You are on page 1of 10

MAJOR ARTICLE

Effect of Aminoglycoside and b-Lactam


Combination Therapy versus b-Lactam Monotherapy
on the Emergence of Antimicrobial Resistance:
A Meta-analysis of Randomized, Controlled Trials
Ioannis A. Bliziotis,1 George Samonis,3 Konstantinos Z. Vardakas,1 Stavroula Chrysanthopoulou,1
and Matthew E. Falagas1,2,4

Background. The addition of an aminoglycoside to a b-lactam therapy regimen has been suggested to have
a beneficial effect in delaying or preventing the development of antimicrobial resistance. We studied the effect of
aminoglycoside/b-lactam combination therapy versus b-lactam monotherapy on the emergence of resistance.
Methods. We performed a meta-analysis of randomized, controlled trials (RCTs) that compared aminoglycoside/b-lactam combination therapy with b-lactam monotherapy and that reported data regarding the emergence
of resistance (primary outcome) and/or development of superinfection, treatment failure, treatment failure attributable to emergence of resistance, treatment failure attributable to superinfection, all-cause mortality during
treatment, and mortality due to infection. Data for this meta-analysis were identified from the PubMed database,
Current Contents database, Cochrane central register of controlled trials, and references in relevant articles.
Results. A total of 8 RCTs were included in the analysis. b-Lactam monotherapy was not associated with a
greater emergence of resistance than was the aminoglycoside/b-lactam combination (odds ratio [OR], 0.90; 95%
confidence interval [CI], 0.561.47). Actually, b-lactam monotherapy was associated with fewer superinfections
(OR, 0.62; 95% CI, 0.420.93) and fewer treatment failures (OR, 0.62; 95% CI, 0.381.01). Rates of treatment
failure attributable to emergence of resistance (OR, 3.09; 95% CI, 0.7512.82), treatment failure attributable to
superinfection (OR, 0.60; 95% CI, 0.331.10), all-cause mortality during treatment (OR, 0.70; 95% CI, 0.401.25),
and mortality due to infection (OR, 0.74; 95% CI, 0.461.21) did not differ significantly between the 2 regimens.
Conclusions. Compared with b-lactam monotherapy, the aminoglycoside/b-lactam combination was not
associated with a beneficial effect on the development of antimicrobial resistance among initially antimicrobialsusceptible isolates.
Combination antimicrobial therapy has been tried for
the management of a variety of infectious diseases, primarily because drugs with different mechanisms of action may act synergistically (i.e., a 14-fold decrease in
the MIC of each drug when tested together against a
pathogen in vitro). Combination therapy may also allow the use of smaller doses of each of the combined
drugs. The initial successful example of clinical use of

Received 6 December 2004; accepted 19 February 2005; electronically published


31 May 2005.
Reprints or correspondence: Dr. Matthew E. Falagas, 9 Neapoleos St., Marousi
151 23, Greece (matthew.falagas@tufts.edu).
Clinical Infectious Diseases 2005; 41:14958
 2005 by the Infectious Diseases Society of America. All rights reserved.
1058-4838/2005/4102-0002$15.00

combinations of antimicrobial agents was the treatment


of infections due to Mycobacterium tuberculosis, for
which combination therapy decreased the rate of development of resistance to rifampin [1]. Later, the same
approach was used for the management of prosthetic
valve endocarditis due to Staphylococcus species [2].
The rationale of using aminoglycoside/b-lactam
combinations to delay or prevent the emergence of antimicrobial resistance was developed from data from
animal studies in the early and mid-1980s [35]. Since
then, the practice of using aminoglycoside/b-lactam
combination therapy has become very popular for the
management of infectious diseases in various conditions and patient populations. However, recent data
have questioned the postulated beneficial effect of the
Aminoglycoside/b-Lactam Resistance Emergence CID 2005:41 (15 July) 149

Downloaded from http://cid.oxfordjournals.org/ by guest on April 27, 2016

1
Alfa Institute of Biomedical Sciences (AIBS) and 2Department of Medicine, Henry Dunant Hospital, Athens, and 3Department of Medicine,
University of Crete, School of Medicine, Iraklion, Greece; and 4Department of Medicine, Tufts University School of Medicine, Boston,
Massachusetts

Table 1. Reported data on the emergence of antimicrobial resistance and superinfection from the randomized, controlled trials
included in the analysis

Treatment regimen
Mono arm

Combo arm

Duration of
treatment

Time of
clinical
evaluation

Year

Patient population

Setting

[61]

1995

Adults with serious hospitalacquired infection

Multicenter study from


North and South
America, Asia, and
Europe

Czid, 2 g q12h

Ctri, 2 g q.d., plus Tm, 35


a
mg/kg

No limit

At end of treatment
and 14 days AT

[62]

1994

Patients aged 116 years with


serious infections (e.g.,
pneumonia, sepsis, and
peritonitis)

73% of the patients


were in the ICU in 3
Swiss hospitals

Imi, 500 mg q6h

Imi, 500 mg q6h, plus Net,


150 mg q12h

No limit

[74]

1993

Adults and trauma patients


with pneumonia

ICUs in a US hospital

Cper, 2 g q12h, or Czid,


2 g q8h

Either Cper, 2 g q12h, or


Czid, 2 g q8h, plus Gm,
1 mg/kg q8h

No limit

[67]

1992

Patients aged 118 years with


nonlife-threatening infections (e.g., biliary infection,
gynecological infection, intra-abdominal infection,
LRTI, or septicemia)

12 German and 5
Austrian hospitals

Imi/Cil, 500/500 mg q8h

Ctax, 2 g q8h, plus Gm,


23 mg/kg q.d.

