You are on page 1of 56

Cochrane Database of Systematic Reviews

Prophylactic antibiotics to prevent surgical site infection after
breast cancer surgery (Review)

Jones DJ, Bunn F, Bell-Syer SV

Jones DJ, Bunn F, Bell-Syer SV.
Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery.
Cochrane Database of Systematic Reviews 2014, Issue 3. Art. No.: CD005360.
DOI: 10.1002/14651858.CD005360.pub4.

www.cochranelibrary.com

Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review)
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

TABLE OF CONTENTS
HEADER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
PLAIN LANGUAGE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Figure 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
AUTHORS’ CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
CHARACTERISTICS OF STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
DATA AND ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
Analysis 1.1. Comparison 1 Preoperative antibiotics versus none or placebo, Outcome 1 Wound infections. . . . . 41
Analysis 1.2. Comparison 1 Preoperative antibiotics versus none or placebo, Outcome 2 Wound infection cefonicid. 42
Analysis 1.3. Comparison 1 Preoperative antibiotics versus none or placebo, Outcome 3 Wound infection cefazolin. . 42
Analysis 1.4. Comparison 1 Preoperative antibiotics versus none or placebo, Outcome 4 Infection rates in those who
received neo-adjuvant chemo. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
Analysis 1.8. Comparison 1 Preoperative antibiotics versus none or placebo, Outcome 8 Readmission to hospital. . 45
Analysis 2.1. Comparison 2 Perioperative antibiotics compared with no antibiotic, Outcome 1 Wound infection. . . 46
APPENDICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
WHAT’S NEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
HISTORY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
CONTRIBUTIONS OF AUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
DECLARATIONS OF INTEREST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
SOURCES OF SUPPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
INDEX TERMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) i
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

[Intervention Review]

Prophylactic antibiotics to prevent surgical site infection after
breast cancer surgery

Daniel J Jones1 , Frances Bunn2 , Sophie V Bell-Syer3

1 Department of Health Sciences, University of York, York, UK. 2 Centre for Research in Primary and Community Care, University of

Hertfordshire, Hatfield, UK. 3 Royal United Hospital, Bath, UK

Contact address: Daniel J Jones, Department of Health Sciences, University of York, Seebohm Rowntree Building, York, YO10 5DD,
UK. ugm4djj@gmail.com.

Editorial group: Cochrane Wounds Group.
Publication status and date: New search for studies and content updated (no change to conclusions), published in Issue 3, 2014.
Review content assessed as up-to-date: 5 December 2013.

Citation: Jones DJ, Bunn F, Bell-Syer SV. Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery. Cochrane
Database of Systematic Reviews 2014, Issue 3. Art. No.: CD005360. DOI: 10.1002/14651858.CD005360.pub4.

Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

ABSTRACT

Background

Surgery has been used as part of breast cancer treatment for centuries; however any surgical procedure has the potential risk of infection.
Infection rates for surgical treatment of breast cancer are documented at between 3% and 15%, higher than average for a clean surgical
procedure. Pre- and perioperative antibiotics have been found to be useful in lowering infection rates in other surgical groups, yet there
is no consensus on the use of prophylactic antibiotics for breast cancer surgery.

Objectives

To determine the effects of prophylactic (pre- or perioperative) antibiotics on the incidence of surgical site infection (SSI) after breast
cancer surgery.

Search methods

For this third update we searched the Cochrane Wounds Group Specialised Register (5 December 2013); the Cochrane Central Register
of Controlled Trials (CENTRAL) (The Cochrane Library); the Database of Abstracts of Reviews of Effects (DARE) (The Cochrane
Library); Ovid MEDLINE; Ovid MEDLINE (In-Process & Other Non-Indexed Citations); Ovid EMBASE; and EBSCO CINAHL.
We applied no language or date restrictions.

Selection criteria

Randomised controlled trials of pre- and perioperative antibiotics for patients undergoing surgery for breast cancer were included.
Primary outcomes were rates of surgical site infection (SSI) and adverse reactions.

Data collection and analysis

Two review authors independently examined the title and abstracts of all studies identified by the search strategy, then assessed study
quality and extracted data from those that met the inclusion criteria.
Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 1
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Main results

A total of eleven studies (2867 participants) were included in the review. Ten studies evaluated preoperative antibiotic compared with
no antibiotic or placebo. One study evaluated perioperative antibiotic compared with no antibiotic. Pooling of the results demonstrated
that prophylactic antibiotics administered preoperatively significantly reduce the incidence of SSI for patients undergoing breast cancer
surgery without reconstruction (pooled risk ratio (RR) 0.67, 95% confidence interval (CI) 0.53 to 0.85). Analysis of the single study
comparing perioperative antibiotic with no antibiotic found no statistically significant effect of antibiotics on the incidence of SSI
(RR 0.11, 95% CI 0.01 to 1.95). No studies presented separate data for patients who underwent reconstructive surgery at the time of
removal of the breast tumour.

Authors’ conclusions

Prophylactic antibiotics administered preoperatively reduce the risk of SSI in patients undergoing surgery for breast cancer. Further
studies involving patients undergoing immediate breast reconstruction are needed as studies have identified this group as being at higher
risk of infection than those who do not undergo immediate breast reconstruction.

PLAIN LANGUAGE SUMMARY

Antibiotics to prevent surgical site infection after breast cancer surgery

Breast cancer accounts for one in 10 of all new cancer cases diagnosed and surgical removal of the breast is a common treatment approach.
An infection of the surgical wound is often a complication of surgery and taking antibiotics just before the operation significantly
reduces the chances of developing an infection. The review is not able to establish which antibiotic is most appropriate. No trials were
found which considered the effect of antibiotics when the operation involved immediate breast reconstruction.

BACKGROUND (Mangram 1999), the incidence of SSI in those being treated for
breast cancer is thought to range between 3% (Lefebvre 2000)
Breast cancer accounts for one in 10 of all new cancer cases di- and 15% (Witt 2003). This is a higher incidence of infection
agnosed around the world each year (Bray 2004) and is the lead- than the 3.4% SSI rate associated with clean surgical techniques
ing cause of cancer death in women (Pisani 1999). Surgery for (Vazquez-Aragon 2003). A recent review (Pittet 2005) found that
removal of breast cancer has been common practice for centuries women who had been treated for breast cancer and who had im-
(Donegan 1995) and this is normally used as part of a multi- mediate reconstruction had a SSI rate of between 0% and 53%,
faceted approach to care with the aim of curing the patient of their whilst non-cancer patients undergoing the same reconstructive
cancer in early stage tumours or prolonging life for others (NICE surgery had an average rate of 2.5%. There are several factors that
2002). Surgical intervention ranges from removing the breast and are documented as increasing the risk of infection for surgical pa-
associated axillary lymph nodes, to lumpectomy with or without tients generally. These include: patient risk factors, e.g. diabetes,
sentinel node biopsy (Harris 2004). Whilst the risk of breast can- obesity or smoking (Haley 1985; Mangram 1999); surgical tech-
cer for men is only 1%, treatment for men is very similar to that of nique, e.g. aseptic technique (Ritter 1988); and type of surgery,
women (Harris 2004). As with all surgical procedures, breast can- e.g. whether the wound is contaminated (Gruendemann 2001).
cer surgery runs the risk of complications. One such risk is post- In addition, surgery for breast cancer has several risk factors that
operative surgical site infection (SSI), even though breast cancer make this patient group more susceptible to infection, including
surgery is considered a ’clean surgical procedure’. Clean surgical use of chemotherapy prior to surgery (neo-adjuvant chemother-
procedures, as defined by Haley 1985, are those which have a low apy); technique of diagnostic biopsy; re-operation for recurrence
risk of bacterial contamination during the surgery.Some women or to achieve better tumour margins; reconstructive surgery with
have immediate breast reconstruction; however this group of pa- implants and seroma accumulation and drainage (Morris 1988;
tients has a higher risk of SSI (Spauwen 2000). Tran 2003). Infection may lead to significant morbidity for the
Despite internationally recognised infection control guidelines patient, delay in adjuvant treatment, such as radiotherapy, and

Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 2
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

increased cost of care if the patient requires supplementary treat- Types of interventions
ment due to infection (Coello 1993). Any pre- or perioperative antibiotics used as prophylaxis where
Pre- and perioperative antibiotics have been shown to reduce there was no known infection and where the use of antibiotics
the risk of postoperative infection in several patient groups (the was the only systematic treatment difference between comparison
term “perioperative” refers to administration between induc- groups.
tion of anaesthetic and the patient leaving the recovery room) We only included trials of one antibiotic compared with another
(Gruendemann 2001; Majoribanks 2004; SIGN 2008a). In col- if there was a control or placebo arm, as benefit from prophylactic
orectal surgery antibiotic prophylaxis has been found to reduce antibiotics has not yet been established in this patient group.
long and short-term morbidity, decrease length of hospital stay Definitions of key terms:
and lower the overall cost of care (SIGN 2008a). However, the use • ’Antibiotic regimen’ describes the characteristics of the
of prophylactic antibiotics in preventing infection is still a contro- antibiotic treatment, i.e. type of antibiotic, route, dose, number
versial issue and their routine use is not common in breast can- of doses and timing of administration.
cer surgery. Some feel that a clean surgical procedure should not • ’Preoperative antibiotic prophylaxis’ is antibiotic therapy
require prophylactic antibiotics (Sheridan 1994) and that the use given within 24 hours prior to surgery, solely for prophylaxis (i.e.
of pre- or perioperative antibiotics merely masks the symptoms of not for an infection that is already suspected).
infection until after the patient is discharged (Wagman 1990). In • ’Perioperative antibiotic prophylaxis’ is antibiotic therapy
addition increased antibiotic use may lead to antibiotic resistance administered between commencement of induction of surgery
(PHLS 2000) and adverse effects such as clostridium difficile in- and the patient leaving the recovery room.
fection that causes gastro-intestinal problems (SIGN 2008a). In Comparisons of interest were as follows.
order to clarify the situation, this systematic review evaluated the • Preoperative antibiotic compared with no antibiotic or
effectiveness of pre- or perioperative antibiotics in reducing the placebo.
incidence of postoperative infections in patients undergoing breast • Perioperative antibiotics compared with no antibiotic or
cancer surgery. placebo.
• Head to head comparisons of antibiotics.

OBJECTIVES Types of outcome measures
To determine the effects of prophylactic antibiotics on SSI after
breast cancer surgery.
Primary outcomes
1. Incidence of postsurgical breast surgical site (wound)
infection (SSI)*. Where possible, this should be reported as the
METHODS
number of participants in each group with a clinically significant
infection. Research demonstrates that 98% of acute SSIs related
to non-implant breast surgery occur within 28 days (Mitchell
Criteria for considering studies for this review 1999). However, where there is surgical re-construction,
guidelines recommend that this time is increased to one year post
surgery (Mangram 1999). Therefore we included all studies that
Types of studies present data on acute SSI within one year of surgery.
2. Adverse reactions (e.g. anaphylaxis, gastro-intestinal or skin
Randomised controlled trials (RCTs) and controlled clinical trials rash).
(where patients were allocated by quasi-random methods such as *Surgical site infection: ideally this will be defined using outcomes
alternation, case records numbers or days of the week). from a validated assessment tool such as ASEPSIS (Wilson 1986)
which are based on CDC definitions (Mangram 1999).

Types of participants
People with breast cancer undergoing breast surgery with or with- Secondary outcomes
out immediate re-construction as part of their treatment. 1. Death.
We included studies that involved mixed patient groups (i.e. cancer 2. Delay in adjuvant cancer treatment because of breast
and non-cancer, other surgeries or breast implants not as part of wound infection.
cancer treatment) as long as it was possible to extract separate data 3. Time to wound healing.
for those undergoing surgery primarily to treat breast cancer. 4. Time to infection.

Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 3
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

kw 1043 of citations identified by the search. #23 (antibiotic* or clindamycin or cefuroxime or cefuroxim or ceftazidime or ofloxacin or levofloxacin or azithromycin or sul- bactam or ampicillin or mezlocillin or oxacillin or vancomycin or Search methods for identification of studies tobramycin or ciprofloxacin) 21709 #24 #22 or #23 25029 #25 #12 and #21 and #24 33 The search strategies for Ovid MEDLINE.kw 551 independently applied the inclusion criteria. Appendix 3 and See Appendix 1 for the search strategy used for the second update Appendix 4 respectively.ab. 2013). We combined the MEDLINE search with of this review. We contacted experts in Citations 03 December.kw 4427 Selection of studies #5 (surg* near/5 site*):ti. length of follow-up and outcomes.kw 756 study risk of bias (as defined below).ab. In addition. We obtained all reports of po- #7 (surg* near/5 dehisc*):ti.SU] 97 included study using the Cochrane Collaboration tool for assess- #21 {or #13 #20} 3148 ing risk of bias (Higgins 2011).kw 4032 #4 (surg* near/5 wound*):ti. Ovid format (Lefebvre 2011). ters developed by the Scottish Intercollegiate Guidelines Network • the Cochrane Central Register of Controlled Trials (SIGN) (SIGN 2008b). source of #16 ((breast next carcinoma*) near/5 surg*):ti.ab.ab.ab. dose route.kw 47 funding.ab. Published by John Wiley & Sons. 5. Cost of care (should be a comparison between the 8601 treatment and control group). duration of treatment).kw 2029 Assessment of risk of bias in included studies #20 MeSH descriptor: [Breast] explode all trees and with qualifiers: For this update two review authors independently assessed each [Surgery .e.kw 385 tentially eligible trials as full-text articles and two review authors #8 (wound* near/5 dehisc*):ti. #1 MeSH descriptor: [Surgical Wound Infection] explode all trees 2643 #2 MeSH descriptor: [Surgical Wound Dehiscence] explode all Data collection and analysis trees 346 #3 (surg* near/5 infect*):ti.ab.kw 919 drug name. We followed up conference proceedings and grey literature that was considered to be potentially eligible for We used the following search strategy in the Cochrane Central inclusion by both authors by contacting the study authors for Register of Controlled Trials (CENTRAL): further information. we screened references in all articles found by the • Ovid MEDLINE (In-Process & Other Non-Indexed above search strategy for further studies. We extracted data on #14 ((breast next cancer*) near/5 surg*):ti. Issue 11). #17 MeSH descriptor: [Mastectomy] explode all trees 1212 #18 MeSH descriptor: [Mammaplasty] explode all trees 218 #19 (mastectomy or mammaplasty):ti.SU] 1627 specifically designed data extraction tool. antibiotic intervention (i. #10 (wound near/5 disruption*):ti.ab. Issue 11) • Ovid MEDLINE (1946 to November Week 3 2013). We • the Cochrane Wounds Group Specialised Register (searched combined the EMBASE and CINAHL searches with the trial fil- 5 December 2013). setting.kw 42 #11 (wound next complication*):ti. Ovid EMBASE and Electronic searches EBSCO CINAHL can be found in Appendix 2. resolving disagree- #9 (wound* near/5 infect*):ti. .kw 1016 Two review authors independently examined the title and abstract #6 (surg* near/5 incision*):ti.ab.ab. • The Database of Abstracts of Reviews of Effects (DARE) (The Cochrane Library 2013.and precision-maximis- and searched the following electronic databases over all years: ing version (2008 revision).kw 420 #12 {or #1-#11} 8341 Data extraction and management #13 MeSH descriptor: [Breast Neoplasms] explode all trees and Two review authors independently extracted trial data using a with qualifiers: [Surgery .kw 4440 ments by discussion. the field and interest groups to try and obtain access to unpublished • EBSCO CINAHL (1982 to 20 December 2013). or ongoing work. Searching other resources • Ovid EMBASE (1974 to 2013 Week 48). This tool addresses six specific Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 4 Copyright © 2014 The Cochrane Collaboration. the Cochrane Highly Sensitive Search Strategy for identifying ran- For the third update of this review we revised the search strategies domised trials in MEDLINE: sensitivity. #22 MeSH descriptor: [Anti-Bacterial Agents] explode all trees 6. Ltd.ab.ab. tions.ab. We applied no language or date restric- (CENTRAL) (The Cochrane Library 2013. #15 ((breast next neoplasm*) near/5 surg*):ti. Readmission to hospital.

