Z = 0. Analysis 1.0. 20.59).0 % 0.55 ] Total (95% CI) 133 127 100. Chi2 = 0.Random. df = 1 (P = 0.1 1 10 100 Azithromycin Erythromycin Analysis 2.93 [ 0. Ltd. Comparison 1 Azithromycin versus erythromycin.62 [ 0. M- H.08.0% Test for overall effect: Z = 0. Published by John Wiley & Sons. 1. Outcome 9 Failure rate in radiographically confirmed pneumonia. M- H.9. .2 0. Outcome 1 Age below 5 years.20.95% H.Random.95% n/N n/N CI CI Kogan 2003 13/33 14/26 87.1 % 1.95% H.56. 1.63 ] Total events: 14 (Azithromycin).5 1 2 5 10 Clarithromycin Erythromycin Antibiotics for community-acquired pneumonia in children (Review) 72 Copyright © 2013 The Cochrane Collaboration.1 0. 1.Random. 15 (Erythromycin) Heterogeneity: Tau2 = 0.1.33) Test for subgroup differences: Not applicable 0.56.79) Test for subgroup differences: Not applicable 0. Comparison 2 Clarithromycin versus erythromycin.9 % 0.

3.84.Random.0 % 1.84.08 ] Total events: 104 (Clarithromycin).61 [ 0.2.15) Test for subgroup differences: Not applicable 0. 3.5 1 2 5 10 Clarithromycin Erythromycin Antibiotics for community-acquired pneumonia in children (Review) 73 Copyright © 2013 The Cochrane Collaboration. 84 (Erythromycin) Heterogeneity: not applicable Test for overall effect: Z = 1. Comparison 2 Clarithromycin versus erythromycin. Ltd.0 % 1. Published by John Wiley & Sons. Analysis 2.2 0.61 [ 0.08 ] Total (95% CI) 124 110 100.95% n/N n/N CI CI Block 1995 104/124 84/110 100. . Review: Antibiotics for community-acquired pneumonia in children Comparison: 2 Clarithromycin versus erythromycin Outcome: 2 Cure rates Study or subgroup Clarithromycin Erythromycin Odds Ratio Weight Odds Ratio M. M- H.Random.1 0.44 (P = 0.95% H. Outcome 2 Cure rates.

38) Test for subgroup differences: Not applicable 0.1 0.23 ] Total (95% CI) 124 110 100. .2 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 2 Clarithromycin versus erythromycin Outcome: 3 Clinical success rate Study or subgroup Clarithromycin Erythromycin Odds Ratio Weight Odds Ratio M.Random. Published by John Wiley & Sons.Random.45.3. Comparison 2 Clarithromycin versus erythromycin.12. 2. Comparison 2 Clarithromycin versus erythromycin.23 ] Total (95% CI) 124 110 100. Outcome 3 Clinical success rate. M- H.88 (P = 0.95% n/N n/N CI CI Block 1995 121/124 105/110 100.38) Test for subgroup differences: Not applicable 0.0 % 1.52 [ 0. 5 (Erythromycin) Heterogeneity: not applicable Test for overall effect: Z = 0.92 [ 0.95% H.95% H. 8.52 [ 0. Analysis 2.95% n/N n/N CI CI Block 1995 3/124 5/110 100.2 0. Outcome 4 Failure rate.0 % 0. 2.0 % 0.92 [ 0.0 % 1.12. Review: Antibiotics for community-acquired pneumonia in children Comparison: 2 Clarithromycin versus erythromycin Outcome: 4 Failure rate Study or subgroup Clarithromycin Erythromycin Odds Ratio Weight Odds Ratio M.45. Ltd.5 1 2 5 10 Clarithromycin Erythromycin Analysis 2.Random.5 1 2 5 10 Clarithromycin Erythromycin Antibiotics for community-acquired pneumonia in children (Review) 74 Copyright © 2013 The Cochrane Collaboration. 105 (Erythromycin) Heterogeneity: not applicable Test for overall effect: Z = 0.88 (P = 0.4.23 ] Total events: 121 (Clarithromycin). M- H.23 ] Total events: 3 (Clarithromycin). 8.Random.1 0.

02.16 ] Total events: 103 (Clarithromycin).2 0.95% n/N n/N CI CI Block 1995 103/111 82/98 100. M- H.51 [ 1.044) Test for subgroup differences: Not applicable 0.16 ] Total (95% CI) 111 98 100.1 0.17 [ 0. 6. Ltd. 82 (Erythromycin) Heterogeneity: not applicable Test for overall effect: Z = 2. Analysis 2.Random.Random.17 [ 0.95% H. 1.45 ] Total events: 1 (Clarithromycin). Comparison 2 Clarithromycin versus erythromycin.0 % 0.95% n/N n/N CI CI Block 1995 1/121 5/105 100.45 ] Total (95% CI) 121 105 100.1 1 10 200 Clarithromycin Erythromycin Analysis 2.01 (P = 0.5.5 1 2 5 10 Clarithromycin Erythromycin Antibiotics for community-acquired pneumonia in children (Review) 75 Copyright © 2013 The Cochrane Collaboration. Review: Antibiotics for community-acquired pneumonia in children Comparison: 2 Clarithromycin versus erythromycin Outcome: 6 Radiologic resolution Study or subgroup Clarithromycin Erythromycin Odds Ratio Weight Odds Ratio M.51 [ 1.10) Test for subgroup differences: Not applicable 0. 6.02.Random.0 % 2.005 0. Outcome 5 Relapse rate.02. 5 (Erythromycin) Heterogeneity: not applicable Test for overall effect: Z = 1.6. 1. M- H.Random.0 % 2.0 % 0.95% H.62 (P = 0. .02. Published by John Wiley & Sons. Review: Antibiotics for community-acquired pneumonia in children Comparison: 2 Clarithromycin versus erythromycin Outcome: 5 Relapse rate Study or subgroup Clarithromycin Erythromycin Odds Ratio Weight Odds Ratio M. Outcome 6 Radiologic resolution. Comparison 2 Clarithromycin versus erythromycin.

2 0.06.34 [ 0.1 0.13) Test for subgroup differences: Not applicable 0.95% H.55 [ 0. 18. 5 (Erythromycin) Heterogeneity: not applicable Test for overall effect: Z = 1.0 % 3.06. Review: Antibiotics for community-acquired pneumonia in children Comparison: 2 Clarithromycin versus erythromycin Outcome: 8 Radiologic failure Study or subgroup Clarithromycin Erythromycin Odds Ratio Weight Odds Ratio M.5 1 2 5 10 Clarithromycin Erythromycin Analysis 2.95% n/N n/N CI CI Block 1995 109/111 92/98 100.0 % 3. 92 (Erythromycin) Heterogeneity: not applicable Test for overall effect: Z = 1.95% n/N n/N CI CI Block 1995 2/111 5/98 100.80 ] Total events: 2 (Clarithromycin). .04 ] Total (95% CI) 111 98 100. M- H. 1. 18.0 % 0.21) Test for subgroup differences: Not applicable 0. Outcome 7 Radiologic success. Comparison 2 Clarithromycin versus erythromycin.70. Published by John Wiley & Sons. Outcome 8 Radiologic failure. Review: Antibiotics for community-acquired pneumonia in children Comparison: 2 Clarithromycin versus erythromycin Outcome: 7 Radiologic success Study or subgroup Clarithromycin Erythromycin Odds Ratio Weight Odds Ratio M. Ltd.5 1 2 5 10 Clarithromycin Erythromycin Antibiotics for community-acquired pneumonia in children (Review) 76 Copyright © 2013 The Cochrane Collaboration.Random.27 (P = 0.70. Analysis 2.8.04 ] Total events: 109 (Clarithromycin).80 ] Total (95% CI) 111 98 100.Random.Random.1 0.2 0.7.34 [ 0.53 (P = 0.95% H.0 % 0. 1.55 [ 0. Comparison 2 Clarithromycin versus erythromycin. M- H.Random.

0 % 1.95% n/N n/N CI CI Block 1995 24/27 16/18 100.Random. 6.07 [ 0.0 % 1.9.2 0. Comparison 2 Clarithromycin versus erythromycin. Published by John Wiley & Sons.95% H.0) Test for subgroup differences: Not applicable 0.15.60.5 1 2 5 10 Clarithromycin Erythromycin Analysis 2. Review: Antibiotics for community-acquired pneumonia in children Comparison: 2 Clarithromycin versus erythromycin Outcome: 9 Adverse events Study or subgroup Clarithromycin Erythromycin Odds Ratio Weight Odds Ratio M.10.07 [ 0.95% n/N n/N CI CI Block 1995 32/133 29/127 100.95% H.5 1 2 5 10 Clarithromycin Erythromycin Antibiotics for community-acquired pneumonia in children (Review) 77 Copyright © 2013 The Cochrane Collaboration. 6.Random.15. Ltd.Random. Comparison 2 Clarithromycin versus erythromycin. . M- H. Outcome 9 Adverse events.82) Test for subgroup differences: Not applicable 0.Random.0 % 1. 1. Review: Antibiotics for community-acquired pneumonia in children Comparison: 2 Clarithromycin versus erythromycin Outcome: 10 Bacteriologic response Study or subgroup Clarithromycin Erythromycin Odds Ratio Weight Odds Ratio M. Outcome 10 Bacteriologic response.0 (P = 1.60.23 (P = 0.1 0. 16 (Erythromycin) Heterogeneity: not applicable Test for overall effect: Z = 0.67 ] Total events: 24 (Clarithromycin).90 ] Total (95% CI) 133 127 100.90 ] Total events: 32 (Clarithromycin).2 0.1 0. 29 (Erythromycin) Heterogeneity: not applicable Test for overall effect: Z = 0.00 [ 0.67 ] Total (95% CI) 27 18 100.0 % 1. 1. M- H.00 [ 0. Analysis 2.

0.53 ] Total events: 12 (Azithromycin).5 1 2 5 10 Azithromycin Co-amoxyclavulanic acid Analysis 3. Comparison 3 Azithromycin versus co-amoxyclavulanic acid. . 3. 42 (Co-amoxyclavulanic acid) Heterogeneity: not applicable Test for overall effect: Z = 0.55).95% H. df = 1 (P = 0.08 ] Total (95% CI) 164 112 100.54. Analysis 3.36. Review: Antibiotics for community-acquired pneumonia in children Comparison: 3 Azithromycin versus co-amoxyclavulanic acid Outcome: 2 Failure rate Co- amoxyclavulanic Study or subgroup Azithromycin acid Odds Ratio Weight Odds Ratio M.94) Test for subgroup differences: Not applicable 0.36 (P = 0.0 % 1. 1.02 [ 0.Random. Published by John Wiley & Sons. 7. 1.Random.07 (P = 0.95% H. Chi2 = 0.2 % 0. M- H. Ltd.95 ] Total (95% CI) 125 63 100.02 [ 0.Random.72) Test for subgroup differences: Not applicable 0.1 0. I2 =0.5 1 2 5 10 Azithromycin Co-amoxyclavulanic acid Antibiotics for community-acquired pneumonia in children (Review) 78 Copyright © 2013 The Cochrane Collaboration.42 [ 0. 4.2.54.2 0.42. M- H. Outcome 1 Cure rate. Review: Antibiotics for community-acquired pneumonia in children Comparison: 3 Azithromycin versus co-amoxyclavulanic acid Outcome: 1 Cure rate Co- amoxyclavulanic Study or subgroup Azithromycin acid Odds Ratio Weight Odds Ratio M.0 % 1.21 [ 0. Outcome 2 Failure rate.95% n/N n/N CI CI Harris 1998 84/125 42/63 100.95% n/N n/N CI CI Harris 1998 11/125 4/63 80.8 % 1.1 0.1. Comparison 3 Azithromycin versus co-amoxyclavulanic acid.43.62 [ 0.0% Test for overall effect: Z = 0.0 % 1.66 ] Wubbel 1999 1/39 2/49 19. 6 (Co-amoxyclavulanic acid) Heterogeneity: Tau2 = 0.2 0.05.Random.95 ] Total events: 84 (Azithromycin).

66) Test for subgroup differences: Not applicable 0. 1.3.85 [ 0. Published by John Wiley & Sons.85 [ 0. 17 (Co-amoxyclavulanic acid) Heterogeneity: not applicable Test for overall effect: Z = 0.71 ] Total (95% CI) 125 63 100.Random.95% H.43. Comparison 3 Azithromycin versus co-amoxyclavulanic acid. Analysis 3. .43.0 % 0. M- H. Review: Antibiotics for community-acquired pneumonia in children Comparison: 3 Azithromycin versus co-amoxyclavulanic acid Outcome: 3 Improved Co- amoxyclavulanic Study or subgroup Azithromycin acid Odds Ratio Weight Odds Ratio M.5 1 2 5 10 Azithromycin Co-amoxyclavulanic acid Antibiotics for community-acquired pneumonia in children (Review) 79 Copyright © 2013 The Cochrane Collaboration.1 0.0 % 0. 1.95% n/N n/N CI CI Harris 1998 30/125 17/63 100.71 ] Total events: 30 (Azithromycin). Ltd. Outcome 3 Improved.Random.45 (P = 0.2 0.

95% n/N n/N CI CI Harris 1998 5/125 2/63 100.0078) Test for subgroup differences: Not applicable 0.Random.0 % 0.5 1 2 5 10 Azithromycin Co-amoxyclavulanic acid Analysis 3.0 % 1. 0.0 % 1.28 (P = 0.61 ] Total events: 19 (Azithromycin).24.74 ] Total events: 5 (Azithromycin).78) Test for subgroup differences: Not applicable 0.95% n/N n/N CI CI Harris 1998 14/125 19/63 54.04.76.Random.22 ] Total (95% CI) 164 112 100. df = 1 (P = 0.Random.63 ] Wubbel 1999 5/39 33/49 45. Outcome 4 Side effects.95% H.15 [ 0. Published by John Wiley & Sons.07 [ 0.27 [ 0. Outcome 5 Organisms isolated. Ltd.24.8 % 0.2 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 3 Azithromycin versus co-amoxyclavulanic acid Outcome: 5 Organisms isolated Co- amoxyclavulanic Study or subgroup Azithromycin acid Odds Ratio Weight Odds Ratio M. 6.95% H.04). Comparison 3 Azithromycin versus co-amoxyclavulanic acid.27 [ 0. 2 (Co-amoxyclavulanic acid) Heterogeneity: not applicable Test for overall effect: Z = 0. M- H. 0. 0.13.4. I2 =76% Test for overall effect: Z = 2.5.2 0.Random. Review: Antibiotics for community-acquired pneumonia in children Comparison: 3 Azithromycin versus co-amoxyclavulanic acid Outcome: 4 Side effects Co- amoxyclavulanic Study or subgroup Azithromycin acid Odds Ratio Weight Odds Ratio M.5 1 2 5 10 Azithromycin Co-amoxyclavulanic acid Antibiotics for community-acquired pneumonia in children (Review) 80 Copyright © 2013 The Cochrane Collaboration. Chi2 = 4.74 ] Total (95% CI) 125 63 100. .19. M- H.29 [ 0.1 0. Analysis 3.2 % 0. 52 (Co-amoxyclavulanic acid) Heterogeneity: Tau2 = 0.66 (P = 0. 6.1 0.02. Comparison 3 Azithromycin versus co-amoxyclavulanic acid.

64. .Random.0 % 1. Review: Antibiotics for community-acquired pneumonia in children Comparison: 3 Azithromycin versus co-amoxyclavulanic acid Outcome: 6 Mycoplasma serology positive Co- amoxyclavulanic Study or subgroup Azithromycin acid Odds Ratio Weight Odds Ratio M.19 [ 0.22 ] Total events: 85 (Azithromycin).95% H. 2.6. M- H.64.0 % 1. Comparison 3 Azithromycin versus co-amoxyclavulanic acid. 39 (Co-amoxyclavulanic acid) Heterogeneity: not applicable Test for overall effect: Z = 0.54 (P = 0.2 0. Outcome 6 Mycoplasma serology positive.1 0.5 1 2 5 10 Azithromycin Co-amoxyclavulanic acid Antibiotics for community-acquired pneumonia in children (Review) 81 Copyright © 2013 The Cochrane Collaboration.19 [ 0.59) Test for subgroup differences: Not applicable 0. 2. Ltd.22 ] Total (95% CI) 129 63 100.Random. Published by John Wiley & Sons.95% n/N n/N CI CI Harris 1998 85/129 39/63 100. Analysis 3.

39 (P = 0.00001) Test for subgroup differences: Not applicable -100 -50 0 50 100 Azithromycin Amoxycillin Antibiotics for community-acquired pneumonia in children (Review) 82 Copyright © 2013 The Cochrane Collaboration. Outcome 7 Failure rates in radiographically confirmed pneumonia. .08 ] Total events: 1 (Azithromycin).01 0. 7.1 1 10 100 Azithromycin Co-amoxyclavulanic acid Analysis 4.0 % 58. Comparison 4 Azithromycin versus amoxycillin.3) 100. Outcome 1 Age in months. Review: Antibiotics for community-acquired pneumonia in children Comparison: 3 Azithromycin versus co-amoxyclavulanic acid Outcome: 7 Failure rates in radiographically confirmed pneumonia Co- amoxyclavulanic Study or subgroup Azithromycin acid Odds Ratio Weight Odds Ratio M.95% CI Kogan 2003 23 64. 7. 80.70) Test for subgroup differences: Not applicable 0. 80.05. M- H.08 ] Total (95% CI) 39 49 100.7.95% n/N n/N CI CI Wubbel 1999 1/39 2/49 100.10 [ 35.Random.59.59.0 % 0.Random.0 % 0.06 (P < 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 4 Azithromycin versus amoxycillin Outcome: 1 Age in months Mean Mean Study or subgroup Azithromycin Amoxycillin Difference Weight Difference N Mean(SD) N Mean(SD) IV. 2 (Co-amoxyclavulanic acid) Heterogeneity: not applicable Test for overall effect: Z = 0.Random.61 ] Heterogeneity: not applicable Test for overall effect: Z = 5. Published by John Wiley & Sons.1 (44) 53 6 (50.95% H.10 [ 35.95% CI IV.0 % 58.1. Analysis 3.05. Comparison 3 Azithromycin versus co-amoxyclavulanic acid.62 [ 0.Random.62 [ 0. Ltd.61 ] Total (95% CI) 23 53 100.

96 ] Total events: 17 (Azithromycin).6 (2.30 ] Heterogeneity: not applicable Test for overall effect: Z = 0. Outcome 3 Wheezing present. Comparison 4 Azithromycin versus amoxycillin.2. Review: Antibiotics for community-acquired pneumonia in children Comparison: 4 Azithromycin versus amoxycillin Outcome: 3 Wheezing present Study or subgroup Azithromycin Amoxycillin Odds Ratio Weight Odds Ratio M. 1.95% n/N n/N CI CI Kogan 2003 17/23 14/24 100.12 (P = 0.10 [ -1. Published by John Wiley & Sons.3) 24 3.30 ] Total (95% CI) 23 24 100.14 (P = 0.26) Test for subgroup differences: Not applicable 0.96 ] Total (95% CI) 23 24 100.10 [ -1.3.0 % 2.59. 14 (Amoxycillin) Heterogeneity: not applicable Test for overall effect: Z = 1.50. Analysis 4.59. M- H.50.6) 100.Random.02 [ 0.7 (2.01 0. Comparison 4 Azithromycin versus amoxycillin.02 [ 0. Outcome 2 Duration of illness.0 % -0. 1. Review: Antibiotics for community-acquired pneumonia in children Comparison: 4 Azithromycin versus amoxycillin Outcome: 2 Duration of illness Mean Mean Study or subgroup Azithromycin Amoxycillin Difference Weight Difference N Mean(SD) N Mean(SD) IV.95% CI IV. . 6.Random.1 1 10 100 Azithromycin Amoxycillin Antibiotics for community-acquired pneumonia in children (Review) 83 Copyright © 2013 The Cochrane Collaboration. Ltd.89) Test for subgroup differences: Not applicable -10 -5 0 5 10 Azithromycin Amoxycillin Analysis 4.0 % -0. 6.Random.Random.95% H.95% CI Kogan 2003 23 3.0 % 2.

5 1 2 5 10 Azithromycin Amoxycillin Antibiotics for community-acquired pneumonia in children (Review) 84 Copyright © 2013 The Cochrane Collaboration.95% H. Review: Antibiotics for community-acquired pneumonia in children Comparison: 4 Azithromycin versus amoxycillin Outcome: 5 Cure rate radiological Study or subgroup Azithromycin Amoxycillin Odds Ratio Weight Odds Ratio M.2 0. Comparison 4 Azithromycin versus amoxycillin.13) Test for subgroup differences: Not applicable 0.85 [ 0.1 0.Random. Outcome 5 Cure rate radiological. Ltd.0 % 2.Random.95% H.09 ] Total (95% CI) 23 24 100.2 0.51 (P = 0.95% n/N n/N CI CI Kogan 2003 23/23 24/24 Not estimable Total (95% CI) 23 24 Not estimable Total events: 23 (Azithromycin). Comparison 4 Azithromycin versus amoxycillin.85 [ 0. M- H.0 % 2.1 0.5 1 2 5 10 Azithromycin Amoxycillin Analysis 4. . M- H.09 ] Total events: 19 (Azithromycin).Random. 11.73. Published by John Wiley & Sons. 11. 15 (Amoxycillin) Heterogeneity: not applicable Test for overall effect: Z = 1. Analysis 4.Random.5.95% n/N n/N CI CI Kogan 2003 19/23 15/24 100.4. 24 (Amoxycillin) Heterogeneity: not applicable Test for overall effect: not applicable Test for subgroup differences: Not applicable 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 4 Azithromycin versus amoxycillin Outcome: 4 Cure rate clinical Study or subgroup Azithromycin Amoxycillin Odds Ratio Weight Odds Ratio M. Outcome 4 Cure rate clinical.73.

M- H.95% CI IV.5 1 2 5 10 Azithromycin Amoxycillin Analysis 5. Ltd.95% CI Tsarouhas 1998 77 2.Random.37 [ 0.12 ] Heterogeneity: not applicable Test for overall effect: Z = 0.8) 93 2.6.95% n/N n/N CI CI Kogan 2003 16/23 15/24 100. 4.6) 100.2 0.37 [ 0.0 % 0.72 (P = 0.0 % 1.12 ] Total (95% CI) 77 93 100.0 % 1.6 (2. Comparison 4 Azithromycin versus amoxycillin. Outcome 1 Median age. . 1.41. Review: Antibiotics for community-acquired pneumonia in children Comparison: 4 Azithromycin versus amoxycillin Outcome: 6 Fever day 7 Study or subgroup Azithromycin Amoxycillin Odds Ratio Weight Odds Ratio M. Analysis 4. Published by John Wiley & Sons.95% H.1 0. 1.52. Outcome 6 Fever day 7.9 (2.Random. Review: Antibiotics for community-acquired pneumonia in children Comparison: 5 Amoxycillin versus procaine penicillin Outcome: 1 Median age Mean Mean Study or subgroup Amoxycillin Procaine penicillin Difference Weight Difference N Mean(SD) N Mean(SD) IV.47) Test for subgroup differences: Not applicable -10 -5 0 5 10 Amoxycillin Procaine penicillin Antibiotics for community-acquired pneumonia in children (Review) 85 Copyright © 2013 The Cochrane Collaboration.51 (P = 0.30 [ -0.41.30 [ -0.52. 15 (Amoxycillin) Heterogeneity: not applicable Test for overall effect: Z = 0.Random. Comparison 5 Amoxycillin versus procaine penicillin. 4.61 ] Total events: 16 (Azithromycin).61 ] Total (95% CI) 23 24 100.61) Test for subgroup differences: Not applicable 0.1.0 % 0.Random.

Random.0 % 0. Ltd.0 % 0.17. 0.08 [ 0. 3. Review: Antibiotics for community-acquired pneumonia in children Comparison: 6 Co-amoxyclavulanic acid versus amoxycillin Outcome: 1 Poor or no response Co- amoxyclavulanic Study or subgroup acid Amoxycillin Odds Ratio Weight Odds Ratio M.Random. Analysis 5.01. 5 (Procaine penicillin) Heterogeneity: not applicable Test for overall effect: Z = 0.019) Test for subgroup differences: Not applicable 0.1 0.01.95% n/N n/N CI CI Tsarouhas 1998 3/68 5/86 100.95% H. M- H.25 ] Total (95% CI) 68 86 100.1.75 [ 0. Comparison 6 Co-amoxyclavulanic acid versus amoxycillin. 0.2 0. Outcome 1 Poor or no response.95% H.Random.67 ] Total (95% CI) 50 50 100.5 1 2 5 10 Co-amoxyclavulanic acid Amoxycillin Antibiotics for community-acquired pneumonia in children (Review) 86 Copyright © 2013 The Cochrane Collaboration.67 ] Total events: 1 (Co-amoxyclavulanic acid). Published by John Wiley & Sons.0 % 0. M- H.34 (P = 0.95% n/N n/N CI CI Jibril 1989 1/50 10/50 100.0 % 0.2. Outcome 2 Failure rate. 3.1 0.75 [ 0.25 ] Total events: 3 (Amoxycillin).2 0.17. 10 (Amoxycillin) Heterogeneity: not applicable Test for overall effect: Z = 2. Comparison 5 Amoxycillin versus procaine penicillin. .5 1 2 5 10 Amoxycillin Procaine penicillin Analysis 6.08 [ 0.Random.39 (P = 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 5 Amoxycillin versus procaine penicillin Outcome: 2 Failure rate Study or subgroup Amoxycillin Procaine penicillin Odds Ratio Weight Odds Ratio M.70) Test for subgroup differences: Not applicable 0.

Random.1 0.00039) Test for subgroup differences: Not applicable 0.5 1 2 5 10 Co-amoxyclavulanic acid Amoxycillin Analysis 6.5 1 2 5 10 Co-amoxyclavulanic acid Amoxycillin Antibiotics for community-acquired pneumonia in children (Review) 87 Copyright © 2013 The Cochrane Collaboration.Random.0 % 10.44 [ 2.0 % 10.21 ] Total events: 47 (Co-amoxyclavulanic acid).85.06 (P = 0. Published by John Wiley & Sons.21 [ 0. 38.29) Test for subgroup differences: Not applicable 0. 111.44 [ 2.21 ] Total (95% CI) 50 50 100.Random.2. 0 (Amoxycillin) Heterogeneity: not applicable Test for overall effect: Z = 1.2 0. Comparison 6 Co-amoxyclavulanic acid versus amoxycillin. Review: Antibiotics for community-acquired pneumonia in children Comparison: 6 Co-amoxyclavulanic acid versus amoxycillin Outcome: 3 Complications Co- amoxyclavulanic Study or subgroup acid Amoxycillin Odds Ratio Weight Odds Ratio M. 38. M- H.Random.0 % 5.95% n/N n/N CI CI Jibril 1989 47/50 30/50 100. 111.24.0 % 5.3.85. Ltd.21 [ 0.95% H.55 (P = 0.24. Comparison 6 Co-amoxyclavulanic acid versus amoxycillin.24 ] Total events: 2 (Co-amoxyclavulanic acid). Analysis 6. 30 (Amoxycillin) Heterogeneity: not applicable Test for overall effect: Z = 3. Outcome 3 Complications. Review: Antibiotics for community-acquired pneumonia in children Comparison: 6 Co-amoxyclavulanic acid versus amoxycillin Outcome: 2 Cure rate Co- amoxyclavulanic Study or subgroup acid Amoxycillin Odds Ratio Weight Odds Ratio M.95% n/N n/N CI CI Jibril 1989 2/50 0/50 100. Outcome 2 Cure rate. M- H. .1 0.95% H.24 ] Total (95% CI) 50 50 100.2 0.

48) 100.6 (32.95% CI IV.0 % 4. Review: Antibiotics for community-acquired pneumonia in children Comparison: 6 Co-amoxyclavulanic acid versus amoxycillin Outcome: 4 Age (months) Co- amoxyclavulanic Mean Mean Study or subgroup acid Amoxycillin Difference Weight Difference N Mean(SD) N Mean(SD) IV. 17. Analysis 6. .8 (33.47) Test for subgroup differences: Not applicable -10 -5 0 5 10 Co-amoxyclavulanic acid Amoxycillin Antibiotics for community-acquired pneumonia in children (Review) 88 Copyright © 2013 The Cochrane Collaboration.28) 50 58.Random.0 % 4. 17.95% CI Jibril 1989 50 63. Outcome 4 Age (months). Ltd.80 [ -8.69 ] Total (95% CI) 50 50 100. Comparison 6 Co-amoxyclavulanic acid versus amoxycillin.80 [ -8.69 ] Heterogeneity: not applicable Test for overall effect: Z = 0.73 (P = 0.Random.09. Published by John Wiley & Sons.09.4.