172 h

At end of treatment

[12]

1989

Adults with cholangitis

US hospital

Mez, 4 g q6h

Amp, 1 g q4h, plus Gm,


1.5 mg/kg q8h

15 days and

At end of treatment
and after 2
months

Either Cfaz, 1.5 g q8h, or


Tic, 3 g q4h, plus Tm,
1.7 mg/kg q8h

15 days and

Tic, 3 g q4h, plus Tm, 1


mg/kg q8h

14 days and

Carb, 41.583.5 mg/kg q4h,


d
or Tic plus either Gm,
d
4.5 mg/kg q.d., or Tm

No limit; treatment
stopped after 3
afebrile days

[21]

[30]

[33]

1987

1985

1983

Patients aged 116 years with


pneumonia

Canadian hospitals

Patients with pneumonia


and/or bacteremia

US hospital

Adults with various serious


infections

Canadian hospital

Czid, 2 g q8h

Czid, 2 g q8h

Pip, 2550 mg/kg q4h

!30 days

At end of treatment

!21 days

!16 days

NOTE. Amp, ampicillin; AT, after treatment; Carb, carboxypenicillin; Cfaz, cefazolin; Cil, cilastatin; combo arm, aminoglycoside/b-lactam combination therapy
arm; Cper, cefoperazone; Ctax, cefotaxime; Ctri, ceftriaxone; Czid, ceftazidime; Gm, gentamicin; ICU, intensive care unit; Imi, imipenem; LRTI, lower respiratory
tract infection; Mez, mezlocillin; mono arm, b-lactam monotherapy arm; ND, data not reported; Net, netilmicin; Pip, piperacillin; Tic, ticarcillin; Tm, tobramycin.
a
b
c
d

Modified dose/dosing according to serum levels.


Mean duration of therapy, 8 days.
Data are no. of pathogens.
Dosage not reported

aminoglycoside/b-lactam combinations on several outcomes,


including their effectiveness for curing infection and decreasing
mortality [6].
We tried to evaluate whether the addition of an aminoglycoside to a b-lactam therapy regimen has an effect on the
development of antimicrobial resistance among hospitalized,
nonneutropenic patients with serious infections. Thus, we performed a meta-analysis of the available randomized, controlled
trials (RCTs) that examined the use of b-lactam monotherapy
versus aminoglycoside/b-lactam combination therapy and presented data regarding the emergence of antimicrobial-resistant
organisms during treatment or during the follow-up period.
METHODS
Data sources. Relevant studies for our meta-analysis were
identified from searches of the PubMed database, Current Con150 CID 2005:41 (15 July) Bliziotis et al.

tents database, Cochrane central register of controlled trials,


and references from articles, including review papers. Search
terms included monotherapy, combination therapy, resistance, susceptibility, synergy, b-lactam, aminoglycoside, clinical trial, and randomized controlled trial.
Study selection. Two independent reviewers performed literature searches and examined the identified relevant studies
for further evaluation of data on effectiveness and toxicity. A
study was considered eligible if (1) it was an RCT, (2) it compared b-lactam monotherapy with the combination of an aminoglycoside with a b-lactam for the treatment of infection in
hospitalized patients, (3) it examined the development of emergence of resistance (i.e., a change to a more-resistant phenotype) of bacterial isolates, (4) it studied the effectiveness of
therapy and the mortality rate, (5) it was performed with adult
patients, (6) it was written in the English language, and (7) it

Downloaded from http://cid.oxfordjournals.org/ by guest on April 27, 2016

Study

No. of
clinically
evaluable
patients

No. of
microbiologically
evaluable
patients

Predominant organisms
isolated before treatment
(% of patients)

No. of
patients with
reisolation of
initial pathogen /
no. of patients

No. of
patients with
emergence of
resistant isolates /
no. of patients

No. of
patients with
super-infections /
no. of patients

Frequency of
follow-up cultures

Mono
arm

Combo
arm

Mono
arm

Combo
arm

Mono
arm

Combo
arm

Mono
arm

Combo
arm

Mono
arm

Combo
arm

3045 days AT

261

230

197

188

Pseudomonas aeruginosa,
Klebsiella species, Escherichia coli, Acinetobacter
species, Staphylococcus
aureus

24-day intervals
during treatment
and 14 days AT

ND

ND

12/197

14/188

14/306

20/274

None

142

138

98

101

E. coli, P. aeruginosa, Enterococcus faecalis, Klebsiella


species

Not specified

ND

ND

6/98

9/101

8/142

11/138

None

39

70

39

70

S. aureus (21), Enterobacter


species (14), Haemophilus
influenzae (13), Acinetobacter species (11)