come separately. Cabaluna 2012. Yetim 2010) included women only. Of the eleven studies. We completed a ’Risk of bias’ table for each el- igible study. we examined 5 for details of criteria on which the judgement was based). In the absence of clinical and statistical hetero. concealment in a prespecified sensitivity analysis. these are now excluded as the study authors have not responded to our requests for further clarification (Kumar 2005). Gupta 2000. All included studies had been published. We discussed any disagreement amongst all review authors to achieve a consensus. Where possible for each trial we calculated the risk ratio (RR) of Gulluoglu 2013.e. Chow geneity we applied a fixed-effect model to pool data. as mean difference (MD) with Included studies 95% CI. one almost entirely women (Hall 2006) and two (Amland 1995. Cabaluna 2012. Bold 1998. We for this third update (Cabaluna 2012. United • patients undergoing immediate reconstruction with States of America (Bold 1998. were single-centre trials and were • patients undergoing immediate reconstruction without conducted in eight different countries.e. although this could not be established from the data presented in the report Subgroup analysis and investigation of heterogeneity or by contacting the authors. We identified two further studies which met the inclusion criteria tion across studies due to heterogeneity rather than chance. We explored both clinical and statis. ies were: Australia (Hall 2006). Wagman 1990. Japan implants (i. All of these studies included breast As patients undergoing reoperation. Yetim 2010). This display RESULTS of internal validity indicates the weight the reader may give the results of each study. where possible. Where syn. A further four abstracts which geneity and would have resulted in a random-effects model being were considered to be multiple publications of the same study were applied or not pooling results. Country of origin for stud- implants (i. Platt 1990. out implants and patients receiving neo-adjuvant chemotherapy The studies were conducted between 1990 and 2013. In total eleven studies met the inclusion criteria for this version of the Data synthesis review (Amland 1995. allocation concealment. Wagman 1990). Description of studies Assessment of heterogeneity Results of the search We assessed heterogeneity between study results using the I2 statis- tic (Higgins 2003). Gulluoglu 2013) and ex- considered values of I2 over 75% to indicate a high level of hetero. eight (Bold 1998. number of days to infection). The proposed ment arms and 1428 in control arms. dom (Gupta 2000). 1439 in treat- factors where there were sufficient data available. Wagman thesis was inappropriate we have presented a narrative overview. ratio of greater than one indicates a higher risk of infection in the first group named.e. Participants tical heterogeneity. Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 5 Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons. We reported continuous data (i. extreme baseline imbalance) (see Appendix particularly affects the result of studies (Schulz 1995). Norway (Amland 1995). The Philipines (Cabaluna • patients who have received chemotherapy (excluding 2012). Platt infection and 95% confidence interval (95% CI). . Paajanen 2009.domains. reconstruction with or with. cluded one study (Nicholas 2007). Gulluoglu 2013. design and heterogeneity. Hall 2006. Gupta 2000. namely sequence generation. This examined the percentage of total varia. silicone or saline). cancer patients as one of multiple patient groups being analysed. We presented assessment of risk of bias using a ’Risk of bias’ summary figure. such that a risk 1990. 1990. and (Chow 2000). selective outcome reporting Since there is evidence that the quality of allocation concealment and other issues (e. Sensitivity analysis blinding. Turkey (Gulluoglu 2013. Methods of synthesising the studies were dependent on trial qual- ity. These studies were con- subgroups were: ducted in the hospital setting. We the effect of excluding studies judged to have inadequate allocation assessed blinding and completeness of outcome data for each out. In total we conducted a prespecified subgroup analysis of each of these 2867 participants were included for meta-analysis. Finland (Paajanen 2009). Ltd. Paajanen 2009.g. which presents all of the judgements in a cross-tabulation of study by entry. Study sizes are documented as having a higher risk of infection (Tran 2003) ranged between 44 (Yetim 2010) and 618 (Hall 2006). Chow 2000. TRAM flap). Platt 1990) may have contained male and female breast surgery participants. Yetim 2010) and United King- hormone treatment) prior to surgery. 2000. awaiting assessment pending clarification from the study authors. incomplete outcome data.

isting implants (Baker 2000). non-reconstructive surgery for breast cancer (Cabaluna the surgical wound prior to surgical closure (Yetim 2010). Five studies The antibiotics evaluated included: did not provide discrete data for breast cancer patients and two • azithromycin. Studies were judged 2009). (Gulluoglu 2013) illary lymph node dissection for breast cancer (Bold 1998) and • collagen plus gentamycin sulphate (200mg) inserted under primary. respectively). control or placebo arm. Franchelli 1994) non-reconstructive surgery for breast cancer. Platt 1990) stated that they were sponsored by a pharmaceutical company (Pfizer AS. Platt 1990). Wagman 1990). Four abstracts were excluded as Antibiotics used they may have been multiple publications of the same study and clarification could not be obtained from the authors. single dose decided according to body were found to be studies focused on other types of surgery (see weight. Wagman 1990) are very similar in terms of length of follow- rather than infective episodes. The four remain. Characteristics of excluded studies table). • intravenous augmentin (1. • a single dose of intravenous flucloxacillin (2g) (Hall 2006). Excluded studies Smith Kline & Beecham and Smith Kline & French laboratories. Published by John Wiley & Sons. clusion. Ltd. Platt One study (Chow 2000) was designed to look at inflammatory 1990. Gupta 2000. herniorrhaphy or breast surgery (Platt 1990). We excluded a total of 27 studies for the following reasons: two ican Cancer Society and another (Paajanen 2009) by the Finnish were reviews. • oral clarithromycin (500mg) for 10 doses (Chow 2000).2g) (Gupta 2000). as unclear or at high risk of bias in the majority of the domains. Platt 1990). Bold 1998. Two studies looked at axillary lymph node dissection as part of breast cancer treatment (Bold 1998. Gulluoglu 2013. choice of antibiotic and type of surgery undertaken. Baker 2000. 2010) looked solely at breast cancer patients receiving primary. Whilst three studies (Amland 1995.Types of surgery • a single dose of intravenous cefazolin (1g) (Cabaluna 2012) Types of participants included patients undergoing plastic surgery • a single dose of intravenous ampicillin-sulbactam (1g) (Amland 1995). No eligible studies evaluating prophylactic antibiotics for re- ing studies (Cabaluna 2012. Gulluoglu 2013. Neo-adjuvant chemotherapy Two studies included patients who had received neo-adjuvant che- Source of funding motherapy (Bold 1998. . Immediate reconstruction with or without implants non-reconstructive surgery for breast cancer. One study (Gupta 2000) fol. ax. we excluded the studies following scrutiny. One compared different regimens and doses. one was a mul- cultural foundation. Gupta 2000. however discrete data on infection up. 2012. Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 6 Copyright © 2014 The Cochrane Collaboration. All studies rates were presented and therefore the study was eligible for in. We excluded one study as it could not be obtained from the British Library. 11 were not RCTs or quasi RCTs. taken 8 pm the night before surgery (Amland 1995). • cefazolin (six doses) (Wagman 1990). Risk of bias in included studies • a single dose of intravenous dicloxacillin (1g) (Paajanen See ’Risk of bias’ summary figure: Figure 1. Three studies (Amland 1995. but did not document the length of hospital stay for these patients. Hall 2006. Five studies (Bold 1998. but had no the other studies. to be at overall unclear or high risk of bias if they were described • a single dose of cefonicid (1g) (Bold 1998. included patients undergoing reconstructive surgery. Franchelli 1994) whilst one study was excluded because the Length of follow-up from surgery ranged from five days (Chow research was addressing the needs of dental patients who have ex- 2000) to six months (Yetim 2010). lowed up patients between 10 and 14 days post discharge. had similar inclusion and exclusion criteria. Cabaluna 2012. Yetim constructive surgery (with or without implants) were identified. Gulluoglu 2013). One study (Wagman 1990) was funded by the Amer. The source of funding was not reported in tiple drug comparison excluded as there was no placebo or con- trol arm. Wagman 1990. One study (Paajanen 2009) looked at core needle biopsy and primary. It was not clear that the patients had undergone surgery as part of breast cancer treatment (Amland Length of follow-up 1995.

Methodological quality summary: review authors’ judgements about each methodological quality item for each included study. Ltd. Published by John Wiley & Sons. Figure 1. Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 7 Copyright © 2014 The Cochrane Collaboration. .

Platt 1990) there was some funding reported from adequate but as there was no statement by the study authors we pharmaceutical companies but it was unclear the extent of the judged this to be unclear. Hall 2006.all come measures stated in the methods section are reported in the outcomes) results and therefore we judged this domain to be at low risk of Adequate blinding of participants and treatment providers was bias for all studies. In one were at low risk of bias for this domain (Amland 1995. Cabaluna the method of allocation concealment was not described (Amland 2012. Hall blinded (Bold 1998) and in another (Yetim 2010) it was judged 2006. Hall 2006. Bold 1998. Incomplete outcome data Allocation concealment In ten studies we judged the loss to follow-up to be low. Chow 2000. Gulluoglu 2013. Wagman reported by Hall 2006 the antibiotic was administered after the 1990. Gulluoglu 2013. 2000) however the method of allocation was not described and We judged six studies to have undertaken an ITT analysis ei- two studies used sealed opaque sequentially numbered envelopes ther because they explicitly reported this or because there were (Gupta 2000. Chow 2000. Platt 1990. Platt 1990. the groups analysed at the final follow up of the study were the bered envelopes (Paajanen 2009). Gupta 2000. opaque. Chow 2000. 1995. Cabaluna 2012. Hall 2006). Three of these studies used the hospital 2006. Gulluoglu 2013. the studies appeared free from other forms of bias (Cabaluna 2012. Bold 1998. Wagman 1990). Gupta 2000. One study reported the use of both no drop outs from the study and the numbers of participants in hospital pharmacy as well as sealed. Paajanen 2009). . Paajanen 2009. Paajanen 2009. Wagman 1990). Platt 1990. Three studies were classified as unclear as that the nature of the collagen implants under the wound site the authors failed to report the method by which randomisation would unblind the outcome assessors. Wagman 1990). Chow 2000. clearly reported in seven trials and therefore these were at low risk of bias (Amland 1995. Published by John Wiley & Sons.all outcomes) Nine studies described adequate blinding of outcome assessors and these were at low risk of measurement bias. Gupta 2000. ies (Bold 1998. All antibiotic compared Sequence generation with placebo studies stated that the key physician was unaware Eleven studies were described as RCTs. Gulluoglu 2013. Blinding Selective reporting The study protocols were not available but all the important out- Blinding (participants and treatment providers . One study stated that consecutive because the authors stated that patients would be followed up for patients were allocated to group by a computer program (Chow six months post surgery but only reported data at seven days. Gupta 2000. Yetim 2010). Whilst blinding was not specifically Chow 2000. Ltd. Cabaluna 2012. Cabaluna 2012. industry involvement and we have adopted a cautious approach by Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 8 Copyright © 2014 The Cochrane Collaboration. Wagman 1990). computer-generated numbers or sequences of blocks of 10 and Hall 2006. but only eight adequately of patient allocation until data collection was complete (Amland generated the randomisation sequence by reporting the use of 1995. Wagman 1990. (Yetim 2010) the study was judged to be unclear for this domain Platt 1990. Wagman 1990) and they were Cabaluna 2012. Gupta 2000. Yetim 2010) and therefore they are classified Intention-to-treat analysis was not reported in the other five stud- as at unclear risk of bias. Three trials Other potential sources of bias were classified as having inadequate blinding of both participants We judged seven trials to be at low risk of bias for this domain and treatment providers mainly because the control groups were because there was no imbalance in the baseline characteristics and not blinded as they were not given any treatment. Gupta 2000. Bold 1998. Hall therefore at low risk of bias. Paajanen 2009. study it remained unclear if the outcome assessors were adequately Cabaluna 2012. Platt 1990. Gupta 2000. Yetim 2010). Hall ment groups and valid reasons given (Amland 1995. sequence was generated (Paajanen 2009. In one study pharmacy to generate the allocation for participants (Bold 1998. In three remaining studies (Amland 1995. Wagman 1990). Platt 1990. Gulluoglu 2013. sequentially num. with Adequate allocation concealment was described for eight studies similar numbers of participants lost in both control and treat- (Bold 1998. Gulluoglu 2013. Chow 2000. Paajanen 2009. Yetim 2010). (Chow 2000. Yetim 2010).Allocation Blinding (outcome assessors . 2006. Paajanen 2009. induction of anaesthesia therefore it is possible that blinding was Bold 1998. Platt 1990. In the remaining three studies same as those randomised at the outset (Amland 1995.

one com.e. they did not reply. One study (Gulluoglu 2013) little detail is given regarding what the cost (Chow 2000) reported no infections in either group but in the included.6). Two studies (Bold 1998. mitted to hospital for wound complications. Ltd. the cost of operation or associated stay in hospital. Pooling the two studies which compared cefazolin tibiotic versus none or placebo) (Analysis 1. one compared flu. Another study provided details of the number of patients who had previously received chemotherapy (Platt 1990) Preoperative antibiotics compared with placebo or but did not report separate data on infection rates for these pa- no antibiotic (ten trials. The study reported that the control group had a signifi- treated with antibiotics. Paajanen 2009. Cabaluna 2012. as this study had short follow. . stating that flammation rather than infection as its primary outcome. Results are presented as risk ratio (RR) where the risk ratio is the Cost of care risk of infection in the intervention group divided by the risk of infection in the control group. one compared the cost of any additional care or medications (i. 2823 participants) tients. We carried out a sensitivity analysis events. mention of adverse events in the study report. antibiotic pro- augmentin with placebo (Gupta 2000). Hall 2006. Gupta 2000. Wagman We pooled all the trials using a fixed-effect model as there was no 1990) and one study (Gupta 2000) reported 41 adverse events evidence of heterogeneity (I2 = 0%).95) to other data tables for adverse effects from antibiotics under an- (Analysis 1. only compared antibiotics with no antibiotic and reported in- highlighted a difference in the baseline characteristics. Gulluoglu 2013) compared preoperative antibiotics ment in a prespecified sensitivity analysis. this was statistically significant in one cantly higher cost (USD 20. 95% confidence interval (CI) 0.67.48).judging there to be a high risk of bias. Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 9 Copyright © 2014 The Cochrane Collaboration. Platt 1990. Removing these studies from the meta-analysis resulted in a pooled RR of 0.80 in the antibiotic cloxacillin with no treatment (Hall 2006).67 (95% CI 0.12) (Analysis 1.01 to 4. Bold 1998. Seven studies re- with placebo (Cabaluna 2012.33 to 0. 95% CI. The remaining three studies made no to exclude one study (Chow 2000). one compared ampi. Platt 1990) showed a statistically sig. In addition pooling the two studies which compared cefonicid Adverse reactions to treatment with placebo (Bold 1998. prophylaxis group and USD 364. that giving preoperative antibiotics significantly reduces the risk but no details were reported on type of adverse events. but calculated pared azithromycin with placebo (Amland 1995). Although of wound infection after breast cancer surgery (RR 0. (Analysis 1. Wagman 1990) reported adverse events (please refer tibiotics (RR 0.85) (Analysis 1. breast cancer patients. 95% CI we contacted authors for clarification about the nature of these 0. icant reduction in infection (RR 0. Gulluoglu 2013) reported the cost of care cates fewer infections in the intervention group. They found with placebo (Cabaluna 2012. ported there were no adverse events (Bold 1998. Gupta Since there is evidence that the quality of allocation concealment 2000. In the other eight trials the results were not statistically significant (Analysis 1. this study was judged at a high risk of bias for this domain. Chow 2000) to have unclear allocation conceal- ment. Platt 1990. The report states it calculated “SSI-related treatment remaining nine trials there were fewer infections in the groups cost”. Hall 2006.21. Platt 1990). that the average cost per patient was USD 49. majority of this cost difference was accounted for in patients read- pared dicloxacillin with placebo (Paajanen 2009) and one com.5). 0. Published by John Wiley & Sons.52 to 0.85) which was still signif- Incidence of postoperative wound infection icantly in favour of prophylactic antibiotics. Platt 1990. The cillin-sublactam with no treatment (Gulluoglu 2013). Paajanen 2009. Three studies (Chow 2000. A risk ratio of less than one indi. Paajanen 2009. Sensi- patients in the control group had significantly more frequent open tivity analysis demonstrated no effect from removing Chow from surgical biopsies than those in the prophylaxis group. Seven studies (Amland 1995. We judged two studies with no treatment. influences study results (Schulz 1995) we examined the effect of operative antibiotics with placebo.47 to 1. In the other study pared clarithromycin with no treatment (Chow 2000).26) when compared with the treat- study (Gulluoglu 2013) which considered overweight and obese ment group (USD 8. (Amland 1995.1).42) Gulluoglu 2013.87 in the control group. One study (Bold 1998) documented infection rates in those who received neo-adjuvant chemotherapy. Six studies (Bold 1998. Cabaluna 2012.2). All ten trials recorded incidence of wound infection as an outcome. as a result the pooled analysis.4). In one study (Bold 1998) the cost did not include pared cefonicid with placebo (Bold 1998. 95% CI 0. One study (Gulluoglu 2013) up.56. two compared cefazolin phylaxis. one com. Wagman 1990) compared pre. there was no statistically sig- nificant difference between the groups treated with cefonicid com- Effects of interventions pared with the placebo group (RR 0. nificant reduction in infection associated with preoperative an. Wagman 1990) showed no signif. Two studies com. The pooled risk ratio shows (23%) in the treatment group and 33 (18%) in the control group.1).82. Wagman 1990).53 to 0. postoperative antibiotics or wound care). excluding studies judged to have inadequate allocation conceal- Hall 2006.