31 [ 0.00 (P = 0.52) Test for subgroup differences: Not applicable 0. .95% CI IV.5.71) 50 14.10 [ -1.95% H.64 (P = 0.03 ] Total events: 35 (Co-amoxyclavulanic acid). Published by John Wiley & Sons.Random. 3.32) Test for subgroup differences: Not applicable -10 -5 0 5 10 Co-amoxyclavulanic acid Amoxycillin Analysis 6. Ltd. Comparison 6 Co-amoxyclavulanic acid versus amoxycillin.57.95% n/N n/N CI CI Jibril 1989 35/50 32/50 100.6. Review: Antibiotics for community-acquired pneumonia in children Comparison: 6 Co-amoxyclavulanic acid versus amoxycillin Outcome: 5 Weight Co- amoxyclavulanic Mean Mean Study or subgroup acid Amoxycillin Difference Weight Difference N Mean(SD) N Mean(SD) IV. M- H.95% CI Jibril 1989 50 16 (5.0 % 1.06.0 % 1.06.10 [ -1.Random. Review: Antibiotics for community-acquired pneumonia in children Comparison: 6 Co-amoxyclavulanic acid versus amoxycillin Outcome: 6 Male sex Co- amoxyclavulanic Study or subgroup acid Amoxycillin Odds Ratio Weight Odds Ratio M. 32 (Amoxycillin) Heterogeneity: not applicable Test for overall effect: Z = 0.5 1 2 5 10 Co-amoxyclavulanic acid Amoxycillin Antibiotics for community-acquired pneumonia in children (Review) 89 Copyright © 2013 The Cochrane Collaboration.26 ] Total (95% CI) 50 50 100. Outcome 5 Weight. 3.Random.31 [ 0.0 % 1.0 % 1. 3.57.2 0.9 (5.26 ] Heterogeneity: not applicable Test for overall effect: Z = 1. Comparison 6 Co-amoxyclavulanic acid versus amoxycillin.28) 100. Analysis 6.Random. 3.1 0. Outcome 6 Male sex.03 ] Total (95% CI) 50 50 100.

1 0.92 ] Total events: 5 (Co-amoxyclavulanic acid). Outcome 8 Side effects.95% n/N n/N CI CI Jibril 1989 2/50 0/50 100.5 1 2 5 10 Co-amoxyclavulanic acid Amoxycillin Analysis 6. Published by John Wiley & Sons. 111. Review: Antibiotics for community-acquired pneumonia in children Comparison: 6 Co-amoxyclavulanic acid versus amoxycillin Outcome: 7 Wheeze present Co- amoxyclavulanic Study or subgroup acid Amoxycillin Odds Ratio Weight Odds Ratio M.24.7. Comparison 6 Co-amoxyclavulanic acid versus amoxycillin. Review: Antibiotics for community-acquired pneumonia in children Comparison: 6 Co-amoxyclavulanic acid versus amoxycillin Outcome: 8 Side effects Co- amoxyclavulanic Study or subgroup acid Amoxycillin Odds Ratio Weight Odds Ratio M.Random.8. Comparison 6 Co-amoxyclavulanic acid versus amoxycillin.Random.24. 0 (Amoxycillin) Heterogeneity: not applicable Test for overall effect: Z = 1. 1.06 (P = 0.95% n/N n/N CI CI Jibril 1989 5/50 8/50 100.2 0.18. 111.29) Test for subgroup differences: Not applicable 0.18.95% H.Random.1 0. Ltd.5 1 2 5 10 Co-amoxyclavulanic acid Amoxycillin Antibiotics for community-acquired pneumonia in children (Review) 90 Copyright © 2013 The Cochrane Collaboration. Outcome 7 Wheeze present. 8 (Amoxycillin) Heterogeneity: not applicable Test for overall effect: Z = 0.21 [ 0. . 1.2 0.95% H.Random.24 ] Total (95% CI) 50 50 100. M- H.24 ] Total events: 2 (Co-amoxyclavulanic acid).92 ] Total (95% CI) 50 50 100.58 [ 0.21 [ 0.0 % 5. Analysis 6.0 % 5.0 % 0. M- H.88 (P = 0.0 % 0.38) Test for subgroup differences: Not applicable 0.58 [ 0.

M- H.18 (P = 0. .1 0. Ltd. Chi2 = 4.29 ] Total events: 642 (Co-trimoxazole).90. 525 (Amoxycillin) Heterogeneity: Tau2 = 0.13).12.95% n/N n/N CI CI Awasthi 2008 43/138 38/135 20.Random.74.5 1 2 5 10 Co-trimoxazole Amoxycillin Antibiotics for community-acquired pneumonia in children (Review) 91 Copyright © 2013 The Cochrane Collaboration.0 % 0. Comparison 7 Co-trimoxazole versus amoxycillin.97 [ 0.03. 1.94 ] CATCHUP 2002 372/734 360/745 48. Outcome 1 Age less than 1 year.35 ] Straus 1998 227/398 127/197 31.04 ] Total (95% CI) 1270 1077 100.10 [ 0.1 % 1.0 % 1.73 [ 0.2 0. Analysis 7.9 % 0.16 [ 0.86) Test for subgroup differences: Not applicable 0. 1. df = 2 (P = 0.1.Random. Published by John Wiley & Sons.95% H.69.51. 1. Review: Antibiotics for community-acquired pneumonia in children Comparison: 7 Co-trimoxazole versus amoxycillin Outcome: 1 Age less than 1 year Study or subgroup Co-trimoxazole Amoxycillin Odds Ratio Weight Odds Ratio M. 1. I2 =51% Test for overall effect: Z = 0.

70 [ 0. Ltd.0% Test for overall effect: Z = 4.18 (P = 0. Outcome 3 Mean Z score for weight.0 % -0. Published by John Wiley & Sons. df = 2 (P = 0.0 % 0. Analysis 7.1 0. Outcome 2 Male sex.Random. 1.Random.59. 1.5 1 2 5 10 Co-trimoxazole Amoxycillin Analysis 7.000029) Test for subgroup differences: Not applicable 0.57.49.02 ] Total (95% CI) 1261 1057 100. Chi2 = 1.83 ] Total events: 679 (Co-trimoxazole). 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 7 Co-trimoxazole versus amoxycillin Outcome: 3 Mean Z score for weight Mean Mean Study or subgroup Co-trimoxazole Amoxycillin Difference Weight Difference N Mean(SD) N Mean(SD) IV. 646 (Amoxycillin) Heterogeneity: Tau2 = 0. 0.27.2.Random.81 ] Straus 1998 246/398 137/197 21.95% H.51 ] CATCHUP 2002 362/734 435/730 66.56 (P = 0. Comparison 7 Co-trimoxazole versus amoxycillin.15 ] Heterogeneity: not applicable Test for overall effect: Z = 0.5 % 0.0 % -0.06 [ -0.95% n/N n/N CI CI Awasthi 2008 71/129 74/130 11.2 0. 0.94 (0) 730 -1.11 (0) Not estimable Straus 1998 398 -0.27.15 ] Total (95% CI) 1139 927 100.45 (1.46). .28) 197 -0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 7 Co-trimoxazole versus amoxycillin Outcome: 2 Male sex Study or subgroup Co-trimoxazole Amoxycillin Odds Ratio Weight Odds Ratio M.95% CI IV.58) Test for subgroup differences: Not applicable -10 -5 0 5 10 Co-trimoxazole Amoxycillin Antibiotics for community-acquired pneumonia in children (Review) 92 Copyright © 2013 The Cochrane Collaboration.54.39 (1.57.93 [ 0. 0. I2 =0.21) 100.Random.06 [ -0.0. M- H. Comparison 7 Co-trimoxazole versus amoxycillin.3.95% CI CATCHUP 2002 741 -0.7 % 0.66 [ 0.8 % 0.71 [ 0.

86) Test for subgroup differences: Not applicable 0.69.Random.0 % 0. Analysis 7.4. Comparison 7 Co-trimoxazole versus amoxycillin. M- H. 98 (Amoxycillin) Heterogeneity: not applicable Test for overall effect: Z = 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 7 Co-trimoxazole versus amoxycillin Outcome: 4 Non-severe pneumonia Study or subgroup Co-trimoxazole Amoxycillin Odds Ratio Weight Odds Ratio M. 1.97 [ 0.1 0.69.Random.5 1 2 5 10 Co-trimoxazole Amoxycillin Antibiotics for community-acquired pneumonia in children (Review) 93 Copyright © 2013 The Cochrane Collaboration.0 % 0.97 [ 0.2 0.37 ] Total events: 195 (Co-trimoxazole).17 (P = 0.37 ] Total (95% CI) 398 197 100.95% n/N n/N CI CI Straus 1998 195/398 98/197 100. Published by John Wiley & Sons. . Outcome 4 Non-severe pneumonia. Ltd.95% H. 1.

46. Review: Antibiotics for community-acquired pneumonia in children Comparison: 7 Co-trimoxazole versus amoxycillin Outcome: 5 Received antibiotics in previous week Study or subgroup Co-trimoxazole Amoxycillin Odds Ratio Weight Odds Ratio M.5 1 2 5 10 Co-trimoxazole Amoxycillin Antibiotics for community-acquired pneumonia in children (Review) 94 Copyright © 2013 The Cochrane Collaboration. 0.13 (P = 0.Random.5. Published by John Wiley & Sons.5 1 2 5 10 Co-trimoxazole Amoxycillin Analysis 7.Random. Comparison 7 Co-trimoxazole versus amoxycillin. .03 [ 0.86) Test for subgroup differences: Not applicable 0. Outcome 6 Severe pneumonia. 1.1 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 7 Co-trimoxazole versus amoxycillin Outcome: 6 Severe pneumonia Study or subgroup Co-trimoxazole Amoxycillin Odds Ratio Weight Odds Ratio M.67 [ 0.45 ] Total events: 203 (Co-trimoxazole).0 % 1.95% n/N n/N CI CI Straus 1998 203/398 99/197 100. M- H.73.1 0.73.95% H. Comparison 7 Co-trimoxazole versus amoxycillin.6.0 % 0.95% n/N n/N CI CI Straus 1998 102/398 67/197 100.Random.17 (P = 0.0 % 0. 1.2 0.033) Test for subgroup differences: Not applicable 0. 99 (Amoxycillin) Heterogeneity: not applicable Test for overall effect: Z = 0.97 ] Total events: 102 (Co-trimoxazole). Analysis 7.46.0 % 1. 67 (Amoxycillin) Heterogeneity: not applicable Test for overall effect: Z = 2. 0.03 [ 0.Random. Outcome 5 Received antibiotics in previous week.45 ] Total (95% CI) 398 197 100.97 ] Total (95% CI) 398 197 100.2 0.67 [ 0. Ltd.95% H. M- H.

2 0. 1.5 1 2 5 10 Co-trimoxazole Amoxycillin Antibiotics for community-acquired pneumonia in children (Review) 95 Copyright © 2013 The Cochrane Collaboration.15 ] Total (95% CI) 948 839 100.21 [ 0.Random.91. Chi2 = 0. M- H. 1.21) Test for subgroup differences: Not applicable 0.93.1 0. 132 (Amoxycillin) Heterogeneity: Tau2 = 0.49. I2 =0. . Ltd.01 ] CATCHUP 2002 139/734 117/725 86. Outcome 7 Failure rate in non-severe pneumonia. Published by John Wiley & Sons.95% n/N n/N CI CI Awasthi 2008 2/19 3/18 1.71).Random.0 % 1.18 [ 0.26 (P = 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 7 Co-trimoxazole versus amoxycillin Outcome: 7 Failure rate in non-severe pneumonia Study or subgroup Co-trimoxazole Amoxycillin Odds Ratio Weight Odds Ratio M.09.0% Test for overall effect: Z = 1.03 [ 0.59 ] Straus 1998 25/195 12/96 11.95% H.7.68.7 % 0. Analysis 7.59 [ 0.0. Comparison 7 Co-trimoxazole versus amoxycillin.6 % 1.7 % 1. 4. 2.51 ] Total events: 166 (Co-trimoxazole). df = 2 (P = 0.

2 0.71 [ 0. 18 (Amoxycillin) Heterogeneity: not applicable Test for overall effect: Z = 1. Comparison 7 Co-trimoxazole versus amoxycillin.2 0. 10 (Amoxycillin) Heterogeneity: not applicable Test for overall effect: Z = 1.96.0 % 2. 3. M- H.1 0.95% n/N n/N CI CI Straus 1998 56/203 18/99 100. Outcome 9 Failure rate radiological positive pneumonia.063) Test for subgroup differences: Not applicable 0.076) Test for subgroup differences: Not applicable 0.94. 4. Review: Antibiotics for community-acquired pneumonia in children Comparison: 7 Co-trimoxazole versus amoxycillin Outcome: 9 Failure rate radiological positive pneumonia Study or subgroup Co-trimoxazole Amoxycillin Odds Ratio Weight Odds Ratio M.78 ] Total events: 35 (Co-trimoxazole).0 % 1.71 [ 0.78 ] Total (95% CI) 102 51 100.95% H.Random. Analysis 7.Random.8.5 1 2 5 10 Co-trimoxazole Amoxycillin Antibiotics for community-acquired pneumonia in children (Review) 96 Copyright © 2013 The Cochrane Collaboration.9.14 [ 0. Published by John Wiley & Sons.96.0 % 2.86 (P = 0. Ltd.Random. Review: Antibiotics for community-acquired pneumonia in children Comparison: 7 Co-trimoxazole versus amoxycillin Outcome: 8 Failure rate severe pneumonia clinical diagnosis Study or subgroup Co-trimoxazole Amoxycillin Odds Ratio Weight Odds Ratio M.1 0.11 ] Total (95% CI) 203 99 100.Random. Comparison 7 Co-trimoxazole versus amoxycillin. Outcome 8 Failure rate severe pneumonia clinical diagnosis.0 % 1. 4. 3.14 [ 0.5 1 2 5 10 Co-trimoxazole Amoxycillin Analysis 7. M- H.94. .95% n/N n/N CI CI Straus 1998 35/102 10/51 100.77 (P = 0.95% H.11 ] Total events: 56 (Co-trimoxazole).

3.1 0.81 (P = 0.72 [ 0. Analysis 7. Outcome 10 Failure rate radiological negative pneumonia.09 ] Total (95% CI) 283 141 100. .Random.96.96. Ltd.Random.0 % 1.070) Test for subgroup differences: Not applicable 0.2 0.10. M- H. Comparison 7 Co-trimoxazole versus amoxycillin. 3. Published by John Wiley & Sons.95% n/N n/N CI CI Straus 1998 54/283 17/141 100.95% H. Review: Antibiotics for community-acquired pneumonia in children Comparison: 7 Co-trimoxazole versus amoxycillin Outcome: 10 Failure rate radiological negative pneumonia Study or subgroup Co-trimoxazole Amoxycillin Odds Ratio Weight Odds Ratio M.0 % 1.5 1 2 5 10 Co-trimoxazole Amoxycillin Antibiotics for community-acquired pneumonia in children (Review) 97 Copyright © 2013 The Cochrane Collaboration.09 ] Total events: 54 (Co-trimoxazole). 17 (Amoxycillin) Heterogeneity: not applicable Test for overall effect: Z = 1.72 [ 0.

df = 1 (P = 0.06. M- H. 2.63 (P = 0.4 % 1.Random.Random. Outcome 12 Lost to follow-up.86 [ 0.0.1 0.70).5 1 2 5 10 Co-trimoxazole Amoxycillin Analysis 7.Random.2 0. M- H.06 ] Total events: 2 (Co-trimoxazole).16 (P = 0.12.0 % 2. I2 =0.11.88 [ 0. Ltd.96 ] Straus 1998 1/398 0/193 50. I2 =0. 0 (Amoxycillin) Heterogeneity: Tau2 = 0. Analysis 7.0% Test for overall effect: Z = 0. 33 (Amoxycillin) Heterogeneity: Tau2 = 0.Random.0% Test for overall effect: Z = 0. 4.71 ] Total (95% CI) 1270 1055 100. 20.95% H.22.0 % 2. Review: Antibiotics for community-acquired pneumonia in children Comparison: 7 Co-trimoxazole versus amoxycillin Outcome: 11 Death rate Study or subgroup Co-trimoxazole Amoxycillin Odds Ratio Weight Odds Ratio M.87) Test for subgroup differences: Not applicable 0.6 % 0.0.39.95% n/N n/N CI CI CATCHUP 2002 1/734 0/725 50. . Review: Antibiotics for community-acquired pneumonia in children Comparison: 7 Co-trimoxazole versus amoxycillin Outcome: 12 Lost to follow-up Study or subgroup Co-trimoxazole Amoxycillin Odds Ratio Weight Odds Ratio M.09.96 [ 0. 72.2 0.00 ] CATCHUP 2002 14/734 16/725 46.71.46 [ 0.0 % 0.50 [ 0.95% n/N n/N CI CI Awasthi 2008 12/138 13/133 35. 36. 1.08 [ 0. Outcome 11 Death rate.0 % 1.1 0.97 [ 0. 1. Published by John Wiley & Sons.01 ] Total (95% CI) 1132 918 100. Comparison 7 Co-trimoxazole versus amoxycillin. Chi2 = 0.0 % 0.57 ] Total events: 38 (Co-trimoxazole). Chi2 = 0.42.78 ] Straus 1998 12/398 4/197 18.95% H. Comparison 7 Co-trimoxazole versus amoxycillin.48.5 1 2 5 10 Co-trimoxazole Amoxycillin Antibiotics for community-acquired pneumonia in children (Review) 98 Copyright © 2013 The Cochrane Collaboration.76).53) Test for subgroup differences: Not applicable 0.12.59. df = 2 (P = 0.

0 % 0.19 (P = 0.5 1 2 5 10 Co-trimoxazole Amoxycillin Antibiotics for community-acquired pneumonia in children (Review) 99 Copyright © 2013 The Cochrane Collaboration. 1.13.76 [ 0. Outcome 13 Wheeze positive.2 0. 47 (Amoxycillin) Heterogeneity: not applicable Test for overall effect: Z = 1.23) Test for subgroup differences: Not applicable 0.76 [ 0. Comparison 7 Co-trimoxazole versus amoxycillin. Ltd. Published by John Wiley & Sons.19 ] Total events: 37 (Co-trimoxazole).Random.95% H. 1.49.49.0 % 0. Analysis 7.19 ] Total (95% CI) 741 730 100. .Random. Review: Antibiotics for community-acquired pneumonia in children Comparison: 7 Co-trimoxazole versus amoxycillin Outcome: 13 Wheeze positive Study or subgroup Co-trimoxazole Amoxycillin Odds Ratio Weight Odds Ratio M. M- H.1 0.95% n/N n/N CI CI CATCHUP 2002 37/741 47/730 100.

08 ] Total (95% CI) 872 860 100.Random.0 % 1.11). . Comparison 7 Co-trimoxazole versus amoxycillin. 1. M- H.95% H.0 % 1. Chi2 = 2. Review: Antibiotics for community-acquired pneumonia in children Comparison: 7 Co-trimoxazole versus amoxycillin Outcome: 15 Change of antibiotics Study or subgroup Co-trimoxazole Amoxycillin Odds Ratio Weight Odds Ratio M.57 [ 0.10 (P = 0.11) Test for subgroup differences: Not applicable 0.95% n/N n/N CI CI Awasthi 2008 125/138 116/135 34.Random.95% n/N n/N CI CI CATCHUP 2002 121/734 98/725 100. df = 1 (P = 0.69 ] Total events: 121 (Co-trimoxazole).15.69 ] Total (95% CI) 734 725 100. Ltd.89 ] Total events: 720 (Co-trimoxazole).Random.2 0.9 % 1.82 [ 0.54.14.95% H. Review: Antibiotics for community-acquired pneumonia in children Comparison: 7 Co-trimoxazole versus amoxycillin Outcome: 14 Cure rate Study or subgroup Co-trimoxazole Amoxycillin Odds Ratio Weight Odds Ratio M. 1.0 % 1.58 (P = 0.63.13. Outcome 14 Cure rate.1 0.5 1 2 5 10 Co-trimoxazole Amoxycillin Analysis 7.1 0.Random. 1. 98 (Amoxycillin) Heterogeneity: not applicable Test for overall effect: Z = 1.03 [ 0.5 1 2 5 10 Co-trimoxazole Amoxycillin Antibiotics for community-acquired pneumonia in children (Review) 100 Copyright © 2013 The Cochrane Collaboration.56. Published by John Wiley & Sons.1 % 0.74.26 [ 0.2 0. Outcome 15 Change of antibiotics. 3.33 ] CATCHUP 2002 595/734 608/725 65. Analysis 7.95.92) Test for subgroup differences: Not applicable 0. Comparison 7 Co-trimoxazole versus amoxycillin. I2 =61% Test for overall effect: Z = 0. 724 (Amoxycillin) Heterogeneity: Tau2 = 0. M- H.95.26 [ 0. 1.

Review: Antibiotics for community-acquired pneumonia in children Comparison: 7 Co-trimoxazole versus amoxycillin Outcome: 16 Failure rates after excluding study by Awasthi 2008 Study or subgroup Co-trimoxazole Amoxycillin Odds Ratio Weight Odds Ratio M.96.086) Test for subgroup differences: Not applicable 0.60 ] Total (95% CI) 349 374 100.53 ] Total events: 164 (Co-trimoxazole). I2 =0.8 % 1.26 [ 0.Random.23.Random.1.18) Test for subgroup differences: Not applicable 0. 1. Ltd.68. Review: Antibiotics for community-acquired pneumonia in children Comparison: 8 Co-trimoxazole versus procaine penicillin Outcome: 1 Age less than 1 year Study or subgroup Co-trimoxazole Procaine penicillin Odds Ratio Weight Odds Ratio M.17.1 % 1.92.95% n/N n/N CI CI Keeley 1990 148/303 134/311 87. Comparison 8 Co-trimoxazole versus procaine penicillin.95% H.16.0% Test for overall effect: Z = 1.75 ] Total events: 164 (Co-trimoxazole).0% Test for overall effect: Z = 1. 150 (Procaine penicillin) Heterogeneity: Tau2 = 0.92.2 % 1.59 ] Straus 1998 25/195 12/96 11.63).93.0 % 1. 1.95% n/N n/N CI CI CATCHUP 2002 139/734 117/725 88.2 0. 1.19 [ 0.57 [ 0. df = 1 (P = 0.68).0.Random.01 0. Outcome 16 Failure rates after excluding study by Awasthi 2008.03 [ 0. 3.9 % 1.95% H.1 0.5 1 2 5 10 Co-trimoxazole Procaine penicillin Antibiotics for community-acquired pneumonia in children (Review) 101 Copyright © 2013 The Cochrane Collaboration. df = 1 (P = 0. Comparison 7 Co-trimoxazole versus amoxycillin.15 ] Total (95% CI) 929 821 100. Analysis 7.30 [ 0.0. Chi2 = 0. Outcome 1 Age less than 1 year. I2 =0. 129 (Amoxycillin) Heterogeneity: Tau2 = 0.71 (P = 0. Chi2 = 0.34 (P = 0.Random. M- H.21 [ 0.49. . 1.0 % 1.1 1 10 100 Co-trimoxazole Amoxycillin Analysis 8. M- H.73 ] Sidal 1994 16/46 16/63 12. 2. Published by John Wiley & Sons.

0 % 0.5 1 2 5 10 Co-trimoxazole Procaine penicillin Antibiotics for community-acquired pneumonia in children (Review) 102 Copyright © 2013 The Cochrane Collaboration. 159 (Procaine penicillin) Heterogeneity: not applicable Test for overall effect: Z = 1. Comparison 8 Co-trimoxazole versus procaine penicillin.95% n/N n/N CI CI Keeley 1990 142/303 159/311 100.06 (P = 0. Published by John Wiley & Sons. Analysis 8.29) Test for subgroup differences: Not applicable 0. 1.0 % 0.Random.61.Random. Ltd. .95% H. 1. Outcome 2 Age 1 to 5 years. Review: Antibiotics for community-acquired pneumonia in children Comparison: 8 Co-trimoxazole versus procaine penicillin Outcome: 2 Age 1 to 5 years Study or subgroup Co-trimoxazole Procaine penicillin Odds Ratio Weight Odds Ratio M.1 0.61.84 [ 0.2 0.16 ] Total events: 142 (Co-trimoxazole).84 [ 0.16 ] Total (95% CI) 303 311 100.2. M- H.

49 ] Total (95% CI) 349 374 100.84 (P = 0. Comparison 8 Co-trimoxazole versus procaine penicillin.95% CI IV.Random.37).4 % 1.20 ] Heterogeneity: Chi2 = 2. Outcome 4 Duration of illness in days. 0.6 % -0. M- H. Published by John Wiley & Sons.0% Test for overall effect: Z = 0. I2 =0.63.4. 3.47 (6) 3.33 ] Sidal 1994 13/46 17/63 43. Analysis 8.46.92) 63 5. I2 =62% Test for overall effect: Z = 0.42) Test for subgroup differences: Not applicable 0.79 [ 0. .40) Test for subgroup differences: Not applicable -10 -5 0 5 10 Co-trimoxazole Procaine penicillin Antibiotics for community-acquired pneumonia in children (Review) 103 Copyright © 2013 The Cochrane Collaboration. Chi2 = 0.24 ] Total (95% CI) 349 374 100.0.20 [ -0.Fixed.5 1 2 5 10 Co-trimoxazole Procaine penicillin Analysis 8. Ltd. 36 (Procaine penicillin) Heterogeneity: Tau2 = 0.50.30.95% CI Keeley 1990 303 3. 2.2 % 1.07 [ 0.4 (2) 311 3.0 % -0.1 0.Random.82. df = 1 (P = 0.39 ] Total events: 25 (Co-trimoxazole).10).4) 96.6 (2.63 [ 0. Comparison 8 Co-trimoxazole versus procaine penicillin.95% H. df = 1 (P = 0. Outcome 3 Age 5 to 12 years.84 (3.37 [ -0.2 0. 1. 1. 0.45.15 ] Sidal 1994 46 6.49.15 [ -0.8 % 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 8 Co-trimoxazole versus procaine penicillin Outcome: 3 Age 5 to 12 years Study or subgroup Co-trimoxazole Procaine penicillin Odds Ratio Weight Odds Ratio M. Review: Antibiotics for community-acquired pneumonia in children Comparison: 8 Co-trimoxazole versus procaine penicillin Outcome: 4 Duration of illness in days Mean Mean Study or subgroup Co-trimoxazole Procaine penicillin Difference Weight Difference N Mean(SD) N Mean(SD) IV.95% n/N n/N CI CI Keeley 1990 12/303 19/311 56.3.0 % 0.55.Fixed.81 (P = 0.

95% H.93 [ 0.5.95% n/N n/N CI CI Keeley 1990 153/303 163/311 100. Comparison 8 Co-trimoxazole versus procaine penicillin.27 ] Total events: 153 (Co-trimoxazole). . 1. Published by John Wiley & Sons. Review: Antibiotics for community-acquired pneumonia in children Comparison: 8 Co-trimoxazole versus procaine penicillin Outcome: 5 Male sex Study or subgroup Co-trimoxazole Procaine penicillin Odds Ratio Weight Odds Ratio M.63) Test for subgroup differences: Not applicable 0.0 % 0.93 [ 0.0 % 0.Random. Outcome 5 Male sex. Analysis 8. 163 (Procaine penicillin) Heterogeneity: not applicable Test for overall effect: Z = 0. 1.48 (P = 0.67. M- H.27 ] Total (95% CI) 303 311 100.2 0. Ltd.Random.67.5 1 2 5 10 Co-trimoxazole Procaine penicillin Antibiotics for community-acquired pneumonia in children (Review) 104 Copyright © 2013 The Cochrane Collaboration.1 0.