At day 5, day 7, or if
symptoms
persisted

6/39

14/70

5/39

13/70

11/39

34/70

None

144

124

94

68

E. coli, Pseudomonas species,


enterococci, Bacteroides
species, S. aureus

During therapy (not


specified) and 13
days AT

7/144

6/124

1/94

1/68

ND

ND

Up to 3 years

24

22

24

22

E. coli, anaerobic gram-positive


bacilli, Klebsiella pneumoniae

Days 1, 3, and 5
and 48 h, 2
weeks, and 8
weeks AT

4/24

9/22

0/24

0/22

0/24

3/22

30 Days AT

52

58

37

38

Klebsiella species, P. aeruginosa, E. coli, S. aureus, H.


influenzae

Not specified

12/57c

14/64c

2/37

0/38

3/52

5/58

None

21

19

21

19

Staphylococcus species, P.
aeruginosa, Streptococcus
pneumoniae, K. pneumoniae

Not specified

ND

ND

0/21

0/19

3/21

0/19

2 Weeks or up
to hospital
discharge

26

24

21

17

P. aeruginosa (40), S. aureus,


enterococci

Every 4 days and at


2472 h AT

8/21

5/17

6/21

2/17

5/26

4/24

was published after 1 January 1980. Both studies with blinded


designs and those with unblinded designs were included in the
analysis. Experimental trials and studies focusing on pharmacokinetic and/or pharmacodynamic parameters were excluded.
In addition, studies that focused on patients with neutropenia
or patients with cystic fibrosis were excluded.
Data extraction. The following data were extracted from
each study: year of publication, clinical setting, patient population, number of patients, antimicrobial agents and doses used,
mortality, and clinical and microbiological outcomes. Data were
extracted by 2 independent reviewers. Any disagreement between the 2 reviewers was resolved by consensus in meetings
that involved all authors.
Outcomes. The primary outcome of the analysis was emergence of antimicrobial resistancethat is, the proportion of
patients in whom bacterial isolates became resistant to the
drug(s) administered to them, during treatment or during the
follow-up study period. An organism was considered to have
developed resistance only if it was clear from the study that it
was reisolated and its susceptibility had changed to a moreresistant phenotype (i.e., from initially susceptible to inter-

mediate or to resistant, or from initially intermediate to resistant). Secondary outcomes of the analysis were development
of superinfection (i.e., isolation of a pathogen thought to be
responsible for an infection that was of a different species from
the initially isolated pathogen), treatment failure (no response
to treatment, no clinical improvement, clinical deterioration,
or death from infection), treatment failure attributable to emergence of resistance, treatment failure attributable to superinfection, all-cause mortality during treatment, and mortality due
to infection.
Data analysis and statistical methods. Statistical analyses
were performed using Meta-analyst software (Joseph Lau; Tufts
University School of Medicine, Boston, MA). The heterogeneity
between studies was assessed by using a x2 test; a P value of
!.10 was used to note statistical significance in the analysis of
heterogeneity [7]. Publication bias was assessed by the funnel
plot method using Eggers test [8]. Pooled ORs and 95% CIs
for all primary and secondary outcomes were calculated by use
of both the Mantel-Haenszel fixed-effects and the DerSimonian-Laird random-effects models [9, 10]. For all analyses,
results from the fixed-effects model are presented only when

Aminoglycoside/b-Lactam Resistance Emergence CID 2005:41 (15 July) 151

Downloaded from http://cid.oxfordjournals.org/ by guest on April 27, 2016

Duration of
follow-up

Table 2. Reported data on mortality and treatment failure from the randomized, controlled trials included
in the analysis.

All-cause
mortality

Mortality
attributable
to infection

Treatment
failure

Treatment
failure due
to emergence
of resistance

Treatment
failure due to
superinfections

Study

Mono
arm

Combo
arm

Mono
arm

Combo
arm

Mono
arm

Combo
arm

Mono
arm

Combo
arm

Mono
arm

Combo
arm

[61]
[62]

20/261
ND

21/230
ND

8/261
18/142

14/230
13/138

34/261
29/142

51/230
19/138

4/261
ND

0/230
ND

ND
3/142

ND
5/138

[74]
[67]
[12]

ND
2/144
ND

ND
5/124
ND

2/39
0/144
0/24

7/70
0/124
3/22

17/39
11/144
4/24

48/70
12/124
13/22

ND
1/144
0/24

ND
0/124
0/22

11/39
ND
0/24

34/70
ND
3/22

[21]
[30]
[33]

ND
1/21
ND

ND
2/19
ND

ND
ND
2/26

ND
ND
3/24

5/52
3/21
6/26

8/58
4/19
6/24

ND
0/21
1/26

ND
0/19
0/24

ND
3/21
2/26

ND
0/19
0/24

there was no heterogeneity between studies; otherwise, results


from the random-effects model are presented. The reported
results of outcomes of the analyzed studies were weighted by
the inverse of their variance with the fixed-effects model.
RESULTS
Study selection. We identified 65 relevant studies that compared outcomes for patients receiving b-lactam monotherapy
with outcomes for those receiving an aminoglycoside/b-lactam
combination. Thirty-six evaluable studies were published during 19801989 [1146], and 29 studies were published from
1990 through October 2004 [4775]. Thirteen studies were
excluded from further analysis for various reasons: 7 studies
were excluded because they examined perioperative anti-infective prophylaxis [16, 19, 27, 55, 58, 63, 68], 3 were excluded
because they were performed with immunocompromised patients [22, 69, 73], 1 was excluded because patients were stratified and received combination therapy or monotherapy according to the severity of disease [75], 1 was excluded because
it was a comparative but not randomized trial [46], and 1 was
excluded because the studied aminoglycoside was administered
endotracheally [43].
Of the remaining 52 studies, 42 were excluded after careful
review of the reported data because they did not present specific
data regarding the emergence of resistance to the study antibiotics. Two more studies were excluded because it was not
clearly stated whether the antimicrobial-resistant isolates were
of the same species as the initial pathogens (i.e., there was
emergence of resistance) or whether they were resistant isolates
of different species (i.e., superinfection occurred) [54, 56]. Subsequently, 8 RCTs were evaluable for further analysis of data
regarding the primary or secondary outcomes of our metaanalysis (tables 1 and 2) [12, 21, 30, 33, 61, 62, 67, 74].
152 CID 2005:41 (15 July) Bliziotis et al.