8%) we did not pool results.6 days in the control cancer treatment caused by SSI. Platt 1990) documented sim- ilar mean times to onset of infection: 12 and 11 days in the in- tervention group and 11 and 10 days in the control group respec. the other study found no reduction in read. One study (Bold 1998) reported statistically significantly lower readmission rates in those treated with prophylactic an- DISCUSSION tibiotics (RR 0. Wagman 1990 documented mean time of onset of infection The study did not report any information on delays in adjuvant of 17. Four studies reported time to onset of infection (Analysis 1. . 95% CI 0. due to heterogeneity (I2 = 70.11.01 to 1. including deaths.7).11.1). cost and readmissions were One study (Yetim 2010) compared perioperative antibiotics with reported by few of the included studies. analysis. The study compared gentamycin-infused collagen (Gentacoll) inserted perioperatively with no antibiotic. Platt 1990) reported readmitted for parenteral antibiotics as a result of wound infection.8). Adverse reactions to treatment Time to onset of infection The study did not report any adverse reactions to treatment. Whilst the study au- No studies presented information on time to wound healing. infection. readmission rates following treatment. however detailed information about the nature of the adverse events was not given and adverse events were generally poorly reported across the included stud- Perioperative antibiotics compared with placebo or ies. Delay in adjuvant cancer treatment caused by SSI Cost of care No studies presented information on delays in adjuvant cancer treatments due to SSI. duce the risk of SSI in people undergoing surgery for breast cancer mission rates (RR 1. This small study at overall high risk of bias presented wound in- fection as an outcome. As when compared with placebo or no treatment.0 days).88) (Analysis 1. Of the nine studies such no conclusions can be drawn on this outcome.0. no antibiotic (one trial. 2013) no patients required readmission to hospital.29 to 3. 95% CI 0. In the other two studies. 95% CI 0. Gulluoglu 2013. however they all provided the mean time to onset of infection and Deaths not a range and therefore we have not combined this in a meta- The study did not report any information on deaths. The study did not report the cost of care. Ltd. evaluated perioperative antibiotics compared with no antibiotic. that reported data on adverse events only one found an increase of events in the intervention group. prophy. Readmission to hospital Readmission to hospital The study reports that two patients in the control group had to be Three studies (Bold 1998.42) (Analysis 1. In addition data for some of the outcomes. Delay in adjuvant cancer treatment caused by SSI tively. group. Gulluoglu 2013 states that the time to onset of infection was similar in both the control and intervention group. thor stated this to be significantly better in favour of the antibiotic group this was not replicated in our analysis (RR 0. Time to onset of infection The majority of infections (62%) were detected between 3 and 7 The study did not report any information on the time to onset of days postoperatively.Death Incidence of postoperative wound infection No studies presented information on deaths. The study provides a table of data with a range of onset of infection times. Published by John Wiley & Sons. Two studies (Gupta 2000.7 days in the intervention group and 9.9 days. We found one study that no antibiotic.95) (Analysis 2. This review found that preoperative antibiotics significantly re- laxis group 3.01 to 0.8) and a shorter duration of readmission (placebo group 5. In one study (Gulluoglu No patients in the antibiotic group were readmitted. 44 participants) delays in adjuvant cancer treatments. Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 10 Copyright © 2014 The Cochrane Collaboration. There were no infections in the antibiotic-treated group compared Time to wound healing with four infections in the control group.

In addition trials need to evaluate the use of antibiotics Cabaluna 2012. recommended that antibiotic prophylaxis should generally be administered as a single dose preoperatively in order to maximise benefit and minimise ACKNOWLEDGEMENTS adverse effects from treatment (SIGN 2008a). One study (Chow both contributed as authors of the original review but have not 2000) had a follow-up of only five days. therefore we were in women undergoing immediate breast reconstruction. 17. cer group (Sharon Parker and Liz Lostumbo) for their comments to Chow 2000. high-quality randomised controlled trials are needed included studies had been published. In addition. Sanchez-Manuel 2007). Andrew Jull. Another systematic review (Sajid 2012) also considered non breast Overall. excluded them as we were unable to obtain discrete data specifically for breast cancer patients. although we found some trials people undergoing breast cancer surgery. Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 11 Copyright © 2014 The Cochrane Collaboration. However. therefore there is potential to establish the most effective prophylactic antibiotic protocols. Melanie Cunningham and Karen Hanscombe analysis made little difference to the result. Published by John Wiley & Sons. however. Mieke Flour. Other recent research has. Platt 1990. surgery. would aid the Although there was no statistical heterogeneity only four studies development of well considered and useful protocols and standards compared the same antibiotic using the same regimen (Bold 1998. we judged one study (Yetim 2010) which compared perioperative antibiotics with no antibiotic to be at high risk of bias overall due We found one systematic review and meta analysis on the effects to a failure of blinding and insufficient information given regard- of antibiotic prophylaxis for breast cancer surgery which also con- ing selection bias. We found no studies evalu. Yetim 2010) and excluding these studies from the improve the review. Whilst it is encouraging that a statistically significant result was found it is possible that the numbers are not adequate Implications for practice to evaluate fully the risks and benefits of antibiotic prophylaxis for Preoperative prophylactic antibiotics reduce the risk of a SSI in breast cancer surgery. Wagman 1990). for practice. for publication bias. not many considering the number of people affected globally by AUTHORS’ CONCLUSIONS breast cancer.7 days it may have been appropriate to specify in the protocol nificantly reduce the incidence of SSI. produced varied results (Gillespie 2010. excluding data from ating antibiotics for breast reconstruction at the time of the initial this study made no difference to the overall outcomes. However. ther research would be required to establish the best protocols for tematic reviews in other types of clean surgery are scarce and have practice. However.this small study found that perioperative antibiotics did not sig. a minimum length for follow-up.Two non-systematic reviews (D’Amico dergoing non-reconstructive breast cancer surgery. Implications for research Whilst all efforts were made to obtain unpublished data. in adjuvant cancer treatments and costs of care. Similar sys. Richardson and Allyson Lipp) and from the Cochrane Breast Can- Three studies had unclear allocation concealment (Amland 1995. such as adverse events. . delays to the small number of studies obtained. We found only eleven studies with a total of 2867 participants. this review that included people having immediate breast reconstruction we does not establish the most effective antibiotic regimen to use. participated in any of the subsequent updates and are no longer set of infection in the other included studies ranged between 11 to involved. unable to make conclusions about the most effective antibiotic and regimen. Testing for publication bias was not done due Analysis of secondary outcomes. The authors would like to thank referees from the Cochrane In general the included trials were at low risk of bias for the Wounds Group (Nicky Cullum. However fur- 2001. there are sufficient data from this review to suggest that cancer patients and concluded that prophylactic antibiotics re. all the Further large. Rachel main domains of sequence generation and allocation concealment. antibiotic prophylaxis reduces surgical site infections in those un- duce the incidence of SSIs. Hall 2000) did not draw any firm conclusions. Ltd. As the average time to on. cluded that prophylactic antibiotics reduce SSIs (Tejirian 2006).

Hopkins CC.93(11): Neuherberg GmbH Oberschleissheim (DE)]. et al. Dervisoglu A. World Journal of Surgery conference. Hall 2006 {published data only} Impfstoff gegen krebs–Infektionsabwehr und Hall JC. Sinnett D. American Journal of Surgery 1998. Bertin 1998 {published data only} Chow 2000 {published data only} Bertin ML.26(4):363–6. infections and intraluminal antibiotics. SIGLE. et al.257 D’Amico 2001 {published data only} (1):37–43.28(35):620–6. Zaleznik DF. Lang E.322(3):153–60. 2002. Clinical and Diagnostic Laboratory Immunology 2000. randomised Constantini M. Cuniato V. Wei W. Gynecology and Obstetrics 1990. Melling A. Cafierio F. D’Amico DF.Suppl 567:9–12. Royle GT. Baker E. 2012. Mansfield PF. Erzurumlu K. Noviello S.7(6):925–31.13(1):108–11. GSF– single-dose antibiotic prophylaxis for non-reconstructive Forschungszentrum fuer Umwelt und Gesundhiet breast surgery. Ames FC. response. Hermunen H. British Journal of Surgery 2006. Ugurlu U. Antibiotic prophylaxis for selected implants and node dissection. Germany 2005.26(1):61–5. Vigna E. A prospective. Ltd. Canbolant M. Marvaso Paajanen H. Antibiotic prophylaxis for post-operative wound infection Erfle 2002 {published data only} in clean elective breast surgery. Esposito 2006 {published data only} Paajanen 2009 {published data only} Esposito S. Effect of local gentamycin application on healing placebo-controlled trial of a single dose of azithromycin on and wound infection in patients with modified radical postoperative wound infections in plastic surgery. double-blinded placebo-controlled clinical trial of the Does warming effect wound healing profiles following routine use of preoperative antibiotic prophylaxis in breast surgery?. Abschlussbericht. Perioperative antibiotic prophylaxis tolerability and pharmokinetics of teicoplanin in patients for hernioplasty and breast surgery. Fried M. Strategiefond ’Infektabwehr und Krebspaevention’. Journal of Chemotherapy 2001. 239–43. Prospective randomised. Kokal Franchelli S. Schnur PL. Published by John Wiley & Sons. Randomized clinical trial of Krebspraevention.176(3): devices. Rapsosio E. Riihimaki DU. Sandsmark Yetim I. Guler S. Galicia RM.68(1):43–7. Ruffolo C. Ianniello F. Surgery. Does preoperative core needle A. Linaas E. Determinants of surgical Chow LW. et al. A randomized. Amland 1995 {published data only} Yetim 2010 {published data only} Amland PF.38(4):1442–7. Karchmer Exner K. Wagman 1990 {published data only} Franchelli 1994 {published data only} Wagman LD. Dellinger EP. Gulluoglu B. Schoen A. Tegtmeier B. European Journal of Surgical Erfle V. Stuttgart. et al. Clarithromycin site infection after breast surgery. Carpenter R. Annals of Surgery 2013. Antibiotic prophylaxis in hernia biopsy increase surgical site infections in breast cancer repair and breast surgery: a prospective randomised study surgery? Randomized study of antibiotic prophylaxis. Plastic mastectomy: a prospective randomised study. Yuen KY. overweight or obese patients. International Medical Research 2010. Sinletary References to studies excluded from this review SE. Leone S. Whetter D. 1342–6. Plastic antibiotic administration for breast cancer surgery in and Reconstructive Surgery 1981. European wound management association modified radical mastectomy. Cabaluna 2012 {published data only} Baker 2005 {published and unpublished data} Cabaluna ND. Marquart KH. Leone MS. Journal of and Reconstructive Surgery 1995. . Efficacy. CDA Journal 2000. Berger DH. Borsche A. AW. Leaper D.37:59–66.170 (1):12–6. 15-17 September. American Journal of attenuates mastectomy-induced acute inflammatory Infection Control 1998. Ozkan OV. Hall JL. Preece PE. Willsher PC. double- Baker 2000 {published data only} blind study of prophylactic antibiotics in axillary lymph Baker KA. Yray MDS.18(3):278–84. 2005. 1992. WA. Antibiotic Gupta 2000 {published data only} prophylaxis in clean surgery: breast surgery and hernia Gupta R. et al. Abeyholm F. et al. Bold 1998 {published data only} Bold RJ. Kumar S. E. A prospective double-blind. Platt 1990 {published data only} Exener 1992 {published data only} Platt R. Parimbelli P. Crowe J. Burkhardt MD. comparing piperacillin/tazobactam versus placebo. New England Journal of undergoing breast surgery. Pawlita M. Woo PC. repair. Efficacy of prophylactic Capsules. Boyd 1981 {published data only} Gulluoglu 2013 {published data only} Boyd R. Todfield JJ. Lemperle G. Kloth DD.10(4):317–21. Gordan S.96(6):1378–83. REFERENCES References to studies included in this review mastectomy. Beatty JD. Andenaes K. Samdal F. European Journal of Surgery Medicine 1990. 2:68. Pollock RE. Perioperative teicoplanin prophylaxis double-blind study of the use of antibiotics at the time of in patients undergoing breast reconstruction with the Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 12 Copyright © 2014 The Cochrane Collaboration. [HGF– Oncology 2000. Bryan CS. of Chemotherapy 2006. Journal Surgical Infections 2009.