2 0.5 1 2 5 10 Co-trimoxazole Procaine penicillin Antibiotics for community-acquired pneumonia in children (Review) 105 Copyright © 2013 The Cochrane Collaboration.88.95% n/N n/N CI CI Keeley 1990 12/303 5/311 100. Analysis 8. 319 (Procaine penicillin) Heterogeneity: Tau2 = 1.58 [ 0.0 % 2.86.6.01).47. M- H. Chi2 = 6.Random.01 ] Total (95% CI) 349 374 100. 2.02.4 % 0. Outcome 6 Cure rate.0 % 1.69 ] Total events: 328 (Co-trimoxazole).57 [ 0.7.50 (P = 0. 6. 9.52 [ 0. Comparison 8 Co-trimoxazole versus procaine penicillin. I2 =85% Test for overall effect: Z = 0.88.62) Test for subgroup differences: Not applicable 0.25 ] Total events: 12 (Co-trimoxazole). Outcome 7 Hospitalisation rate.52 [ 0.0 % 2.Random.26.68 [ 2.2 0. Ltd.16. Published by John Wiley & Sons.1 0.71 ] Sidal 1994 40/46 58/63 45.95% H.Random.72 (P = 0.1 0. .6 % 3. M- H. df = 1 (P = 0. Comparison 8 Co-trimoxazole versus procaine penicillin. 7. 5 (Procaine penicillin) Heterogeneity: not applicable Test for overall effect: Z = 1.5 1 2 5 10 Co-trimoxazole Procaine penicillin Analysis 8.95% H. Review: Antibiotics for community-acquired pneumonia in children Comparison: 8 Co-trimoxazole versus procaine penicillin Outcome: 6 Cure rate Study or subgroup Co-trimoxazole Procaine penicillin Odds Ratio Weight Odds Ratio M.086) Test for subgroup differences: Not applicable 0. 7. Review: Antibiotics for community-acquired pneumonia in children Comparison: 8 Co-trimoxazole versus procaine penicillin Outcome: 7 Hospitalisation rate Study or subgroup Co-trimoxazole Procaine penicillin Odds Ratio Weight Odds Ratio M.Random.25 ] Total (95% CI) 303 311 100.95% n/N n/N CI CI Keeley 1990 288/303 261/311 54.

0 % 0.49) Test for subgroup differences: Not applicable 0.13 ] Total events: 1 (Co-trimoxazole).9. M- H. Review: Antibiotics for community-acquired pneumonia in children Comparison: 8 Co-trimoxazole versus procaine penicillin Outcome: 8 Well at end of follow-up Study or subgroup Co-trimoxazole Procaine penicillin Odds Ratio Weight Odds Ratio M. .09 [ 0. 0 (Procaine penicillin) Heterogeneity: not applicable Test for overall effect: Z = 0.13.51.13 ] Total (95% CI) 303 311 100.90 [ 0.1 0.0 % 3.69 (P = 0.0 % 3. 1.95% n/N n/N CI CI Keeley 1990 275/303 285/311 100. Outcome 9 Death.1 0. Analysis 8.95% H.8. 76. Outcome 8 Well at end of follow-up.5 1 2 5 10 Co-trimoxazole Procaine penicillin Analysis 8.Random.Random.09 [ 0.57 ] Total events: 275 (Co-trimoxazole). Comparison 8 Co-trimoxazole versus procaine penicillin.70) Test for subgroup differences: Not applicable 0. 285 (Procaine penicillin) Heterogeneity: not applicable Test for overall effect: Z = 0.2 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 8 Co-trimoxazole versus procaine penicillin Outcome: 9 Death Study or subgroup Co-trimoxazole Procaine penicillin Odds Ratio Weight Odds Ratio M.95% H. Comparison 8 Co-trimoxazole versus procaine penicillin. Ltd. M- H.51. 76.Random.0 % 0.5 1 2 5 10 Co-trimoxazole Procaine penicillin Antibiotics for community-acquired pneumonia in children (Review) 106 Copyright © 2013 The Cochrane Collaboration.95% n/N n/N CI CI Keeley 1990 1/303 0/311 100.13.90 [ 0.57 ] Total (95% CI) 303 311 100.39 (P = 0. Published by John Wiley & Sons.Random. 1.2 0.

95% CI Campbell 1988 66 22 (0) 68 21. 3 (Procaine penicillin) Heterogeneity: not applicable Test for overall effect: Z = 0.27 ] Total (95% CI) 303 311 100.74 (P = 0.1 0.5 1 2 5 10 Co-trimoxazole Procaine penicillin Analysis 9.10.0 % 1.0 % 1.Random. M- H.95% CI IV. Analysis 8. Published by John Wiley & Sons.Random. Comparison 8 Co-trimoxazole versus procaine penicillin.41. Review: Antibiotics for community-acquired pneumonia in children Comparison: 8 Co-trimoxazole versus procaine penicillin Outcome: 10 Treatment failure Study or subgroup Co-trimoxazole Procaine penicillin Odds Ratio Weight Odds Ratio M. Comparison 9 Co-trimoxazole versus procaine penicillin and ampicillin. Outcome 10 Treatment failure. . Outcome 1 Mean age in months.95% H.46) Test for subgroup differences: Not applicable 0.41.72 [ 0.2 0. 7.Random. Ltd.1.8 (0) Not estimable Total (95% CI) 66 68 Not estimable Heterogeneity: not applicable Test for overall effect: not applicable Test for subgroup differences: Not applicable -10 -5 0 5 10 Co-trimoxazole Procaine penicillin and a Antibiotics for community-acquired pneumonia in children (Review) 107 Copyright © 2013 The Cochrane Collaboration.27 ] Total events: 5 (Co-trimoxazole). Review: Antibiotics for community-acquired pneumonia in children Comparison: 9 Co-trimoxazole versus procaine penicillin and ampicillin Outcome: 1 Mean age in months Procaine penicillin Mean Mean Study or subgroup Co-trimoxazole and a Difference Weight Difference N Mean(SD) N Mean(SD) IV.72 [ 0. 7.Random.95% n/N n/N CI CI Keeley 1990 5/303 3/311 100.

Published by John Wiley & Sons.0 % 0.5 1 2 5 10 Co-trimoxazole Procaine penicillin and a Antibiotics for community-acquired pneumonia in children (Review) 108 Copyright © 2013 The Cochrane Collaboration.Random. 1.3. 2.0 % 1. Review: Antibiotics for community-acquired pneumonia in children Comparison: 9 Co-trimoxazole versus procaine penicillin and ampicillin Outcome: 3 Male sex Procaine penicillin Study or subgroup Co-trimoxazole and a Odds Ratio Weight Odds Ratio M.58 ] Total events: 41 (Co-trimoxazole).64 ] Total events: 20 (Co-trimoxazole).54) Test for subgroup differences: Not applicable 0.95% n/N n/N CI CI Campbell 1988 20/66 24/68 100.2 0.80 [ 0.64 ] Total (95% CI) 66 68 100.39.58 ] Total (95% CI) 66 68 100.80 [ 0.65.0 % 0.5 1 2 5 10 Co-trimoxazole Procaine penicillin and a Analysis 9.Random.39. 1. Comparison 9 Co-trimoxazole versus procaine penicillin and ampicillin.2.1 0.73 (P = 0. 2. Outcome 3 Male sex. 24 (Procaine penicillin and a) Heterogeneity: not applicable Test for overall effect: Z = 0.61 (P = 0. Ltd.46) Test for subgroup differences: Not applicable 0. Analysis 9. M- H. Comparison 9 Co-trimoxazole versus procaine penicillin and ampicillin. M- H. 38 (Procaine penicillin and a) Heterogeneity: not applicable Test for overall effect: Z = 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 9 Co-trimoxazole versus procaine penicillin and ampicillin Outcome: 2 Age less than 1 year Procaine penicillin Study or subgroup Co-trimoxazole and a Odds Ratio Weight Odds Ratio M.0 % 1.Random. Outcome 2 Age less than 1 year.65.95% H.Random.2 0. .1 0.29 [ 0.95% n/N n/N CI CI Campbell 1988 41/66 38/68 100.95% H.29 [ 0.

3.15 [ 0. 61 (Procaine penicillin and a) Heterogeneity: not applicable Test for overall effect: Z = 0.61 ] Total events: 60 (Co-trimoxazole). Ltd.5 1 2 5 10 Co-trimoxazole Procaine penicillin and a Antibiotics for community-acquired pneumonia in children (Review) 109 Copyright © 2013 The Cochrane Collaboration.36.61 ] Total (95% CI) 66 68 100. Published by John Wiley & Sons.95% n/N n/N CI CI Campbell 1988 60/66 61/68 100.0 % 1.4.1 0. Outcome 4 Cure rate. Analysis 9.Random.81) Test for subgroup differences: Not applicable 0. 3.36.15 [ 0. M- H.2 0.Random.24 (P = 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 9 Co-trimoxazole versus procaine penicillin and ampicillin Outcome: 4 Cure rate Procaine penicillin Study or subgroup Co-trimoxazole and a Odds Ratio Weight Odds Ratio M. .95% H.0 % 1. Comparison 9 Co-trimoxazole versus procaine penicillin and ampicillin.

M- H.57 [ 0.72 ] Total (95% CI) 66 68 100.20 [ 0.25 ] Total (95% CI) 66 68 100. Outcome 5 Hospitalisation rate.2 0. Analysis 9.63) Test for subgroup differences: Not applicable 0.95% n/N n/N CI CI Campbell 1988 3/66 2/68 100.2 0.20 [ 0. Published by John Wiley & Sons. Outcome 6 Death rate.72 ] Total events: 3 (Co-trimoxazole).1 0.Random.6.0 % 0.01. 4. .25.03 (P = 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 9 Co-trimoxazole versus procaine penicillin and ampicillin Outcome: 5 Hospitalisation rate Procaine penicillin Study or subgroup Co-trimoxazole and a Odds Ratio Weight Odds Ratio M.25 ] Total events: 0 (Co-trimoxazole). 2 (Procaine penicillin and a) Heterogeneity: not applicable Test for overall effect: Z = 0. Comparison 9 Co-trimoxazole versus procaine penicillin and ampicillin. 9.57 [ 0.25.5 1 2 5 10 Co-trimoxazole Procaine penicillin and a Antibiotics for community-acquired pneumonia in children (Review) 110 Copyright © 2013 The Cochrane Collaboration. Comparison 9 Co-trimoxazole versus procaine penicillin and ampicillin. Review: Antibiotics for community-acquired pneumonia in children Comparison: 9 Co-trimoxazole versus procaine penicillin and ampicillin Outcome: 6 Death rate Procaine penicillin Study or subgroup Co-trimoxazole and a Odds Ratio Weight Odds Ratio M.0 % 1.0 % 1.5. 4.95% n/N n/N CI CI Campbell 1988 0/66 2/68 100. 2 (Procaine penicillin and a) Heterogeneity: not applicable Test for overall effect: Z = 1. 9. Ltd.Random.30) Test for subgroup differences: Not applicable 0.Random.Random.01.49 (P = 0.95% H.1 0.0 % 0.5 1 2 5 10 Co-trimoxazole Procaine penicillin and a Analysis 9.95% H. M- H.

18.0 % 0. Analysis 10.5 1 2 5 10 Cefpodoxime Co-amoxyclavulanic acid Antibiotics for community-acquired pneumonia in children (Review) 111 Copyright © 2013 The Cochrane Collaboration. Outcome 1 Cure rate (response rate) at end of treatment.95% n/N n/N CI CI Klein 1995 179/188 87/90 100.56 (P = 0.58) Test for subgroup differences: Not applicable 0.1 0.Random.95% H.69 [ 0. . 2. Ltd. 87 (Co-amoxyclavulanic acid) Heterogeneity: not applicable Test for overall effect: Z = 0. Published by John Wiley & Sons.60 ] Total (95% CI) 188 90 100. Review: Antibiotics for community-acquired pneumonia in children Comparison: 10 Cefpodoxime versus co-amoxyclavulanic acid Outcome: 1 Cure rate (response rate) at end of treatment Co- amoxyclavulanic Study or subgroup Cefpodoxime acid Odds Ratio Weight Odds Ratio M.69 [ 0. M- H.60 ] Total events: 179 (Cefpodoxime). Comparison 10 Cefpodoxime versus co-amoxyclavulanic acid.2 0.18.Random.0 % 0.1. 2.

Comparison 10 Cefpodoxime versus co-amoxyclavulanic acid. Outcome 3 Adverse effects.46 [ 0. M- H. Comparison 10 Cefpodoxime versus co-amoxyclavulanic acid. 7 (Co-amoxyclavulanic acid) Heterogeneity: not applicable Test for overall effect: Z = 1.2 0.Random.2 (0) Not estimable Total (95% CI) 234 114 Not estimable Heterogeneity: not applicable Test for overall effect: not applicable Test for subgroup differences: Not applicable -10 -5 0 5 10 Cefpodoxime Co-amoxyclavulanic acid Analysis 10.16.Random.0 % 0.35 ] Total (95% CI) 188 90 100.95% CI Klein 1995 234 31.35 ] Total events: 7 (Cefpodoxime).16) Test for subgroup differences: Not applicable 0.46 [ 0.42 (P = 0.Random. Outcome 2 Mean age (months).32 (0) 114 37.95% CI IV.95% n/N n/N CI CI Klein 1995 7/188 7/90 100.0 % 0.95% H. 1.1 0.3. 1. Analysis 10.2. Review: Antibiotics for community-acquired pneumonia in children Comparison: 10 Cefpodoxime versus co-amoxyclavulanic acid Outcome: 2 Mean age (months) Co- amoxyclavulanic Mean Mean Study or subgroup Cefpodoxime acid Difference Weight Difference N Mean(SD) N Mean(SD) IV. Ltd. Published by John Wiley & Sons.5 1 2 5 10 Cefpodoxime Co-amoxyclavulanic acid Antibiotics for community-acquired pneumonia in children (Review) 112 Copyright © 2013 The Cochrane Collaboration.Random. . Review: Antibiotics for community-acquired pneumonia in children Comparison: 10 Cefpodoxime versus co-amoxyclavulanic acid Outcome: 3 Adverse effects Co- amoxyclavulanic Study or subgroup Cefpodoxime acid Odds Ratio Weight Odds Ratio M.16.

Random. Analysis 10. 87 (Co-amoxyclavulanic acid) Heterogeneity: not applicable Test for overall effect: Z = 1.11. Review: Antibiotics for community-acquired pneumonia in children Comparison: 10 Cefpodoxime versus co-amoxyclavulanic acid Outcome: 5 Follow-up Co- amoxyclavulanic Study or subgroup Cefpodoxime acid Odds Ratio Weight Odds Ratio M.Random.1 0.95% H.95% CI IV.95% n/N n/N CI CI Klein 1995 172/188 87/90 100.0 % 0.1 (0) Not estimable Total (95% CI) 234 114 Not estimable Heterogeneity: not applicable Test for overall effect: not applicable Test for subgroup differences: Not applicable -10 -5 0 5 10 Cefpodoxime Co-amoxyclavulanic acid Analysis 10.31 ] Total (95% CI) 188 90 100. Comparison 10 Cefpodoxime versus co-amoxyclavulanic acid. . Outcome 5 Follow-up.5 1 2 5 10 Cefpodoxime Co-amoxyclavulanic acid Antibiotics for community-acquired pneumonia in children (Review) 113 Copyright © 2013 The Cochrane Collaboration.5. Published by John Wiley & Sons.Random.11. Outcome 4 Age in years. Ltd.4. Review: Antibiotics for community-acquired pneumonia in children Comparison: 10 Cefpodoxime versus co-amoxyclavulanic acid Outcome: 4 Age in years Co- amoxyclavulanic Mean Mean Study or subgroup Cefpodoxime acid Difference Weight Difference N Mean(SD) N Mean(SD) IV. Comparison 10 Cefpodoxime versus co-amoxyclavulanic acid.31 ] Total events: 172 (Cefpodoxime).2 0.8 (0) 114 3.95% CI Klein 1995 234 1.Random.0 % 0. 1.37 [ 0.37 [ 0.54 (P = 0.12) Test for subgroup differences: Not applicable 0. 1. M- H.

123 (Penicillin+gentamycin) Heterogeneity: not applicable Test for overall effect: Z = 1.0 % 1.76.88 (P = 0.07 ] Total events: 36 (Chloramphenicol).1. Analysis 11. Review: Antibiotics for community-acquired pneumonia in children Comparison: 11 Chloramphenicol versus penicillin plus gentamicin Outcome: 1 Adverse events Study or subgroup Chloramphenicol Penicillin+gentamycin Odds Ratio Weight Odds Ratio M.0 % 1.Random.26 [ 0.0 % 1.5 1 2 5 10 Chloramphenicol Penicillin+gentamycin Analysis 11.96.38) Test for subgroup differences: Not applicable 0.Random.95% H.5 1 2 5 10 Chloramphenicol Penicillin+gentamycin Antibiotics for community-acquired pneumonia in children (Review) 114 Copyright © 2013 The Cochrane Collaboration. Outcome 2 Death.26 [ 0.25 [ 0. Published by John Wiley & Sons. 1.66 ] Total events: 147 (Chloramphenicol).64 (P = 0.76.25 [ 0. Comparison 11 Chloramphenicol versus penicillin plus gentamicin.96.Random.2. Ltd. 2. Outcome 1 Adverse events. M- H.95% H.2 0.1 0.95% n/N n/N CI CI Duke 2002 147/559 123/557 100.07 ] Total (95% CI) 559 557 100. Comparison 11 Chloramphenicol versus penicillin plus gentamicin. 29 (Penicillin+gentamycin) Heterogeneity: not applicable Test for overall effect: Z = 0.2 0. M- H.10) Test for subgroup differences: Not applicable 0. .1 0.95% n/N n/N CI CI Duke 2002 36/559 29/557 100.0 % 1. 1.66 ] Total (95% CI) 559 557 100. Review: Antibiotics for community-acquired pneumonia in children Comparison: 11 Chloramphenicol versus penicillin plus gentamicin Outcome: 2 Death Study or subgroup Chloramphenicol Penicillin+gentamycin Odds Ratio Weight Odds Ratio M.Random. 2.

Random. M- H. 1.02. Published by John Wiley & Sons.Random.1 0.13 (P = 0. 1. Comparison 11 Chloramphenicol versus penicillin plus gentamicin. 60 (Penicillin+gentamycin) Heterogeneity: not applicable Test for overall effect: Z = 1. Review: Antibiotics for community-acquired pneumonia in children Comparison: 11 Chloramphenicol versus penicillin plus gentamicin Outcome: 4 Readmission before 30 days Study or subgroup Chloramphenicol Penicillin+gentamycin Odds Ratio Weight Odds Ratio M.26) Test for subgroup differences: Not applicable 0. Ltd.95% n/N n/N CI CI Duke 2002 50/559 32/557 100.03 (P = 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 11 Chloramphenicol versus penicillin plus gentamicin Outcome: 3 Change of antibiotics Study or subgroup Chloramphenicol Penicillin+gentamycin Odds Ratio Weight Odds Ratio M.61 [ 1.95% n/N n/N CI CI Duke 2002 49/559 60/557 100.54.0 % 1.02. Outcome 4 Readmission before 30 days.4.80 [ 0. Comparison 11 Chloramphenicol versus penicillin plus gentamicin.18 ] Total (95% CI) 559 557 100. . Analysis 11.Random.5 1 2 5 10 Chloramphenicol Penicillin+gentamycin Antibiotics for community-acquired pneumonia in children (Review) 115 Copyright © 2013 The Cochrane Collaboration. 2.1 0.2 0. 32 (Penicillin+gentamycin) Heterogeneity: not applicable Test for overall effect: Z = 2.0 % 1.61 [ 1.0 % 0.80 [ 0.3.55 ] Total events: 50 (Chloramphenicol).54. 2.Random.95% H.18 ] Total events: 49 (Chloramphenicol).042) Test for subgroup differences: Not applicable 0.55 ] Total (95% CI) 559 557 100.0 % 0.95% H.5 1 2 5 10 Chloramphenicol Penicillin+gentamycin Analysis 11. M- H.2 0. Outcome 3 Change of antibiotics.

Comparison 11 Chloramphenicol versus penicillin plus gentamicin. Analysis 11. 2.Random.09 ] Total (95% CI) 559 557 100.5 1 2 5 10 Chloramphenicol Penicillin+gentamycin Antibiotics for community-acquired pneumonia in children (Review) 116 Copyright © 2013 The Cochrane Collaboration.0 % 1.15 (P = 0. M- H.Random.5. 34 (Penicillin+gentamycin) Heterogeneity: not applicable Test for overall effect: Z = 1.1 0. .31 [ 0.09 ] Total events: 44 (Chloramphenicol).95% H. 2.0 % 1. Ltd.95% n/N n/N CI CI Duke 2002 44/559 34/557 100. Review: Antibiotics for community-acquired pneumonia in children Comparison: 11 Chloramphenicol versus penicillin plus gentamicin Outcome: 5 Absconded Study or subgroup Chloramphenicol Penicillin+gentamycin Odds Ratio Weight Odds Ratio M.31 [ 0.25) Test for subgroup differences: Not applicable 0.83. Outcome 5 Absconded.83.2 0. Published by John Wiley & Sons.

Comparison 11 Chloramphenicol versus penicillin plus gentamicin.Random. Review: Antibiotics for community-acquired pneumonia in children Comparison: 11 Chloramphenicol versus penicillin plus gentamicin Outcome: 6 Age (months) Mean Mean Study or subgroup Chloramphenicol Penicillin+gentamycin Difference Weight Difference N Mean(SD) N Mean(SD) IV. Published by John Wiley & Sons.6.7. Ltd.92 (P = 0.Random. Comparison 11 Chloramphenicol versus penicillin plus gentamicin. 1.21 ] Total events: 67 (Chloramphenicol).95% H.Random. 77 (Penicillin+gentamycin) Heterogeneity: not applicable Test for overall effect: Z = 0. Outcome 6 Age (months).95% CI IV.9 (0) Not estimable Total (95% CI) 559 557 Not estimable Heterogeneity: not applicable Test for overall effect: not applicable Test for subgroup differences: Not applicable -10 -5 0 5 10 Chloramphenicol Penicillin+gentamycin Analysis 11.85 [ 0.0 % 0. Analysis 11.21 ] Total (95% CI) 559 557 100. Review: Antibiotics for community-acquired pneumonia in children Comparison: 11 Chloramphenicol versus penicillin plus gentamicin Outcome: 7 Culture positive Study or subgroup Chloramphenicol Penicillin+gentamycin Odds Ratio Weight Odds Ratio M.95% n/N n/N CI CI Duke 2002 67/559 77/557 100.95% CI Duke 2002 559 6.5 1 2 5 10 Chloramphenicol Penicillin+gentamycin Antibiotics for community-acquired pneumonia in children (Review) 117 Copyright © 2013 The Cochrane Collaboration.0 % 0.36) Test for subgroup differences: Not applicable 0. 1.60. M- H. .85 [ 0. Outcome 7 Culture positive.60.1 (0) 557 5.1 0.2 0.Random.

95% n/N n/N CI CI Duke 2002 324/559 340/557 100.Random. M- H. Outcome 9 Received antibiotics in previous 1 week.75.0 % 0.Random.88 [ 0.69.95% H.29) Test for subgroup differences: Not applicable 0.72) Test for subgroup differences: Not applicable 0.22 ] Total (95% CI) 559 557 100.88 [ 0.35 (P = 0.Random.69.Random.0 % 0.0 % 0.96 [ 0.1 0.2 0.1 0.96 [ 0. 217 (Penicillin+gentamycin) Heterogeneity: not applicable Test for overall effect: Z = 0. M- H. Comparison 11 Chloramphenicol versus penicillin plus gentamicin.0 % 0.2 0. Outcome 8 Male sex.12 ] Total (95% CI) 559 557 100. Published by John Wiley & Sons. 1.95% H.5 1 2 5 10 Chloramphenicol Penicillin+gentamycin Antibiotics for community-acquired pneumonia in children (Review) 118 Copyright © 2013 The Cochrane Collaboration. 1.12 ] Total events: 324 (Chloramphenicol). Review: Antibiotics for community-acquired pneumonia in children Comparison: 11 Chloramphenicol versus penicillin plus gentamicin Outcome: 8 Male sex Study or subgroup Chloramphenicol Penicillin+gentamycin Odds Ratio Weight Odds Ratio M.22 ] Total events: 212 (Chloramphenicol). 340 (Penicillin+gentamycin) Heterogeneity: not applicable Test for overall effect: Z = 1. Analysis 11.5 1 2 5 10 Chloramphenicol Penicillin+gentamycin Analysis 11. Ltd. .9.8.75. 1.05 (P = 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 11 Chloramphenicol versus penicillin plus gentamicin Outcome: 9 Received antibiotics in previous 1 week Study or subgroup Chloramphenicol Penicillin+gentamycin Odds Ratio Weight Odds Ratio M. Comparison 11 Chloramphenicol versus penicillin plus gentamicin. 1.95% n/N n/N CI CI Duke 2002 212/559 217/557 100.

09 ] Total (95% CI) 559 557 100.10 [ -1.Random. Outcome 1 Mean age.95% n/N n/N CI CI Duke 2002 44/559 34/557 100.Random. Comparison 11 Chloramphenicol versus penicillin plus gentamicin.03) 479 8 (8.13) 100.12.25) Test for subgroup differences: Not applicable 0.2 0. Ltd.0 % 1. 0.31 [ 0. Outcome 10 Lost to follow-up.92 ] Heterogeneity: not applicable Test for overall effect: Z = 0.0 % -0. 0.0 % -0. Published by John Wiley & Sons.83.92 ] Total (95% CI) 479 479 100.0 % 1.12.09 ] Total events: 44 (Chloramphenicol). 2. Review: Antibiotics for community-acquired pneumonia in children Comparison: 12 Chloramphenicol with ampicillin and gentamicin Outcome: 1 Mean age Mean Mean Study or subgroup Chloramphenicol Ampicillin+gentamycin Difference Weight Difference N Mean(SD) N Mean(SD) IV.95% CI Asghar 2008 479 7.85) Test for subgroup differences: Not applicable -100 -50 0 50 100 Chloramphenicol Ampicillin+gentamycin Antibiotics for community-acquired pneumonia in children (Review) 119 Copyright © 2013 The Cochrane Collaboration.95% H.9 (8. Analysis 11.15 (P = 0. 2. Comparison 12 Chloramphenicol with ampicillin and gentamicin. M- H. Review: Antibiotics for community-acquired pneumonia in children Comparison: 11 Chloramphenicol versus penicillin plus gentamicin Outcome: 10 Lost to follow-up Study or subgroup Chloramphenicol Penicillin+gentamycin Odds Ratio Weight Odds Ratio M.19 (P = 0.10. 34 (Penicillin+gentamycin) Heterogeneity: not applicable Test for overall effect: Z = 1.1 0.5 1 2 5 10 Chloramphenicol Penicillin+gentamycin Analysis 12. .Random.Random.1.83.95% CI IV.31 [ 0.10 [ -1.

67.19 (P = 0.Random.0 % 0. 303 (Ampicillin+gentamycin) Heterogeneity: not applicable Test for overall effect: Z = 1. Analysis 12. 1.1 1 10 100 Chloramphenicol Ampicillin+gentamycin Analysis 12. Review: Antibiotics for community-acquired pneumonia in children Comparison: 12 Chloramphenicol with ampicillin and gentamicin Outcome: 2 Male sex Study or subgroup Chloramphenicol Ampicillin+gentamycin Odds Ratio Weight Odds Ratio M.95% n/N n/N CI CI Asghar 2008 285/479 303/479 100. Review: Antibiotics for community-acquired pneumonia in children Comparison: 12 Chloramphenicol with ampicillin and gentamicin Outcome: 3 Number received antibiotics in past 7 days Study or subgroup Chloramphenicol Ampicillin+gentamycin Odds Ratio Weight Odds Ratio M.85 [ 0. . 167 (Ampicillin+gentamycin) Heterogeneity: not applicable Test for overall effect: Z = 0.Random.23) Test for subgroup differences: Not applicable 0.0 % 0.2.01 0.66.Random.95% H.0 % 0.85 [ 0.87 [ 0.14 ] Total (95% CI) 477 473 100.11 ] Total events: 285 (Chloramphenicol). Outcome 3 Number received antibiotics in past 7 days.66.98 (P = 0.14 ] Total events: 154 (Chloramphenicol).3.Random. M- H. Comparison 12 Chloramphenicol with ampicillin and gentamicin. 1. Ltd.95% n/N n/N CI CI Asghar 2008 154/477 167/473 100. 1. Outcome 2 Male sex.33) Test for subgroup differences: Not applicable 0.87 [ 0.0 % 0. Comparison 12 Chloramphenicol with ampicillin and gentamicin.11 ] Total (95% CI) 479 479 100.01 0. Published by John Wiley & Sons.95% H. M- H.1 1 10 100 Chloramphenicol Ampicillin+gentamycin Antibiotics for community-acquired pneumonia in children (Review) 120 Copyright © 2013 The Cochrane Collaboration. 1.67.