Emergence of resistance. Analysis of data from 8 RCTs that


were included in our study showed no difference in the emergence of antimicrobial resistance between b-lactam monotherapy and aminoglycoside/b-lactam combination therapy (OR,
0.90; 95% CI, 0.561.47). A subset analysis of 2 RCTs that used
the same b-lactam in both trial arms did not show different
results (OR, 0.66, 95% CI, 0.301.42). Figure 1 shows the individual study and pooled ORs, as well as their 95% CIs, for
emergence of antimicrobial resistance during b-lactam monotherapy, compared with aminoglycoside/b-lactam combination
therapy.
Five RCTs reported data regarding development of resistance
in different microbial species. The respective noted rates of
emergence of resistance for the main pathogens in the monotherapy and the combination therapy group were as follows:
Pseudomonas aeruginosa, 8 (20.5%) of 39 isolates versus 10
(20.8%) of 48 isolates (P p .97); Pseudomonas species, 10
(16.4%) of 61 isolates versus 10 (14.9%) of 67 isolates (P p
.82); Klebsiella species, 0 (0%) of 22 isolates versus 1 (3.8%)
of 26 isolates (P p 1); Proteus species, 1 (6.3%) of 16 isolates
versus 0 (0%) of 17 isolates (P p .48); Acinetobacter species, 0
(0%) of 11 isolates versus 1 (4.6%) of 22 isolates (P p 1); and
Staphylococcus aureus, 2 (5.4%) of 37 isolates versus 7 (13.7%)
of 51 isolates (P p .29).
Development of superinfection. Seven RCTs reported data
regarding superinfections during the study period in each study
group. The results of the analysis of superinfections for the
individual studies and the combined data are presented in figure
2. Treatment with b-lactam monotherapy was associated with
a lower number of superinfections than was aminoglycoside/
b-lactam combination treatment (OR, 0.62; 95% CI, 0.42
0.93). A subset analysis of 2 RCTs that used the same b-lactam
in both trial arms also showed results in favor of b-lactam

Downloaded from http://cid.oxfordjournals.org/ by guest on April 27, 2016

NOTE. Data are no. of patients with finding or result/no. of clinically evaluable patients. Combo arm, aminoglycoside/b-lactam
combination therapy arm; mono arm, b-lactam monotherapy arm; ND, data not reported.

Figure 1. Emergence of antimicrobial resistance in randomized controlled trials comparing b-lactam monotherapy with aminoglycoside/b;-lactam
combination therapy. The size of each square denotes the proportion of information given by each trial. Vertical line, no difference point in emergence
of antimicrobial resistance between the 2 regimens; horizontal lines, 95% CIs; squares, ORs; diamond, pooled OR for all studies. GA Study Group,
German and Austrian Imipenem/Cilastin Study Group.

3B) or treatment failure due to superinfection (OR, 0.60; 95%


CI, 0.331.10; figure 3C).
Mortality. All-cause mortality during treatment was reported in 3 RCTs, whereas mortality attributable to infection
was presented in 6 RCTs. No difference between b-lactam
monotherapy and aminoglycoside/b-lactam combination therapy was found for either all-cause mortality (OR, 0.70; 95%
CI, 0.401.25; figure 4A) or mortality due to infection (OR,
0.74; 95% CI, 0.461.21; figure 4B).
DISCUSSION
The results of several observational prospective and retrospective studies with different designs that examined risk factors

Figure 2. Development of superinfections in randomized, controlled trials comparing b-lactam monotherapy with aminoglycoside/b-lactam combination therapy. The size of each square denotes the proportion of information given by each trial. Vertical line, no difference point in development
of superinfections between the 2 regimens; horizontal lines, 95% CIs; squares, ORs; diamond, pooled OR for all studies.
Aminoglycoside/b-Lactam Resistance Emergence CID 2005:41 (15 July) 153