Smolianskaia AZ. Given KS. abdominal wall. Meyer Melling A.92(1):67. Pizzorusso M. Cavallaro G.a (2):135–41. Japanese Journal of Antibiotics 1988. Fereres J.80(4):275–9.766 patients. Efficacy of prophylactic antibiotics in reduction Melling 2005 {published and unpublished data} mammoplasty. Zaleznik DF.96(2):124–9. Davenport MS.166(3):556–60. Bray F. Prophylaxis against wound Coello 1993 infection following hernioplasty or breast surgery.9(1):9–14. Sylvia G. McCarron P. Chemotherapy 1998. 16th conference of the european wound umsetzung einer perioperativen antibiotikapropylaxe an management association. Infectious Diseases 1992. prospective trial of various regimens. Daniele L. Leaper A.33(5):476–80. A case control study. Lefebvres D. Whetter D. Germany 2005. Ruef Ch. A prospective Additional references evaluation of antibiotic prophylaxis for breast surgery following previous chemotherapy.466: Lewis RT. Nawaz M. Accardo R. Leigh D. Once daily use of ofloxacin for Sultan S.46(3):165–70. 2006:88. The antibiotic prophylaxis rate of surgical site infection after breast surgery. A Shamilov 1991 {published data only} comparison of warming with antibiotics or both on the Shamilov AK. Annals of Plastic Surgery 1994. Zurich]. Platt R. Antibiotic prophylaxis in breast DJ. Pezzolla Kumar S. et al. Sedgwick J. Uchino J. Zaleznik DF. Journal of Pakistan Medical Association 1989. Spicher 2003 {published data only} Melling 2006 {published and unpublished data} Spicher I. Ahmed R. Czech der klinik fur wiederstellungschirugie des universitatsspitals Republic. Karchmer AW. 15-17 Antibiotiki i Khimioterapiia 1991. Stuttgart. Minder J. Hall JL. Dellinger Breast Cancer Research 2004. et al. LeRoy 1991 {published data only} LeRoy J. Nicholas 2007 {published data only} Nicholas P. Serletti JM. The cost of infection in surgical patients: Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 13 Copyright © 2014 The Cochrane Collaboration. Marie Pierre C.45(3):217–23. randomizzato su 1400 casi trattati]. Zucker JR.39(1):9–12. British Senologica Risultati preliminari di uno studio multicentrico Journal of Surgery 2005.15(4):303–5. Kumar S.6(6):229–39. 2006. Bray 2004 91(5):445–8. Kumar S. EP. Mufti T. Melling A. surgery: a randomised double-blind trial involving 1. Baker E. Pennel 2004 {published data only} Pennel N. Parkin DM. Abstract no. Yazdan Y. Whetter D. Antibiotic prophylaxis for patients and wound infection following breast surgery. Hopkins CC. Wound infection in breast Sasaki 1988 {published data only} augmentation: the role of prophylactic perioperative Sasaki F. Cortino GR. Sanguinetti 2009 {published data only} Kumar 2005 {published and unpublished data} Sanguinetti A. et al.36(6):43–4. 2005. Cirocchi R. Annali Italiani di Chirurgia 2009.31(Suppl B):43–8. Journal of Coello R. Giard S. Zucker JR. EH. Prague. Caldwell 118(4):742–7. Whetter D. Minerva Chirurgica Platt 1993 {published data only} 1994. Barberini F. Hata Y. et al. Platt R. Morimoto 1998 {published data only} Sultan 1989 {published data only} Morimoto K. Journal of cephalosporins for preoperative prophylaxis in breast Surgical Oncology 2007. Rosato L. Leaper D. wound management association conference. Alivino P. 2000. Long acting versus short acting in patients at high risk of surgical site infection. Takada N. et al. Local VE. Hopkins CC. Tataryn D. prophylaxis in breast cancer surgery. Perioperative antibiotic prophylaxis Hall JC. Herrera HR. Should antibiotic prophylaxis be used routinely Serletti 1994 {published data only} in clean surgical procedures: a tentative yes. Journal of Hospital Infection Antimicrobial Chemotherapy 1993.44 Use of prophylactic antibiotics in clean surgical cases . Fournier C. The changing global Platt 1992 {published data only} pattern of female breast cancer incidence and mortality. Bulletin du Cancer 2004. Prevention of surgical site infection Thomas R. Bartlett C. Beer GM. . Journal of undergoing breast surgery.Thursday 14:00-15: 30. Dellinger Hall 2000 {published data only} EP. Aesthetic Plastic Surgery 1991. September. Kyunoh K. Jean Charles Thomas 1999 {published data only} N. Swiss Surgery 2003. Glenister H. Preliminary results of a multicenter randomised randomised controlled trial comparing the efficacy of study on 1400 cases [Antibiotico profilassi in Chirurgia antibiotics and perioperative wound warming. Mamazza J. Surgery 1995.109. A antibiotics. 18-20 May. European of postoperative complications in breast cancer patients.49(1):59–60. Melling A. Ltd. Perioperative antibiotic prophylaxis at the clinic of warming may enhance the effect of prophylactic antibiotics reconstructive surgery of the university hospital Zurich [Zur in breast surgery. Leaper D. 114–20. Mohammad G. Weigand FM. Szucs T. et al. Wound infection prophylaxis in breast surgery: a surgery. study on the transfer of cefbuperazone into postoperative exudate in patients subjected to cancer mastectomy or Lewis 1995 {published data only} thyroidectomy. Lloyd-Smith W. Wong PF. Chemotherapy 1999. Baker L. Rinaldo after breast cancer surgery by targeted prophylaxis antibiotic M. Karchmer AW. V25-3:67. Kinoshita H. Published by John Wiley & Sons.

Jarvis wound infection. Cochrane Handbook for Systematic Reviews Ritter 1988 of Interventions Version 5. [DOI: 10. Callis K. Available from Schulz 1995 www. Issue 3. outcomes in breast cancer: Manual update. BMJ 2003. Higgins JPT. Year Book Medical. Cochrane Database of Systematic Reviews 2007. Green S implants. Antimicrobial Harris JR (Ed).1. Deberles MF. Bray F. of Interventions Version 5.83(6):18–29. surgery for proximal femoral and other closed long bone Morris 1988 fractures.1002/ Sheridan 1994 14651858. Hayes RJ. Journal of Gruendemann BJ. Culver DH. International Journal of Cancer Higgins 2011 1999. Issue 1. [DOI: 10. Missouri: Mitchell DH. Ferlay J. Antibiotic prophylaxis for 1999. Pearson ML. Fournier C.25 Infection Control and Hospital Epidemiology 1999. a case control study. Identifying patients at high risk of 2002:http://www. Swift G. Pittet D.5(2):94–106.CD004637] Sheridan RL. Silver LC. An updated meta-analysis on the effectiveness of surgery.uk/web/HPAwebFile/HPAweb˙C/ 1194947317696 2000. Kalra L.0 [updated March 2011]. Cancer of the Breast. /www. Chapter Sanchez-Manuel 2007 6: Searching for studies. Lefebvre 2000 Sajid 2012 Lefebvre D. Montandon D. Altman DG. Seco-Gil JL. [DOI: 10. Philadelphia: resistance in 2000.pub2] and timing of mastectomy on the development of post Gruendemann 2001 mastectomy non sociocomial wound infection. Ritter MA. Horan TC.18(4): Lefebvre C. 2001.1002/14651858. Donegan 1995 Mitchell 1999 Donegan WL. Altman DG. Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 14 Copyright © 2014 The Cochrane Collaboration. Empirical Majoribanks 2004 evidence of bias. Prophylactic Mangram 1999 antibiotics and their role in the prevention of surgical Mangram AJ. preoperative prophylactic antibiotics in patients undergoing Lefebvre 2011 breast surgical procedures. Thompson SG. on behalf of the 312–7. Mead Morgan W. J. Published by John Wiley & Sons.CD003769. Surgical wound Elsevier Science. Hooton TM. Int. ANZ Journal of Surgery Gillespie WJ. London: W.121(2):206–15. B. Presse Medicale 2000.0 [updated March 2011].CD000244. Improving Haley RW.cochrane-handbook. 18-29 (1999). Walenkamp G. White JW.cochrane-handbook. Cochrane Database of Systematic Reviews 2010. Spratt JS. Schulz KF. Tompkins RG. Robinson K.327 of the worldwide mortality from 25 cancers in 1990. Orthopaedic www. Cancer.org. Journal of Hospital infection 1993. Diseases of the Breast. 2004. The exogenous sources and controls The Cochrane Collaboration.1002/14651858. Deeks JJ. (editors). Ltd. (7414):557–60. 1994:43–65. Chapter 8: Assessing risk Pittet B. Marmion P. Pisani P. http:/ Lippincott Williams & Wilkins. 1995. Manheimer E.uk/userfiles/documents/ surgical wound infection. Advances in Surgery.org. Glanville J. on behalf of the Cochrane Statistical Methods Group and the Cochrane Bias Pittet 2005 Methods Group (Editors).29(35):1927–32. Altman DG. Infection in breast of bias in included studies. 1985. 2011. Dimensions of methodological quality Majoribanks J. Rapisarda IF. The Breast Journal 2012. Akhter N.hpa.pdf. In: Higgins JPT.7(4):23–8. and surgical wound infection risk factors in breast cancer Bonomi R. Hutson K.mccn. Saunders. Burke JF.140(37):363–6. surgical setting. 83. Cochrane Information Retrieval Methods Group. Nursing 1988. Parkin DM. Penel N. . NICE 2002 Haley 1985 National Institute for Clinical Excellence. The effect of method of biopsy Issue 3. PHLS 2000 Harris 2004 Public Health Laboratory Service (PHLS). JAMA 1995. Morris DM. England and Wales. Higgins 2003 Pisani 1999 Higgins JP. Antibiotic prophylaxis for (editors).org. Infection prevention in the the Louisiana State Medical Society 1988.20(4): (4):239–50. In: Higgins JPT. 247–78. 2011. infection surveillance: the importance of infections that Gillespie 2010 develop after hospital discharge. Lancet Infectious Diseases 2005. Jordan V. Guideline for the prevention of surgical site infection.1. Systematic Reviews 2004. Cochrane Database of trials. Antibiotic prophylaxis associated with estimates of treatment effects in controlled for elective hysterectomy. Erratum: Estimates Measuring inconsistency in meta-analyses. Incidence Sajid MS. Magnum SS.pub3] The Cochrane Collaboration.273(5):408–12. Cochrane Handbook for Systematic Reviews hernia repair. Chalmers I.69(2):117–20. NICE Emori GT. Available from of microorganisms in the operating room.nhs. WR. Gilberts GL. American Journal of Epidemiology Improving˙outcomes˙breastcancer˙manual. Green S Sanchez-Manuel FJ.

sign. .uk/methodology/filters. random (accessed 14 August 2008). Wilson 1986 Spauwen 2000 Wilson APR.203(5):729–34. Brodwick-Villa G. http://www. Walchetseder C. Langer S.69(10):852–6. Benefits and risks of antibiotic prophylaxis. American Surgery 2003. College of Surgeons 2006. Cascales-Sanchez P. Tran CL. Strohmer H. Journal of Infection 2003. Wobbes T. Preoperative core needle biopsy as an independent risk for preventing wound infection after breast surgery: a factor for wound infection after breast surgery. Lancet 1986. Scottish Intercollegiate Guidelines Network.23:211–3. Witt 2003 Tejirian 2006 Witt A.sign. Ltd. Obstetrics systematic review and meta analysis. Journal of the American and Gynecology 2003. Antibiotic prophylaxis E. ∗ Indicates the major publication for the study Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 15 Copyright © 2014 The Cochrane Collaboration. Haigh PI. . European infections for use in clinical trials of antibiotic prophylaxis. Journal of Plastic Surgery 2000.SIGN 2008a Tran 2003 Scottish Intercollegiate Guidelines Network (SIGN). Nosocomial infection Scottish Intercollegiate Guidelines Network (SIGN). Immediate A scoring method (ASEPSIS) for postoperative wound breast reconstruction: results and satisfaction. Vazquez-Aragon 2003 Vazquez-Aragon P. Gruneberg RN. DiFronzo LA. Treasure T.101(4):745–50. infection in women undergoing operation for breast cancer? Available at http://www. Sturridge MF.uk/pdf/sign104.ac. SIGN 2008b Villar-Canovas MT. Van Der Sluis RF. Antibiotic Does re-operation predispose to post-operative wound prophylaxis in surgery: a national clinical guideline. Lizan-Garcia M.1(8476):311–3.html# prospective study. Published by John Wiley & Sons. 2008:9–12.ac. Search and related risk factors in a general surgery setting: a filters.46(1):17–22. Garcia-Olmo D. DiFronzo A. Kubista Tejirian T. Spauwen PHM.pdf. Yavuz D.

history of mental illness Total breast excision participants: 76 Study included breast reconstruction and implants. bias) double-blind study (no further detail Participants given) Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 16 Copyright © 2014 The Cochrane Collaboration. Age 6 years or above. Admitted for plastic surgery and able to give informed consent. .single dose. serious underlying disease. Method of generating the random schedule reported Allocation concealment (selection bias) Unclear risk Comment: not reported Blinding (performance bias and detection Low risk Comment: reported as placebo-controlled.CHARACTERISTICS OF STUDIES Characteristics of included studies [ordered by study ID] Amland 1995 Methods RCT: randomisation via computer-generated blocks of 10 Loss to follow-up: < 20% Intention-to-treat: unclear Power calculation unclear as not stated by the author Reliable primary outcome: done Participants Male and female. received antibiotics in the 2 weeks prior to surgery. pregnant or breast feeding. Ltd. malabsorption illnesses. which have not been included in this analysis as the author could not be contacted to find out if reconstruction was secondary to cancer treatment Interventions I) Azithromycin . Trial exclusion criteria: intolerance to trial drug. Dose according to body weight. Published by John Wiley & Sons. Dose taken 8 pm the night before surgery (n = 42) C) Placebo used but no details provided (n = 34) Outcomes Infection rates Adverse effects Notes Length of follow-up: 30 days Funding organisation not stated Country of origin: Norway Risk of bias Bias Authors’ judgement Support for judgement Random sequence generation (selection Low risk Quote: “randomisation was performed in bias) blocks of 10 patients using a randomised chart” Comment: computer-generated blocks of 10. renal or hepatic impairment. receiving carbamazepine or cyclosporins. terminal illness or immunosuppression.

Amland 1995 (Continued) Blinding (performance bias and detection Low risk Comment: reported as placebo-controlled. stated as under-powered Clear definition of infection Participants All female. The study states “there were 8 wound in- fections in the azithromycin group and 32 in the placebo group.” Comment: only 1 patient was lost to fol- low-up (placebo group) Selective outcome data Low risk Comment: the study protocol was not available. Comment: the wound was assessed “by the physician” using a “specifically designed wound assessment chart”. . recent antibiotic use or infection. however. It was judged that the physician undertaking the wound assessment was likely blinded Incomplete outcome data (attrition bias) Low risk Quote: “the inclusion of these patients in All outcomes the final analysis after the intention to treat principle did not alter the end result signif- icantly. the results section clearly reports the incidence of wound in- fection using a prespecified scoring system. 18 years old or above undergoing axillary lymph node dissection Excluded if: there was history of allergy to cephalosporin. inflammatory breast cancer. Ltd. aspirin use within 5 days. diabetes. wound infection from surgery in the past 4 weeks. However. bias) double-blind study (no further detail Treatment Provider given) Blinding (performance bias and detection Low risk Quote: “Blinding was maintained until ev- bias) ery patient had completed follow-up and Outcome assessor all diagnosis of wound infection had been made”. the study appears to be free of any other source of bias Bold 1998 Methods RCT: randomisation using computer-generated blocks Loss to follow-up: < 20% Intention-to-treat: unclear Power calculation: unclear. Published by John Wiley & Sons. concomitant isolated limb perfusion or those undergoing immediate breast re- Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 17 Copyright © 2014 The Cochrane Collaboration.” Other sources of potential bias High risk Comment: in the acknowledgements the authors state “the present work was sup- ported by Pfizer AS.” This is a pharmaceu- tical company. pregnancy or breast feeding. hepatic or renal impairment.

Method of generating the random schedule reported Allocation concealment (selection bias) Low risk Comment: hospital pharmacy performed randomisation and provided placebo or an- tibiotic in identical IV bags Blinding (performance bias and detection Low risk Quote: ”Blinding of antibiotic administra- bias) tion was accomplished through the hospi- Participants tal pharmacy.” The authors go on to state “[pharmacy] provided the placebo or ce- fonicid in identical intravenous bags” Comment: treatment provider likely blinded Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 18 Copyright © 2014 The Cochrane Collaboration. intravenously 60 minutes prior to operation (n = 88) C) Placebo used was normal saline as per antibiotic regime (n = 90) Outcomes Infection rates Re-hospitalisation rates Cost of care Adverse events Notes Length of follow-up: 4 weeks post surgery Funded by SmithKline Beecham Country of origin: USA Risk of bias Bias Authors’ judgement Support for judgement Random sequence generation (selection Low risk Quote: “Randomisation was accomplished bias) with a computer-generated block randomi- sation table”. 141 were confirmed breast cancer patients Interventions I) Cefonicid 1 g.Bold 1998 (Continued) construction Total number of patients randomised = 200 22 excluded after randomisation Of these. . Ltd. Published by John Wiley & Sons.” The authors go on to state “[pharmacy] provided the placebo or ce- fonicid in identical intravenous bags” Comment: participants likely blinded Blinding (performance bias and detection Low risk Quote: “Blinding of antibiotic administra- bias) tion was accomplished through the hospi- Treatment Provider tal pharmacy. Comment: computer-generated blocks used.