031) Test for subgroup differences: Not applicable 0.04. Ltd. Outcome 4 Failure rates on day 5.1 1 10 100 Chloramphenicol Ampicillin+gentamycin Antibiotics for community-acquired pneumonia in children (Review) 121 Copyright © 2013 The Cochrane Collaboration. .4. 67 (Ampicillin+gentamycin) Heterogeneity: not applicable Test for overall effect: Z = 2.0 % 1. Review: Antibiotics for community-acquired pneumonia in children Comparison: 12 Chloramphenicol with ampicillin and gentamicin Outcome: 4 Failure rates on day 5 Study or subgroup Chloramphenicol Ampicillin+gentamycin Odds Ratio Weight Odds Ratio M. 2.Random.0 % 1. Published by John Wiley & Sons. 2.04.Random.06 ] Total (95% CI) 479 479 100. Comparison 12 Chloramphenicol with ampicillin and gentamicin. M- H.0 % 1. Analysis 12.19 ] Total (95% CI) 479 479 100.51 [ 1.95% H. Review: Antibiotics for community-acquired pneumonia in children Comparison: 12 Chloramphenicol with ampicillin and gentamicin Outcome: 5 Failure rates on day 10 Study or subgroup Chloramphenicol Ampicillin+gentamycin Odds Ratio Weight Odds Ratio M.Random. Outcome 5 Failure rates on day 10.06 ] Total events: 92 (Chloramphenicol).19 ] Total events: 77 (Chloramphenicol).95% n/N n/N CI CI Asghar 2008 77/479 54/479 100. 2.15 (P = 0.46 [ 1.01 0.04.46 [ 1.0 % 1. 2.51 [ 1.04. 54 (Ampicillin+gentamycin) Heterogeneity: not applicable Test for overall effect: Z = 2.1 1 10 100 Chloramphenicol Ampicillin+gentamycin Analysis 12.16 (P = 0.5.Random. Comparison 12 Chloramphenicol with ampicillin and gentamicin.95% H.01 0. M- H.95% n/N n/N CI CI Asghar 2008 92/479 67/479 100.031) Test for subgroup differences: Not applicable 0.

14 (P = 0.032) Test for subgroup differences: Not applicable 0. Ltd.03.6.98 ] Total events: 103 (Chloramphenicol).43 [ 1.1 1 10 100 Chloramphenicol Ampicillin+gentamycin Antibiotics for community-acquired pneumonia in children (Review) 122 Copyright © 2013 The Cochrane Collaboration.01 0. . M- H.43 [ 1.0 % 1.0 % 1. 1.Random. Review: Antibiotics for community-acquired pneumonia in children Comparison: 12 Chloramphenicol with ampicillin and gentamicin Outcome: 6 Failure rates on day 21 Study or subgroup Chloramphenicol Ampicillin+gentamycin Odds Ratio Weight Odds Ratio M.95% H. Analysis 12.98 ] Total (95% CI) 479 479 100. 1. Outcome 6 Failure rates on day 21. Comparison 12 Chloramphenicol with ampicillin and gentamicin.03.95% n/N n/N CI CI Asghar 2008 103/479 77/479 100.Random. 77 (Ampicillin+gentamycin) Heterogeneity: not applicable Test for overall effect: Z = 2. Published by John Wiley & Sons.

017) Test for subgroup differences: Not applicable 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 12 Chloramphenicol with ampicillin and gentamicin Outcome: 7 Need for change in antibiotics (day 5) Study or subgroup Chloramphenicol Ampicillin+gentamycin Odds Ratio Weight Odds Ratio M. Comparison 12 Chloramphenicol with ampicillin and gentamicin.7.0 % 1.95% n/N n/N CI CI Asghar 2008 45/479 26/479 100. 35 (Ampicillin+gentamycin) Heterogeneity: not applicable Test for overall effect: Z = 2.021) Test for subgroup differences: Not applicable 0.98 ] Total (95% CI) 479 479 100.39 (P = 0.8. . M- H. Outcome 8 Need for change in antibiotics (day 10).10.Random.32 (P = 0. 2. 2. Review: Antibiotics for community-acquired pneumonia in children Comparison: 12 Chloramphenicol with ampicillin and gentamicin Outcome: 8 Need for change in antibiotics (day 10) Study or subgroup Chloramphenicol Ampicillin+gentamycin Odds Ratio Weight Odds Ratio M.95% H.66 ] Total (95% CI) 479 479 100.0 % 1. Ltd.10.1 1 10 100 Chloramphenicol Ampicillin+gentamycin Analysis 12.71 [ 1.01 0.10. Outcome 7 Need for change in antibiotics (day 5).Random.01 0.Random.95% n/N n/N CI CI Asghar 2008 57/479 35/479 100.81 [ 1. 26 (Ampicillin+gentamycin) Heterogeneity: not applicable Test for overall effect: Z = 2.95% H.98 ] Total events: 45 (Chloramphenicol). Analysis 12.81 [ 1.10.0 % 1. 2.Random. Published by John Wiley & Sons.71 [ 1.66 ] Total events: 57 (Chloramphenicol).0 % 1. Comparison 12 Chloramphenicol with ampicillin and gentamicin.1 1 10 100 Chloramphenicol Ampicillin+gentamycin Antibiotics for community-acquired pneumonia in children (Review) 123 Copyright © 2013 The Cochrane Collaboration. 2. M- H.

1 1 10 100 Chloramphenicol Ampicillin+gentamycin Antibiotics for community-acquired pneumonia in children (Review) 124 Copyright © 2013 The Cochrane Collaboration. Review: Antibiotics for community-acquired pneumonia in children Comparison: 12 Chloramphenicol with ampicillin and gentamicin Outcome: 9 Need for change in antibiotics (day 21) Study or subgroup Chloramphenicol Ampicillin+gentamycin Odds Ratio Weight Odds Ratio M. 2.49 ] Total events: 64 (Chloramphenicol).0 % 1.01 0. Ltd.65 [ 1.09. Analysis 12.9. .0 % 1. Outcome 9 Need for change in antibiotics (day 21).09.Random. Comparison 12 Chloramphenicol with ampicillin and gentamicin. M- H.018) Test for subgroup differences: Not applicable 0.49 ] Total (95% CI) 479 479 100.95% H.36 (P = 0. 41 (Ampicillin+gentamycin) Heterogeneity: not applicable Test for overall effect: Z = 2. Published by John Wiley & Sons.65 [ 1. 2.Random.95% n/N n/N CI CI Asghar 2008 64/479 41/479 100.

Random.056) Test for subgroup differences: Not applicable 0.1 0.91 (P = 0.99. Comparison 12 Chloramphenicol with ampicillin and gentamicin.99.57 (P = 0. Comparison 13 Chloramphenicol plus penicillin versus ceftriaxone.0 % 1. Published by John Wiley & Sons.Random. M- H. Outcome 1 Cure rates.65 [ 0.10.36 [ 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 12 Chloramphenicol with ampicillin and gentamicin Outcome: 10 Death rates Study or subgroup Chloramphenicol Ampicillin+gentamycin Odds Ratio Weight Odds Ratio M.5 1 2 5 10 Chloramphenicol+penicilli Ceftriaxone Antibiotics for community-acquired pneumonia in children (Review) 125 Copyright © 2013 The Cochrane Collaboration.36 [ 0.Random.95% H.77 ] Total (95% CI) 479 479 100.2 0. 25 (Ampicillin+gentamycin) Heterogeneity: not applicable Test for overall effect: Z = 1.0 % 1. . 2.01 0.0 % 1.93 ] Total events: 39 (Cholramphenicol+penicilli).1 1 10 100 Chloramphenicol Ampicillin+gentamycin Analysis 13. 41 (Ceftriaxone) Heterogeneity: not applicable Test for overall effect: Z = 0.95% n/N n/N CI CI Asghar 2008 40/479 25/479 100.47. Review: Antibiotics for community-acquired pneumonia in children Comparison: 13 Chloramphenicol plus penicillin versus ceftriaxone Outcome: 1 Cure rates Study or subgroup Cholramphenicol+penicilli Ceftriaxone Odds Ratio Weight Odds Ratio M. Outcome 10 Death rates.77 ] Total events: 40 (Chloramphenicol).Random. Analysis 12.95% H. 3.0 % 1.47.95% n/N n/N CI CI Cetinkaya 2004 39/46 41/51 100. Ltd. M- H.93 ] Total (95% CI) 46 51 100.1.57) Test for subgroup differences: Not applicable 0. 2.65 [ 0. 3.

0 % 0.09 ] Total events: 48 (Chloramphenicol).49 [ 0.0 % 0.5 1 2 5 10 Chloramphenicol Chloramphenicol+penicilli Antibiotics for community-acquired pneumonia in children (Review) 126 Copyright © 2013 The Cochrane Collaboration.95% H.12.2.Random.48.1 0.01 (P = 0.48.2 0.0 % 0.1.2 0. 1.73 [ 0.73 [ 0. 62 (Chloramphenicol+penicilli) Heterogeneity: not applicable Test for overall effect: Z = 1.97 ] Total (95% CI) 377 371 100. M- H. Comparison 14 Chloramphenicol versus chloramphenicol plus penicillin. Analysis 14.49 [ 0.53 (P = 0. Published by John Wiley & Sons.Random.31) Test for subgroup differences: Not applicable 0.1 0. Ltd. 1.Random.12. 1. 1. . Outcome 2 Death rates.95% n/N n/N CI CI Shann 1985 48/377 62/371 100. M- H.95% H.0 % 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 14 Chloramphenicol versus chloramphenicol plus penicillin Outcome: 2 Death rates Study or subgroup Chloramphenicol Chloramphenicol+penicilli Odds Ratio Weight Odds Ratio M.09 ] Total (95% CI) 377 371 100.97 ] Total events: 3 (Chloramphenicol).5 1 2 5 10 Chloramphenicol Chloramphenicol+penicilli Analysis 14.Random.95% n/N n/N CI CI Shann 1985 3/377 6/371 100. Review: Antibiotics for community-acquired pneumonia in children Comparison: 14 Chloramphenicol versus chloramphenicol plus penicillin Outcome: 1 Need for change of antibiotics Study or subgroup Chloramphenicol Chloramphenicol+penicilli Odds Ratio Weight Odds Ratio M. Comparison 14 Chloramphenicol versus chloramphenicol plus penicillin.13) Test for subgroup differences: Not applicable 0. Outcome 1 Need for change of antibiotics. 6 (Chloramphenicol+penicilli) Heterogeneity: not applicable Test for overall effect: Z = 1.

Analysis 14. M- H.21) Test for subgroup differences: Not applicable 0.80.1 0.1.0 % 1. M- H.0 % 0.95% n/N n/N CI CI Deivanayagam 1996 42/52 44/49 100.61 (P = 0.15. Comparison 14 Chloramphenicol versus chloramphenicol plus penicillin.53 ] Total (95% CI) 377 371 100.Random. 44 (Penicillin+chloramphenico) Heterogeneity: not applicable Test for overall effect: Z = 1.2 0.51 ] Total (95% CI) 52 49 100.Random. Outcome 1 Cure rates.51 ] Total events: 42 (Ampicillin).Random. Outcome 3 Lost to follow-up. 1.95% H.5 1 2 5 10 Chloramphenicol Chloramphenicol+penicilli Analysis 15. Comparison 15 Ampicillin alone versus penicillin with chloramphenicol. 1.26 (P = 0.15.3. 1. .53 ] Total events: 105 (Chloramphenicol). Published by John Wiley & Sons.1 0. 1. Review: Antibiotics for community-acquired pneumonia in children Comparison: 14 Chloramphenicol versus chloramphenicol plus penicillin Outcome: 3 Lost to follow-up Study or subgroup Chloramphenicol Chloramphenicol+penicilli Odds Ratio Weight Odds Ratio M.48 [ 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 15 Ampicillin alone versus penicillin with chloramphenicol Outcome: 1 Cure rates Study or subgroup Ampicillin Penicillin+chloramphenico Odds Ratio Weight Odds Ratio M.48 [ 0.54) Test for subgroup differences: Not applicable 0.95% n/N n/N CI CI Shann 1985 105/377 96/371 100.0 % 1.Random. 96 (Chloramphenicol+penicilli) Heterogeneity: not applicable Test for overall effect: Z = 0.2 0.80. Ltd.11 [ 0.0 % 0.11 [ 0.95% H.5 1 2 5 10 Ampicillin Penicillin+chloramphenico Antibiotics for community-acquired pneumonia in children (Review) 127 Copyright © 2013 The Cochrane Collaboration.

2 (9. 1.93 ] Total (95% CI) 52 49 100.54.Random.0 % 0.2.31 (P = 0.95% CI IV.69 [ -5.16 ] Heterogeneity: not applicable Test for overall effect: Z = 0.88 [ 0.69 [ -5.41.Random. Ltd. Published by John Wiley & Sons.41.0 % 0. Outcome 3 Male sex.3.5 1 2 5 10 Ampicillin Penicillin+chloramphenico Antibiotics for community-acquired pneumonia in children (Review) 128 Copyright © 2013 The Cochrane Collaboration.16) 49 15.47) 100.95% H. 26 (Penicillin+chloramphenico) Heterogeneity: not applicable Test for overall effect: Z = 0. Comparison 15 Ampicillin alone versus penicillin with chloramphenicol.39) Test for subgroup differences: Not applicable -10 -5 0 5 10 Ampicillin Penicillin+chloramphenico Analysis 15. Review: Antibiotics for community-acquired pneumonia in children Comparison: 15 Ampicillin alone versus penicillin with chloramphenicol Outcome: 2 Age (months) Mean Mean Study or subgroup Ampicillin Penicillin+chloramphenico Difference Weight Difference N Mean(SD) N Mean(SD) IV. Comparison 15 Ampicillin alone versus penicillin with chloramphenicol.Random.Random.86 (P = 0.95% CI Deivanayagam 1996 52 14.88 [ 0. Outcome 2 Age (months).95% n/N n/N CI CI Deivanayagam 1996 26/52 26/49 100. Analysis 15. 1.16 ] Total (95% CI) 52 49 100. M- H. .1 0.89 (10.0 % -1.2 0.93 ] Total events: 26 (Ampicillin).0 % -1.54.76) Test for subgroup differences: Not applicable 0. 2. 2. Review: Antibiotics for community-acquired pneumonia in children Comparison: 15 Ampicillin alone versus penicillin with chloramphenicol Outcome: 3 Male sex Study or subgroup Ampicillin Penicillin+chloramphenico Odds Ratio Weight Odds Ratio M.

0.95% CI Deivanayagam 1996 52 6. Review: Antibiotics for community-acquired pneumonia in children Comparison: 15 Ampicillin alone versus penicillin with chloramphenicol Outcome: 4 Duration of hospital stay Mean Mean Study or subgroup Ampicillin Penicillin+chloramphenico Difference Weight Difference N Mean(SD) N Mean(SD) IV.5.93 ] Total events: 26 (Ampicillin).88 [ 0.5 1 2 5 10 Ampicillin Penicillin+chloramphenico Antibiotics for community-acquired pneumonia in children (Review) 129 Copyright © 2013 The Cochrane Collaboration.78) 49 6. Comparison 15 Ampicillin alone versus penicillin with chloramphenicol.Random.2 0.1 0.4.0 % 0.Random. Comparison 15 Ampicillin alone versus penicillin with chloramphenicol.19 (2.95% H.5) 100.76) Test for subgroup differences: Not applicable 0. Outcome 5 Grade 2 to 4 malnutrition.19 (P = 0.93 ] Heterogeneity: not applicable Test for overall effect: Z = 0.0 % -0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 15 Ampicillin alone versus penicillin with chloramphenicol Outcome: 5 Grade 2 to 4 malnutrition Study or subgroup Ampicillin Penicillin+chloramphenico Odds Ratio Weight Odds Ratio M.13. 1.95% n/N n/N CI CI Deivanayagam 1996 26/52 26/49 100. Analysis 15.85) Test for subgroup differences: Not applicable -10 -5 0 5 10 Ampicillin Penicillin+chloramphenico Analysis 15. .29 (2. 0. Outcome 4 Duration of hospital stay.41. Ltd.0 % -0.0 % 0.10 [ -1. M- H.93 ] Total (95% CI) 52 49 100.95% CI IV.31 (P = 0.93 ] Total (95% CI) 52 49 100. 26 (Penicillin+chloramphenico) Heterogeneity: not applicable Test for overall effect: Z = 0. Published by John Wiley & Sons. 1.41.Random.10 [ -1.13.Random.88 [ 0.

0 % 0. 1. M- H.08. 113 (Procaine penicillin) Heterogeneity: Tau2 = 0.053) Test for subgroup differences: Not applicable 0.78).Random.2 % 1.80 [ 1.86 ] Sidal 1994 16/42 35/63 67. .2 0. 5. Review: Antibiotics for community-acquired pneumonia in children Comparison: 16 Benzathine penicillin versus procaine penicillin Outcome: 2 Failure rate Study or subgroup Benzathin penicillin Procaine penicillin Odds Ratio Weight Odds Ratio M.11 ] Total events: 21 (Benzathin penicillin). Comparison 16 Benzathine penicillin versus procaine penicillin.53 [ 0.14).0 % 3.80 (P = 0. I2 =0.01 ] Total events: 100 (Benzathin penicillin). 11.95% H.61 [ 0.0.22.09 ] Total (95% CI) 135 146 100.Random. 9 (Procaine penicillin) Heterogeneity: Tau2 = 0.45.0 % 0.5 1 2 5 10 Benzathin penicillin Procaine penicillin Antibiotics for community-acquired pneumonia in children (Review) 130 Copyright © 2013 The Cochrane Collaboration. Analysis 16.Random.1. Ltd.8 % 5. Outcome 1 Cure rate.27. Chi2 = 2. 1. Review: Antibiotics for community-acquired pneumonia in children Comparison: 16 Benzathine penicillin versus procaine penicillin Outcome: 1 Cure rate Study or subgroup Benzathin penicillin Procaine penicillin Odds Ratio Weight Odds Ratio M.94 (P = 0.1 0.19.71 ] Total (95% CI) 135 146 100. M- H.95% H.95% n/N n/N CI CI Camargos 1997 84/93 78/83 33. Published by John Wiley & Sons.60 [ 0.45.90. Comparison 16 Benzathine penicillin versus procaine penicillin.17 [ 0.2 0.1 0. 17. Outcome 2 Failure rate. I2 =55% Test for overall effect: Z = 1. df = 1 (P = 0. 1. Chi2 = 0.0 % 0.90.70 ] Sidal 1994 14/42 5/63 52.49 [ 0. df = 1 (P = 0.5 1 2 5 10 Benzathin penicillin Procaine penicillin Analysis 16.0% Test for overall effect: Z = 1.Random.2.95% n/N n/N CI CI Camargos 1997 7/93 4/83 47.22.072) Test for subgroup differences: Not applicable 0.

M- H. df = 1 (P = 0.9 % 1.95% H.43. 1. .12.09 [ 0. Published by John Wiley & Sons. Outcome 3 Male sex.11 ] Sidal 1994 27/42 41/63 35.5 1 2 5 10 Benzathin penicillin Procaine penicillin Antibiotics for community-acquired pneumonia in children (Review) 131 Copyright © 2013 The Cochrane Collaboration. Review: Antibiotics for community-acquired pneumonia in children Comparison: 16 Benzathine penicillin versus procaine penicillin Outcome: 3 Male sex Study or subgroup Benzathin penicillin Procaine penicillin Odds Ratio Weight Odds Ratio M.19 ] Total (95% CI) 135 146 100. Chi2 = 0.0% Test for overall effect: Z = 0.1 0.67.76 ] Total events: 83 (Benzathin penicillin).73) Test for subgroup differences: Not applicable 0. 88 (Procaine penicillin) Heterogeneity: Tau2 = 0.16 [ 0. Comparison 16 Benzathine penicillin versus procaine penicillin. 2. I2 =0.Random. 2. Analysis 16.1 % 0.64.0 % 1.Random.0. Ltd.34 (P = 0.72).2 0.95% n/N n/N CI CI Camargos 1997 56/93 47/83 64.97 [ 0.3.

Comparison 16 Benzathine penicillin versus procaine penicillin. M- H.47.48 ] Total events: 82 (Benzathin penicillin). 2. Chi2 = 3.Random.08 [ 0.41 ] Total (95% CI) 155 146 100.95% H.24. df = 1 (P = 0.69 [ 0.Random.1 0. 1. Analysis 16.95 ] Sidal 1994 24/62 30/63 47. Published by John Wiley & Sons.34.95% n/N n/N CI CI Camargos 1997 58/93 42/83 52.19 (P = 0. Ltd.5 1 2 5 10 Benzathin penicillin Procaine penicillin Antibiotics for community-acquired pneumonia in children (Review) 132 Copyright © 2013 The Cochrane Collaboration. Outcome 4 Age between 2 to 6 years.89. Review: Antibiotics for community-acquired pneumonia in children Comparison: 16 Benzathine penicillin versus procaine penicillin Outcome: 4 Age between 2 to 6 years Study or subgroup Benzathin penicillin Procaine penicillin Odds Ratio Weight Odds Ratio M. 2.4.4 % 0.2 0.0 % 1. I2 =68% Test for overall effect: Z = 0.08).6 % 1. 72 (Procaine penicillin) Heterogeneity: Tau2 = 0. .62 [ 0.85) Test for subgroup differences: Not applicable 0.17.

95% n/N n/N CI CI Camargos 1997 2/93 1/83 100.52 [ 0.0 % 1.90 ] Total (95% CI) 155 146 100.95% H.95% H.2 0. df = 1 (P = 0.95% n/N n/N CI CI Camargos 1997 35/93 41/83 71.Random.5 % 0.6.80 [ 0.5 % 0.13.0 % 1. Review: Antibiotics for community-acquired pneumonia in children Comparison: 16 Benzathine penicillin versus procaine penicillin Outcome: 5 Age between 7 to 12 years Study or subgroup Benzathin penicillin Procaine penicillin Odds Ratio Weight Odds Ratio M.31). Analysis 16.25 ] Total (95% CI) 93 83 100. Review: Antibiotics for community-acquired pneumonia in children Comparison: 16 Benzathine penicillin versus procaine penicillin Outcome: 6 Lost to follow-up Study or subgroup Benzathin penicillin Procaine penicillin Odds Ratio Weight Odds Ratio M.34. . 0.02.Random. Outcome 5 Age between 7 to 12 years. Outcome 6 Lost to follow-up. Comparison 16 Benzathine penicillin versus procaine penicillin. M- H. 1. Chi2 = 1.2 0. 0. 20.45 (P = 0.25 ] Total events: 2 (Benzathin penicillin).014) Test for subgroup differences: Not applicable 0. 1 (Procaine penicillin) Heterogeneity: not applicable Test for overall effect: Z = 0. 20. 58 (Procaine penicillin) Heterogeneity: Tau2 = 0. I2 =2% Test for overall effect: Z = 2.63) Test for subgroup differences: Not applicable 0. Comparison 16 Benzathine penicillin versus procaine penicillin.1 0.00. M- H.48 (P = 0.5 1 2 5 10 Benzathin penicillin Procaine penicillin Antibiotics for community-acquired pneumonia in children (Review) 133 Copyright © 2013 The Cochrane Collaboration. Published by John Wiley & Sons.16.31.1 0.88 ] Total events: 42 (Benzathin penicillin).0 % 0.Random.80 [ 0.Random.62 [ 0.5 1 2 5 10 Benzathin penicillin Procaine penicillin Analysis 16.5.16. Ltd.34 [ 0.13 ] Sidal 1994 7/62 17/63 28.

46) Test for subgroup differences: Not applicable 0.61 [ 0. 4 (Procaine penicillin) Heterogeneity: not applicable Test for overall effect: Z = 0.70 ] Total (95% CI) 93 83 100. Comparison 16 Benzathine penicillin versus procaine penicillin. 5.70 ] Total events: 7 (Benzathin penicillin).Random. 5.95% n/N n/N CI CI Camargos 1997 7/93 4/83 100.74 (P = 0. .1 1 10 100 Benzathin penicillin Procaine penicillin Antibiotics for community-acquired pneumonia in children (Review) 134 Copyright © 2013 The Cochrane Collaboration. Review: Antibiotics for community-acquired pneumonia in children Comparison: 16 Benzathine penicillin versus procaine penicillin Outcome: 7 Failure rates in radiographically confirmed pneumonia Study or subgroup Benzathin penicillin Procaine penicillin Odds Ratio Weight Odds Ratio M. Ltd.Random. Analysis 16. Outcome 7 Failure rates in radiographically confirmed pneumonia. Published by John Wiley & Sons.45.61 [ 0.0 % 1.7.45.95% H.0 % 1.01 0. M- H.

1.2 0.90 [ 0. M- H. Review: Antibiotics for community-acquired pneumonia in children Comparison: 17 Amoxycillin versus penicillin Outcome: 2 Age less than 1 year Study or subgroup Amoxycillin Penicillin Odds Ratio Weight Odds Ratio M.0 % 0. 513 (Penicillin) Heterogeneity: not applicable Test for overall effect: Z = 0.1 0.95% H.95% n/N n/N CI CI Addo-Yobo 2004 201/743 217/743 100. pneumoniae.0 % 0.87.1. Comparison 17 Amoxycillin versus penicillin.06 [ 0.Random.72. Analysis 17.29 ] Total (95% CI) 857 845 100. pneumoniae Study or subgroup Amoxycillin Penicillin Odds Ratio Weight Odds Ratio M.72.95% H.29 ] Total events: 532 (Amoxycillin).Random. Outcome 2 Age less than 1 year.36) Test for subgroup differences: Not applicable 0.06 [ 0.5 1 2 5 10 Amoxycilin Penicillin Analysis 17. M- H.2.0 % 1.13 ] Total (95% CI) 743 743 100. 1.56) Test for subgroup differences: Not applicable 0.92 (P = 0.58 (P = 0.Random. . Comparison 17 Amoxycillin versus penicillin.90 [ 0. Outcome 1 Nasopharyngeal aspirates for S.1 0.95% n/N n/N CI CI Addo-Yobo 2004 532/857 513/845 100.0 % 1. 1.13 ] Total events: 201 (Amoxycillin).Random. 217 (Penicillin) Heterogeneity: not applicable Test for overall effect: Z = 0.2 0.87. Review: Antibiotics for community-acquired pneumonia in children Comparison: 17 Amoxycillin versus penicillin Outcome: 1 Nasopharyngeal aspirates for S.5 1 2 5 10 Amoxycillin Penicillin Antibiotics for community-acquired pneumonia in children (Review) 135 Copyright © 2013 The Cochrane Collaboration. Ltd. 1. Published by John Wiley & Sons.

70. Ltd. 1.95% n/N n/N CI CI Addo-Yobo 2004 124/845 133/841 100. M- H.25 ] Total events: 590 (Amoxycillin).8 % 1.87. Chi2 = 0.19 ] Total (95% CI) 845 841 100.Random.86. Outcome 3 Male sex. 575 (Penicillin) Heterogeneity: Tau2 = 0.71 ] Total (95% CI) 957 948 100. Review: Antibiotics for community-acquired pneumonia in children Comparison: 17 Amoxycillin versus penicillin Outcome: 4 Weight below 2 Z score (indicating severe malnutrition) Study or subgroup Amoxycillin Penicillin Odds Ratio Weight Odds Ratio M.5 1 2 5 10 Amoxycillin Penicillin Antibiotics for community-acquired pneumonia in children (Review) 136 Copyright © 2013 The Cochrane Collaboration. Comparison 17 Amoxycillin versus penicillin. Published by John Wiley & Sons.04 [ 0.65 (P = 0.43 (P = 0. df = 1 (P = 0.0 % 1.57. 1. Review: Antibiotics for community-acquired pneumonia in children Comparison: 17 Amoxycillin versus penicillin Outcome: 3 Male sex Study or subgroup Amoxycillin Penicillin Odds Ratio Weight Odds Ratio M. 1. Comparison 17 Amoxycillin versus penicillin.Random.95% H.0 % 0.95% n/N n/N CI CI Addo-Yobo 2004 537/857 520/845 88.28 ] Atkinson 2007 53/100 55/103 11. I2 =0.0 % 0.05.Random.0% Test for overall effect: Z = 0.67) Test for subgroup differences: Not applicable 0.Random.0.92 [ 0. 1. Outcome 4 Weight below 2 Z score (indicating severe malnutrition).83). 133 (Penicillin) Heterogeneity: not applicable Test for overall effect: Z = 0.1 0.70.98 [ 0.5 1 2 5 10 Amoxycillin Penicillin Analysis 17. .3.2 0.2 % 0.19 ] Total events: 124 (Amoxycillin).95% H. M- H.1 0.52) Test for subgroup differences: Not applicable 0.05 [ 0. Analysis 17. 1.2 0.4.92 [ 0.