Downloaded from http://cid.oxfordjournals.org/ by guest on April 27, 2016

monotherapy, compared with the aminoglycoside/b-lactam


combination (OR, 0.52; 95% CI, 0.280.97).
Treatment failure. Treatment failure in each treatment arm
was reported in all 8 RCTs that were included in our study.
Treatment failure was less common in the b-lactam monotherapy arm than in the aminoglycoside/b-lactam combination
arm, as shown in figure 3A (OR, 0.62; 95% CI, 0.381.01).
Additional analysis of data regarding treatment failure was
performed in 5 studies that presented information about treatment failure attributable to emergence of resistance, as well as
in 5 studies that presented information about treatment failure
attributable to superinfection. No statistically significant difference was found regarding either treatment failure due to
emergence of resistance (OR, 3.09; 95% CI, 0.7512.82; figure

and outcomes related to the emergence of antimicrobial resistance have not been consistent. For example, 3 studies that
examined the emergence of antimicrobial resistance in initially
antimicrobial-susceptible Enterobacter clinical isolates reported
different results. Chow et al. [76] (in a prospective, noninterventional study) and Lee et al. [77] (in a retrospective study)
reported that emergence of resistance to the administered blactam therapy did not occur less frequently among patients
receiving aminoglycoside/b-lactam combination therapy than
among patients receiving b-lactam monotherapy. On the other
hand, in a retrospective study, Kaye et al. [78] reported a lower
rate of emergence of antimicrobial resistance with the use of
aminoglycoside/b-lactam combination therapy, compared with
b-lactam monotherapy, although this finding was not statistically significant. Furthermore, 2 studies that examined the
emergence of resistance (among other outcomes) in patients
with pneumonia who received combination antimicrobial therapy or monotherapy also reported conflicting results. In a large
retrospective study, Kosmidis et al. [79] reported that a lower
rate of emergence of resistance was noted among patients with
154 CID 2005:41 (15 July) Bliziotis et al.

nosocomial pneumonia and among patients in the intensive


care unit (ICU) who were receiving various antibiotic combinations (including b-lactams and aminoglycosides), compared with those who were receiving monotherapy. In contrast,
in a more recent, large, prospective study, Alvarez-Lerma et al.
[80] did not find any protective role for combination antimicrobial therapy against the emergence of resistance in ICU
patients with pneumonia.
In an attempt to clarify the controversial issue of the comparative effect of aminoglycoside/b-lactam combination therapy versus b-lactam monotherapy on the emergence of antimicrobial resistance, we used data from RCTs that offer the
most methodologically rigorous evidence. Previous metaanalyses have studied various outcomes in neutropenic patients
and in immunocompetent patients with sepsis who received
aminoglycoside/b-lactam combination therapy or b-lactam
monotherapy [6, 81, 82]. However, these studies mainly examined effectiveness and safety outcomes without focusing on
the emergence of resistance. The results from our meta-analysis
regarding effectiveness and mortality are in accordance with

Downloaded from http://cid.oxfordjournals.org/ by guest on April 27, 2016

Figure 3. Treatment failure due to any reason (A), due to emergence of resistance (B), or due to superinfections (C) in randomized, controlled trials
comparing b-lactam monotherapy with aminoglycoside/b-lactam combination therapy. The size of each square denotes the proportion of information
given by each trial. Vertical line, no difference point in development of superinfections between the 2 regimens; horizontal lines, 95% CIs; squares,
ORs; diamond, pooled OR for all studies. GA Study Group, German and Austrian Imipenem/Cilastin Study Group.

the results from these studies; aminoglycoside/b-lactam combination therapy is associated with more clinical failures, but
no difference in mortality rates has been found between the 2
regimens.
Our study has several limitations that should be taken into
account when interpreting the results. The most important limitation is that most of the RCTs tested a different b-lactam in
the compared treatment arms. Therefore, one may claim that
development of resistance was influenced by the different blactams in the compared study arms, thus confounding the
effect of aminoglycosides on this outcome, as well as other
outcomes. However, the results of the analysis of data on the
emergence of resistance from 2 studies in which the same blactam was administered in both study groups showed similar
results with those of the full set of the studied RCTs.
Another limitation of our study is that the available RCTs
comparing b-lactam monotherapy versus aminoglycoside/b-

Aminoglycoside/b-Lactam Resistance Emergence CID 2005:41 (15 July) 155

Downloaded from http://cid.oxfordjournals.org/ by guest on April 27, 2016

Figure 4. All-cause mortality (A) and mortality attributable to infection


(B) in randomized, controlled trials comparing b-lactam monotherapy with
aminoglycoside/b-lactam combination therapy. The size of each square
denotes the proportion of information given by each trial. Vertical line,
no difference point in development of superinfections between the 2
regimens; horizontal lines, 95% CIs; squares, ORs; diamond, pooled OR
for all studies. GA Study Group, German and Austrian Imipenem/Cilastin
Study Group.