This left 90 patients in the placebo group and 88 patients in the cefonicid group. However. predefined clinical indicators Participants All females aged between 18 and 80 with histologically diagnosed breast cancer who were schedules for modified radical mastectomy (MRM) Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 19 Copyright © 2014 The Cochrane Collaboration.Bold 1998 (Continued) Blinding (performance bias and detection Unclear risk Quote: “patients were followed up in an bias) outpatient clinic and monitored for signs of Outcome assessor symptoms of infection. .10 were from the placebo group and 12 from the treatment group. Comment: the reasons for exclusion seem valid and are unlikely to introduce bias.” The authors go on to say “a research nurse also contacted the patient and referring physician for wound follow up for 4 weeks after surgery” Comment: no comment is made as to whether the assessors remained blinded Incomplete outcome data (attrition bias) Low risk Quote: “Twenty-two patients were ex- All outcomes cluded from the analysis because of pro- tocol violations”. Published by John Wiley & Sons. the study appears to be free of any other source of bias Cabaluna 2012 Methods RCT: randomisation sequence generated by computer Loss to follow-up: < 20% Intention-to-treat: not stated. overall the loss to follow-up was less than 20% Selective outcome data Low risk Quote: in the introduction to the study the authors state “the study was under- taken to determine whether a single dose of cephalosporin could decrease the incidence of post operative wound infection”. They go on to state “the results would be subject to a cost benefit analysis” Comment: the results clearly document the incidence of wound infection in table II as well as a cost benefit analysis in table III Other sources of potential bias High risk Comment: the paper states “the study was sponsored in part by a grant from Smith Kline & Beecham laboratories”. no patients excluded from the analysis Power calculation: not stated Clear definition of infection: done. Ltd. This is a pharmaceutical company.

Method of generat- ing the random schedule reported Allocation concealment (selection bias) Low risk Quote: “The table of random numbers was only available to a single nurse research as- sistant. . surgeon. anesthesiologist and all other reporting room staff were blinded as to treatment allocation” Comment: Treatment provider likely blinded. Participants Patient. patients with existing local infection Interventions I) Cefazolin 1g given intravenously at the time of anaesthesia C) Normal Saline 10ml Outcomes Primary outcome: incidence of surgical site infection (SSI) within 30 days of MRM Secondary outcomes: presence of haematoma and seroma. Blinding (performance bias and detection Low risk Quote: “A nurse research assistant pre- bias) pared the antibiotic and placebo solutions. anesthesiologist and all other reporting room staff were blinded as to treatment allocation” Comment: Participants likely blinded. patients who had had antibiotics in the last week. surgeon. patients receiving steroids. patients planned for imme- diate reconstruction. Published by John Wiley & Sons. Ltd. patients with diabetes or severe malnutrition. Treatment Provider Patient. recurrent breast cancer.Cabaluna 2012 (Continued) Total number: 254 No patients excluded after randomisation Exclusion criteria: males. Notes Length of follow up: 30 days post surgery Source of funding: not stated Country of origin: The Philippines Risk of bias Bias Authors’ judgement Support for judgement Random sequence generation (selection Low risk Quote: “patients were randomised with a bias) computer-generated randomization list” Comment: computer generated random- ization lists were used. patients with allergy to cephalosporins. who was responsible for preparing the treatment and placebo solutions” Comment: Allocation concealment achieved using a research nurse Blinding (performance bias and detection Low risk Quote: “A nurse research assistant pre- bias) pared the antibiotic and placebo solutions. previous radiotherapy. Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 20 Copyright © 2014 The Cochrane Collaboration.

myasthenia gravis. tendency to bleeding. Incomplete outcome data (attrition bias) Low risk Quote: “ No patent withdrew from the All outcomes study”. Treatment continued twice daily for 3 days post surgery. diabetic. C) Control group received no placebo (n = 24) Outcomes Inflammatory responses Infection rates Flap necrosis (stated as minor in both groups) Pain Range of movement Notes Length of follow-up 5 days post surgery Country of origin: Japan Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 21 Copyright © 2014 The Cochrane Collaboration. hepatic or renal impairment. Comment: Low risk of attrition bias Selective outcome data Low risk Quote: “the primary outcome was occur- rence of SSI within 30 days of MRM” Comment: The authors state the primary outcome of the study was occurrence of surgical site infection (SSI). The authors clearly state the definition of a SSI and re- port the incidence of SSIs in the results sec- tion Other sources of potential bias Low risk Comment: the study appears to be free of other sources of bias Chow 2000 Methods RCT: computer-generated sequence Loss to follow-up: < 20% Intention-to-treat: not done Power calculation: unclear Clear definition of infection: unclear. immunosuppression or antibiotics within the preceding 2 weeks Interventions I) Clarithromycin 500 mg orally first dose commenced the day prior to surgery (n = 28) . . Addressed inflammation rather than infection Participants All females diagnosed with breast cancer and undergoing mastectomy Total patients randomised: 56 with 2 being excluded after randomisation Excluded if: pregnant. Published by John Wiley & Sons.Cabaluna 2012 (Continued) Blinding (performance bias and detection Low risk Quote: “the outcome evaluator was blinded bias) as to what study arm the patient belonged Outcome assessor to” Comment: Outcome assessor likely blinded. Ltd.

” Comment: control group received no treat- ment Blinding (performance bias and detection High risk Quote: ”Patients in the study group were bias) given oral clarithromycin. Patients in the Treatment Provider control group did not receive any clar- ithromycin. Ltd.” Comment: the number lost to follow-up is low and the reason was valid Selective outcome data Low risk Comment: the study protocol was not available.Chow 2000 (Continued) Risk of bias Bias Authors’ judgement Support for judgement Random sequence generation (selection Low risk Quote: “consecutive patients (except those bias) excluded) were enrolled and randomised into two groups by computer”. This was documented in the introduction to the Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 22 Copyright © 2014 The Cochrane Collaboration. Method of generating the ran- dom schedule reported Allocation concealment (selection bias) Unclear risk Comment: no further information is given on the randomisation process Blinding (performance bias and detection High risk Quote: “Patients in the study group were bias) given oral clarithromycin. Published by John Wiley & Sons. . Comment: randomised into 2 groups by computer.” Comment: control group received no treat- ment Blinding (performance bias and detection Low risk Quote: “All surgeons and medical staff re- bias) sponsible for assessing the outcome were Outcome assessor unaware of the randomisation results be- cause separate prescription sheets were given for the clarithromycin prescription”. Comment: blinding of outcome assessor achieved Incomplete outcome data (attrition bias) Low risk Quote: “fifty six patients with breast can- All outcomes cer were recruited for the randomised trial. however. the outcomes of this study included postoperative wound infec- tion as well as evidence of the systemic in- flammatory response syndrome. Two patients in the control group dropped out due to refusal of venepuncture. Patients in the Participants control group did not receive any clar- ithromycin.

predefined clinical indicators Participants All females with primary non recurrent operable breast cancer Total number: 369 3 excluded after randomisation Exclusion criteria: males. patients with an allergy to beta-lactam or cephalosporin an- tibiotics. patients having neoadjuvant chemotherapy. Ltd. immunosuppression. locally advanced. adverse reactions. and the amount of parenteral fluid administered during the perioperative pe- riod”. blood loss. metastatic or bi- lateral breast cancer. Comment: there was no imbalance in the baseline characteristics and the study seems to be free from other forms of bias Gulluoglu 2013 Methods RCT: randomisation sequence generated by computer Loss to follow-up: < 20% Intention-to-treat: done. Cushings disease. patients with diabetes mellitus. time to SSI development.Chow 2000 (Continued) study and in the outcomes. area of dissection. . or history of antibiotics in the last month Interventions I) Ampicillin-sulbactam (Ampisid) by intravenous bolus at the time of anaesthesia C) No intervention Outcomes Primary outcome: incidence of surgical site infection (SSI) in patients with a BMI over 25 Secondary outcomes: incidence of SSI in patients with a BMI under 25. Published by John Wiley & Sons. culture results. renal disease. cost Notes Length of follow up: 30 days post surgery Source of funding: not stated Country of origin: Turkey Risk of bias Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 23 Copyright © 2014 The Cochrane Collaboration. 3 patients excluded from the analysis Power calculation: done and adequately powered Clear definition of infection: done. patients undergoing immediate reconstruction. operation time. The authors state “no patient devel- oped a wound infection” Other sources of potential bias Low risk Quote: “There were no significant differ- ences between the two groups in terms of age. HIV. The results dis- cuss the changes in several inflammatory markers and the results of blood culture tests. ductal carcinoma in situ. blood count abnormalities. addition cancers.

Incomplete outcome data (attrition bias) Low risk Quote: “Three patients were excluded from All outcomes the primary analysis because they under- went a second surgery within 1 month: 2 in the prophylaxis group and 1 in the con- trol group”. Ltd.Gulluoglu 2013 (Continued) Bias Authors’ judgement Support for judgement Random sequence generation (selection Low risk Quote: “Allocation of patients was done by bias) simple randomization using a computer- generated table of random numbers” Comment: computer generated random- ization lists were used. Those in the con- trol group received no intervention” Comment: control group received no treat- ment Blinding (performance bias and detection High risk Quote: ”Patients in the prophylaxis group bias) received 1 g ampicillin-sulbactam by intra- Treatment Provider venous bolus injection. with a legitimate reason Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 24 Copyright © 2014 The Cochrane Collaboration. Those in the con- trol group received no intervention” Comment: control group received no treat- ment Blinding (performance bias and detection Low risk Quote: “Wounds were inspected before bias) their discharge and examined by care-givers Outcome assessor blinded to the allocation” Comment: outcome assessor likely blinded. Method of generat- ing the random schedule reported Allocation concealment (selection bias) Low risk Quote: “Allocations were concealed from patients and observers. . The computer gen- erated randomization list was prepared by an investigator with no involvement in the trial” Comment: allocation concealment achieved using a research assistant Blinding (performance bias and detection High risk Quote: “Patients in the prophylaxis group bias) received 1 g ampicillin-sulbactam by intra- Participants venous bolus injection. “Data from the randomized patients were analyzed on an intention-to- treat basis” Comment: small number of loss to follow up. Published by John Wiley & Sons.

C) Placebo: normal saline as per treatment regime Outcomes Infection rate Adverse events Time to wound healing Notes Follow-up for 10 to 14 days post discharge Funding not stated Country of origin: UK Risk of bias Bias Authors’ judgement Support for judgement Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 25 Copyright © 2014 The Cochrane Collaboration. 18 years of age or above Total number: 357 44 excluded after randomisation Exclusion criteria: known penicillin allergy. This is reported clearly as the first outcome in the results Other sources of potential bias High risk Quote: “patients in the control group had significantly more open surgical biopsies than those in the prophylaxis group (16% vs 6%)” Comment: this could potentially be a source of bias as patients who had an open surgical biopsy may be more likely to have an infection Gupta 2000 Methods RCT: randomisation sequence generated by computer Loss to follow-up: < 20% Intention-to-treat: not done. 6 patients excluded from the analysis Power calculation: done. infection within 72 hours pre-surgery. preg- nant. Ltd.Gulluoglu 2013 (Continued) Selective outcome data Low risk Comment: primary outcome measure clearly stated on page 38 (SSI incidence). Receiving mastectomy or wide local excision with or without axillary Interventions I) Augmentin 1.2 g intravenous. Given perioperatively (after induction but before first incision). . Single dose. on other antibiotics or with hepatic or renal impairment Treatment group: 177 Placebo group: 180 Diagnosis of breast cancer. Published by John Wiley & Sons. predefined clinical indicators Participants All female. but under-powered Clear definition of infection: done.

9% sterile saline) by reference to a computer generated list”. Published by John Wiley & Sons. It was then ad- ministered to the patients as a single intra- venous bolus injection through a peripher- ally placed 22 gauge intravenous cannula.9% Treatment Provider sterile saline). . It was then ad- ministered to the patients as a single intra- venous bolus injection through a peripher- ally placed 22 gauge intravenous cannula. Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 26 Copyright © 2014 The Cochrane Collaboration. Finally the author state “neither the patient nor any of the staff involved with this study were aware of the allocation of treatment until after the study had been completed”.” Comment: sealed. Participants were blinded Blinding (performance bias and detection Low risk Quote: ”Patients were randomized to re- bias) ceive the antibiotic or placebo (20 ml 0. shortly after the induction of anaesthesia”. observer blind. sequentially- numbered envelopes Blinding (performance bias and detection Low risk Quote: “Patients were randomized to re- bias) ceive the antibiotic or placebo (20 ml 0. Comment: the study is described as ”a prospective. randomised.” The administration of an- tibiotic is then described “Where the study agent was administered the anaesthetist was instructed to reconstitute the antibiotic from vials of sterile powder.” The administration of an- tibiotic is then described “Where the study agent was administered the anaesthetist was instructed to reconstitute the antibiotic from vials of sterile powder. placebo-controlled study“. Neither the patient nor any of the staff involved with this study were aware of the allocation of treatment until after the study had been completed. opaque. The randomization codes were kept in sealed envelopes. Ltd.Gupta 2000 (Continued) Random sequence generation (selection Low risk Quote: “Patients were randomized to re- bias) ceive the antibiotic or placebo (20 ml 0. Codes were sequentially allocated to randomized patients.9% Participants sterile saline). Comment: computer-generated list used. Method of generating the random schedule reported Allocation concealment (selection bias) Low risk Quote: ”The randomization list was gen- erated by computer.

so making this study ‘ob- server blind”. Comment: 3 patients were lost from each group for the efficacy analysis. Comment: healthcare providers blinded. . Comment: outcome assessors were blinded Incomplete outcome data (attrition bias) Low risk Quote: “Protocol violations resulted in six All outcomes patients being excluded from the intention- to-treat group”. The number of secondary endpoints is also documented Other sources of potential bias Low risk Comment: the study appears to be free of other sources of bias Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 27 Copyright © 2014 The Cochrane Collaboration. duration of post-oper- ative hospital stay.Gupta 2000 (Continued) shortly after the induction of anaesthesia”. The table also states that 313 pa- tients “completed study”. Table 1 shows that 357 pa- tients were randomised and screened and 351 patients were “valid for efficacy anal- ysis”. “The pri- mary end point was the incidence of wound infection. Finally the authors state “neither the patient nor any of the staff involved with this study were aware of the allocation of treatment until after the study had been completed”. but the study reports that an intention-to-treat analysis was undertaken on 351 patients Selective outcome data Low risk Quote: the methods section details the pri- mary and secondary outcomes. The anaesthetist was not blinded but took no further part in the study Blinding (performance bias and detection Low risk Quote: “At no time until the breaking of bias) the code was the investigator made aware Outcome assessor of whether the active agent or the placebo was administered. No information is given on these 44 patients who did not complete the study. Secondary endpoints included febrile morbidity. delay in progressing to chemotherapy radiotherapy or surgical cosmesis due to wound infection and the incidence of chest or urinary infection. sep- ticaemia or other infections” Comment: in the results the incidence of wound infections are clearly shown in table 6. Published by John Wiley & Sons. Ltd.

Comment: allocation concealment achieved Blinding (performance bias and detection Unclear risk Comment: the study had no placebo for the bias) control group. Ltd. antibiotics within 72 hours. warfarin therapy. Method of generating the random schedule reported Allocation concealment (selection bias) Low risk Quote: “Concealment was achieved by placing the group allocation into opaque. the authors state Participants “Patients in the study group received flu- cloxacillin 2 g administered intravenously. reconstructive surgery. serially numbered envelopes that were monitored to detect breaches of the ran- domisation protocol”. immediately after the induction of general anaesthesia” Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 28 Copyright © 2014 The Cochrane Collaboration. Only 2 patients (one in each group) had received preoperative chemotherapy Interventions I) Single IV dose of 2 g flucloxacillin administered over at least 5 minutes immediately after the induction of general anaesthesia C) No treatment Outcomes Infection rates Adverse effects Cellulitis Wound scores Notes Follow-up: 42 days Country of origin: Australia Risk of bias Bias Authors’ judgement Support for judgement Random sequence generation (selection Low risk Quote: “Patients were allocated to a group bias) using computer-generated random num- bers arranged into blocks with a cell size of 10”.Hall 2006 Methods RCT: computer-generated random numbers arranged into blocks of 10 Intention-to-treat analysis: done Power calculation: done Reliable primary outcome: done No loss to follow-up Participants 618 (616 women and 2 men). phenytoin therapy or existing infection. Comment: computer-generated list used. . Excluded if penicillin hypersensitivity. over at least 5 min. Published by John Wiley & Sons. however. Scheduled for non-reconstructive breast surgery.