12 [ 0.Random. . 1.92.95% n/N n/N CI CI Addo-Yobo 2004 561/857 531/845 100.13 (P = 0.5 1 2 5 10 Amoxycillin Penicillin Antibiotics for community-acquired pneumonia in children (Review) 137 Copyright © 2013 The Cochrane Collaboration.5. 531 (Penicillin) Heterogeneity: not applicable Test for overall effect: Z = 1. Review: Antibiotics for community-acquired pneumonia in children Comparison: 17 Amoxycillin versus penicillin Outcome: 5 Breast fed Study or subgroup Amoxycillin Penicillin Odds Ratio Weight Odds Ratio M. M- H. Outcome 5 Breast fed.26) Test for subgroup differences: Not applicable 0. Ltd. Comparison 17 Amoxycillin versus penicillin.0 % 1.37 ] Total (95% CI) 857 845 100.Random.12 [ 0.2 0. Published by John Wiley & Sons.0 % 1. Analysis 17.1 0.92.95% H. 1.37 ] Total events: 561 (Amoxycillin).

38 ] Total events: 93 (Amoxycillin). Outcome 6 Received antibiotics in last 7 days.01. Analysis 17.69. M- H.6.23 (P = 0.2 0.1 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 17 Amoxycillin versus penicillin Outcome: 7 Failure rate at 48 hours Study or subgroup Amoxycillin Penicillin Odds Ratio Weight Odds Ratio M.2 0.0 % 1.95% H.15 (P = 0.82) Test for subgroup differences: Not applicable 0.95% H.24 ] Atkinson 2007 18/100 14/103 19. Comparison 17 Amoxycillin versus penicillin. 1.95% n/N n/N CI CI Addo-Yobo 2004 167/857 161/845 100.Random.89 [ 0.81. Ltd. 2.7.64. .88) Test for subgroup differences: Not applicable 0. 1. 96 (Penicillin) Heterogeneity: Tau2 = 0.5 1 2 5 10 Amoxycillin Penicillin Analysis 17.31 ] Total events: 167 (Amoxycillin).1 0. df = 1 (P = 0.40 [ 0.81.5 1 2 5 10 Amoxycillin Penicillin Antibiotics for community-acquired pneumonia in children (Review) 138 Copyright © 2013 The Cochrane Collaboration.0 % 0. 1.29). 1. 161 (Penicillin) Heterogeneity: not applicable Test for overall effect: Z = 0.03 [ 0.97 [ 0.95% n/N n/N CI CI Addo-Yobo 2004 75/857 82/845 80. Chi2 = 1.0 % 1. Outcome 7 Failure rate at 48 hours.03 [ 0.65.6 % 0.12.98 ] Total (95% CI) 957 948 100.Random. I2 =11% Test for overall effect: Z = 0. Published by John Wiley & Sons.31 ] Total (95% CI) 857 845 100.Random.4 % 1.Random. Review: Antibiotics for community-acquired pneumonia in children Comparison: 17 Amoxycillin versus penicillin Outcome: 6 Received antibiotics in last 7 days Study or subgroup Amoxycillin Penicillin Odds Ratio Weight Odds Ratio M. M- H. Comparison 17 Amoxycillin versus penicillin.

39 ] Total (95% CI) 960 945 100.40 (P = 0.95% n/N n/N CI CI Addo-Yobo 2004 185/857 187/845 80. df = 1 (P = 0.84.59.77.95% H.5 % 2. Review: Antibiotics for community-acquired pneumonia in children Comparison: 17 Amoxycillin versus penicillin Outcome: 8 Failure rate on day 5 Study or subgroup Amoxycillin Penicillin Odds Ratio Weight Odds Ratio M.Random. . 1. Ltd.30 ] Total events: 192 (Amoxycillin).04 [ 0.58.5 1 2 5 10 Amoxycillin Penicillin Antibiotics for community-acquired pneumonia in children (Review) 139 Copyright © 2013 The Cochrane Collaboration. 190 (Penicillin) Heterogeneity: Tau2 = 0.0 % 1. 2.69) Test for subgroup differences: Not applicable 0. Comparison 17 Amoxycillin versus penicillin.21).15 [ 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 17 Amoxycillin versus penicillin Outcome: 9 Failure rate on day 14 Study or subgroup Amoxycillin Penicillin Odds Ratio Weight Odds Ratio M. I2 =35% Test for overall effect: Z = 0.36 [ 0.84.0 % 1.97 [ 0.005 0. 221 (Penicillin) Heterogeneity: not applicable Test for overall effect: Z = 0.04 [ 0. Chi2 = 1.5 % 0.14. Comparison 17 Amoxycillin versus penicillin.Random. 9.22 ] Atkinson 2007 7/103 3/100 19.95% H.37 (P = 0.1 1 10 200 Amoxycillin Penicillin Analysis 17.95% n/N n/N CI CI Addo-Yobo 2004 231/857 221/845 100. Outcome 8 Failure rate on day 5. Published by John Wiley & Sons.9.2 0.71) Test for subgroup differences: Not applicable 0.0 % 1.1 0.55.Random.29 ] Total events: 231 (Amoxycillin). M- H. 1.29 ] Total (95% CI) 857 845 100. Outcome 9 Failure rate on day 14.Random. Analysis 17. M- H.8. 1.

0 % 0. .00 [ 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 17 Amoxycillin versus penicillin Outcome: 11 Nasopharyngeal H.85 (P = 0. 145 (Penicillin) Heterogeneity: not applicable Test for overall effect: Z = 0.00 [ 0. 1.95% n/N n/N CI CI Addo-Yobo 2004 0/845 7/857 100. Analysis 17.10.29 ] Total (95% CI) 743 739 100. influenzae Study or subgroup Amoxycillin Penicillin Odds Ratio Weight Odds Ratio n/N n/N M-H.29 ] Total events: 146 (Amoxycillin).1 0.Random. 1.0 % 0.95% CI M-H. Published by John Wiley & Sons.01 0.07 [ 0. influenzae. Comparison 17 Amoxycillin versus penicillin.07 [ 0.0 % 1. Outcome 11 Nasopharyngeal H. Review: Antibiotics for community-acquired pneumonia in children Comparison: 17 Amoxycillin versus penicillin Outcome: 10 Death rates Study or subgroup Amoxycillin Penicillin Odds Ratio Weight Odds Ratio M. Outcome 10 Death rates.Fixed.01 (P = 0. Comparison 17 Amoxycillin versus penicillin.0 % 1.00. M- H.Fixed.99) Test for subgroup differences: Not applicable 0.11.Random.1 1 10 100 1000 Amoxycillin Penicillin Analysis 17.5 1 2 5 10 Amoxycillin Penicillin Antibiotics for community-acquired pneumonia in children (Review) 140 Copyright © 2013 The Cochrane Collaboration.95% H.064) Test for subgroup differences: Not applicable 0. 7 (Penicillin) Heterogeneity: not applicable Test for overall effect: Z = 1.2 0.00. Ltd.95% CI Addo-Yobo 2004 146/743 145/739 100. 1.78.18 ] Total events: 0 (Amoxycillin).78.18 ] Atkinson 2007 0/100 0/103 Not estimable Total (95% CI) 945 960 100.001 0. 1.

9 % 0. Analysis 17.0.0 % 1. Published by John Wiley & Sons.95% CI IV. Outcome 12 Respiratory syncytial virus (RSV) in nasopharyngeal swabs.95% CI Atkinson 2007 100 2.74) Test for subgroup differences: Not applicable 0. df = 1 (P = 0.95% H.2 0.33 (P = 0.21. Ltd. Review: Antibiotics for community-acquired pneumonia in children Comparison: 17 Amoxycillin versus penicillin Outcome: 13 Mean age Mean Mean Study or subgroup Amoxycillin Penicillin Difference Weight Difference N Mean(SD) N Mean(SD) IV.Random. 1. 3. 188 (Penicillin) Heterogeneity: Tau2 = 0.32 ] Atkinson 2007 4/100 5/103 2.12. Review: Antibiotics for community-acquired pneumonia in children Comparison: 17 Amoxycillin versus penicillin Outcome: 12 Respiratory syncytial virus (RSV) in nasopharyngeal swabs Study or subgroup Amoxycillin Penicillin Odds Ratio Weight Odds Ratio M.Random.13. 1.13.82 [ 0.0% Test for overall effect: Z = 0. Comparison 17 Amoxycillin versus penicillin. Outcome 13 Mean age.5 (0) Not estimable Total (95% CI) 100 103 Not estimable Heterogeneity: not applicable Test for overall effect: not applicable Test for subgroup differences: Not applicable -100 -50 0 50 100 Amoxycillin Penicillin Antibiotics for community-acquired pneumonia in children (Review) 141 Copyright © 2013 The Cochrane Collaboration. I2 =0.05 [ 0.04 [ 0.83.5 1 2 5 10 Amoxycillin Penicillin Analysis 17. . Comparison 17 Amoxycillin versus penicillin.72).1 % 1.31 ] Total events: 196 (Amoxycillin). M- H.1 0.Random.13 ] Total (95% CI) 869 862 100. Chi2 = 0.4 (0) 103 2.83.95% n/N n/N CI CI Addo-Yobo 2004 192/769 183/759 97.Random.

0 % 0. Ltd.Random. Outcome 14 Blood culture positive for S. M- H.1 1 10 100 Amoxycillin Penicillin Antibiotics for community-acquired pneumonia in children (Review) 142 Copyright © 2013 The Cochrane Collaboration. 3 (Penicillin) Heterogeneity: not applicable Test for overall effect: Z = 0. pneumoniae.94 (P = 0.01 0. 3.95% n/N n/N CI CI Atkinson 2007 7/103 3/100 100.59. 3 (Penicillin) Heterogeneity: not applicable Test for overall effect: Z = 1.39 ] Total (95% CI) 103 100 100. 3. 9.22) Test for subgroup differences: Not applicable 0.36 [ 0.1 1 10 100 Amoxycillin Penicillin Analysis 17.Random.0 % 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 17 Amoxycillin versus penicillin Outcome: 15 Failure rate on day 5 in radiographically confirmed pneumonia Study or subgroup Amoxycillin Penicillin Odds Ratio Weight Odds Ratio M. .22 (P = 0.36 [ 0.35) Test for subgroup differences: Not applicable 0. Comparison 17 Amoxycillin versus penicillin.34 [ 0. pneumoniae Study or subgroup Amoxycillin Penicillin Odds Ratio Weight Odds Ratio M. Analysis 17.29 ] Total events: 1 (Amoxycillin).95% n/N n/N CI CI Atkinson 2007 1/100 3/103 100.Random.14.59.15.03.01 0.0 % 2.Random.95% H. 9.29 ] Total (95% CI) 100 103 100.03. Published by John Wiley & Sons. Review: Antibiotics for community-acquired pneumonia in children Comparison: 17 Amoxycillin versus penicillin Outcome: 14 Blood culture positive for S.95% H. M- H.0 % 2.34 [ 0. Comparison 17 Amoxycillin versus penicillin.39 ] Total events: 7 (Amoxycillin). Outcome 15 Failure rate on day 5 in radiographically confirmed pneumonia.

1.1 1 10 100 Amoxycillin Ampicillin Antibiotics for community-acquired pneumonia in children (Review) 143 Copyright © 2013 The Cochrane Collaboration.62 (P = 0.Random. Outcome 1 Age below one year. Analysis 18. Published by John Wiley & Sons.79. M- H.94 [ 0. 1. .13 ] Total (95% CI) 1025 1012 100.0 % 0.0 % 0.95% n/N n/N CI CI Hazir 2008 653/1025 658/1012 100.13 ] Total events: 653 (Amoxycillin).Random.54) Test for subgroup differences: Not applicable 0.95% H.79.1. Review: Antibiotics for community-acquired pneumonia in children Comparison: 18 Amoxycillin with IV ampicillin Outcome: 1 Age below one year Study or subgroup Amoxycillin Ampicillin Odds Ratio Weight Odds Ratio M. Comparison 18 Amoxycillin with IV ampicillin. Ltd.94 [ 0. 658 (Ampicillin) Heterogeneity: not applicable Test for overall effect: Z = 0.01 0.

Outcome 2 Male sex.91 (P = 0. M- H. 1. Outcome 3 Wheezing in infants.1 1 10 100 Amoxycillin Ampicillin Analysis 18.0 % 1.95% n/N n/N CI CI Hazir 2008 630/1025 602/1012 100.36) Test for subgroup differences: Not applicable 0.95% H. M- H.30 ] Total (95% CI) 1025 1012 100.91.1 1 10 100 Amoxycillin Ampicillin Antibiotics for community-acquired pneumonia in children (Review) 144 Copyright © 2013 The Cochrane Collaboration. 1. 602 (Ampicillin) Heterogeneity: not applicable Test for overall effect: Z = 0.0 % 1.37 ] Total events: 535 (Amoxycillin).0 % 1.Random.01 0.78.37 ] Total (95% CI) 653 658 100.03 [ 0.Random. Ltd.22 (P = 0. 536 (Ampicillin) Heterogeneity: not applicable Test for overall effect: Z = 0. Comparison 18 Amoxycillin with IV ampicillin.91.09 [ 0.2.3.03 [ 0.0 % 1. Published by John Wiley & Sons.95% n/N n/N CI CI Hazir 2008 535/653 536/658 100.83) Test for subgroup differences: Not applicable 0. .78. Review: Antibiotics for community-acquired pneumonia in children Comparison: 18 Amoxycillin with IV ampicillin Outcome: 3 Wheezing in infants Study or subgroup Amoxycillin Ampicillin Odds Ratio Weight Odds Ratio M.01 0.Random.Random. Review: Antibiotics for community-acquired pneumonia in children Comparison: 18 Amoxycillin with IV ampicillin Outcome: 2 Male sex Study or subgroup Amoxycillin Ampicillin Odds Ratio Weight Odds Ratio M. Analysis 18.30 ] Total events: 630 (Amoxycillin). Comparison 18 Amoxycillin with IV ampicillin. 1.95% H. 1.09 [ 0.

Analysis 18.4. Comparison 18 Amoxycillin with IV ampicillin, Outcome 4 Wheezing in age group one to five
years.

Review: Antibiotics for community-acquired pneumonia in children

Comparison: 18 Amoxycillin with IV ampicillin

Outcome: 4 Wheezing in age group one to five years

Study or subgroup Amoxycillin Ampicillin Odds Ratio Weight Odds Ratio
M- M-
H,Random,95% H,Random,95%
n/N n/N CI CI
Hazir 2008 232/372 242/354 100.0 % 0.77 [ 0.56, 1.04 ]

Total (95% CI) 372 354 100.0 % 0.77 [ 0.56, 1.04 ]
Total events: 232 (Amoxycillin), 242 (Ampicillin)
Heterogeneity: not applicable
Test for overall effect: Z = 1.69 (P = 0.090)
Test for subgroup differences: Not applicable

0.01 0.1 1 10 100
Amoxycillin Ampicillin

Analysis 18.5. Comparison 18 Amoxycillin with IV ampicillin, Outcome 5 Failure rates.

Review: Antibiotics for community-acquired pneumonia in children

Comparison: 18 Amoxycillin with IV ampicillin

Outcome: 5 Failure rates

Study or subgroup Amoxycillin Ampicillin Odds Ratio Weight Odds Ratio
M- M-
H,Random,95% H,Random,95%
n/N n/N CI CI
Hazir 2008 77/1025 87/1012 100.0 % 0.86 [ 0.63, 1.19 ]

Total (95% CI) 1025 1012 100.0 % 0.86 [ 0.63, 1.19 ]
Total events: 77 (Amoxycillin), 87 (Ampicillin)
Heterogeneity: not applicable
Test for overall effect: Z = 0.90 (P = 0.37)
Test for subgroup differences: Not applicable

0.01 0.1 1 10 100
Amoxycillin Ampicillin

Antibiotics for community-acquired pneumonia in children (Review) 145
Copyright © 2013 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Analysis 18.6. Comparison 18 Amoxycillin with IV ampicillin, Outcome 6 Relapse rates.

Review: Antibiotics for community-acquired pneumonia in children

Comparison: 18 Amoxycillin with IV ampicillin

Outcome: 6 Relapse rates

Study or subgroup Amoxycillin Ampicillin Odds Ratio Weight Odds Ratio
M- M-
H,Random,95% H,Random,95%
n/N n/N CI CI
Hazir 2008 25/948 31/925 100.0 % 0.78 [ 0.46, 1.33 ]

Total (95% CI) 948 925 100.0 % 0.78 [ 0.46, 1.33 ]
Total events: 25 (Amoxycillin), 31 (Ampicillin)
Heterogeneity: not applicable
Test for overall effect: Z = 0.91 (P = 0.37)
Test for subgroup differences: Not applicable

0.01 0.1 1 10 100
Amoxycillin Ampicillin

Analysis 18.7. Comparison 18 Amoxycillin with IV ampicillin, Outcome 7 Death rates.

Review: Antibiotics for community-acquired pneumonia in children

Comparison: 18 Amoxycillin with IV ampicillin

Outcome: 7 Death rates

Study or subgroup Amoxycillin Ampicillin Odds Ratio Weight Odds Ratio
M- M-
H,Random,95% H,Random,95%
n/N n/N CI CI
Hazir 2008 1/1025 4/1012 100.0 % 0.25 [ 0.03, 2.21 ]

Total (95% CI) 1025 1012 100.0 % 0.25 [ 0.03, 2.21 ]
Total events: 1 (Amoxycillin), 4 (Ampicillin)
Heterogeneity: not applicable
Test for overall effect: Z = 1.25 (P = 0.21)
Test for subgroup differences: Not applicable

0.01 0.1 1 10 100
Amoxycillin Ampicillin

Antibiotics for community-acquired pneumonia in children (Review) 146
Copyright © 2013 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Analysis 18.8. Comparison 18 Amoxycillin with IV ampicillin, Outcome 8 Lost to follow-up.

Review: Antibiotics for community-acquired pneumonia in children

Comparison: 18 Amoxycillin with IV ampicillin

Outcome: 8 Lost to follow-up

Study or subgroup Amoxycillin Ampicillin Odds Ratio Weight Odds Ratio
M- M-
H,Random,95% H,Random,95%
n/N n/N CI CI
Hazir 2008 6/1025 13/1012 100.0 % 0.45 [ 0.17, 1.20 ]

Total (95% CI) 1025 1012 100.0 % 0.45 [ 0.17, 1.20 ]
Total events: 6 (Amoxycillin), 13 (Ampicillin)
Heterogeneity: not applicable
Test for overall effect: Z = 1.60 (P = 0.11)
Test for subgroup differences: Not applicable

0.01 0.1 1 10 100
Amoxycillin Ampicillin

Analysis 18.9. Comparison 18 Amoxycillin with IV ampicillin, Outcome 9 Protocol violation.

Review: Antibiotics for community-acquired pneumonia in children

Comparison: 18 Amoxycillin with IV ampicillin

Outcome: 9 Protocol violation

Study or subgroup Amoxycillin Ampicillin Odds Ratio Weight Odds Ratio
M- M-
H,Random,95% H,Random,95%
n/N n/N CI CI
Hazir 2008 13/1025 14/1012 100.0 % 0.92 [ 0.43, 1.96 ]

Total (95% CI) 1025 1012 100.0 % 0.92 [ 0.43, 1.96 ]
Total events: 13 (Amoxycillin), 14 (Ampicillin)
Heterogeneity: not applicable
Test for overall effect: Z = 0.23 (P = 0.82)
Test for subgroup differences: Not applicable

0.01 0.1 1 10 100
Amoxycillin Ampicillin

Antibiotics for community-acquired pneumonia in children (Review) 147
Copyright © 2013 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Analysis 19.1. Comparison 19 Amoxycillin with cefuroxime, Outcome 1 Mean age in months.

Review: Antibiotics for community-acquired pneumonia in children

Comparison: 19 Amoxycillin with cefuroxime

Outcome: 1 Mean age in months

Mean Mean
Study or subgroup Amoxycillin Cefuroxime Difference Weight Difference
N Mean(SD) N Mean(SD) IV,Random,95% CI IV,Random,95% CI

Aurangzeb 2003 43 12.79 (16.47) 41 8.32 (10.72) 100.0 % 4.47 [ -1.45, 10.39 ]

Total (95% CI) 43 41 100.0 % 4.47 [ -1.45, 10.39 ]
Heterogeneity: not applicable
Test for overall effect: Z = 1.48 (P = 0.14)
Test for subgroup differences: Not applicable

-100 -50 0 50 100
Amoxycillin Cefuroxime

Analysis 19.2. Comparison 19 Amoxycillin with cefuroxime, Outcome 2 Male sex.

Review: Antibiotics for community-acquired pneumonia in children

Comparison: 19 Amoxycillin with cefuroxime

Outcome: 2 Male sex

Study or subgroup Amoxycillin Cefuroxime Odds Ratio Weight Odds Ratio
M- M-
H,Random,95% H,Random,95%
n/N n/N CI CI
Aurangzeb 2003 35/43 41/42 100.0 % 0.11 [ 0.01, 0.90 ]

Total (95% CI) 43 42 100.0 % 0.11 [ 0.01, 0.90 ]
Total events: 35 (Amoxycillin), 41 (Cefuroxime)
Heterogeneity: not applicable
Test for overall effect: Z = 2.06 (P = 0.039)
Test for subgroup differences: Not applicable

0.01 0.1 1 10 100
Amoxycillin Cefuroxime

Antibiotics for community-acquired pneumonia in children (Review) 148
Copyright © 2013 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Analysis 19.3. Comparison 19 Amoxycillin with cefuroxime, Outcome 3 Cure rates.

Review: Antibiotics for community-acquired pneumonia in children

Comparison: 19 Amoxycillin with cefuroxime

Outcome: 3 Cure rates

Study or subgroup Amoxycillin Cefuroxime Odds Ratio Weight Odds Ratio
M- M-
H,Random,95% H,Random,95%
n/N n/N CI CI
Aurangzeb 2003 41/42 40/42 100.0 % 2.05 [ 0.18, 23.51 ]

Total (95% CI) 42 42 100.0 % 2.05 [ 0.18, 23.51 ]
Total events: 41 (Amoxycillin), 40 (Cefuroxime)
Heterogeneity: not applicable
Test for overall effect: Z = 0.58 (P = 0.56)
Test for subgroup differences: Not applicable

0.01 0.1 1 10 100
Amoxycillin Cefuroxime

Analysis 19.4. Comparison 19 Amoxycillin with cefuroxime, Outcome 4 Failure rates.

Review: Antibiotics for community-acquired pneumonia in children

Comparison: 19 Amoxycillin with cefuroxime

Outcome: 4 Failure rates

Study or subgroup Amoxycillin Cefuroxime Odds Ratio Weight Odds Ratio
M- M-
H,Random,95% H,Random,95%
n/N n/N CI CI
Aurangzeb 2003 1/42 2/42 100.0 % 0.49 [ 0.04, 5.59 ]

Total (95% CI) 42 42 100.0 % 0.49 [ 0.04, 5.59 ]
Total events: 1 (Amoxycillin), 2 (Cefuroxime)
Heterogeneity: not applicable
Test for overall effect: Z = 0.58 (P = 0.56)
Test for subgroup differences: Not applicable

0.01 0.1 1 10 100
Amoxycillin Cefuroxime

Antibiotics for community-acquired pneumonia in children (Review) 149
Copyright © 2013 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Analysis 20.1. Comparison 20 Amoxycillin with clarithromycin, Outcome 1 Mean age.

Review: Antibiotics for community-acquired pneumonia in children

Comparison: 20 Amoxycillin with clarithromycin

Outcome: 1 Mean age

Mean Mean
Study or subgroup Amoxycillin Clarithromycin Difference Weight Difference
N Mean(SD) N Mean(SD) IV,Random,95% CI IV,Random,95% CI

Aurangzeb 2003 43 12.79 (16.47) 42 15.95 (17.12) 100.0 % -3.16 [ -10.30, 3.98 ]

Total (95% CI) 43 42 100.0 % -3.16 [ -10.30, 3.98 ]
Heterogeneity: not applicable
Test for overall effect: Z = 0.87 (P = 0.39)
Test for subgroup differences: Not applicable

-100 -50 0 50 100
Amoxycillin Clarithromycin

Analysis 20.2. Comparison 20 Amoxycillin with clarithromycin, Outcome 2 Male sex.

Review: Antibiotics for community-acquired pneumonia in children

Comparison: 20 Amoxycillin with clarithromycin

Outcome: 2 Male sex

Study or subgroup Amoxycillin Clarithromycin Odds Ratio Weight Odds Ratio
M- M-
H,Random,95% H,Random,95%
n/N n/N CI CI
Aurangzeb 2003 35/43 31/42 100.0 % 1.55 [ 0.55, 4.35 ]

Total (95% CI) 43 42 100.0 % 1.55 [ 0.55, 4.35 ]
Total events: 35 (Amoxycillin), 31 (Clarithromycin)
Heterogeneity: not applicable
Test for overall effect: Z = 0.84 (P = 0.40)
Test for subgroup differences: Not applicable

0.01 0.1 1 10 100
Amoxycillin Clarithromycin

Antibiotics for community-acquired pneumonia in children (Review) 150
Copyright © 2013 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Analysis 20.3. Comparison 20 Amoxycillin with clarithromycin, Outcome 3 Cure rates.

Review: Antibiotics for community-acquired pneumonia in children

Comparison: 20 Amoxycillin with clarithromycin

Outcome: 3 Cure rates

Study or subgroup Amoxycillin Clarithromycin Odds Ratio Weight Odds Ratio
M- M-
H,Random,95% H,Random,95%
n/N n/N CI CI
Aurangzeb 2003 41/42 39/40 100.0 % 1.05 [ 0.06, 17.40 ]

Total (95% CI) 42 40 100.0 % 1.05 [ 0.06, 17.40 ]
Total events: 41 (Amoxycillin), 39 (Clarithromycin)
Heterogeneity: not applicable
Test for overall effect: Z = 0.03 (P = 0.97)
Test for subgroup differences: Not applicable

0.01 0.1 1 10 100
Amoxycillin Clarithromycin

Analysis 20.4. Comparison 20 Amoxycillin with clarithromycin, Outcome 4 Failure rates.

Review: Antibiotics for community-acquired pneumonia in children

Comparison: 20 Amoxycillin with clarithromycin

Outcome: 4 Failure rates

Study or subgroup Amoxycillin Clarithromycin Odds Ratio Weight Odds Ratio
M- M-
H,Random,95% H,Random,95%
n/N n/N CI CI
Aurangzeb 2003 1/42 1/40 100.0 % 0.95 [ 0.06, 15.74 ]

Total (95% CI) 42 40 100.0 % 0.95 [ 0.06, 15.74 ]
Total events: 1 (Amoxycillin), 1 (Clarithromycin)
Heterogeneity: not applicable
Test for overall effect: Z = 0.03 (P = 0.97)
Test for subgroup differences: Not applicable

0.01 0.1 1 10 100
Amoxycillin Clarithromycin

Antibiotics for community-acquired pneumonia in children (Review) 151
Copyright © 2013 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Analysis 21.1. Comparison 21 Penicillin and gentamycin with co-amoxyclavulanic acid, Outcome 1 Number
of children less than 1 year age.

Review: Antibiotics for community-acquired pneumonia in children

Comparison: 21 Penicillin and gentamycin with co-amoxyclavulanic acid

Outcome: 1 Number of children less than 1 year age

Co-
amoxyclavulanic
Study or subgroup Penicillin+gentamycin acid Odds Ratio Weight Odds Ratio
M- M-
H,Random,95% H,Random,95%
n/N n/N CI CI
Bansal 2006 21/38 23/33 100.0 % 0.54 [ 0.20, 1.43 ]

Total (95% CI) 38 33 100.0 % 0.54 [ 0.20, 1.43 ]
Total events: 21 (Penicillin+gentamycin), 23 (Co-amoxyclavulanic acid)
Heterogeneity: not applicable
Test for overall effect: Z = 1.24 (P = 0.21)
Test for subgroup differences: Not applicable

0.01 0.1 1 10 100
Penicillint+gentamycin Co-amoxyclavulanic acid

Analysis 21.2. Comparison 21 Penicillin and gentamycin with co-amoxyclavulanic acid, Outcome 2 Male sex.