lactam combination therapy focused primarily on effectiveness,


toxicity, and mortality outcomes. These RCTs were not designed
specifically to study the emergence of antimicrobial resistance
and did not include the emergence of antimicrobial resistance
as one of the primary outcomes of the analysis. Subsequently,
the methodology to study this outcome was not the same in
the included RCTs. Specifically, there was considerable variation
between RCTs regarding length of the follow-up period and
the frequency of follow-up microbiological tests throughout
the study period.
Furthermore, although the outcomes in patients receiving blactam monotherapy or aminoglycoside/b-lactam combination
therapy were compared in 160 studies, only 8 studies fulfilled
the criteria for inclusion in our analysis. One may also criticize
the fact that all of these studies were published 110 years ago.
It should be emphasized that resistance is a continuously evolving phenomenon, and this fact makes the interpretation of
results from RCTs performed with a 15- or 20-year time difference difficult. In fact, our data seem to indicate that a lower
rate of emergence of antimicrobial resistance was noted in the
monotherapy arm in studies performed after 1989.
Another noteworthy limitation of our study is the fact that
genetic and/or molecular typing of the initial bacterial isolates
and those isolates of the same species that were isolated again
during the study and found to have developed resistance was
not performed in any of the evaluable RCTs. Thus, we cannot
make a definite statement that the isolation of an antimicrobialresistant microorganism represents, in all cases, emergence of
resistance in the same organism and not another isolate (different clone) of the same microbial species. In 2 prospective
studies that examined risk factors for development of resistance
in Enterobacter bacteremia [76] and Pseudomonas infection
[83], molecular typing and PFGE revealed that isolates with
altered susceptibility were clones of the initial isolate in 64%
and 100% of cases, respectively.
Despite these limitations, we believe that our study adds
useful information to the literature regarding the effect of aminoglycoside/b-lactam combination therapy versus b-lactam
monotherapy on the emergence of antimicrobial resistance. The
data do not support a beneficial effect of aminoglycoside/blactam combination therapy, compared with b-lactam monotherapy, with regard to this outcome. In fact, the data from
several studies do not provide evidence to support the widespread belief among clinicians that aminoglycoside/b-lactam
combinations lead to better outcomes than does b-lactam
monotherapy in various populations and settings. All recent
meta-analyses of RCTs that examined various outcomes, including effectiveness and toxicity, favor the use of b-lactam
monotherapy instead of aminoglycoside/b-lactam combination
therapy. Furno et al. [81] and Paul et al. [82] provided evidence
against the safety and efficacy of the aminoglycoside/b-lactam

12.

13.

14.

15.

16.

17.

18.

19.

20.

Acknowledgments
Potential conflicts of interest. All authors: no conflicts.

21.

References
1. Alford RH, Manian FA. Current antimicrobial management of tuberculosis. In: Remington JS, Swartz MN, eds. Current clinical topics in
infectious diseases. New York: Mc-Graw-Hill Books, 1987:20426.
2. Karchmer AW, Archer GL, Dismukes WE. Rifampin treatment of prosthetic valve endocarditis due to Staphylococcus epidermidis. Rev Infect
Dis 1983; 5(Suppl 3):S5438.
3. Gerber AU, Vastola AP, Brandel J, Craig WA. Selection of aminoglycoside-resistant variants of Pseudomonas aeruginosa in an in vivo
model. J Infect Dis 1982; 146:6917.
4. Johnson DE, Thompson B. Efficacy of single-agent therapy with azlocillin, ticarcillin, and amikacin and beta-lactam/amikacin combinations
for treatment of Pseudomonas aeruginosa bacteremia in granulocytopenic rats. Am J Med 1986; 80:538.
5. Pechere JC, Marchou B, Michea-Hamzehpour M, Auckenthaler R.
Emergence of resistance after therapy with antibiotics used alone or
combined in a murine model. J Antimicrob Chemother 1986; 17(Suppl
A):118.
6. Paul M, Benuri-Silbiger I, Soares-Weiser K, Leibovici L. Beta lactam
monotherapy versus beta lactamaminoglycoside combination therapy
for sepsis in immunocompetent patients: systematic review and metaanalysis of randomised trials. BMJ 2004; 328:668.
7. Lau J, Ioannidis JP, Schmid CH. Quantitative synthesis in systematic
reviews. Ann Intern Med 1997; 127:8206.
8. Egger M, Davey SG, Schneider M, Minder C. Bias in meta-analysis
detected by a simple, graphical test. BMJ 1997; 315:62934.
9. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin
Trials 1986; 7:17788.
10. Mantel N, Haenszel W. Statistical aspects of the analysis of data from
retrospective studies of disease. J Natl Cancer Inst 1959; 22:71948.
11. Piccart M, Klastersky J, Meunier F, Lagast H, Van Laethem Y, Weerts
D. Single-drug versus combination empirical therapy for gram-negative