There was no loss to follow-up Selective outcome data Low risk Quote: ”Wound infection was the primary endpoint. immediately after the induction of general anaesthesia” As the antibiotic was given after the induc- tion of anaesthesia by an anaesthetist it may be assumed that treatment personnel was blinded but this was not reported in the study Blinding (performance bias and detection Low risk Quote: “All assessments [of wound infec- bias) tion] were performed without any knowl- Outcome assessor edge of the patient’s allocated group” Comment: outcome assessors blinded Incomplete outcome data (attrition bias) Low risk Comment: Figure 1 shows 618 patients All outcomes randomised to either control or flu- cloxacillin. over at least 5 min. however. the authors state Treatment Provider “Patients in the study group received flu- cloxacillin 2 g administered intravenously.” Comment: the study protocol was not available but the important outcome mea- sures stated in the methods section are re- ported in the results Other sources of potential bias Low risk Comment: the study appears to be free of other sources of bias Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 29 Copyright © 2014 The Cochrane Collaboration. In the results the authors clearly document the incidence of wound infection “Both groups had a similar rate of postoperative wound infection: ten of 311 (3. Wounds were also evaluated using a previously validated scoring system”. It was defined as either the dis- charge of pus.2 per cent) in the flucloxacillin group and 14 of 307 (4. Published by John Wiley & Sons. Ltd. or a serous discharge contain- ing pathogenic organisms.Hall 2006 (Continued) As the antibiotic was given after the induc- tion of anaesthesia by an anaesthetist it may be assumed that participants were blinded but this was not reported in the study Blinding (performance bias and detection Unclear risk Comment: the study had no placebo for the bias) control group. All patients were followed up at 42 days. .6 per cent) in the control group.

penicillin hypersensitivity. . other Participants medical staff.” Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 30 Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons. were blinded to the participant’s allocation.Paajanen 2009 Methods RCT: method of randomisation not reported No loss to follow-up Intention-to-treat analysis: done as all the participants were analysed in the groups to which they were randomised Power calculation: unclear as not stated by the author Reliable primary outcome: done Clear definition of infection: done Participants All females patients undergoing non-reconstructive breast cancer surgery between years 2004 and 2007 were included Total number: 292 Exclusion criteria patients with lack of consent.” Comment: allocation concealed using sealed opaque sequentially numbered en- velopes Blinding (performance bias and detection Low risk Quote: “The research group including the bias) operating surgeon. Confirmed preoperatively by mammographic stereotactic or ultrasound-guided core needle biopsy Interventions I) Intravenous 1 g of dicloxacillin in a 100 ml bottle. research nurses. Single dose 30 minutes before surgery. C) Placebo infusion of 100 ml of saline Outcomes Infection rates Notes Follow-up: 30 days Country of origin: Finland Risk of bias Bias Authors’ judgement Support for judgement Random sequence generation (selection Unclear risk Quote: “The patients were randomised to bias) receive either an intravenous single dose of 1 g of dicloxacillin in a 100 ml or a placebo infusion of 100 ml of saline 30 min prior to surgery. and study participants. Ltd. logistic failure Treatment group: 144 Control group: 148 Diagnosis of breast cancer.” Comment: method of generating the ran- dom schedule not reported Allocation concealment (selection bias) Low risk Quote: “The hospital pharmacy generated allocation using sealed opaque sequentially numbered envelopes.

Ltd.” Comment: outcome assessors were blinded adequately Incomplete outcome data (attrition bias) Low risk Comment: there was no loss to follow-up.” Comment: there was no imbalance in the baseline characteristics and the study seems to be free from other forms of bias Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 31 Copyright © 2014 The Cochrane Collaboration.” Comment: healthcare providers were blinded adequately Blinding (performance bias and detection Low risk Quote: “All assessments were performed bias) without knowledge of the patient’s assigned Outcome assessor group. The results state “The rate of postoperative SSI was 5. research nurses. A clear definition of in- fection is documented.” Comment: the study protocol was not available but the important outcome mea- sures stated in the methods section are re- ported in the results Other sources of potential bias Low risk Quote: “The patient characteristics and risk factors for SSI were similar in the an- tibiotic prophylaxis and placebo groups. . and study participants. Published by John Wiley & Sons. 8% (13/148) in the placebo group. All outcomes Table 3 depicts that all the randomised participants were analysed in the group to which they were allocated Selective outcome data Low risk Quote: the authors state that SSI was the primary endpoint.6% (8/144) in the dicloxacillin group and 8.Paajanen 2009 (Continued) Comment: participants were blinded ade- quately Blinding (performance bias and detection Low risk Quote: “The research group including the bias) operating surgeon. were blinded to the participant’s allocation. other Treatment Provider medical staff.

Included are those who speak English. Ltd. but may be under-powered for the breast group Clear definition of infection: done Participants Included male and female patients aged 18 or above having mastectomy. . Dose regime: single dose. have no recognised infection at the time of surgery. C) Placebo was a mixture of glycerin. axillary node clearance or reduction mammoplasty. lumpectomy. randomisation via blocks of 10 Loss to follow-up: < 20% Intention-to-treat: unclear Power calculation: adequate for the study as a whole.Platt 1990 Methods RCT. excisional breast biopsy. recent antibiotic use or known allergy to beta-lactam antibiotics. Kline and French Laboratories Country of origin: USA Risk of bias Bias Authors’ judgement Support for judgement Random sequence generation (selection Unclear risk Quote: ”patients were randomly assigned bias) separately in blocks of 10 to receive cefoni- cid or placebo.” Comment: allocation concealment achieved Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 32 Copyright © 2014 The Cochrane Collaboration. Within 90 minutes pre-surgery (n = 303).” They go on to state “in- vestigators were required to return these la- bels intact. Total number of participants: 606 18 years old or over Interventions I) Cefonicid 1 g intravenous. Published by John Wiley & Sons. lived within 35 miles of the hospital.” Comment: method of generating the ran- dom schedule not reported Allocation concealment (selection bias) Low risk Quote: “treatment codes were not known by anyone at the participating centres un- less the sealed opaque label attached to each vial was opened. mannitol and riboflavin given as per the treatment regime Outcomes Infection rate Adverse reaction to treatment Time to onset of infection Associated morbidity from wound infection Economic evaluation Other infective episodes Notes Length of follow-up: 4 to 6 weeks Sponsored by Smith.

The authors state “treat- ment codes were not known by anyone at the participating centres unless the sealed opaque label attached to each vial was opened.” They repeat this in the surveil- lance of wound infection paragraph “all in- vestigators were unaware of the treatment codes until the last evaluation was com- pleted. No separate exclusion data are given for the breast cancer patients Overall the loss to follow-up was less than 20% and therefore judged to be adequate Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 33 Copyright © 2014 The Cochrane Collaboration. double-blind trial”. “Cefonicid Treatment Provider and placebo were supplied in identical numbered vials.” They go on to state “investigators were required to return these labels intact”. ”. 50 patients from the treatment group and 51 from the control group were excluded. Comment: participants were blinded ade- quately Blinding (performance bias and detection Low risk Quote: study described as a “Ran- bias) domised. Published by John Wiley & Sons. double-blind trial”. including Outcome assessor non scheduled visits for suspected wound infection.Platt 1990 (Continued) Blinding (performance bias and detection Low risk Quote: study described as a “Ran- bias) domised. Ltd. .” They go on to state “investigators were required to return these labels intact. Similar rea- sons for exclusion were documented for both groups. “Cefonicid Participants and placebo were supplied in identical numbered vials. Outcome assessor blinded Incomplete outcome data (attrition bias) Low risk Comment: table 1 documents numbers of All outcomes and reasons for exclusion of patients from analysis after randomisation. The authors state “treat- ment codes were not known by anyone at the participating centres unless the sealed opaque label attached to each vial was opened. Comment: treatment providers were blinded adequately Blinding (performance bias and detection Low risk Quote: “drug assignments were not known bias) during any follow up evaluations.” Comment: automated data processing and analyses in laboratory.

Dose regime: 6 doses at 6-hour intervals (n = 59). predefined clinical indicators Participants All breast cancer surgery except re-construction Excluded were those with a history of allergy to the study antibiotic or receiving other antibiotics Total number of participants: 118 Interventions I) Cefazolin 25 mg per kg. the incidence of wound infection was the primary outcome mea- sure. C) Placebo: normal saline bolus as per the treatment regime (n = 59) Outcomes Infection rates Adverse events Time to onset of infection Affect of length of surgery Affect of pre-surgery biopsy Notes Length of follow-up: 30 days postoperative Country of origin: USA Sponsored by the American society career development award Risk of bias Bias Authors’ judgement Support for judgement Random sequence generation (selection Low risk Quote: “Randomisation was performed in bias) the Pharmacy using a table of random Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 34 Copyright © 2014 The Cochrane Collaboration. They are separated for breast surgery verses hernia surgery Other sources of potential bias High risk Comment: the paper states “the study was supported by a grant from Smith. the study appears to be free of any other source of bias Wagman 1990 Methods RCT: random numbers table generated by dept of biostatistics Loss to follow-up: < 20% Intention-to-treat: unclear Power calculation: unclear Clear definition of infection: done. Published by John Wiley & Sons. however. This is a pharmaceu- tical company. However. . Ltd. Kline & French laboratories”.Platt 1990 (Continued) Selective outcome data Low risk Comment: the study protocol was not available. The results are displayed in table 4. The authors document the definition of a wound infection clearly. Intravenous. First dose within 30 minutes pre-surgery.

The code was bro- ken after initial data evaluation”. Comment: blinding of treatment providers done Blinding (performance bias and detection Low risk Quote: “The patient. surgeon and Infec- bias) tion Control office had no knowledge of Treatment Provider the patient assignments”. i. Comment: blinding of outcome assessors done Incomplete outcome data (attrition bias) Low risk Quote: “Nine patients were excluded from All outcomes the study after randomisation (one patient did not undergo surgical treatment. surgeon and Infec- bias) tion Control office had no knowledge of Participants the patient assignments”. . phar- macy-controlled Blinding (performance bias and detection Low risk Quote: “The patient. Ltd.Wagman 1990 (Continued) numbers generated by the Department of Biostatistics”. one underwent biopsy only.e. Published by John Wiley & Sons. Comment: random number tables used. Comment: central allocation. five patients failed to receive a complete course of antibiotics and two had antibiotics for another reason” Comment: the number lost to follow-up is low and the reasons were valid Selective outcome data Low risk Comment: the study protocol was not available but the important outcome mea- sures stated in the methods section are re- ported in the results Other sources of potential bias Low risk Comment: the study appears to be free of other sources of bias Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 35 Copyright © 2014 The Cochrane Collaboration. surgeon and Infec- bias) tion Control office had no knowledge of Outcome assessor the patient assignments. Comment: blinding of participants done Blinding (performance bias and detection Low risk Quote: “The patient. Method of generating the random schedule reported Allocation concealment (selection bias) Low risk Quote: “Randomisation was performed in the Pharmacy using a table of random numbers generated by the Department of Biostatistics”.

Published by John Wiley & Sons. chronic diseases.5 cm collagen plus gentamycin sulphate (200 mg). Ltd. each com- prising 10 x 10 x 0. immune suppression. e.g. were excluded Treatment group: 22 Control group: 22 Interventions I) Gentacoll applied to the axillary area and under the flap before closure of the surgical wound. Two pieces of Gen- tacoll were used for each area. Group II Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 36 Copyright © 2014 The Cochrane Collaboration. Gentacoll is 10 cm x 10 cm x 0.Yetim 2010 Methods RCT: patients were randomly allocated into 1 of 2 groups No loss to follow-up Intention-to-treat analysis: done as all the participants were analysed in the groups to which they were randomised Power calculation: not done Reliable primary outcome: done Participants All female patients who were diagnosed with breast cancer and underwent modified radical mastectomy with axillary dissection between June 2006 and June 2009 were included Exclusion criteria: patients with inflammatory breast cancer who had neoadjuvant ra- diotherapy. diabetes.5 cm collagen from equine tendons with 200 mg gentamycin sulphate C) No Gentacoll Outcomes Wound infection Seroma formation Drain removal time Total drainage volumes Duration of hospital stay Notes Length of follow-up: 6 months Country of origin: Turkey Risk of bias Bias Authors’ judgement Support for judgement Random sequence generation (selection Unclear risk Quote: “patients were randomly allocated bias) in to one of two groups” Comment: no further information regard- ing randomisation is given Allocation concealment (selection bias) Unclear risk Comment: no information is given regard- ing the concealment of randomisation Blinding (performance bias and detection High risk Quote: “Group I underwent modified rad- bias) ical mastectomy during which Gentacoll Participants was applied to the axiliary area and under the flap area of the breast before closure of the surgical wound. Two pieces of Gentacoll were used in each area. .

5 cm collagen plus gentamycin sulphate (200 mg). they may be aware of four 10 x 10 x 0. however. 3 Outcome assessor and 6 months after surgery” Comment: no information is given as to who performed the follow-up and whether or not they were blinded. However. the only information given in the paper is for the first 7 days Selective outcome data Low risk Comment: the study protocol was not available. At follow-up would infection and seroma formation was assessed as well as drain information and duration of hospital stay. how- ever.Yetim 2010 (Continued) underwent modified radical mastectomy without the application of Gentacoll. drain removal time. 44 patients were enrolled and randomised and the results ta- bles given follow-up data for all 44 partic- ipants.” Comment: as the surgeons were responsible for applying the Gentacoll they could not be blinded in the study Blinding (performance bias and detection High risk Quote: “patients were followed up 7 days bias) after discharge from hospital and at 1.5 cm collagen placed under the skin and were therefore unable to be blinded Blinding (performance bias and detection High risk Quote: “Group I underwent modified rad- bias) ical mastectomy during which Gentacoll Treatment Provider was applied to the axiliary area and under the flap area of the breast before closure of the surgical wound. It could be con- sidered that the healthcare professional as- sessing for wound infection would be able to see if collagen implants had been inserted Incomplete outcome data (attrition bias) Unclear risk Comment: there was no loss to follow-up All outcomes documented in the study. . each com- prising 10 x 10 x 0. the authors state that pa- tients would be followed up for 6 months post surgery.” Comment: it is not clear whether the par- ticipants were blinded in the study. seroma formation. Two pieces of Gen- tacoll were used for each area. Group II underwent modified radical mastectomy without the application of Gentacoll. total drainage volumes and duration of hospital stay were recorded and Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 37 Copyright © 2014 The Cochrane Collaboration. wound infection. Ltd. Published by John Wiley & Sons.

. Unable to separate data for breast patients Exener 1992 Unable to obtain through the British Library Franchelli 1994 Although the data were on reconstructive surgery. Full paper not published. Full paper not published. not infection risk as an acute surgical complication Baker 2005 Full paper not published. Unable to ascertain if this is a duplicate of Kumar 2005 Bertin 1998 Not a RCT or quasi-RCT Boyd 1981 Not a RCT.Yetim 2010 (Continued) displayed in results tables 2 and 3 Other sources of potential bias Low risk Comment: the study appears to be free from other sources of bias C: control I: intervention IV: intravenous RCT: randomised controlled trial Characteristics of excluded studies [ordered by study ID] Study Reason for exclusion Baker 2000 This study was from the perspective of dentists managing risk in patients undergoing dental work who are at risk of remote infection due to implants. Unable to ascertain if this is a duplicate LeRoy 1991 Excluded as retrospective analysis Lewis 1995 Excluded as unable to obtain separate data for breast patients despite writing to the author Melling 2005 Abstract only. retrospective analysis D’Amico 2001 Review Erfle 2002 Not a RCT or quasi-RCT Esposito 2006 Study includes hernia repair and breast cancer surgery. It also did not state whether the surgery was immediate or delayed reconstruction Hall 2000 Review Kumar 2005 Abstract only. the paper did not state being secondary to breast cancer treatment. Ltd. Unable to ascertain if this is a duplicate of Kumar 2005 Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 38 Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons.