Review: Antibiotics for community-acquired pneumonia in children

Comparison: 21 Penicillin and gentamycin with co-amoxyclavulanic acid

Outcome: 2 Male sex

Co-
amoxyclavulanic
Study or subgroup Penicillin+gentamycin acid Odds Ratio Weight Odds Ratio
M- M-
H,Random,95% H,Random,95%
n/N n/N CI CI
Bansal 2006 8/30 7/33 100.0 % 1.35 [ 0.42, 4.32 ]

Total (95% CI) 30 33 100.0 % 1.35 [ 0.42, 4.32 ]
Total events: 8 (Penicillin+gentamycin), 7 (Co-amoxyclavulanic acid)
Heterogeneity: not applicable
Test for overall effect: Z = 0.51 (P = 0.61)
Test for subgroup differences: Not applicable

0.01 0.1 1 10 100
Penicillint+gentamycin Co-amoxyclavulanic acid

Antibiotics for community-acquired pneumonia in children (Review) 152
Copyright © 2013 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Random.0 % 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 21 Penicillin and gentamycin with co-amoxyclavulanic acid Outcome: 3 Failure rates Co- amoxyclavulanic Study or subgroup Penicillin+gentamycin acid Odds Ratio Weight Odds Ratio M. 1 (Co-amoxyclavulanic acid) Heterogeneity: not applicable Test for overall effect: Z = 0.0 % 0.39 ] Total (95% CI) 38 33 100.86 [ 0.05.39 ] Total events: 1 (Penicillin+gentamycin). Ltd. 14.Random. .86 [ 0.92) Test for subgroup differences: Not applicable 0. 14.3.05.1 1 10 100 Penicillint+gentamycin Co-amoxyclavulanic acid Antibiotics for community-acquired pneumonia in children (Review) 153 Copyright © 2013 The Cochrane Collaboration.95% n/N n/N CI CI Bansal 2006 1/38 1/33 100.01 0. Published by John Wiley & Sons.95% H. Comparison 21 Penicillin and gentamycin with co-amoxyclavulanic acid. Analysis 21.10 (P = 0. M- H. Outcome 3 Failure rates.

74 ] Heterogeneity: not applicable Test for overall effect: Z = 0. 92 (Comparator) Heterogeneity: not applicable Test for overall effect: Z = 0.0 % 0.14 (P = 0.Random.05 [ -0.64. Comparison 22 Levofloxacin with comparator (co-amoxyclavulanic acid/ceftriaxone). 0. M- H.64.97 [ 0.95% n/N n/N CI CI Bradley 2007 266/529 92/180 100.95% H.52 (4.2.01 0.Random.36 ] Total events: 266 (Levofloxacin).12) 100.19 (P = 0.36 ] Total (95% CI) 529 180 100.69. Review: Antibiotics for community-acquired pneumonia in children Comparison: 22 Levofloxacin with comparator (co-amoxyclavulanic acid/ceftriaxone) Outcome: 1 Mean age Mean Mean Study or subgroup Levofloxacin Comparator Difference Weight Difference N Mean(SD) N Mean(SD) IV. 0.Random. 1.74 ] Total (95% CI) 529 180 100. Outcome 1 Mean age. Outcome 2 Male sex.89) Test for subgroup differences: Not applicable -100 -50 0 50 100 Levofloxacin Comparator Analysis 22.0 % 0. Published by John Wiley & Sons. Comparison 22 Levofloxacin with comparator (co-amoxyclavulanic acid/ceftriaxone).0 % 0.Random.0 % 0.1.01) 180 5.1 1 10 100 Levofloxacin Comparator Antibiotics for community-acquired pneumonia in children (Review) 154 Copyright © 2013 The Cochrane Collaboration. .05 [ -0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 22 Levofloxacin with comparator (co-amoxyclavulanic acid/ceftriaxone) Outcome: 2 Male sex Study or subgroup Levofloxacin Comparator Odds Ratio Weight Odds Ratio M.85) Test for subgroup differences: Not applicable 0.47 (4.69. 1.95% CI IV. Analysis 22.97 [ 0.95% CI Bradley 2007 529 5. Ltd.

51 (Comparator) Heterogeneity: not applicable Test for overall effect: Z = 0.35 ] Total (95% CI) 529 180 100. 1.Random.42 ] Total (95% CI) 405 134 100.93 [ 0.0 % 0. M- H.1 1 10 100 Levofloxacin Comparator Antibiotics for community-acquired pneumonia in children (Review) 155 Copyright © 2013 The Cochrane Collaboration.46.95% H.35 ] Total events: 142 (Levofloxacin).4.Random.46. 2. 1. Outcome 3 Numbers received antibiotics in past 1 week.90) Test for subgroup differences: Not applicable 0. .42 ] Total events: 382 (Levofloxacin). Comparison 22 Levofloxacin with comparator (co-amoxyclavulanic acid/ceftriaxone). Ltd.01 0.39 (P = 0. 126 (Comparator) Heterogeneity: not applicable Test for overall effect: Z = 0.3.05 [ 0. 2.0 % 1.01 0.64.Random.95% n/N n/N CI CI Bradley 2007 382/405 126/134 100.64. M- H.Random. Outcome 4 Cure rates.1 1 10 100 Levofloxacin Comparator Analysis 22.0 % 0.93 [ 0.95% H. Analysis 22.95% n/N n/N CI CI Bradley 2007 142/529 51/180 100. Review: Antibiotics for community-acquired pneumonia in children Comparison: 22 Levofloxacin with comparator (co-amoxyclavulanic acid/ceftriaxone) Outcome: 3 Numbers received antibiotics in past 1 week Study or subgroup Levofloxacin Comparator Odds Ratio Weight Odds Ratio M.13 (P = 0.05 [ 0.0 % 1. Comparison 22 Levofloxacin with comparator (co-amoxyclavulanic acid/ceftriaxone). Review: Antibiotics for community-acquired pneumonia in children Comparison: 22 Levofloxacin with comparator (co-amoxyclavulanic acid/ceftriaxone) Outcome: 4 Cure rates Study or subgroup Levofloxacin Comparator Odds Ratio Weight Odds Ratio M. Published by John Wiley & Sons.70) Test for subgroup differences: Not applicable 0.

12) 100. . M- H.78.012) Test for subgroup differences: Not applicable 0.0 % -7. -0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 23 Cefuroxime with clarithromycin Outcome: 2 Male sex Study or subgroup Cefuroxime Clarithromycin Odds Ratio Weight Odds Ratio M. Comparison 23 Cefuroxime with clarithromycin.03 [ -13. 118.76 ] Total (95% CI) 42 42 100.90 ] Total (95% CI) 41 42 100. -0.0 % 14.Random. 31 (Clarithromycin) Heterogeneity: not applicable Test for overall effect: Z = 2.78.16.1 1 10 100 Cefuroxime Clarithromycin Antibiotics for community-acquired pneumonia in children (Review) 156 Copyright © 2013 The Cochrane Collaboration. Published by John Wiley & Sons.Random. Outcome 2 Male sex.0 % 14.50 (P = 0.Random.32 (10.72) 42 15.55 [ 1.025) Test for subgroup differences: Not applicable -100 -50 0 50 100 Cefuroxime Clarithromycin Analysis 23.95% CI IV.0 % -7. Comparison 23 Cefuroxime with clarithromycin.2.01 0.03 [ -13. Analysis 23.95% H. Ltd.95% n/N n/N CI CI Aurangzeb 2003 41/42 31/42 100.76 ] Total events: 41 (Cefuroxime).16.1.55 [ 1.35 (17.Random.95% CI Aurangzeb 2003 41 8. Review: Antibiotics for community-acquired pneumonia in children Comparison: 23 Cefuroxime with clarithromycin Outcome: 1 Mean age Mean Mean Study or subgroup Cefuroxime Clarithromycin Difference Weight Difference N Mean(SD) N Mean(SD) IV. 118.25 (P = 0. Outcome 1 Mean age.90 ] Heterogeneity: not applicable Test for overall effect: Z = 2.

0 % 0. Analysis 23.1 1 10 100 Cefuroxime Clarithromycin Analysis 23.05 [ 0.51 [ 0.01 0.04. Published by John Wiley & Sons. 23. Ltd.Random. M- H. Outcome 4 Failure rates.3. 23.18.89 ] Total events: 40 (Cefuroxime).51 ] Total events: 2 (Cefuroxime).51 ] Total (95% CI) 42 42 100.05 [ 0.95% n/N n/N CI CI Aurangzeb 2003 40/42 39/40 100.95% H.0 % 2. Comparison 23 Cefuroxime with clarithromycin. 5. 1 (Clarithromycin) Heterogeneity: not applicable Test for overall effect: Z = 0. M- H.0 % 0. .0 % 2.Random. 5. Comparison 23 Cefuroxime with clarithromycin. Review: Antibiotics for community-acquired pneumonia in children Comparison: 23 Cefuroxime with clarithromycin Outcome: 4 Failure rates Study or subgroup Cefuroxime Clarithromycin Odds Ratio Weight Odds Ratio M.18. Outcome 3 Cure rates. 39 (Clarithromycin) Heterogeneity: not applicable Test for overall effect: Z = 0.51 [ 0.01 0.54 (P = 0.59) Test for subgroup differences: Not applicable 0.Random.4.04. Review: Antibiotics for community-acquired pneumonia in children Comparison: 23 Cefuroxime with clarithromycin Outcome: 3 Cure rates Study or subgroup Cefuroxime Clarithromycin Odds Ratio Weight Odds Ratio M.Random.56) Test for subgroup differences: Not applicable 0.89 ] Total (95% CI) 42 40 100.1 1 10 100 Cefuroxime Clarithromycin Antibiotics for community-acquired pneumonia in children (Review) 157 Copyright © 2013 The Cochrane Collaboration.95% n/N n/N CI CI Aurangzeb 2003 2/42 1/42 100.95% H.58 (P = 0.

Review: Antibiotics for community-acquired pneumonia in children Comparison: 24 Co-trimoxazole versus chloramphenicol Outcome: 1 Age in months Mean Mean Study or subgroup Co-trimoxazole Chloramphenicol Difference Weight Difference N Mean(SD) N Mean(SD) IV.31 (P = 0.95% CI Mulholland 1995 56 18. Analysis 24.89 [ 0.Random.0 % 1.2 0.Random. .4) 100. 1.89 ] Total events: 22 (Co-trimoxazole).95% H.5 1 2 5 10 Co-trimoxazole Chloramphenicol Antibiotics for community-acquired pneumonia in children (Review) 158 Copyright © 2013 The Cochrane Collaboration.44 ] Heterogeneity: not applicable Test for overall effect: Z = 1. 1.64.0 % 1. Comparison 24 Co-trimoxazole versus chloramphenicol.90 [ -0.1 0. 4. 4.6 (6.Random.95% CI IV.0 % 0.89 ] Total (95% CI) 55 56 100.95% n/N n/N CI CI Mulholland 1995 22/55 24/56 100.90 [ -0. Outcome 2 Male sex.14) Test for subgroup differences: Not applicable -10 -5 0 5 10 Co-trimoxazole Chloramphenicol Analysis 24. Review: Antibiotics for community-acquired pneumonia in children Comparison: 24 Co-trimoxazole versus chloramphenicol Outcome: 2 Male sex Study or subgroup Co-trimoxazole Chloramphenicol Odds Ratio Weight Odds Ratio M. Published by John Wiley & Sons.2. 24 (Chloramphenicol) Heterogeneity: not applicable Test for overall effect: Z = 0.42. Outcome 1 Age in months. Ltd.Random.1. Comparison 24 Co-trimoxazole versus chloramphenicol.47 (P = 0.76) Test for subgroup differences: Not applicable 0.0 % 0. M- H.44 ] Total (95% CI) 56 55 100.64.89 [ 0.21) 55 16.5 (7.42.

15 ] Total (95% CI) 55 56 100.3) 56 60.11 ] Total (95% CI) 55 56 100.0 (P = 1.Random.67 [ 0.0 % 0.4) 100. Review: Antibiotics for community-acquired pneumonia in children Comparison: 24 Co-trimoxazole versus chloramphenicol Outcome: 4 Wheezing positive Study or subgroup Co-trimoxazole Chloramphenicol Odds Ratio Weight Odds Ratio M. Review: Antibiotics for community-acquired pneumonia in children Comparison: 24 Co-trimoxazole versus chloramphenicol Outcome: 3 Weight for age Mean Mean Study or subgroup Co-trimoxazole Chloramphenicol Difference Weight Difference N Mean(SD) N Mean(SD) IV.Random.Random.95% CI Mulholland 1995 55 60.5 1 2 5 10 Co-trimoxazole Chloramphenicol Antibiotics for community-acquired pneumonia in children (Review) 159 Copyright © 2013 The Cochrane Collaboration.95% CI IV.11.3.2 0.11. 4.0) Test for subgroup differences: Not applicable -10 -5 0 5 10 Co-trimoxazole Chloramphenicol Analysis 24. 3.0 % 0.11. Outcome 4 Wheezing positive.11 ] Heterogeneity: not applicable Test for overall effect: Z = 0.43 (P = 0. M- H.15 ] Total events: 2 (Co-trimoxazole).0 % 0.4.0 % 0.66) Test for subgroup differences: Not applicable 0.0 [ -3. Ltd.2 (8.67 [ 0. 3.Random. Analysis 24.1 0. Outcome 3 Weight for age.95% H. Comparison 24 Co-trimoxazole versus chloramphenicol.2 (8.0 [ -3. Comparison 24 Co-trimoxazole versus chloramphenicol. 3 (Chloramphenicol) Heterogeneity: not applicable Test for overall effect: Z = 0. Published by John Wiley & Sons.95% n/N n/N CI CI Mulholland 1995 2/55 3/56 100. . 4.11.

95% H.0 % 1.95) Test for subgroup differences: Not applicable 0. 2.95% n/N n/N CI CI Mulholland 1995 16/55 16/56 100.Random. Comparison 24 Co-trimoxazole versus chloramphenicol. 16 (Chloramphenicol) Heterogeneity: not applicable Test for overall effect: Z = 0.40 ] Total (95% CI) 55 56 100.33 ] Total events: 16 (Co-trimoxazole). Outcome 5 Cure rate.88) Test for subgroup differences: Not applicable 0.06 (P = 0. Ltd.0 % 1.06 [ 0.Random.03 [ 0.2 0.0 % 1. 2.5 1 2 5 10 Co-trimoxazole Chloramphenicol Analysis 24. Analysis 24.5.1 0. 39 (Chloramphenicol) Heterogeneity: not applicable Test for overall effect: Z = 0.Random. Outcome 6 Failure rate.40 ] Total events: 39 (Co-trimoxazole).5 1 2 5 10 Co-trimoxazole Chloramphenicol Antibiotics for community-acquired pneumonia in children (Review) 160 Copyright © 2013 The Cochrane Collaboration.03 [ 0.33 ] Total (95% CI) 55 56 100. 2.47. M- H.47. M- H. 2.95% n/N n/N CI CI Mulholland 1995 39/55 39/56 100.1 0.2 0.45. Published by John Wiley & Sons.0 % 1.95% H. Comparison 24 Co-trimoxazole versus chloramphenicol.45. .Random. Review: Antibiotics for community-acquired pneumonia in children Comparison: 24 Co-trimoxazole versus chloramphenicol Outcome: 5 Cure rate Study or subgroup Co-trimoxazole Chloramphenicol Odds Ratio Weight Odds Ratio M.6. Review: Antibiotics for community-acquired pneumonia in children Comparison: 24 Co-trimoxazole versus chloramphenicol Outcome: 6 Failure rate Study or subgroup Co-trimoxazole Chloramphenicol Odds Ratio Weight Odds Ratio M.15 (P = 0.06 [ 0.

0 % 1. 4.30 ] Total (95% CI) 55 56 100. Published by John Wiley & Sons.03 (P = 0.02 [ 0.98) Test for subgroup differences: Not applicable 0.2 0.Random.2 0. Outcome 8 Relapse rate. 4.88) Test for subgroup differences: Not applicable 0. .0 % 1.24.5 1 2 5 10 Co-trimoxazole Chloramphenicol Analysis 24.8. Analysis 24. Outcome 7 Excluded.Random. 17 (Chloramphenicol) Heterogeneity: not applicable Test for overall effect: Z = 0. M- H. Ltd.12 ] Total (95% CI) 55 56 100.1 0.Random. Comparison 24 Co-trimoxazole versus chloramphenicol.42.02 [ 0. 2.12 ] Total events: 16 (Co-trimoxazole). Review: Antibiotics for community-acquired pneumonia in children Comparison: 24 Co-trimoxazole versus chloramphenicol Outcome: 8 Relapse rate Study or subgroup Co-trimoxazole Chloramphenicol Odds Ratio Weight Odds Ratio M.95% n/N n/N CI CI Mulholland 1995 16/55 17/56 100.Random. M- H. 4 (Chloramphenicol) Heterogeneity: not applicable Test for overall effect: Z = 0.24.95% H.94 [ 0. 2.15 (P = 0.1 0. Comparison 24 Co-trimoxazole versus chloramphenicol.0 % 0.94 [ 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 24 Co-trimoxazole versus chloramphenicol Outcome: 7 Excluded Study or subgroup Co-trimoxazole Chloramphenicol Odds Ratio Weight Odds Ratio M.5 1 2 5 10 Co-trimoxazole Chloramphenicol Antibiotics for community-acquired pneumonia in children (Review) 161 Copyright © 2013 The Cochrane Collaboration.7.95% H.95% n/N n/N CI CI Mulholland 1995 4/55 4/56 100.0 % 0.42.30 ] Total events: 4 (Co-trimoxazole).

6 (Chloramphenicol) Heterogeneity: not applicable Test for overall effect: Z = 0. 4.40 ] Total (95% CI) 55 56 100.1 0.Random. Comparison 24 Co-trimoxazole versus chloramphenicol. Outcome 9 Need for change in antibiotics. .46.0 % 1.0 % 1.Random.21 [ 0. 7. M- H.42 [ 0.63. 7. Ltd.Random.95% H.0 % 2.40 ] Total events: 8 (Co-trimoxazole).10.0 % 2.21 [ 0. 4 (Chloramphenicol) Heterogeneity: not applicable Test for overall effect: Z = 1.46.23 (P = 0.5 1 2 5 10 Co-trimoxazole Chloramphenicol Antibiotics for community-acquired pneumonia in children (Review) 162 Copyright © 2013 The Cochrane Collaboration.63.9.95% H.Random. Review: Antibiotics for community-acquired pneumonia in children Comparison: 24 Co-trimoxazole versus chloramphenicol Outcome: 10 Death rate Study or subgroup Co-trimoxazole Chloramphenicol Odds Ratio Weight Odds Ratio M.2 0. 4. Comparison 24 Co-trimoxazole versus chloramphenicol. Review: Antibiotics for community-acquired pneumonia in children Comparison: 24 Co-trimoxazole versus chloramphenicol Outcome: 9 Need for change in antibiotics Study or subgroup Co-trimoxazole Chloramphenicol Odds Ratio Weight Odds Ratio M.83 ] Total events: 8 (Co-trimoxazole).54) Test for subgroup differences: Not applicable 0.22) Test for subgroup differences: Not applicable 0. M- H.2 0.42 [ 0.83 ] Total (95% CI) 55 56 100. Analysis 24. Outcome 10 Death rate.95% n/N n/N CI CI Mulholland 1995 8/55 4/56 100.61 (P = 0. Published by John Wiley & Sons.95% n/N n/N CI CI Mulholland 1995 8/55 6/56 100.1 0.5 1 2 5 10 Co-trimoxazole Chloramphenicol Analysis 24.

Comparison 25 Ceftibuten versus cefuroxime. 9 (Chloramphenicol) Heterogeneity: not applicable Test for overall effect: Z = 0.0 % 1.Random.Random.01 0.0 % 0. Outcome 1 Male sex. . 39 (Cefuroxime) Heterogeneity: not applicable Test for overall effect: Z = 0.45 (P = 0.95% H. 1.79 [ 0.69 (P = 0. Ltd.30 ] Total events: 11 (Co-trimoxazole).95% n/N n/N CI CI Nogeova 1997 36/71 39/69 100.41. Analysis 24. Comparison 24 Co-trimoxazole versus chloramphenicol. 1.2 0.54 ] Total events: 36 (Ceftibuten).41.Random.30 ] Total (95% CI) 56 55 100.25 [ 0.54 ] Total (95% CI) 71 69 100. M- H. Review: Antibiotics for community-acquired pneumonia in children Comparison: 25 Ceftibuten versus cefuroxime Outcome: 1 Male sex Study or subgroup Ceftibuten Cefuroxime Odds Ratio Weight Odds Ratio M.49) Test for subgroup differences: Not applicable 0. M- H. 3.47. Published by John Wiley & Sons.5 1 2 5 10 Co-trimoxazole Chloramphenicol Analysis 25. Review: Antibiotics for community-acquired pneumonia in children Comparison: 24 Co-trimoxazole versus chloramphenicol Outcome: 11 Organisms isolated on blood culture or lung puncture Study or subgroup Co-trimoxazole Chloramphenicol Odds Ratio Weight Odds Ratio M.47.0 % 0.95% n/N n/N CI CI Mulholland 1995 11/56 9/55 100.65) Test for subgroup differences: Not applicable 0.95% H. Outcome 11 Organisms isolated on blood culture or lung puncture.11.25 [ 0.0 % 1.1 1 10 100 Ceftibuten Cefuroxime Antibiotics for community-acquired pneumonia in children (Review) 163 Copyright © 2013 The Cochrane Collaboration.1.1 0.79 [ 0.Random. 3.

35. 11.01 0. 11.87.00 [ 0.83 ] Total (95% CI) 71 69 100. Comparison 25 Ceftibuten versus cefuroxime.0 % 3. Comparison 25 Ceftibuten versus cefuroxime. 7.95% CI Nogeova 1997 53/71 30/69 100.95% CI M-H. 2 (Cefuroxime) Heterogeneity: not applicable Test for overall effect: Z = 0.Fixed.1 1 10 100 Ceftibuten Cefuroxime Analysis 25.2.00024) Test for subgroup differences: Not applicable 0.83 [ 1. Analysis 25.Fixed.78 (P = 0. 30 (Cefuroxime) Heterogeneity: not applicable Test for overall effect: Z = 3. Outcome 2 Positive for microbial agent.1 1 10 100 Ceftibuten Cefuroxime Antibiotics for community-acquired pneumonia in children (Review) 164 Copyright © 2013 The Cochrane Collaboration.0 % 3.95% CI Nogeova 1997 4/71 2/69 100.83 ] Total events: 53 (Ceftibuten).00 [ 0. 7.0 % 2.0 % 2.35. Review: Antibiotics for community-acquired pneumonia in children Comparison: 25 Ceftibuten versus cefuroxime Outcome: 2 Positive for microbial agent Study or subgroup Ceftibuten Cefuroxime Odds Ratio Weight Odds Ratio n/N n/N M-H.68 (P = 0.Fixed.43) Test for subgroup differences: Not applicable 0. Ltd.3. Outcome 3 Adverse reaction. Review: Antibiotics for community-acquired pneumonia in children Comparison: 25 Ceftibuten versus cefuroxime Outcome: 3 Adverse reaction Study or subgroup Ceftibuten Cefuroxime Odds Ratio Weight Odds Ratio n/N n/N M-H.95% CI M-H. .87.Fixed.83 [ 1. Published by John Wiley & Sons.29 ] Total events: 4 (Ceftibuten).29 ] Total (95% CI) 71 69 100.01 0.

81 [ 1. Outcome 4 Cure rate.95% CI Nogeova 1997 57/71 64/69 100.11.46. Analysis 25.94 ] Total events: 57 (Ceftibuten).95% CI M-H.95% CI M-H.Fixed.95% CI Nogeova 1997 12/71 2/69 100. 64 (Cefuroxime) Heterogeneity: not applicable Test for overall effect: Z = 2. Comparison 25 Ceftibuten versus cefuroxime.11. 2 (Cefuroxime) Heterogeneity: not applicable Test for overall effect: Z = 2.014) Test for subgroup differences: Not applicable 0.Fixed.01 0.4.Fixed.70 ] Total events: 12 (Ceftibuten).5. . Review: Antibiotics for community-acquired pneumonia in children Comparison: 25 Ceftibuten versus cefuroxime Outcome: 5 Failure rate Study or subgroup Ceftibuten Cefuroxime Odds Ratio Weight Odds Ratio n/N n/N M-H.46. Comparison 25 Ceftibuten versus cefuroxime.0 % 0.0 % 0. 0. Ltd.32 [ 0.0 % 6.94 ] Total (95% CI) 71 69 100. Review: Antibiotics for community-acquired pneumonia in children Comparison: 25 Ceftibuten versus cefuroxime Outcome: 4 Cure rate Study or subgroup Ceftibuten Cefuroxime Odds Ratio Weight Odds Ratio n/N n/N M-H.70 ] Total (95% CI) 71 69 100.08 (P = 0. Published by John Wiley & Sons.038) Test for subgroup differences: Not applicable 0. Outcome 5 Failure rate. 0. 31.Fixed.45 (P = 0.1 1 10 100 Ceftibuten Cefuroxime Antibiotics for community-acquired pneumonia in children (Review) 165 Copyright © 2013 The Cochrane Collaboration.01 0. 31.0 % 6.81 [ 1.1 1 10 100 Ceftibuten Cefuroxime Analysis 25.32 [ 0.

95% n/N n/N CI CI Ribeiro 2011 29/56 29/48 100.5 (2 to 60) Analysis 26.Random. 1.Random.32. .0 % 0. Published by John Wiley & Sons. Comparison 26 Oxacillin ceftriaxone versus co-amoxyclavulanic acid.95% H.88 (P = 0.70 [ 0.5 (3 to 60) 10.01 0. Comparison 26 Oxacillin ceftriaxone versus co-amoxyclavulanic acid.54 ] Total (95% CI) 56 48 100.1 1 10 100 Co-amoxtclavulanic acid Oxacillin ceftrioxone Antibiotics for community-acquired pneumonia in children (Review) 166 Copyright © 2013 The Cochrane Collaboration. Review: Antibiotics for community-acquired pneumonia in children Comparison: 26 Oxacillin ceftriaxone versus co-amoxyclavulanic acid Outcome: 2 Male sex Co- amoxtclavulanic Study or subgroup acid Oxacillin ceftrioxone Odds Ratio Weight Odds Ratio M.32. 1. M- H. 29 (Oxacillin ceftrioxone) Heterogeneity: not applicable Test for overall effect: Z = 0.0 % 0. Outcome 2 Male sex.54 ] Total events: 29 (Co-amoxtclavulanic acid).70 [ 0. Median age (months) with IQR Study Amoxicillin clavulanic acid group Oxacillin+ ceftriaxone group Ribeiro 2011 11.2. Outcome 1 Median age (months) with IQR.38) Test for subgroup differences: Not applicable 0.1. Ltd. Analysis 26.

Review: Antibiotics for community-acquired pneumonia in children Comparison: 26 Oxacillin ceftriaxone versus co-amoxyclavulanic acid Outcome: 3 Mean number of days before admission Co- amoxtclavulanic Mean Mean Study or subgroup acid Oxacillin ceftrioxone Difference Weight Difference N Mean(SD) N Mean(SD) IV. M- H.90 [ -2.76 ] Total events: 17 (Co-amoxtclavulanic acid).0 % 1.8 (4) 100.Random. 2. .Random.71) Test for subgroup differences: Not applicable 0.28 (P = 0.17 [ 0.50.48 ] Total (95% CI) 56 48 100.0 % -0.95% n/N n/N CI CI Ribeiro 2011 17/56 13/48 100. Outcome 3 Mean number of days before admission. 13 (Oxacillin ceftrioxone) Heterogeneity: not applicable Test for overall effect: Z = 0. 0.20) Test for subgroup differences: Not applicable -20 -10 0 10 20 Co-amoxtclavulanic acid Oxacillin ceftrioxone Analysis 26. Outcome 4 Received antibiotics before enrolment.9 (3) 48 5. Comparison 26 Oxacillin ceftriaxone versus co-amoxyclavulanic acid.3.95% CI IV.28.28.17 [ 0.Random.95% CI Ribeiro 2011 56 4. 2.1 1 10 100 Co-amoxtclavulanic acid Oxacillin ceftrioxone Antibiotics for community-acquired pneumonia in children (Review) 167 Copyright © 2013 The Cochrane Collaboration. Published by John Wiley & Sons.37 (P = 0.76 ] Total (95% CI) 56 48 100. Review: Antibiotics for community-acquired pneumonia in children Comparison: 26 Oxacillin ceftriaxone versus co-amoxyclavulanic acid Outcome: 4 Received antibiotics before enrolment Co- amoxtclavulanic Study or subgroup acid Oxacillin ceftrioxone Odds Ratio Weight Odds Ratio M.Random.0 % -0.50. Comparison 26 Oxacillin ceftriaxone versus co-amoxyclavulanic acid. Analysis 26.4.01 0.95% H.90 [ -2. Ltd.48 ] Heterogeneity: not applicable Test for overall effect: Z = 1. 0.0 % 1.