156 CID 2005:41 (15 July) Bliziotis et al.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

bacillary infections in febrile cancer patients with and without granulocytopenia. Antimicrob Agents Chemother 1984; 26:8705.
Gerecht WB, Henry NK, Hoffman WW, et al. Prospective randomized
comparison of mezlocillin therapy alone with combined ampicillin and
gentamicin therapy for patients with cholangitis. Arch Intern Med
1989; 149:127984.
Mondorf AW, Bonsiepe C, Mondorf W. Randomized multi center study
comparing nephrotoxicity of ceftazidime versus the combination of
piperacillin and netilmicin with and without furosemide. Adv Exp Med
Biol 1989; 252:30712.
Bergeron MG, Mendelson J, Harding GK, et al. Cefoperazone compared with ampicillin plus tobramycin for severe biliary tract infections.
Antimicrob Agents Chemother 1988; 32:12316.
Hoepelman IM, Rozenberg-Arska M, Verhoef J. Comparison of once
daily ceftriaxone with gentamicin plus cefuroxime for treatment of
serious bacterial infections. Lancet 1988; 1:13059.
Lauridsen F, Bjoernsen K, Nielsen SA, Hart HO. Short-term prophylaxis with cefotaxime in colorectal surgery: a prospective, randomized
trial. Dis Colon Rectum 1988; 31:257.
Mangi RJ, Greco T, Ryan J, Thornton G, Andriole VT. Cefoperazone
versus combination antibiotic therapy of hospital-acquired pneumonia.
Am J Med 1988; 84:6874.
Limson BM, Navarro-Almario E, Litam P, Que E, Kua LT. Ceftazidime
versus a combination of amikacin and ticarcillin in the treatment of
severe infections. Clin Ther 1988; 10:58993.
Kaiser AB, Petracek MR, Lea JW, et al. Efficacy of cefazolin, cefamandole, and gentamicin as prophylactic agents in cardiac surgery: results
of a prospective, randomized, double-blind trial in 1030 patients. Ann
Surg 1987; 206:7917.
Muller EL, Pitt HA, Thompson JE Jr, Doty JE, Mann LL, Manchester
B. Antibiotics in infections of the biliary tract. Surg Gynecol Obstet
1987; 165:28592.
Mandell LA, Nicolle LE, Ronald AR, et al. A prospective randomized
trial of ceftazidime versus cefazolin/tobramycin in the treatment of
hospitalized patients with pneumonia. J Antimicrob Chemother
1987; 20:95107.
Rodjer S, Alestig K, Bergmark J, et al. Treatment of septicaemia in
immunocompromised patients with ceftazidime or with tobramycin
and cefuroxime, with special reference to renal effects. J Antimicrob
Chemother 1987; 20:10916.
Sage R, Nazareth B, Noone P. A prospective randomised comparison
of cefotaxime vs. netilmicin vs. cefotaxime plus netilmicin in the treatment of hospitalised patients with serious sepsis. Scand J Infect Dis
1987; 19:3317.
Moore RD, Smith CR, Holloway JJ, Lietman PS. Cefotaxime vs. nafcillin and tobramycin for the treatment of serious infection: comparative cost-effectiveness. Arch Intern Med 1986; 146:11537.
Rasmussen D, Bremmelgaard A, Rasmussen F, Thorup J. Treatment of
serious urological infections with cefotaxime compared to ampicillin
plus netilmicin. Dan Med Bull 1986; 33:4951.
Tally FP, Kellum JM, Ho JL, ODonnell TF, Barza M, Gorbach SL.
Randomized prospective study comparing moxalactam and cefoxitin
with or without tobramycin for the treatment of serious surgical infections. Antimicrob Agents Chemother 1986; 29:2449.
OLeary JA, Mullins JH Jr, Andrinopoulos GC. Ampicillin vs. ampicillin-gentamicin prophylaxis in high-risk primary cesarean section. J
Reprod Med 1986; 31:2730.
Solomkin JS, Cocchetto DM. Ceftazidime versus tobramycin plus ticarcillin in the treatment of soft-tissue infections. Clin Ther 1986; 9:
12334.
Sheftel TG, Mader JT. Randomized evaluation of ceftazidime or ticarcillin and tobramycin for the treatment of osteomyelitis caused by
gram-negative bacilli. Antimicrob Agents Chemother 1986; 29:1125.
Cone LA, Woodard DR, Stoltzman DS, Byrd RG. Ceftazidime versus
tobramycin-ticarcillin in the treatment of pneumonia and bacteremia.
Antimicrob Agents Chemother 1985; 28:336.
Felisart J, Rimola A, Arroyo V, et al. Cefotaxime is more effective than

Downloaded from http://cid.oxfordjournals.org/ by guest on April 27, 2016

combination (compared with b-lactam monotherapy) in febrile


neutropenic patients. Paul et al. [6] arrived at the same conclusion for nonneutropenic, immunocompetent patients with
sepsis. In addition, Safdar et al. [84] also reported that there
was no benefit for aminoglycoside/b-lactam combination therapy, compared with b-lactam monotherapy, in a meta-analysis
of observational studies and RCTs that involved patients with
bacteremia due to gram-negative pathogens. It should be noted
that the aminoglycoside/b-lactam combination may have some
value in specific subsets of patients with infections, such as
those with Pseudomonas bacteremia and those with septic shock
due to Klebsiella bacteremia [84, 85].
In conclusion, a meta-analysis of data of relevant RCTs did
not show any protective effect on the emergence of antimicrobial resistance for aminoglycoside/b-lactam combination therapy, compared with b-lactam monotherapy. Several laboratory
and clinical studies have showed that emergence of resistance
to b-lactams with various mechanisms may occur during treatment with this class of antimicrobial agents. However, our data
do not provide support for the belief that the addition of an
aminoglycoside to a b-lactam treatment regimen is a likely
solution to this problem.

32.

33.

34.

35.

36.

37.

39.

40.

41.

42.

43.

44.

45.

46.
47.

48.

49.