.(Continued) Melling 2006 Abstract only. Published by John Wiley & Sons. Shamilov 1991 Not a RCT or quasi-RCT Spicher 2003 Found not to be a RCT following translation. The article analyses the authors experience of implementing guidelines for using antibiotics with patients undergoing reconstructive surgery Sultan 1989 No separate data were obtainable for breast patients Thomas 1999 Addresses long-acting versus short-acting antibiotic comparison rather than antibiotic versus none or placebo Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 39 Copyright © 2014 The Cochrane Collaboration. Sanguinetti 2009 Removal of benign lesions included in study. No comparison made Serletti 1994 Addressed reduction mammoplasty. Ltd. No separate data was obtainable for breast cancer patients Sasaki 1988 Excluded as not a RCT following translation. Surgery not cancer-related. Full paper not published. Platt 1992 was not a RCT and Platt 1990 is an included study in this systematic review. Unable to ascertain if this is a duplicate of Kumar 2005 Morimoto 1998 Excluded as this study was comparing antibiotic dose and regime rather than antibiotic versus placebo/none Nicholas 2007 Not a RCT or quasi-RCT Pennel 2004 Not a RCT or quasi-RCT Platt 1992 Not a RCT or quasi-RCT Platt 1993 This is a meta-analysis of Platt (1990) and Platt (1992).

of No. 1. 0.28.DATA AND ANALYSES Comparison 1. 95% CI) 0. Fixed.1 Preoperative antibiotics 2 784 Risk Ratio (M-H. Fixed.1 Preoperative antibiotic Other data No numeric data versus placebo 8 Readmission to hospital 2 Risk Ratio (M-H.97] versus placebo 1.39 [0.2 Preoperative antibiotics Other data No numeric data versus none 7 Time to onset of infection Other data No numeric data 7.42] 4 Infection rates in those who 1 47 Risk Ratio (M-H. 95% CI) 0. Fixed. 0.48 [0.12] received neo-adjuvant chemo 5 Cost of care Other data No numeric data 6 Adverse effects from antibiotics Other data No numeric data 6.53. 95% CI) Totals not selected Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 40 Copyright © 2014 The Cochrane Collaboration. 95% CI) 0.1 Preoperative antibiotic 7 1784 Risk Ratio (M-H.74 [0.56 [0.49] versus placebo Comparison 2. Preoperative antibiotics versus none or placebo No. Fixed.2 Preoperative antibiotic 3 1039 Risk Ratio (M-H. . 4.47. of Outcome or subgroup title studies participants Statistical method Effect size 1 Wound infections 10 2823 Risk Ratio (M-H.56. 95% CI) 0. Fixed. of Outcome or subgroup title studies participants Statistical method Effect size 1 Wound infection 1 Risk Ratio (M-H.85] 1.01. 0. 95% CI) 0. Published by John Wiley & Sons. Fixed. 95% CI) Subtotals only 8.04.33. Random.95] 3 Wound infection cefazolin 2 336 Risk Ratio (M-H. Fixed. of No.21 [0.1 Preoperative antibiotics Other data No numeric data versus placebo 6.82 [0. Ltd. 0.81] versus none 2 Wound infection cefonicid 2 747 Risk Ratio (M-H. 95% CI) 0. Random. 95% CI) 0. 3. Perioperative antibiotics compared with no antibiotic No.67 [0.

73 (P = 0.93 [ 0.0% Test for overall effect: Z = 3. Published by John Wiley & Sons. df = 6 (P = 0.63 [ 0. .97 ] Total events: 80 (Antibiotic).48.67 [ 0. 2. 0.1 % 0. 1. 1.1.32.29 (P = 0.15.Fixed.95% CI M-H.36.5 % 0. Review: Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery Comparison: 1 Preoperative antibiotics versus none or placebo Outcome: 1 Wound infections Study or subgroup Antibiotic Control Risk Ratio Weight Risk Ratio n/N n/N M-H.1 1 10 100 Favours Antibiotic Favours Control Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 41 Copyright © 2014 The Cochrane Collaboration. 0.18 ] Wagman 1990 3/59 5/59 3. 1.88 [ 0.60 ] Gupta 2000 29/164 32/169 21.15.21 (P = 0.73).28.6 % 0.Fixed.0 % 0. df = 8 (P = 0.71 [ 0.65 [ 0. I2 =0.00099) Test for subgroup differences: Chi2 = 2.4 % 0.09.9 % 0.31 [ 0. Outcome 1 Wound infections. Ltd. df = 1 (P = 0.0% Test for overall effect: Z = 2.48 ] Platt 1990 17/303 26/303 17.5 % 0.81 ] Total events: 19 (Antibiotic).027) 2 Preoperative antibiotic versus none Chow 2000 0/28 0/24 Not estimable Gulluoglu 2013 9/187 25/182 17.7 % 0.01 0.53 ] Bold 1998 3/69 10/72 6.73 ] Hall 2006 10/311 14/307 9.4 % 0. I2 =38% Test for overall effect: Z = 2.17.0064) Total (95% CI) 1417 1406 100. 39 (Control) Heterogeneity: Chi2 = 1.0 % 0.35 [ 0. df = 1 (P = 0. 0.61 [ 0.40 ] Subtotal (95% CI) 891 893 73. 109 (Control) Heterogeneity: Chi2 = 3.21).48 [ 0.59.60.85 ] Total events: 99 (Antibiotic). 0.3 % 0.50).15). I2 =51% 0.62. Comparison 1 Preoperative antibiotics versus none or placebo.36.04.09 ] Cabaluna 2012 17/110 19/108 12. 148 (Control) Heterogeneity: Chi2 = 7.60 [ 0. 1. 1.6 % 0.53. Analysis 1.56 ] Subtotal (95% CI) 526 513 26.47 ] Paajanen 2009 8/144 13/148 8. 2. I2 =0.56.27. 1.74 [ 0.95% CI 1 Preoperative antibiotic versus placebo Amland 1995 3/42 4/34 3.

10.0 % 0.48) Test for subgroup differences: Not applicable 0.95% CI Cabaluna 2012 17/110 19/108 79.032) Test for subgroup differences: Not applicable 0.62). 36 (Control) Heterogeneity: Chi2 = 1.25.95% CI M-H. 0. 1. Analysis 1.0 % 0.60 ] Wagman 1990 3/59 5/59 20.18 ] Total (95% CI) 372 375 100. 24 (Control) Heterogeneity: Chi2 = 0.82 [ 0.01 0.3 % 0. 1. Outcome 3 Wound infection cefazolin. df = 1 (P = 0. Outcome 2 Wound infection cefonicid.42 ] Total events: 20 (Cefazolin).7 % 0. 2.95% CI M-H. Comparison 1 Preoperative antibiotics versus none or placebo.71 (P = 0.Fixed.1 1 10 100 Favours cefazolin Favours control Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 42 Copyright © 2014 The Cochrane Collaboration.1 1 10 100 Favours cefonicid Favours control Analysis 1.88 [ 0.0% Test for overall effect: Z = 0. I2 =9% Test for overall effect: Z = 2.Fixed.15 (P = 0.Fixed.60 [ 0.48.95% CI Bold 1998 3/69 10/72 27.56 [ 0.36.7 % 0.29). 1.3 % 0. . df = 1 (P = 0.33. Review: Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery Comparison: 1 Preoperative antibiotics versus none or placebo Outcome: 2 Wound infection cefonicid Study or subgroup Cefonicid Control Risk Ratio Weight Risk Ratio n/N n/N M-H. Ltd.47.3.2.95 ] Total events: 20 (Cefonicid).40 ] Total (95% CI) 169 167 100. Comparison 1 Preoperative antibiotics versus none or placebo.09. I2 =0.09 ] Platt 1990 17/303 26/303 72.01 0. 1.31 [ 0.65 [ 0.Fixed. Published by John Wiley & Sons.15. Review: Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery Comparison: 1 Preoperative antibiotics versus none or placebo Outcome: 3 Wound infection cefazolin Study or subgroup Cefazolin Control Risk Ratio Weight Risk Ratio n/N n/N M-H.

charges for outpatient treat- ment and charges for inpatient treat- ment Gulluoglu 2013 Average SSI related treatment cost: Average SSI related treatment cost: Little information is given regarding USD 8.001 0.1 1 10 100 1000 Favours treatment Favours control Analysis 1.21 [ 0.12 ] Total (95% CI) 23 24 100.0 % 0.01.80 Average per patient: USD 364.26 what is included in the SSI (surgical site infection) treatment cost Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 43 Copyright © 2014 The Cochrane Collaboration.48 USD 20.838. Outcome 5 Cost of care. Comparison 1 Preoperative antibiotics versus none or placebo. Analysis 1. Comparison 1 Preoperative antibiotics versus none or placebo.95% CI Bold 1998 0/23 2/24 100.57 32.Fixed. Outcome 4 Infection rates in those who received neo-adjuvant chemo. 4.87 tration.5.16 from: cost of prophylaxis adminis- Average per patient: USD 49. . Ltd.30) Test for subgroup differences: Not applicable 0. Review: Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery Comparison: 1 Preoperative antibiotics versus none or placebo Outcome: 4 Infection rates in those who received neo-adjuvant chemo Study or subgroup Antibiotic Placebo Risk Ratio Weight Risk Ratio n/N n/N M-H. Cost of care Study Antibiotic Placebo Cost calculation Bold 1998 Total cost in the treatment group: Total cost in the placebo group: USD Treatment costs were calculated USD 4382.01.0 % 0.03 (P = 0.01 0.4. Published by John Wiley & Sons.Fixed. 2 (Placebo) Heterogeneity: not applicable Test for overall effect: Z = 1.12 ] Total events: 0 (Antibiotic).95% CI M-H.21 [ 0. 4.

Outcome 6 Adverse effects from antibiotics. 2 tected between 15 and 30 days. 1 other 2 GI. 2 skin rash. Comparison 1 Preoperative antibiotics versus none or placebo. Adverse effects from antibiotics Study Antibiotic Control Preoperative antibiotics versus placebo Amland 1995 Side effects considered by the investigator to be related to Side effects considered by the investigator to be related treatment were recorded in 4 of the 171 patients receiving to treatment were present in 5 of the control group (3. 6 (67%) infections 1 (4%) infection detected between 0 and 2 days. details not provided as to whether these were per patient or per event whether these were per patient or per event Paajanen 2009 None recorded None recorded Platt 1990 None recorded None recorded Wagman 1990 Stated as: “no untoward reactions” Stated as: “no untoward reactions” Preoperative antibiotics versus none Chow 2000 No adverse events recorded No adverse events recorded Hall 2006 Stated as ’no side effects observed’ from the flucloxacillin None stated Analysis 1.7. (60%) infections detected between 3 and 7 days. . Ltd. 2 (22%) infections de. details not provided as to 33 adverse events noted. Published by John Wiley & Sons. Analysis 1.15 detected between 3 and 7 days. (28%) infections detected between 8 and 14 days. 1 other Bold 1998 Stated as: “no patient suffered a complication related to None recorded the antibiotic administration” Gupta 2000 41 adverse events noted. Outcome 7 Time to onset of infection. no infections detected (8%) infections detected between 15 and 30 days. 1 (11%) infection de. the antibiotic (2. Time to onset of infection Study Antibiotic Control Preoperative antibiotic versus placebo Gulluoglu 2013 0 infections between 0 and 2 days.6. no beyond 30 days infections detected beyond 30 days Gupta 2000 Mean time to onset of infection 12 days Mean time to onset of infection 11 days Platt 1990 Mean time to onset of infection 11 days Mean time to onset of infection 10 days Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 44 Copyright © 2014 The Cochrane Collaboration. 7 tected between 8 and 14 days. Comparison 1 Preoperative antibiotics versus none or placebo.3%) 0%) 2 GI. 1 skin rash.

Outcome 8 Readmission to hospital. 3. Comparison 1 Preoperative antibiotics versus none or placebo.11 [ 0. . 3.01.88 ] Platt 1990 5/303 5/303 56.04. Ltd.40) 0. I2 =71% Test for overall effect: Z = 0.2 % 0.Time to onset of infection (Continued) Wagman 1990 Mean time to onset of infection 17. Published by John Wiley & Sons.0 % 0.80.42 ] Subtotal (95% CI) 391 393 100.8 % 1. 0. M- H. Chi2 = 3.29.8.95% H.Random.49 ] Total events: 6 (Antibiotic).43.84 (P = 0.7 days Analysis 1.Random.00 [ 0.39 [ 0.001 0. Review: Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery Comparison: 1 Preoperative antibiotics versus none or placebo Outcome: 8 Readmission to hospital Study or subgroup Antibiotic Control Risk Ratio Weight Risk Ratio M. df = 1 (P = 0. 14 (Control) Heterogeneity: Tau2 = 1.01 0.1 1 10 100 1000 Favours Antibiotic Favours Control Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 45 Copyright © 2014 The Cochrane Collaboration.95% n/N n/N CI CI 1 Preoperative antibiotics versus placebo Bold 1998 1/88 9/90 43.06).

Outcome 1 Wound infection.95 ] 0.01. • the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library 2011.95% CI M-H. Search strategy for the second update of this review . • Ovid MEDLINE (In-Process & Other Non-Indexed Citations August 30. • EBSCO CINAHL (2008 to 25 August 2011). Analysis 2.Fixed. Issue 3). • Ovid MEDLINE (2008 to August Week 3 2011).2011 Electronic searches For the second update of this review we searched the following electronic databases: • the Cochrane Wounds Group Specialised Register (searched 31 August 2011). We used the following search strategy in the Cochrane Central Register of Controlled Trials (CENTRAL): #1 MeSH descriptor Surgical Wound Infection explode all trees #2 surg* NEAR/5 infection* #3 surgical NEAR/5 wound* #4 (postoperative or post-operative) NEAR/5 infection* #5 MeSH descriptor Preoperative Care explode all trees #6 (preoperative or pre-operative) NEXT care #7 MeSH descriptor Perioperative Care explode all trees #8 (perioperative or peri-operative) NEXT care #9 (#1 OR #2 OR #3 OR #4 OR #5 OR #6 OR #7 OR #8) #10 MeSH descriptor Breast Neoplasms explode all trees with qualifier: SU #11 (breast NEXT cancer) NEAR/5 surg* #12 (breast NEXT neoplasm*) NEAR/5 surg* #13 (breast NEXT carcinoma*) NEAR/5 surg* #14 MeSH descriptor Mastectomy explode all trees #15 MeSH descriptor Mammaplasty explode all trees Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 46 Copyright © 2014 The Cochrane Collaboration.1 1 10 500 Favours gentamycin Favours no antibiotic APPENDICES Appendix 1. Comparison 2 Perioperative antibiotics compared with no antibiotic. Review: Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery Comparison: 2 Perioperative antibiotics compared with no antibiotic Outcome: 1 Wound infection Study or subgroup Gentamycin No antibiotic Risk Ratio Risk Ratio n/N n/N M-H. Published by John Wiley & Sons.1.Fixed. 1.002 0. Ltd. .11 [ 0. 2011). • Ovid EMBASE (1980 to 2011 Week 34).95% CI Yetim 2010 0/22 4/22 0.

tw. (15479) 20 exp Breast/su [Surgery] (5305) 21 or/13-20 (50942) 22 exp Anti-Bacterial Agents/ (537825) 23 (antibiotic* or clindamycin or cefuroxime or cefuroxim or ceftazidime or ofloxacin or levofloxacin or azithromycin or sulbactam or ampicillin or mezlocillin or oxacillin or vancomycin or tobramycin or ciprofloxacin). We combined the MEDLINE search with the Cochrane Highly Sensitive Search Strategy for identifying randomised trials in MEDLINE: sensitivity. Appendix 3 and Appendix 4 respectively.tw. We applied no language or date restrictions. (350) 11 wound complication*. Published by John Wiley & Sons. (22384) 10 (wound adj5 disrupt*). (609) 17 exp Mastectomy/ (23306) 18 exp Mammaplasty/ (8405) 19 (mastectomy or mammaplasty).tw. (18792) 4 (surg* adj5 wound*). We combined the EMBASE and CINAHL searches with the trial filters developed by the Scottish Intercollegiate Guidelines Network (SIGN).tw.tw. Searching other resources In addition.#16 mastectomy or mammaplasty #17 MeSH descriptor Breast explode all trees with qualifier: SU #18 (#10 OR #11 OR #12 OR #13 OR #14 OR #15 OR #16 OR #17) #19 MeSH descriptor Anti-Bacterial Agents explode all trees #20 (antibiotic* or clindamycin or cefuroxime or cefuroxim or ceftazidime or ofloxacin or levofloxacin or azithromycin or sulbactam or ampicillin or mezlocillin or oxacillin or vancomycin or tobramycin or ciprofloxacin) #21 (#19 OR #20) #22 (#9 AND #18 AND #21) The search strategies for Ovid MEDLINE.tw. (266425) 24 22 or 23 (634056) 25 12 and 21 and 24 (175) 26 Randomized controlled trials/ (103066) 27 Single-Blind Method/ (19669) 28 Double-Blind Method/ (132149) 29 Crossover Procedure/ (0) Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 47 Copyright © 2014 The Cochrane Collaboration. . Ovid EMBASE and EBSCO CINAHL can be found in Appendix 2. Ovid MEDLINE search strategy 1 exp Surgical Wound Infection/ (28294) 2 exp Surgical Wound Dehiscence/ (6260) 3 (surg* adj5 infect*). Appendix 2.tw. We followed up conference proceedings and grey literature that was considered to be potentially eligible for inclusion by both authors by contacting the study authors for further information. Ltd. (2813) 9 (wound* adj5 infect*).tw. (6097) 15 (breast neoplasm* adj5 surg*). (10115) 5 (surg* adj5 site*). Ovid format. (6420) 7 (surg* adj5 dehisc*).tw. (585) 8 (wound* adj5 dehisc*). (36) 16 (breast carcinoma* adj5 surg*). we screened references in all articles found by the above search strategy for further studies.tw.tw. (11417) 6 (surg* adj5 incision*). We contacted experts in the field and interest groups to try and obtain access to unpublished or ongoing work.tw.tw.tw. (2992) 12 or/1-11 (80916) 13 exp Breast Neoplasms/su [Surgery] (30204) 14 (breast cancer* adj5 surg*).and precision-maximising version (2008 revision).