Analysis 26.0 % 0.8 (2.1 1 10 100 Co-amoxtclavulanic acid Oxacillin ceftrioxone Analysis 26.01 0.33. Comparison 26 Oxacillin ceftriaxone versus co-amoxyclavulanic acid.5.6.11 ] Total (95% CI) 56 48 100.89.89.0 % 0. Outcome 6 Mean time for improvement in tachypnoea.0 % -1.95% CI IV.0 % -1.04 (P = 0. Comparison 26 Oxacillin ceftriaxone versus co-amoxyclavulanic acid. -0.2) 48 5. -0. 2. Published by John Wiley & Sons. 2.Random. Ltd. .11 ] Heterogeneity: not applicable Test for overall effect: Z = 2.95% H.Random.00 [ -1. Outcome 5 Failure rates.028) Test for subgroup differences: Not applicable -10 -5 0 5 10 Co-amoxtclavulanic acid Oxacillin ceftrioxone Antibiotics for community-acquired pneumonia in children (Review) 168 Copyright © 2013 The Cochrane Collaboration.Random.95% n/N n/N CI CI Ribeiro 2011 8/56 7/48 100.4) 100. Review: Antibiotics for community-acquired pneumonia in children Comparison: 26 Oxacillin ceftriaxone versus co-amoxyclavulanic acid Outcome: 5 Failure rates Co- amoxtclavulanic Study or subgroup acid Oxacillin ceftrioxone Odds Ratio Weight Odds Ratio M. 7 (Oxacillin ceftrioxone) Heterogeneity: not applicable Test for overall effect: Z = 0.97) Test for subgroup differences: Not applicable 0.00 [ -1.98 [ 0.20 (P = 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 26 Oxacillin ceftriaxone versus co-amoxyclavulanic acid Outcome: 6 Mean time for improvement in tachypnoea Co- amoxtclavulanic Mean Mean Study or subgroup acid Oxacillin ceftrioxone Difference Weight Difference N Mean(SD) N Mean(SD) IV. M- H.92 ] Total events: 8 (Co-amoxtclavulanic acid).33.8 (2.95% CI Ribeiro 2011 56 4.98 [ 0.92 ] Total (95% CI) 56 48 100.Random.

-1.95% n/N n/N CI CI Addo-Yobo 2004 537/857 520/845 40.87 ] Total (95% CI) 2094 2070 100.40 [ -5.33.40 [ -5.98 (P = 0.91. Outcome 1 Male sex.4 (4. Review: Antibiotics for community-acquired pneumonia in children Comparison: 26 Oxacillin ceftriaxone versus co-amoxyclavulanic acid Outcome: 7 Mean length of stay Co- amoxtclavulanic Mean Mean Study or subgroup acid Oxacillin ceftrioxone Difference Weight Difference N Mean(SD) N Mean(SD) IV. 1. M- H.95% CI IV. 1.98 [ 0.86.46.95% H.1 % 1. 2.32) Test for subgroup differences: Not applicable 0.09 [ 0.92). .4 % 1.65.30 ] Sidal 1994 27/46 27/42 2.0 % -3.06 [ 0.94. 1.0 % 1.21 ] Total events: 1288 (Oral treatment).0% Test for overall effect: Z = 0.1 % 0. I2 =0.Random.2) 48 14.3 % 1.Random. Review: Antibiotics for community-acquired pneumonia in children Comparison: 27 Oral versus parenteral antibiotics for treatment of severe pneumonia Outcome: 1 Male sex Study or subgroup Oral treatment Parenteral treatment Odds Ratio Weight Odds Ratio M.92.1 1 10 100 Oral treatment Parenteral treatment Antibiotics for community-acquired pneumonia in children (Review) 169 Copyright © 2013 The Cochrane Collaboration.58 ] Hazir 2008 630/1025 602/1012 49.7.05 [ 0.Random. Published by John Wiley & Sons.71 ] Campbell 1988 41/66 38/68 3. df = 4 (P = 0. Analysis 26. Chi2 = 0.79 [ 0.34 ] Total (95% CI) 56 48 100.28 ] Atkinson 2007 53/100 55/103 5.0.01 0. Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia. Comparison 26 Oxacillin ceftriaxone versus co-amoxyclavulanic acid. 1242 (Parenteral treatment) Heterogeneity: Tau2 = 0.0 % -3. Ltd.46. 1. 1.23 (P = 0.29 [ 0. Outcome 7 Mean length of stay. -1.95% CI Ribeiro 2011 56 11 (6.0012) Test for subgroup differences: Not applicable -20 -10 0 10 20 Co-amoxtclavulanic acid Oxacillin ceftrioxone Analysis 27.34 ] Heterogeneity: not applicable Test for overall effect: Z = 3.57.1 % 0.Random.1.5) 100.

83. Published by John Wiley & Sons. 4.13 ] Sidal 1994 16/46 27/42 10.74) Test for subgroup differences: Not applicable 0.8 % 1.30 ] Total events: 1221 (Oral treatment).2.84 [ 0.6 % 1. I2 =71% Test for overall effect: Z = 0. Outcome 2 Age below 12 months.10 ] Hazir 2008 653/1025 658/1012 39. .06 [ 0.95 [ 0. 1211 (Parenteral treatment) Heterogeneity: Tau2 = 0.69.33 (P = 0.Random.94 [ 0.01 0. Ltd. 1.87.95% H.95% n/N n/N CI CI Addo-Yobo 2004 532/857 513/845 38.12. Chi2 = 10. 1.29 ] Campbell 1988 20/66 13/68 11. Review: Antibiotics for community-acquired pneumonia in children Comparison: 27 Oral versus parenteral antibiotics for treatment of severe pneumonia Outcome: 2 Age below 12 months Study or subgroup Oral treatment Parenteral treatment Odds Ratio Weight Odds Ratio M.Random. df = 3 (P = 0.31. M- H.06.71 ] Total (95% CI) 1994 1967 100.4 % 0.30 [ 0.0 % 0. 0.02).1 1 10 100 Oral treatment Parenteral treatment Antibiotics for community-acquired pneumonia in children (Review) 170 Copyright © 2013 The Cochrane Collaboration. Analysis 27. 1. Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia.79.3 % 0.

M- H. Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia.95% H.71) Test for subgroup differences: Not applicable 0.24 ] Atkinson 2007 18/100 14/103 11.95% n/N n/N CI CI Addo-Yobo 2004 72/857 52/845 43. Review: Antibiotics for community-acquired pneumonia in children Comparison: 27 Oral versus parenteral antibiotics for treatment of severe pneumonia Outcome: 4 Children with wheezing Study or subgroup Oral treatment Parenteral treatment Odds Ratio Weight Odds Ratio M.Random. I2 =47% Test for overall effect: Z = 0. 1.32. 2.6 % 1. df = 1 (P = 0.98 ] Hazir 2008 267/1025 216/1012 52.70.36) Test for subgroup differences: Not applicable 0.0 % 1.97.14 [ 0. Ltd. 312 (Parenteral treatment) Heterogeneity: Tau2 = 0.03 ] Hazir 2008 767/1025 778/1012 56. Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia.Random.79.06.1 1 10 100 Oral treatment Parenteral treatment Analysis 27.03.95% n/N n/N CI CI Addo-Yobo 2004 75/857 82/845 36.8 % 1.4 % 0. 830 (Parenteral treatment) Heterogeneity: Tau2 = 0.4.95% H.Random. 1.89 [ 0. M- H.09 [ 0.68 ] Total events: 839 (Oral treatment). Chi2 = 4.30 [ 1.15).10 ] Total (95% CI) 1882 1857 100. .89 [ 0.59 ] Total (95% CI) 1982 1960 100. Review: Antibiotics for community-acquired pneumonia in children Comparison: 27 Oral versus parenteral antibiotics for treatment of severe pneumonia Outcome: 3 Received antibiotics in the past week Study or subgroup Oral treatment Parenteral treatment Odds Ratio Weight Odds Ratio M. df = 2 (P = 0. 1.01 0. Analysis 27.2 % 0.01 0.73. Outcome 4 Children with wheezing.3. Chi2 = 3. I2 =77% Test for overall effect: Z = 0. Outcome 3 Received antibiotics in the past week. Published by John Wiley & Sons.0 % 1.64.Random. 1.52 ] Total events: 360 (Oral treatment).40 [ 0.08.40 [ 0. 2.92 (P = 0. 1.38 (P = 0.04).86.0 % 1.65.1 1 10 100 Oral treatment Parenteral treatment Antibiotics for community-acquired pneumonia in children (Review) 171 Copyright © 2013 The Cochrane Collaboration.

Published by John Wiley & Sons.2 % 1.95% n/N n/N CI CI Addo-Yobo 2004 192/769 183/759 97. df = 1 (P = 0. . 1. 1. 188 (Parenteral treatment) Heterogeneity: Tau2 = 0.75) Test for subgroup differences: Not applicable 0.Random.05 [ 0.19.1 1 10 100 Oral treatment Parenteral treatment Antibiotics for community-acquired pneumonia in children (Review) 172 Copyright © 2013 The Cochrane Collaboration. Review: Antibiotics for community-acquired pneumonia in children Comparison: 27 Oral versus parenteral antibiotics for treatment of severe pneumonia Outcome: 5 RSV positivity Study or subgroup Oral treatment Parenteral treatment Odds Ratio Weight Odds Ratio M.Random.82.75 [ 0.32 ] Atkinson 2007 4/54 5/52 2.32 (P = 0. Chi2 = 0.5.64).83.95% H. M- H.0% Test for overall effect: Z = 0. 2. I2 =0.0. Analysis 27. Outcome 5 RSV positivity.04 [ 0.0 % 1.01 0.31 ] Total events: 196 (Oral treatment). Ltd. Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia.22.97 ] Total (95% CI) 823 811 100.8 % 0.

95 [ 0. 1. I2 =3% Test for overall effect: Z = 0.42 [ 0.03 [ 0.11. Chi2 = 2.3 % 0. Published by John Wiley & Sons.86 [ 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 27 Oral versus parenteral antibiotics for treatment of severe pneumonia Outcome: 6 Failure rates on day 3 Study or subgroup Oral treatment Parenteral treatment Odds Ratio Weight Odds Ratio M.81.95% H.6.53 (P = 0.60) Test for subgroup differences: Not applicable 0.15 ] Total events: 247 (Oral treatment).78. Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia. 255 (Parenteral treatment) Heterogeneity: Tau2 = 0. df = 2 (P = 0.1 1 10 100 Oral treatment Parenteral treatment Antibiotics for community-acquired pneumonia in children (Review) 173 Copyright © 2013 The Cochrane Collaboration.0 % 0.01 0.0 % 0.7 % 1. M- H. Ltd.69 ] Hazir 2008 77/1025 87/1012 36.Random.19 ] Total (95% CI) 1982 1960 100. 1. 1.07.Random. Analysis 27. .31 ] Atkinson 2007 3/100 7/103 2.36).63.95% n/N n/N CI CI Addo-Yobo 2004 167/857 161/845 61. 1.00. Outcome 6 Failure rates on day 3.

1. Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia. df = 5 (P = 0. Chi2 = 11. 0.Random.89 (P = 0.05).09.6 % 0.98 [ 0.36 [ 0.1 1 10 100 Oral treatment Parenteral treatment Antibiotics for community-acquired pneumonia in children (Review) 174 Copyright © 2013 The Cochrane Collaboration. Outcome 7 Failure rates on day 6.61. Analysis 27.7.3 % 0. 1. I2 =55% Test for overall effect: Z = 0.84 [ 0.82 [ 0.24 ] Total events: 291 (Oral treatment).09 ] Total (95% CI) 2174 2157 100. 3.56.16.0 % 0. 319 (Parenteral treatment) Heterogeneity: Tau2 = 0.7 % 0. . 1.9 % 2. 9.95% n/N n/N CI CI Addo-Yobo 2004 185/857 187/845 38. Review: Antibiotics for community-acquired pneumonia in children Comparison: 27 Oral versus parenteral antibiotics for treatment of severe pneumonia Outcome: 7 Failure rates on day 6 Study or subgroup Oral treatment Parenteral treatment Odds Ratio Weight Odds Ratio M.11 ] Sidal 1994 2/46 14/42 5.57 ] Hazir 2008 89/1025 105/1012 35.59.22 ] Atkinson 2007 7/103 3/100 6.27.39 ] Campbell 1988 5/66 5/65 7.43 ] Tsarouhas 1998 3/77 5/93 6.01 0.77.09 [ 0. 3.37) Test for subgroup differences: Not applicable 0. Published by John Wiley & Sons.Random. M- H.97 [ 0.95% H.02.11.2 % 0. Ltd.71 [ 0.2 % 0.

27. Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia.11 ] Total (95% CI) 1948 1922 100. Published by John Wiley & Sons.9 % 0.32) Test for subgroup differences: Not applicable 0.8.98 [ 0.22 ] Campbell 1988 5/66 5/65 1.01 0.95% n/N n/N CI CI Addo-Yobo 2004 185/857 187/845 61.0 % 0.57 ] Hazir 2008 89/1025 105/1012 36. Chi2 = 0. Outcome 8 Failure rate in children below 5 years of age.76. df = 2 (P = 0.76.00 (P = 0. . 3.77.Random. 1. Analysis 27.1 1 10 100 Oral treatment Parenteral treatment Antibiotics for community-acquired pneumonia in children (Review) 175 Copyright © 2013 The Cochrane Collaboration.82 [ 0.0. 1.91 [ 0.09 ] Total events: 279 (Oral treatment).61.97 [ 0.69). Review: Antibiotics for community-acquired pneumonia in children Comparison: 27 Oral versus parenteral antibiotics for treatment of severe pneumonia Outcome: 8 Failure rate in children below 5 years of age Study or subgroup Oral treatment Parenteral treatment Odds Ratio Weight Odds Ratio M.3 % 0.0% Test for overall effect: Z = 1. Ltd. 1. I2 =0.95% H. M- H. 297 (Parenteral treatment) Heterogeneity: Tau2 = 0.8 % 0.Random.

3. Analysis 27.5 % 0. 1. M- H.7 % 2.59. 1. Chi2 = 2. Review: Antibiotics for community-acquired pneumonia in children Comparison: 27 Oral versus parenteral antibiotics for treatment of severe pneumonia Outcome: 9 Failure rates in children receiving oral amoxicillin or injectable antibiotics Study or subgroup Oral treatment Parenteral treatment Odds Ratio Weight Odds Ratio M.82 [ 0. 9. 300 (Parenteral treatment) Heterogeneity: Tau2 = 0.5 % 0.97 [ 0.0.Random. Ltd. Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia.77.10 ] Total events: 284 (Oral treatment).37) Test for subgroup differences: Not applicable 0.0% Test for overall effect: Z = 0. Outcome 9 Failure rates in children receiving oral amoxicillin or injectable antibiotics.3 % 0. Published by John Wiley & Sons.95% n/N n/N CI CI Addo-Yobo 2004 185/857 187/845 60.95% H.92 [ 0.71 [ 0.61.Random.9.36 [ 0.16.0 % 0.09 ] Total (95% CI) 2062 2050 100. df = 3 (P = 0.11 ] Tsarouhas 1998 3/77 5/93 1.45). 1.22 ] Atkinson 2007 7/103 3/100 1.01 0. .89 (P = 0.1 1 10 100 Oral treatment Parenteral treatment Antibiotics for community-acquired pneumonia in children (Review) 176 Copyright © 2013 The Cochrane Collaboration.65.77. I2 =0.39 ] Hazir 2008 89/1025 105/1012 36.

Analysis 27.71 [ 0.09 ] Total (95% CI) 1214 1212 100. . Outcome 10 Failure rate in children receiving cotrimoxazole or injectable penicillin.27.1 1 10 100 Oral treatment Parenteral treatment Analysis 27.09 [ 0.03. Review: Antibiotics for community-acquired pneumonia in children Comparison: 27 Oral versus parenteral antibiotics for treatment of severe pneumonia Outcome: 10 Failure rate in children receiving cotrimoxazole or injectable penicillin Study or subgroup Oral treatment Parenteral treatment Odds Ratio Weight Odds Ratio M.02.3 % 0.57 ] Sidal 1994 2/46 14/42 48.1 1 10 100 Oral treatment Parenteral treatment Antibiotics for community-acquired pneumonia in children (Review) 177 Copyright © 2013 The Cochrane Collaboration. Chi2 = 5.56 [ 0.24. 3.16.95% n/N n/N CI CI Campbell 1988 5/66 5/65 51. df = 1 (P = 0. M- H. 1.45. Ltd. I2 =82% Test for overall effect: Z = 0. Chi2 = 7.1 % 0. df = 3 (P = 0. Published by John Wiley & Sons. 0. 3. 19 (Parenteral treatment) Heterogeneity: Tau2 = 2.01 0. M- H.98 [ 0.11. I2 =61% Test for overall effect: Z = 1.27. 0.7 % 0.Random. Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia.19) Test for subgroup differences: Not applicable 0. 3.0 % 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 27 Oral versus parenteral antibiotics for treatment of severe pneumonia Outcome: 11 Failure rate in children treated with oral or parenteral antibiotics on ambulatory basis Study or subgroup Oral treatment Parenteral treatment Odds Ratio Weight Odds Ratio M.65.9 % 0.95% H.11 ] Sidal 1994 2/46 14/42 17.Random.Random.82 [ 0.9 % 0.97 (P = 0.0 % 0.05). Outcome 11 Failure rate in children treated with oral or parenteral antibiotics on ambulatory basis.29 ] Total events: 7 (Oral treatment). 3. 1. Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia.09 [ 0.02.98 [ 0.1 % 0.61.33) Test for subgroup differences: Not applicable 0.95% n/N n/N CI CI Campbell 1988 5/66 5/65 21.44.32 ] Total events: 99 (Oral treatment).01 0. 129 (Parenteral treatment) Heterogeneity: Tau2 = 0.57 ] Hazir 2008 89/1025 105/1012 41.95% H.43 ] Total (95% CI) 112 107 100.Random.02).31 [ 0.32 (P = 0.37.43 ] Tsarouhas 1998 3/77 5/93 19.10.

95% n/N n/N CI CI Atkinson 2007 7/103 3/100 28. 2.59.82 [ 0.11 [ 0. M- H.3 % 0.39 ] Hazir 2008 89/1025 105/1012 71. 9. Outcome 12 Failure rate after removing one study. Review: Antibiotics for community-acquired pneumonia in children Comparison: 27 Oral versus parenteral antibiotics for treatment of severe pneumonia Outcome: 12 Failure rate after removing one study Study or subgroup Oral treatment Parenteral treatment Odds Ratio Weight Odds Ratio M.7 % 2.01 0.61.14.Random.Random. I2 =53% Test for overall effect: Z = 0. Ltd.83 ] Total events: 96 (Oral treatment).36 [ 0.11 ] Total (95% CI) 1128 1112 100. . Chi2 = 2.12.30. Analysis 27. 108 (Parenteral treatment) Heterogeneity: Tau2 = 0. df = 1 (P = 0.14). Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia. Published by John Wiley & Sons.44.82) Test for subgroup differences: Not applicable 0. 1.95% H.1 1 10 100 Oral treatment Parenteral treatment Antibiotics for community-acquired pneumonia in children (Review) 178 Copyright © 2013 The Cochrane Collaboration.0 % 1.22 (P = 0.

37 ] Hazir 2008 25/948 31/925 65.1 1 10 100 Oral treatment Parenteral treatment Antibiotics for community-acquired pneumonia in children (Review) 179 Copyright © 2013 The Cochrane Collaboration.72 ] Total (95% CI) 192 266 100.38. Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia.13.13 [ 0.34. 9.96 [ 0.28 [ 0.82 ] Total events: 31 (Oral treatment). Outcome 14 Relapse rates. Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia.15. 16. Outcome 13 Hospitalisation.24. Analysis 27.72) Test for subgroup differences: Not applicable 0.0 % 1. 4. M- H. df = 1 (P = 0.50 [ 0.1 1 10 100 Oral treatment Parenteral treatment Analysis 27. Chi2 = 2.65.Random. Ltd.36 (P = 0.83) Test for subgroup differences: Not applicable 0.0. 7 (Parenteral treatment) Heterogeneity: Tau2 = 0.5 % 1. I2 =62% Test for overall effect: Z = 0. .Random.2 % 0.33 ] Total (95% CI) 1048 1028 100.63. Chi2 = 0. I2 =0. 1.70).25.0 % 1. Published by John Wiley & Sons.95% n/N n/N CI CI Campbell 1988 3/69 2/68 35. 33 (Parenteral treatment) Heterogeneity: Tau2 = 0.95% H.8 % 3. Review: Antibiotics for community-acquired pneumonia in children Comparison: 27 Oral versus parenteral antibiotics for treatment of severe pneumonia Outcome: 14 Relapse rates Study or subgroup Oral treatment Parenteral treatment Odds Ratio Weight Odds Ratio M.78 [ 0. 3. Review: Antibiotics for community-acquired pneumonia in children Comparison: 27 Oral versus parenteral antibiotics for treatment of severe pneumonia Outcome: 13 Hospitalisation Study or subgroup Oral treatment Parenteral treatment Odds Ratio Weight Odds Ratio M.95% n/N n/N CI CI Atkinson 2007 6/100 2/103 34.95% H.14.46. M- H.01 0.Random.22 (P = 0. df = 1 (P = 0.5 % 0.0% Test for overall effect: Z = 0.27 ] Sidal 1994 0/46 0/105 Not estimable Tsarouhas 1998 4/77 5/93 64. 3.63.01 0.22 [ 0.34 ] Total events: 7 (Oral treatment).10).Random.

Random.25 [ 0.47).03.00. Ltd.12 (P = 0. Chi2 = 0. Outcome 15 Death rates.95% n/N n/N CI CI Addo-Yobo 2004 0/845 7/857 37. 11 (Parenteral treatment) Heterogeneity: Tau2 = 0. 1. 0.53.0% Test for overall effect: Z = 2.0 % 0.18 ] Atkinson 2007 0/100 0/103 Not estimable Hazir 2008 1/1025 4/1012 63.21 ] Total (95% CI) 1970 1972 100. . Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia. Analysis 27. 2.01 0.001 0.15. df = 1 (P = 0.87 ] Total events: 1 (Oral treatment). M- H.95% H. I2 =0.07 [ 0.03.0.0 % 0.0 % 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 27 Oral versus parenteral antibiotics for treatment of severe pneumonia Outcome: 15 Death rates Study or subgroup Oral treatment Parenteral treatment Odds Ratio Weight Odds Ratio M.034) Test for subgroup differences: Not applicable 0.Random. Published by John Wiley & Sons.15 [ 0.1 1 10 100 1000 Oral treatment Parenteral treatment Antibiotics for community-acquired pneumonia in children (Review) 180 Copyright © 2013 The Cochrane Collaboration.

df = 1 (P = 0.20 (P = 0.11) Test for subgroup differences: Not applicable 0.Random.64.16.54 [ 0.17.0 % 0. 1. 52.Random.32. 1. Ltd.20 ] Total events: 6 (Oral treatment).19.95% H. M- H. 10.95% n/N n/N CI CI Hazir 2008 6/1025 13/1012 100.52.028) Test for subgroup differences: Not applicable 0.60 (P = 0. 141 (Parenteral treatment) Heterogeneity: Tau2 = 0.95% n/N n/N CI CI Atkinson 2007 123/126 113/120 53. Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia.1 % 2. Review: Antibiotics for community-acquired pneumonia in children Comparison: 27 Oral versus parenteral antibiotics for treatment of severe pneumonia Outcome: 17 Cure rate Study or subgroup Oral treatment Parenteral treatment Odds Ratio Weight Odds Ratio M.9 % 11.33 ] Total events: 167 (Oral treatment).Random.93.06 ] Sidal 1994 44/46 28/42 46. Analysis 27.1 1 10 100 Oral treatment Parenteral treatment Analysis 27. 21. Review: Antibiotics for community-acquired pneumonia in children Comparison: 27 Oral versus parenteral antibiotics for treatment of severe pneumonia Outcome: 16 Lost to follow-up Study or subgroup Oral treatment Parenteral treatment Odds Ratio Weight Odds Ratio M.16). Published by John Wiley & Sons.01 0.20 ] Total (95% CI) 1025 1012 100.Random. 13 (Parenteral treatment) Heterogeneity: not applicable Test for overall effect: Z = 1.0 % 5.01 0.1 1 10 100 Oral treatment Parenteral treatment Antibiotics for community-acquired pneumonia in children (Review) 181 Copyright © 2013 The Cochrane Collaboration.0 % 0.00 [ 2. Outcome 16 Lost to follow-up. M- H.45 [ 0. Chi2 = 1. Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia.45 [ 0.05 [ 1.11 ] Total (95% CI) 172 162 100.95% H.17.17. I2 =48% Test for overall effect: Z = 2. . Outcome 17 Cure rate.

21 ] Total (95% CI) 1125 1115 100.1 1 10 100 Oral treatment Parenteral treatment Analysis 27.39 ] Tsarouhas 1998 3/77 5/93 47.59.35.0 % 0. 2. Outcome 19 Death rates after removing one study. df = 1 (P = 0.33 [ 0. Outcome 18 Failure rates in radiographically confirmed-pneumonia.21) Test for subgroup differences: Not applicable 0.95% CI Atkinson 2007 0/100 0/103 Not estimable Hazir 2008 1/1025 4/1012 100. I2 =26% Test for overall effect: Z = 0.1 1 10 100 Oral treatment Parenteral treatment Antibiotics for community-acquired pneumonia in children (Review) 182 Copyright © 2013 The Cochrane Collaboration.Fixed.0 % 1. Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia.19.95% n/N n/N CI CI Atkinson 2007 7/103 3/100 52.0 % 0.21 ] Total events: 1 (Oral treatment).03. M- H.29 ] Total events: 10 (Oral treatment). 9.03.Fixed.25 [ 0.24). Review: Antibiotics for community-acquired pneumonia in children Comparison: 27 Oral versus parenteral antibiotics for treatment of severe pneumonia Outcome: 19 Death rates after removing one study Study or subgroup Oral treatment Parenteral treatment Odds Ratio Weight Odds Ratio n/N n/N M-H. 4.95% CI M-H.Random.1 % 2.Random.36 [ 0. 4 (Parenteral treatment) Heterogeneity: not applicable Test for overall effect: Z = 1.01 0.95% H. .25 [ 0.48 (P = 0.25 (P = 0. 8 (Parenteral treatment) Heterogeneity: Tau2 = 0.9 % 0. Ltd. 2.63) Test for subgroup differences: Not applicable 0.41.09 ] Total (95% CI) 180 193 100. Published by John Wiley & Sons. Chi2 = 1.19. 3.18.16. Analysis 27.01 0.71 [ 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 27 Oral versus parenteral antibiotics for treatment of severe pneumonia Outcome: 18 Failure rates in radiographically confirmed-pneumonia Study or subgroup Oral treatment Parenteral treatment Odds Ratio Weight Odds Ratio M. Comparison 27 Oral versus parenteral antibiotics for treatment of severe pneumonia.

Ltd.0 % 0.95% n/N n/N CI CI CATCHUP 2002 372/734 0/50 50.47 ] Straus 1998 227/398 0/50 49. 848. Review: Antibiotics for community-acquired pneumonia in children Comparison: 28 Co-trimoxazole versus co-amoxyclavulanic acid Outcome: 2 Male sex Co- amoxyclavulanic Study or subgroup Co-trimoxazole acid Odds Ratio Weight Odds Ratio M.1 1 10 100 1000 Co-trimoxazole Co-amoxyclavulanic acid Analysis 28.00001) Test for subgroup differences: Not applicable 0.95% H. 2186.2.78 ] Straus 1998 246/398 35/50 49.31 ] Total (95% CI) 1132 100 100.01 0. 70 (Co-amoxyclavulanic acid) Heterogeneity: Tau2 = 0. Chi2 = 0.54 [ 0.1 % 103.46 (P = 0.38.014) Test for subgroup differences: Not applicable 0. df = 1 (P = 0. 1.95% H.0.02.37 ] Total events: 599 (Co-trimoxazole).74 (P < 0. Comparison 28 Co-trimoxazole versus co-amoxyclavulanic acid. 1688.69 [ 0.87 ] Total (95% CI) 1132 100 100.Random.0 % 117.90).90 [ 16.5 1 2 5 10 Co-trimoxazole Co-amoxyclavulanic acid Antibiotics for community-acquired pneumonia in children (Review) 183 Copyright © 2013 The Cochrane Collaboration. Published by John Wiley & Sons. 0 (Co-amoxyclavulanic acid) Heterogeneity: Tau2 = 0.Random.1 0.88 ] Total events: 608 (Co-trimoxazole).21.001 0. I2 =0.0% Test for overall effect: Z = 4.03. Review: Antibiotics for community-acquired pneumonia in children Comparison: 28 Co-trimoxazole versus co-amoxyclavulanic acid Outcome: 1 Children below 1 year of age Co- amoxyclavulanic Study or subgroup Co-trimoxazole acid Odds Ratio Weight Odds Ratio M.95% n/N n/N CI CI CATCHUP 2002 362/734 35/50 51.Random. M- H. Analysis 28.98 [ 8.1. 0.42 [ 0.22. Outcome 1 Children below 1 year of age.26).33.39. M- H.0 % 0.79 [ 6.0 % 0.Random. . Comparison 28 Co-trimoxazole versus co-amoxyclavulanic acid.37. df = 1 (P = 0.9 % 133. I2 =20% Test for overall effect: Z = 2.2 0.25. Chi2 = 1. Outcome 2 Male sex. 0.