(116870) 32 (singl$ adj blind$).tw.tw. (637) 8 (wound* adj5 dehisc*).tw.ti.tw.ab.tw.tw.tw. (16) 16 (breast carcinoma* adj5 surg*). (98333) 32 (singl$ adj blind$). (18838) 4 (surg* adj5 wound*).ab. (21393) 10 (wound adj5 disrupt*).tw.ab. .ti. (14647) 20 exp Breast Surgery/ (38660) 21 or/13-20 (55407) 22 exp antibiotic agent/ (589638) 23 (antibiotic* or clindamycin or cefuroxime or cefuroxim or ceftazidime or ofloxacin or levofloxacin or azithromycin or sulbactam or ampicillin or mezlocillin or oxacillin or vancomycin or tobramycin or ciprofloxacin).tw.ab. Published by John Wiley & Sons.ab. (3233) 12 or/1-11 (74860) 13 exp Breast Tumor/su [Surgery] (34682) 14 (breast cancer* adj5 surg*). (7513) 7 (surg* adj5 dehisc*).tw. Ltd.tw. (11180) 33 or/26-32 (1116089) 34 exp animals/ or exp invertebrate/ or animal experiment/ or animal model/ or animal tissue/ or animal cell/ or nonhuman/ (17778990) 35 human/ or human cell/ (13712248) 36 and/34-35 (13712248) 37 34 not 36 (4066742) 38 33 not 37 (992785) 39 25 and 38 (46) Appendix 3.ti. (9077) 5 (surg* adj5 site*).ti.ti.ab. (10782) 33 or/26-32 (1093705) 34 exp animals/ or exp invertebrate/ or animal experiment/ or animal model/ or animal tissue/ or animal cell/ or nonhuman/ (12313060) 35 human/ or human cell/ (9544901) 36 and/34-35 (9544609) 37 34 not 36 (2768451) 38 33 not 37 (946515) Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 48 Copyright © 2014 The Cochrane Collaboration.tw. (1039219) 31 (doubl$ adj blind$).ti. (219686) 24 or/22-23 (658872) 25 12 and 21 and 24 (304) 26 Randomized controlled trials/ (42188) 27 Single-Blind Method/ (17133) 28 Double-Blind Method/ (91694) 29 Crossover Procedure/ (34738) 30 (random$ or factorial$ or crossover$ or cross over$ or cross-over$ or placebo$ or assign$ or allocat$ or volunteer$). (7357) 15 (breast neoplasm* adj5 surg*). (492) 17 exp mastectomy/ (27258) 18 exp breast reconstruction/ (11159) 19 (mastectomy or mammaplasty). Ovid EMBASE search strategy 1 exp surgical infection/ (17291) 2 exp wound dehiscence/ (7784) 3 (surg* adj5 infect*). (1049911) 31 (doubl$ adj blind$).tw. (14367) 6 (surg* adj5 incision*).tw. (311) 11 wound complication*. (3036) 9 (wound* adj5 infect*).30 (random$ or factorial$ or crossover$ or cross over$ or cross-over$ or placebo$ or assign$ or allocat$ or volunteer$).

Ltd. .39 25 and 38 (57) Appendix 4. Published by John Wiley & Sons. EBSCO CINAHL search strategy S36 S23 AND S35 S35 S24 or S25 or S26 or S27 or S28 or S29 or S30 or S31 or S32 or S33 or S34 S34 MH “Quantitative Studies” S33 TI placebo* or AB placebo* S32 MH “Placebos” S31 TI random* allocat* or AB random* allocat* S30 MH “Random Assignment” S29 TI randomi?ed control* trial* or AB randomi?ed control* trial* S28 AB ( singl* or doubl* or trebl* or tripl* ) and AB ( blind* or mask* ) S27 TI ( singl* or doubl* or trebl* or tripl* ) and TI ( blind* or mask* ) S26 TI clinic* N1 trial* or AB clinic* N1 trial* S25 PT Clinical trial S24 MH “Clinical Trials+” S23 S10 AND S18 AND S22 S22 S19 OR S20 OR S21 S21 AB ( antibiotic* or clindamycin or cefuroxime or cefuroxim or ceftazidime or ofloxacin or levofloxacin or azithromycin orsulbactam or ampicillin or mezlocillin or oxacillin or vancomycin or tobramycin or ciprofloxacin ) S20 TI ( antibiotic* or clindamycin or cefuroxime or cefuroxim or ceftazidime or ofloxacin or levofloxacin or azithromycin orsulbactam or ampicillin or mezlocillin or oxacillin or vancomycin or tobramycin or ciprofloxacin ) S19 (MH “Antibiotics+”) S18 S11 OR S12 OR S13 OR S14 OR S15 OR S16 OR S17 S17 (MH “Breast/SU”) S16 TI ( mastectomy or mammaplasty ) or AB ( mastectomy or mammaplasty ) S15 (MH “Mastectomy+”) S14 TI breast carcinoma* N5 surg* or AB breast carcinoma* N5 surg* S13 TI breast neoplasm* N5 surg* or AB breast neoplasm* N5 surg* S12 TI breast cancer* N5 surg* or AB breast cancer* N5 surg* S11 (MH “Breast Neoplasms/SU”) S10 S1 or S2 or S3 or S4 or S5 or S6 or S7 or S8 or S9 S9 TI wound complication* or AB wound complication* S8 TI wound* N5 dehisc* or AB wound* N5 dehisc* S7 TI surg* N5 dehisc* or AB surg* N5 dehisc* S6 TI surg* N5 incision* or AB surg* N5 incision* S5 TI surg* N5 site* or AB surg* N5 site* S4 TI surg* N5 wound* or AB surg* N5 wound* S3 TI surg* N5 infection* or AB surg* N5 infection* S2 (MH “Surgical Wound Infection”) S1 (MH “Surgical Wound Dehiscence”) Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 49 Copyright © 2014 The Cochrane Collaboration.

sequentially-numbered drug containers of identical appearance. High risk of bias The investigators describe a non-random component in the sequence generation process. Unclear Insufficient information about the sequence generation process to permit judgement of low or high risk of bias. for example: sequence generated by odd or even date of birth. 2.Appendix 5. any other explicitly unconcealed procedure. a list of random numbers). and unlikely that the blinding could have been broken. throwing dice. case record number. but the review authors judge that the outcome and the outcome measurement are not likely to be influenced by lack of blinding. . Was the treatment allocation adequately concealed? Low risk of bias Participants and investigators enrolling participants could not foresee assignment because one of the following. 3. using a computer random number generator. alternation or rotation. sealed envelopes. date of birth. but it remains unclear whether envelopes were sequentially numbered. drawing of lots. Risk of bias criteria 1. non-random approach. such as allocation based on: using an open random allocation schedule (e. • Blinding of participants and key study personnel ensured. assignment envelopes were used without appropriate safeguards (e. Blinding . web-based and pharmacy-controlled randomisation). sequence generated by some rule based on hospital or clinic record number. This is usually the case if the method of concealment is not described or not described in sufficient detail to allow a definite judgement. Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 50 Copyright © 2014 The Cochrane Collaboration.g. Usually. sequentially-numbered. or an equivalent method.g. but outcome assessment was blinded and the non-blinding of others unlikely to introduce bias.was knowledge of the allocated interventions adequately prevented during the study? Low risk of bias Any one of the following. Ltd. the description would involve some systematic. High risk of bias Participants or investigators enrolling participants could possibly foresee assignments and thus introduce selection bias. Was the allocation sequence randomly generated? Low risk of bias The investigators describe a random component in the sequence generation process such as: referring to a random number table. opaque and sealed. if envelopes were unsealed or non opaque or not sequentially numbered). for example if the use of assignment envelopes is described. opaque. sequence generated by some rule based on date (or day) of admission. • Either participants or some key study personnel were not blinded. was used to conceal allocation: central allocation (including telephone. • No blinding. coin tossing. shuffling cards or envelopes. Published by John Wiley & Sons. Unclear Insufficient information to permit judgement of low or high risk of bias.

• For dichotomous outcome data. censoring unlikely to be introducing bias). plausible effect size (difference in means or standardised difference in means) among missing outcomes not enough to have a clinically relevant impact on observed effect size. with either imbalance in numbers or reasons for missing data across intervention groups. • The study did not address this outcome. and the outcome or outcome measurement is likely to be influenced by lack of blinding. High risk of bias Any one of the following. • Insufficient information to permit judgement of low or high risk of bias. with similar reasons for missing data across groups. . number randomised not stated. • Either participants or some key study personnel were not blinded. Unclear Any one of the following. no reasons for missing data provided). • ‘As-treated’ analysis done with substantial departure of the intervention received from that assigned at randomisation. and the non-blinding of others likely to introduce bias. Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 51 Copyright © 2014 The Cochrane Collaboration. the proportion of missing outcomes compared with observed event risk enough to induce clinically relevant bias in intervention effect estimate. • For continuous outcome data.High risk of bias Any one of the following. • For continuous outcome data. • Missing data have been imputed using appropriate methods. • For dichotomous outcome data. • Blinding of key study participants and personnel attempted. plausible effect size (difference in means or standardised difference in means) among missing outcomes enough to induce clinically relevant bias in observed effect size. Published by John Wiley & Sons. Ltd. • Potentially inappropriate application of simple imputation. Were incomplete outcome data adequately addressed? Low risk of bias Any one of the following. • No blinding or incomplete blinding. • Insufficient reporting of attrition/exclusions to permit judgement of low or high risk of bias (e. • Missing outcome data balanced in numbers across intervention groups. • The study did not address this outcome.g. 5. • Reason for missing outcome data likely to be related to true outcome. • Reasons for missing outcome data unlikely to be related to true outcome (for survival data. • No missing outcome data. Unclear Any one of the following. Are reports of the study free of suggestion of selective outcome reporting? Low risk of bias Any of the following. the proportion of missing outcomes compared with observed event risk not enough to have a clinically relevant impact on the intervention effect estimate. 4. but likely that the blinding could have been broken.

or • had extreme baseline imbalance. Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 52 Copyright © 2014 The Cochrane Collaboration. or • had some other problem. Unclear There may be a risk of bias. • One or more outcomes of interest in the review are reported incompletely so that they cannot be entered in a meta-analysis. Ltd. Other sources of potential bias Low risk of bias The study appears to be free of other sources of bias. • The study report fails to include results for a key outcome that would be expected to have been reported for such a study. including those that were pre-specified (convincing text of this nature may be uncommon) High risk of bias Any one of the following. or • has been claimed to have been fraudulent. • The study protocol is available and all of the study’s pre-specified (primary and secondary) outcomes that are of interest in the review have been reported in the pre-specified way. Published by John Wiley & Sons. High risk of bias There is at least one important risk of bias. such as an unexpected adverse effect). • Not all of the study’s pre-specified primary outcomes have been reported. analysis methods or subsets of the data (e. • One or more reported primary outcomes were not pre-specified (unless clear justification for their reporting is provided. .g. or • insufficient rationale or evidence that an identified problem will introduce bias. but there is either: • insufficient information to assess whether an important risk of bias exists. the study: • had a potential source of bias related to the specific study design used. It is likely that the majority of studies will fall into this category. 6. • One or more primary outcomes is reported using measurements. Unclear Insufficient information to permit judgement of low or high risk of bias. • The study protocol is not available but it is clear that the published reports include all expected outcomes. subscales) that were not pre-specified. For example.

screened records for eligibility. undertook the risk of bias assessment and commented on the review update. Date Event Description 4 March 2014 New citation required but conclusions have not changed Third update. Sanguinetti 2009). Contributions of editorial base: Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 53 Copyright © 2014 The Cochrane Collaboration. Yetim 2010). new searches. extracted data. 18 December 2005 New citation required and conclusions have changed Substantive amendment CONTRIBUTIONS OF AUTHORS Daniel Jones: contributed to the second and took the lead for the third update of this review. extracted data. Conclusions unchanged. Ltd. 24 October 2008 New search has been performed One new trial added. screened records for eligibility. Published by John Wiley & Sons.WHAT’S NEW Last assessed as up-to-date: 5 December 2013. two studies added (Paajanen 2009. 4 March 2013 New search has been performed New search. . undertook the risk of bias assessment and updated the review text. HISTORY Protocol first published: Issue 2. Sophie Bell-Syer: contributed to the second and third update of this review. 2005 Review first published: Issue 2. Frances Bunn: provided methodological support. 11 August 2009 Amended Contact details updated. extracted data. conclusions remain unchanged. screened records. 28 July 2008 Amended Converted to new review format. helped to write the review and undertook the updating of the review. 2006 Date Event Description 23 September 2011 New citation required but conclusions have not New authors added to the review changed 31 August 2011 New search has been performed Second update. two studies excluded (Esposito 2006. two further studies included (Cabaluna 2012. Gulluoglu 2013) and five studies excluded.

Humans Prophylactic antibiotics to prevent surgical site infection after breast cancer surgery (Review) 54 Copyright © 2014 The Cochrane Collaboration. Ltd. SOURCES OF SUPPORT Internal sources • University of Hertfordshire. University of York. INDEX TERMS Medical Subject Headings (MeSH) ∗ Antibiotic Prophylaxis. External sources • NIHR/Department of Health (England). (Cochrane Wounds Group). Approved the final review and review update prior to submission.Nicky Cullum: edited the review. Advised on methodology. Edited and copy edited the review. • Department of Health Sciences. UK. Sally Bell-Syer: co-ordinated the editorial process. advised on methodology. Ruth Foxlee: designed the search strategy and edited the search methods section for the update. UK. York. Published by John Wiley & Sons. . Surgical Wound Infection [∗ prevention & control] MeSH check words Female. DECLARATIONS OF INTEREST None known. Preoperative Care [methods]. Breast Neoplasms [∗ surgery]. Randomized Controlled Trials as Topic. interpretation and content. interpretation and review content. Screened studies for the updated review and edited the updated review. UK.