Analysis 28.18. Chi2 = 0.1 % 11.1 1 10 500 Co-trimoxazole Co-amoxyclavulanic acid Antibiotics for community-acquired pneumonia in children (Review) 184 Copyright © 2013 The Cochrane Collaboration. 53. Review: Antibiotics for community-acquired pneumonia in children Comparison: 28 Co-trimoxazole versus co-amoxyclavulanic acid Outcome: 3 Failure rate Co- amoxyclavulanic Study or subgroup Co-trimoxazole acid Odds Ratio Weight Odds Ratio M.0% Test for overall effect: Z = 3.Random.9 % 14.002 0. df = 1 (P = 0. Comparison 28 Co-trimoxazole versus co-amoxyclavulanic acid. 83. 108.01.Random.0. Ltd.3. Published by John Wiley & Sons. Outcome 3 Failure rate. I2 =0.06 ] Total events: 231 (Co-trimoxazole). M- H.15 ] Total (95% CI) 1132 100 100.00036) Test for subgroup differences: Not applicable 0.57.45 [ 1.95% H.95% n/N n/N CI CI CATCHUP 2002 139/734 1/50 50. .0 % 12. 2 (Co-amoxyclavulanic acid) Heterogeneity: Tau2 = 0.57 (P = 0.03.73 [ 2.61 ] Straus 1998 92/398 1/50 49.98 [ 3.86).

2 0.6 (0) Not estimable Total (95% CI) 50 234 Not estimable Heterogeneity: not applicable Test for overall effect: not applicable Test for subgroup differences: Not applicable -10 -5 0 5 10 Amoxycillin Cefpodoxime Analysis 29. 44.71 [ 0.5 1 2 5 10 Amoxycillin Cefpodoxime Antibiotics for community-acquired pneumonia in children (Review) 185 Copyright © 2013 The Cochrane Collaboration. Outcome 2 Male sex.1 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 29 Amoxycillin versus cefpodoxime Outcome: 2 Male sex Study or subgroup Amoxycillin Cefpodoxime Odds Ratio Weight Odds Ratio M.95% CI IV.71 [ 0. Ltd. .0 % 1.32 (P = 0. Comparison 29 Amoxycillin versus cefpodoxime. Outcome 1 Age in months.Random.0 % 1.09 ] Total events: 18 (Amoxycillin).07. 0 (Cefpodoxime) Heterogeneity: not applicable Test for overall effect: Z = 0.95% n/N n/N CI CI Jibril 1989 18/50 0/1 100.1.95% H.95% CI Jibril 1989 50 58. Comparison 29 Amoxycillin versus cefpodoxime.Random.2.09 ] Total (95% CI) 50 1 100. Review: Antibiotics for community-acquired pneumonia in children Comparison: 29 Amoxycillin versus cefpodoxime Outcome: 1 Age in months Mean Mean Study or subgroup Amoxycillin Cefpodoxime Difference Weight Difference N Mean(SD) N Mean(SD) IV.8 (0) 234 21. Analysis 29.Random.75) Test for subgroup differences: Not applicable 0. 44. M- H. Published by John Wiley & Sons.Random.07.

0 % -6.95% CI CATCHUP 2002 725 12 (2.54) 55 16. Ltd.2 0.30 (P = 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 29 Amoxycillin versus cefpodoxime Outcome: 3 Response/cure rate Study or subgroup Amoxycillin Cefpodoxime Odds Ratio Weight Odds Ratio M. Outcome 1 Age (mean/median).Random.60 [ -6.0 % 0.001).20 [ 0.52.6 (6. Chi2 = 10.Random. 0.Random.53 ] Total events: 40 (Amoxycillin).45. Review: Antibiotics for community-acquired pneumonia in children Comparison: 30 Amoxycillin versus chloramphenicol Outcome: 1 Age (mean/median) Mean Mean Study or subgroup Amoxycillin Chloramphenicol Difference Weight Difference N Mean(SD) N Mean(SD) IV.9 % -8. df = 1 (P = 0.08.1 % -4. M- H.0011) Test for subgroup differences: Not applicable 0.08.33. Comparison 29 Amoxycillin versus cefpodoxime. 0.6 (6.95% H.60 [ -10.30. 179 (Cefpodoxime) Heterogeneity: not applicable Test for overall effect: Z = 3.95% CI IV.0 % 0. -2. -2.4) 50.00097) Test for subgroup differences: Not applicable -10 -5 0 5 10 Amoxycillin Chloramphenicol Antibiotics for community-acquired pneumonia in children (Review) 186 Copyright © 2013 The Cochrane Collaboration. Outcome 3 Response/cure rate.1. Published by John Wiley & Sons.60 [ -10. Comparison 30 Amoxycillin versus chloramphenicol.53 ] Total (95% CI) 50 188 100.90 ] Mulholland 1995 197 8 (2.87 ] Total (95% CI) 922 110 100.20 [ 0. .4) 49.95% n/N n/N CI CI Jibril 1989 40/50 179/188 100.1 0. Analysis 29.Random.68 ] Heterogeneity: Tau2 = 7.5 1 2 5 10 Amoxycillin Cefpodoxime Analysis 30.54) 55 16.26 (P = 0.23.3. I2 =90% Test for overall effect: Z = 3. -6.

1 % 1.34 [ 1.26 [ 2.08 ] Total events: 608 (Amoxycillin). Published by John Wiley & Sons.60.1 0. M- H. Comparison 30 Amoxycillin versus chloramphenicol.2.11.32).5 1 2 5 10 Amoxycillin Chloramphenicol Antibiotics for community-acquired pneumonia in children (Review) 187 Copyright © 2013 The Cochrane Collaboration.0 % 4.000050) Test for subgroup differences: Not applicable 0. . 48 (Chloramphenicol) Heterogeneity: Tau2 = 0.95% H. Chi2 = 0. df = 1 (P = 0.99. 5.57.37 ] Straus 1998 137/197 24/55 44. Analysis 30.0 % 4.0.60 (P < 0.94 [ 1.5 1 2 5 10 Amoxycillin Chloramphenicol Analysis 30.95 [ 1.45 ] Total (95% CI) 922 110 100.Random.Random. I2 =0.Random.08 ] Total (95% CI) 725 71 100. Outcome 2 Male sex.1 0.06 (P = 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 30 Amoxycillin versus chloramphenicol Outcome: 3 Cure rate Study or subgroup Amoxycillin Chloramphenicol Odds Ratio Weight Odds Ratio M.2 0.95% H.26 [ 2.3.0 % 2. Ltd.95% n/N n/N CI CI CATCHUP 2002 435/725 24/55 55.57. 7. 3.Random.9 % 2.00001) Test for subgroup differences: Not applicable 0. Review: Antibiotics for community-acquired pneumonia in children Comparison: 30 Amoxycillin versus chloramphenicol Outcome: 2 Male sex Study or subgroup Amoxycillin Chloramphenicol Odds Ratio Weight Odds Ratio M. Comparison 30 Amoxycillin versus chloramphenicol.55.2 0. M- H.0% Test for overall effect: Z = 4.95% n/N n/N CI CI CATCHUP 2002 608/725 39/71 100. 39 (Chloramphenicol) Heterogeneity: not applicable Test for overall effect: Z = 5. Outcome 3 Cure rate. 3.53 ] Total events: 572 (Amoxycillin). 7.

64 [ 0.0% Test for overall effect: Z = 1.95% n/N n/N CI CI CATCHUP 2002 117/725 16/71 56.0. 32 (Chloramphenicol) Heterogeneity: Tau2 = 0. df = 1 (P = 0. Chi2 = 0.1 0.31.2 0. Bacterial isolation Study/total tested S. 1. Analysis 30.Random. Review: Antibiotics for community-acquired pneumonia in children Comparison: 30 Amoxycillin versus chloramphenicol Outcome: 4 Failure rates Study or subgroup Amoxycillin Chloramphenicol Odds Ratio Weight Odds Ratio M.1 % 0.19 ] Straus 1998 30/198 16/71 43. I2 =0.41.61 [ 0. 1.96 (P = 0. Outcome 4 Failure rates.66 [ 0.21 ] Total (95% CI) 923 142 100.95% H. pneumoniae H. . influenzae Staphylococcus Others Asghar 2008/958 22 8 47 33 Bansal 2006/71 3 2 0 0 Block 1995/122 2 2 0 0 Bradley 2007/709 21 7 0 3 Camargos 1997/90 6 0 0 0 Duke 2002/1116 4 10 10 36 Harris 1998/351 5 2 0 0 Klein 1995/348 14 28 0 17 Kogan 2003/47 7 0 0 4 Mulholland 1995/111 10 2 0 8 Antibiotics for community-acquired pneumonia in children (Review) 188 Copyright © 2013 The Cochrane Collaboration. Ltd. M- H.4.5 1 2 5 10 Amoxycillin Chloramphenicol ADDITIONAL TABLES Table 1. Comparison 30 Amoxycillin versus chloramphenicol. Published by John Wiley & Sons.37.Random.0 % 0.87). 1.9 % 0.050) Test for subgroup differences: Not applicable 0.00 ] Total events: 147 (Amoxycillin).03.

Table 1.DRR #5 (antibiotic* in ti) or (antibiotic* in ab) #6 #4 or #5 Antibiotics for community-acquired pneumonia in children (Review) 189 Copyright © 2013 The Cochrane Collaboration.DER.0%) 69 (1. Issue 2).236/591 (40%) 150/591 (25%) 69/591 (12%) 136/591 (23%) lates (591) APPENDICES Appendix 1. . Ovid format (Lefebvre 2008). Published by John Wiley & Sons.DRR #2 (pneumonia in ti) or (pneumonia in ab) #3 #1 or #2 #4 ’antibiotic-agent’ / all subheadings in DEM. There were no language or publication restrictions.DER. Bacterial isolation (Continued) Nogeova 1997/140 24 21 11 22 Roord 1996/95 11 19 1 13 Straus 1998/595 79 49 0 0 Wubbel 1999/129 35 0 0 0 Total/4882 (12%) 236 (4.8%) 150 (3. MEDLINE (OVID) 1 exp PNEUMONIA/ 2 pneumonia 3 or/1-2 4 exp Anti-Bacterial Agents/ 5 antibiotic$ 6 or/4-5 7 exp CHILD/ 8 exp INFANT/ 9 (children or infant$ or pediatric or paediatric) 10 or/7-9 11 3 and 6 and 10 EMBASE (WebSPIRS) #1 explode ’pneumonia-’ / all subheadings in DEM. MEDLINE (1966 to September 2009) and EMBASE (1990 to September 2009).and precision-maximising version (2008 revision). Ltd.DRM.8%) Out of total bacterial iso. We combined the MEDLINE search with the Cochrane Highly Sensitive Search Strategy for identifying randomised trials in MEDLINE: sensitivity. Previous search strategy We searched the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library 2009.4%) 136 (2. which contains the Acute Respiratory Infections Group’s Specialised Register.DRM.

ti OR (school* NEAR/1 (nursery OR primary OR secondary OR elementary OR high)):ab.ti OR schoolchild*:ab.ti OR cotrimoxazole: ab.ti OR ampicillin:ab.#7 ’child-’ / all subheadings in DEM.ti OR ’cross-over’:ab.ti OR erythromycin:ab.ti OR ’beta-lactams’:ab.ti #1.ti OR pediatric*:ab.ti OR cephalosporin*: ab.ti OR paediatric*:ab.ti OR babies:ab.ti OR boy*:ab.ti OR genamycin:ab.ti OR cephradine:ab.ti OR bronchopneumon*:ab.ti OR levofloxacin:ab.4 #1.ti OR allocat*:ab.3 #1.2 OR #1.ti OR cefpodixime:ab.ti #4 (school NEAR/2 (age* OR child*)):ab.ti OR sulfamethoxazole:ab.ti OR placebo*:ab.ti OR trimethoprim:ab. Published by John Wiley & Sons.ti #1.ti OR baby*:ab.ti OR adolescen*:ab.8 #1.ti OR benzylpenicillin:ab. .DRM.ti OR augmentin:ab.DER.ti OR ’beta-lactam’:ab.1 ’pneumonia’/exp Antibiotics for community-acquired pneumonia in children (Review) 190 Copyright © 2013 The Cochrane Collaboration.ti OR ’co-trimoxazole’:ab.6 antibiotic*:ab.1 OR #1.ti OR amoxicillin:ab.DRM.ti OR minocyclin:ab.ti OR teen*:ab.ti OR roxithromycin:ab.ti OR clarithromycin:ab.ti OR assign*:ab.DRR #8 (child in ti) or (child in ab) #9 (children in ti) or (children in ab) #10 ’infant-’ / all subheadings in DEM.5 ’antibiotic agent’/exp #1.ti OR kid:ab.ti #1.ti OR factorial*:ab.5 OR #1.ti OR crossover*:ab.ti OR cephalexin:ab.ti OR highschool*:ab.ti OR pleuropneumon*:ab.ti OR chlorampheni- col:ab.ti OR volunteer*: ab.ti OR cefotaxime:ab.ti OR toddler*:ab.2 pneumon*:ab.DER.ti OR cap:ab.ti OR girl*:ab.8 #1.ti #7 #1 AND #6 #6 #2 OR #3 OR #4 OR #5 #5 kindergar*:ab.ti OR minor*:ab.ti OR puberty:ab.7 #1.ti OR cefetanet:ab.ti OR child*:ab.ti #1.ti OR preschool*:ab.7 amoxycillin:ab.ti OR infanc*:ab.ti OR cefaclor:ab.ti OR ceftriaxone:ab.ti OR ((singl* OR doubl*) NEAR/1 blind*):ab.ti OR ’cross over’:ab.ti OR sulphamethoxazole:ab.ti OR cefditoren:ab. EMBASE search strategy #9 #7 AND #8 #8 ’randomized controlled trial’/exp OR ’single blind procedure’/exp OR ’double blind procedure’/exp OR ’crossover procedure’/exp OR (random*:ab.DRR #11 (infant* in ti) or (infant* in ab) #12 #7 or #8 or #9 or #10 or #11 #13 #3 and #6 and #12 #14 explode ’randomized-controlled-trial’ / all subheadings #15 explode ’controlled-study’ / all subheadings #16 explode ’single-blind-procedure’ / all subheadings #17 explode ’double-blind-procedure’ / all subheadings #18 explode ’crossover-procedure’ / all subheadings #19 explode ’phase-3-clinical-trial’ / all subheadings #20 (randomi?ed controlled trial in ti) or (randomi?ed controlled trial in ab) #21 ((random* or placebo* or double-blind*)in ti) or ((random* or placebo* or double-blind*)in ab) #22 (controlled clinical trial* in ti) or (controlled clinical trial* in ab) #23 #14 or #15 or #16 or #17 or 318 or #19 or #290 or #21 or #22 #24 (nonhuman in der) not ((human in der) and (nonhuman in der)) #25 #23 not #24 #26 #13 and #25 Appendix 2.ti #3 infant*:ab.ti OR penicllin*:ab.9 #1. Ltd.ti OR cefuroxime:ab.ti OR azithromycin:ab.3 ’community-acquired-pneumonia’:ab.6 OR #1.ti OR ceftrioxone:ab.ti OR kids:ab.ti #2 ’infant’/exp OR ’child’/exp OR ’adolescent’/exp OR ’pediatrics’/exp OR ’juvenile’/exp OR ’puberty’/exp #1 #1.ti OR gentamicin:ab.ti OR quinolone*:ab.4 AND #1.ti OR ’b-lactam’:ab.ti OR newborn*:ab.ti OR moxifloxacin: ab.

Ltd. Secondary”) OR (MH “Schools. Elementary”) OR (MH “Schools. . Special”) 8229 Edit S32 S31 TI (pediatric* or paediatric*) OR AB (pediatric* or paediatric*) S30 (MH “Pediatrics+”) S29 TI (minor* or pubert* or pubescen*) OR AB (minor* or pubert* or pubescen*) S28 (MH “Puberty”) S27 (adoles* or teen* or boy* or girl*) OR (adoles* or teen* or boy* or girl*) S26 (MH “Adolescence+”) S25 TI (child* or schoolchild* or school age* or preschool* or kid or kids or toddler*) OR AB (child* or schoolchild* or school age* or preschool* or kid or kids or toddler*) S24 (MH “Child+”) S23 TI (infant* or infancy or newborn* or baby* or babies or neonat* or preterm* or prematur*) OR AB (infant* or infancy or newborn* or baby* or babies or neonat* or preterm* or prematur*) 66234 S22 (MH “Infant+”) S21 (S11 and S20) S20 S12 or S13 or S14 or S15 or S16 or S17 or S18 or S19 S19 (MH “Placebos”) S18 (MH “Quantitative Studies”) S17 TI placebo* OR AB placebo* S16 TI random* OR AB random* S15 TI (singl* blind* or doubl* blind* or tripl* blind* or trebl* blind* or singl* mask* or doubl* mask* or trebl* mask* or tripl* mask*) OR AB (singl* blind* or doubl* blind* or tripl* blind* or trebl* blind* or singl* mask* or doubl* mask* or trebl* mask* or tripl* mask*) S14 TI clinic* trial* OR AB clinic* trial* S13 PT clinical trial S12 (MH “Clinical Trials+”) S11 S9 and S10 S10 S5 or S6 or S7 or S8 S9 S1 or S2 or S3 or S4 S8 AB amoxycillin* or amoxicillin* or ampicillin* or azithromycin* or augmentin* or benzylpenicillin* or b-lactam* or beta-lactam* or clarithromycin* or ceftriaxone* or cefuroxime* or cotrimoxazole* or co-trimoxazole* or co-amoxyclavulanic acid or cefotaxime* or ceftriaxone* or ceftrioxone* or cefditoren* or chloramphenicol* or cefpo- dioxime* or cephradine* or cephalexin* or cefaclor* or cefetamet* or cephalosporin* or erythromycin* or gentamicin* or gentamycin* or levofloxacin* or macrolide* or minocyclin* or moxifloxacin* or penicillin* or quinolone* or roxithromycin* or sulphamethoxazole* or sulfamethoxazole* or tetracycline* or trimethoprim* S7 TI amoxycillin* or amoxicillin* or ampicillin* or azithromycin* or augmentin* or benzylpenicillin* or b-lactam* or beta-lactam* or clarithromycin* or ceftriaxone* or cefuroxime* or cotrimoxazole* or co-trimoxazole* or co-amoxyclavulanic acid or cefotaxime* or ceftriaxone* or ceftrioxone* or cefditoren* or chloramphenicol* or cefpo- dioxime* or cephradine* or cephalexin* or cefaclor* or cefetamet* or cephalosporin* or erythromycin* or gentamicin* or gentamycin* or levofloxacin* or macrolide* or minocyclin* or moxifloxacin* or penicillin* or quinolone* or roxithromycin* or sulphamethoxazole* or sulfamethoxazole* or tetracycline* or trimethoprim* S6 TI (antibiotic* or antibacter* or anti-bacter*) OR AB (antibiotic* or antibacter* or anti-bacter*) S5 (MH “Antibiotics+”) S4 TI cap OR AB cap S3 TI (bronchopneumon* or pleuropneumon*) OR AB (bronchopneumon* or pleuropneumon*) S2 TI pneumon* OR AB pneumon* Antibiotics for community-acquired pneumonia in children (Review) 191 Copyright © 2013 The Cochrane Collaboration. Nursery”) OR (MH “Schools.Appendix 3. Middle”) OR (MH “Schools. CINAHL (Ebsco) search strategy S37 S35 and S36 S36 S11 and S20 S35 S22 or S23 or S24 or S25 or S26 or S27 or S28 or S29 or S30 or S31 or S32 or S33 or S34 S34 TI school* OR AB school* S33 TI (nursery school* or kindergar* or primary school* or secondary school* or elementary school* or high school* or highschool*) OR AB (nursery school* or kindergar* or primary school* or secondary school* or elementary school* or high school* or highschool*) S32 (MH “Schools”) OR (MH “Schools. Published by John Wiley & Sons.

414. CPCI-S Timespan=All Years Lemmatization=On Antibiotics for community-acquired pneumonia in children (Review) 192 Copyright © 2013 The Cochrane Collaboration.126 Topic=(infant* or infancy or newborn* or baby or babies or neonat* or preterm* or prematur* or child* or schoolchild* or “school age*” or preschool* or kid or kids or toddler* or adoles* or teen* or boy* or girl* or pediatric* or paediatric*) Databases=SCI-EXPANDED. CPCI-S Timespan=All Years Lemmatization=On #2 1. Ltd. CPCI-S Timespan=All Years Lemmatization=On #1 29. CPCI-S Timespan=All Years Lemmatization=On #5 796 #4 AND #3 Databases=SCI-EXPANDED.498 Topic=(pneumon* or bronchopneumon* or pleuropneumon* or cap) AND Topic=(antibiotic* or amoxycillin* or amoxicillin* or ampicillin* or azithromycin* or augmentin* or benzylpenicillin* or b-lactam* or beta-lactam* or clarithromycin* or ceftriaxone* or cefuroxime* or cotrimoxazole* or co-trimoxazole* or co-amoxyclavulanic acid or cefotaxime* or ceftriaxone* or ceftrioxone* or cefditoren* or chloramphenicol* or cefpodioxime* or cephradine* or cephalexin* or cefaclor* or cefetamet* or cephalosporin* or erythromycin* or gentamicin* or gentamycin* or levofloxacin* or macrolide* or minocyclin* or moxifloxacin* or penicillin* or quinolone* or roxithromycin* or sulphamethoxazole* or sulfamethoxazole* or tetracyclin* or trimethoprim*) Databases=SCI-EXPANDED.302. CPCI-S Timespan=All Years Lemmatization=On #4 1.084 Topic=(random* or placebo* or allocat* or crossover* or “cross over” or ((singl* or doubl*) NEAR/1 blind*)) OR Title=(trial) Databases=SCI-EXPANDED. CPCI-S Timespan=All Years Lemmatization=On #3 5.848 #2 AND #1 Databases=SCI-EXPANDED. Published by John Wiley & Sons.S1 (MH “Pneumonia+”) Appendix 4. . Web of Science (Thomson ISI) search strategy #6 113 #4 AND #3 Refined by: Publication Years=( 2011 OR 2012 ) Databases=SCI-EXPANDED.

381.677$ OR MH:C08.730.954. 2006 Antibiotics for community-acquired pneumonia in children (Review) 193 Copyright © 2013 The Cochrane Collaboration. . LILACS (Brieme) search strategy > Search > (MH:pneumonia OR Neumonía OR MH:C08. Date Event Description 7 November 2012 New search has been performed Searches updated. Ribeiro 2011) and excluded three new trials ( Ambroggio 2012.Appendix 5.505. Ltd. We included two new trials (Nogeova 1997.610$ OR Pulmonía OR “Inflamación Ex- perimental del Pulmón” OR “Inflamación del Pulmón” OR “Neumonía Lobar” OR Neumonitis OR “Inflamación Pulmonar” OR “Inflamação Experimental dos Pulmões” OR “Inflamação do Pulmão” OR “Pneumonia Lobar” OR Pneumonite OR “Inflamação Pulmonar” OR Pulmonia OR bronchopneumon$ OR Bronconeumonía OR Broncopneumonia OR Pleuropneumonia OR Pleuroneu- monía OR Pleuropneumonia) AND (MH:“Anti-Bacterial Agents” OR antibiotic$ OR Antibacterianos OR Antibióticos OR MH: D27. Soofi 2012).122.085$ OR MH:amoxicillin OR amoxicillin$ OR Amoxicilina OR MH:ampicillin OR Ampicilina OR ampicillin$ OR MH:Azithromycin OR Azitromicina OR azithromycin$ OR augmentin$ OR benzylpenillin OR MH:“penicillin g” OR “penicilina g” OR MH:“beta-lactams” OR “beta-lactams” OR “beta-lactamas” OR MH:clarithromycin OR claritromicina OR clarithromycin OR MH:ceftriaxone OR ceftriaxone OR ceftriaxona OR MH:cefroxime OR cefroxime$ OR cefuroxima OR cotrimoxazol$ OR “Trimetho- prim-Sulfamethoxazole Combination” OR “Combinación Trimetoprim-Sulfametoxazol” OR “Combinação Trimetoprima-sulfame- toxazol” OR “co-amoxyclavulanic acid” OR cefotaxime OR MH:cefotaxime OR cefotaxima OR MH:ceftriaxone OR ceftriaxone OR ceftriaxona OR ceftrioxone OR cefditoren$ OR chloramphenicol OR cloranfenicol OR MH:chloramphenicol OR cefpodixime OR MH:cephradine OR cephradin$ OR cefradina OR MH:cephalexin OR cefalexina OR cephalexin$ OR cefaclor OR MH:cefaclor OR cefetamet OR cephalosporin$ OR MH:cephalosporins OR cefalosporinas OR MH:erythromycin OR erythromycin OR eritomicina OR MH:gentamicins OR gentamicin$ OR gentamycin$ OR Gentamicinas OR levofloxacin OR MH:ofloxacin OR ofloxacin$ OR MH:macrolides OR macrolide$ OR Macrólidos OR Macrolídeos OR minocyclin$ OR MH:minocycline OR Minociclina OR moxi- floxacin OR penicillin$ OR MH:penicillins OR penicilinas OR MH:quinolones OR quinolon$ OR roxithromycin OR MH:roxithromycin OR roxitromicina OR MH:sulfamethoxazole OR sul- famethoxazole$ OR sulphamethoxazole OR Sulfametoxazol OR MH:tetracyclines OR tetracycline$ OR Tetraciclinas OR MH: trimethoprim OR trimetoprim OR trimetoprima) > clinical˙trials WHAT’S NEW Last assessed as up-to-date: 7 November 2012. 2004 Review first published: Issue 3. Bari 2011. 7 November 2012 New citation required and conclusions have changed We have added conclusions about treatment of severe pneumonia with oral antibiotics and a comparison of antibiotics in radiographically confirmed pneumonia HISTORY Protocol first published: Issue 3. Published by John Wiley & Sons.

in addition to editing the review. External sources • No sources of support supplied Antibiotics for community-acquired pneumonia in children (Review) 194 Copyright © 2013 The Cochrane Collaboration. Other two authors do not have any competing conflicts of interest. 1 August 2009 Amended Converted to new review format. data analysis. Dr RM Pandey (RP) contributed to the sections on data extraction. . Published by John Wiley & Sons. CONTRIBUTIONS OF AUTHORS Dr Sushil K Kabra (SK) and Dr Rakesh Lodha (RL) jointly prepared and edited the review. DECLARATIONS OF INTEREST One of the authors (Kabra) was co-author of one study (Awasthi 2008) included in the review. Ltd. 6 January 2006 New citation required and major changes Search conducted. India. New Delhi. quality assessment and statistical methods. SOURCES OF SUPPORT Internal sources • All India Institute of Medical Sciences.Date Event Description 4 January 2010 New citation required and conclusions have changed Seven new studies included and we have added new information on ambulatory treatment for severe pneu- monia and the superiority of ampicillin/penicillin with gentamycin instead of chloramphenicol for the treat- ment of very severe pneumonia to the conclusions 18 September 2009 New search has been performed Searches conducted.

We therefore decided to include studies that gave either cure rates or treatment failure rates as one of the outcomes. Child. .DIFFERENCES BETWEEN PROTOCOL AND REVIEW In the protocol we decided to include studies with an outcome in the form of cure rates. Randomized Controlled Trials as Topic. Published by John Wiley & Sons. Bacterial [∗ drug therapy]. there were a few studies that did not report cure rates. Sulfamethoxazole Drug Combination [ther- apeutic use] MeSH check words Adolescent. Humans Antibiotics for community-acquired pneumonia in children (Review) 195 Copyright © 2013 The Cochrane Collaboration. Drug Therapy. Gentamicins [therapeutic use]. INDEX TERMS Medical Subject Headings (MeSH) Amoxicillin [therapeutic use]. Combination [methods]. Chloramphenicol [therapeutic use]. Trimethoprim. Anti-Bacterial Agents [∗ therapeutic use]. However. Community-Acquired Infections [drug therapy]. Pneu- monia. Ltd. Penicillins [therapeutic use].