50. Norrby SR, Geddes AM, Shah PM. Randomized comparative trial of
cefpirome versus ceftazidime in the empirical treatment of suspected
bacteraemia or sepsis. Multicentre Study Group. J Antimicrob Chemother 1998; 42:5039.
51. Gilbert DN, Lee BL, Dworkin RJ, et al. A randomized comparison of
the safety and efficacy of once-daily gentamicin or thrice-daily gentamicin in combination with ticarcillin-clavulanate. Am J Med 1998;
105:18291.
52. Speich R, Imhof E, Vogt M, Grossenbacher M, Zimmerli W. Efficacy,
safety, and tolerance of piperacillin/tazobactam compared to co-amoxiclav plus an aminoglycoside in the treatment of severe pneumonia.
Eur J Clin Microbiol Infect Dis 1998; 17:3137.
53. Wing DA, Hendershott CM, Debuque L, Millar LK. A randomized
trial of three antibiotic regimens for the treatment of pyelonephritis
in pregnancy. Obstet Gynecol 1998; 92:24953.
54. Jaspers CA, Kieft H, Speelberg B, et al. Meropenem versus cefuroxime
plus gentamicin for treatment of serious infections in elderly patients.
Antimicrob Agents Chemother 1998; 42:12338.
55. Kriaras I, Michalopoulos A, Michalis A, et al. Antibiotic prophylaxis
in cardiac surgery. J Cardiovasc Surg (Torino) 1997; 38:60510.
56. Sieger B, Berman SJ, Geckler RW, Farkas SA. Empiric treatment of
hospital-acquired lower respiratory tract infections with meropenem
or ceftazidime with tobramycin: a randomized study. Meropenem
Lower Respiratory Infection Group. Crit Care Med 1997; 25:166370.
57. McCormick PA, Greenslade L, Kibbler CC, Chin JK, Burroughs AK,
McIntyre N. A prospective randomized trial of ceftazidime versus netilmicin plus mezlocillin in the empirical therapy of presumed sepsis
in cirrhotic patients. Hepatology 1997; 25:8336.
58. Sandberg T, Alestig K, Eilard T, et al. Aminoglycosides do not improve
the efficacy of cephalosporins for treatment of acute pyelonephritis in
women. Scand J Infect Dis 1997; 29:1759.
59. Ribera E, Gomez-Jimenez J, Cortes E, et al. Effectiveness of cloxacillin
with and without gentamicin in short-term therapy for right-sided
Staphylococcus aureus endocarditis: a randomized, controlled trial. Ann
Intern Med 1996; 125:96974.
60. Solberg CO, Sjursen H. Safety and efficacy of meropenem in patients
with septicaemia: a randomised comparison with ceftazidime, alone or
combined with amikacin. J Antimicrob Chemother 1995; 36(Suppl A):
15766.
61. Rubinstein E, Lode H, Grassi C. Ceftazidime monotherapy vs. ceftriaxone/tobramycin for serious hospital-acquired gram-negative infections.
Antibiotic Study Group. Clin Infect Dis 1995; 20:121728.
62. Cometta A, Baumgartner JD, Lew D, et al. Prospective randomized
comparison of imipenem monotherapy with imipenem plus netilmicin
for treatment of severe infections in nonneutropenic patients. Antimicrob Agents Chemother 1994; 38:130913.
63. Hall JC, Christiansen K, Carter MJ, et al. Antibiotic prophylaxis in
cardiac operations. Ann Thorac Surg 1993; 56:91622.
64. Thompson JE Jr, Bennion RS, Roettger R, Lally KP, Hopkins JA, Wilson
SE. Cefepime for infections of the biliary tract. Surg Gynecol Obstet
1993; 177(Suppl):304.
65. Yellin AE, Berne TV, Appleman MD, et al. A randomized study of
cefepime versus the combination of gentamicin and mezlocillin as an
adjunct to surgical treatment in patients with acute cholecystitis. Surg
Gynecol Obstet 1993; 177(Suppl):239.
66. Finer N,.Goustas P. Ceftazidime versus aminoglycoside and (ureido)penicillin combination in the empirical treatment of serious infection. J R Soc Med 1992; 85:5303.
67. Randomized multicenter clinical trial with imipenem/cilastatin versus
cefotaxime/gentamicin in the treatment of patients with nonlifethreatening infections. German and Austrian Imipenem/Cilastatin
Study Group. Eur J Clin Microbiol Infect Dis 1992; 11:68392.
68. Percin S, Akbas L, Akbas A. A single preoperative dose of cefazolin
versus postoperative amoxicillin and gentamycin combination in prophylaxis of orthopaedic and traumatologic surgery. Mater Med Pol
1992; 24:247.
69. DAntonio D, Fioritoni G, Iacone A, et al. Randomized comparison

Aminoglycoside/b-Lactam Resistance Emergence CID 2005:41 (15 July) 157

Downloaded from http://cid.oxfordjournals.org/ by guest on April 27, 2016

38.

is ampicillin-tobramycin in cirrhotics with severe infections. Hepatology 1985; 5:45762.


Smith CR, Ambinder R, Lipsky JJ, et al. Cefotaxime compared with
nafcillin plus tobramycin for serious bacterial infections: a randomized,
double-blind trial. Ann Intern Med 1984; 101:46977.
Gribble MJ, Chow AW, Naiman SC, et al. Prospective randomized trial
of piperacillin monotherapy versus carboxypenicillin-aminoglycoside
combination regimens in the empirical treatment of serious bacterial
infections. Antimicrob Agents Chemother 1983; 24:38893.
Vetter N, Feist H, Muhar F, Williams KJ. A comparative study of the
efficacy of ceftazidime versus cefazolin and tobramycin in patients with
acute exacerbations of chronic bronchitis. J Antimicrob Chemother
1983; 12(Suppl A):359.
Fainstein V, Bodey GP, Elting L, et al. A randomized study of ceftazidime compared to ceftazidime and tobramycin for the treatment of
infections in cancer patients. J Antimicrob Chemother 1983; 12(Suppl
A):10110.
Crenshaw C, Glanges E, Webber C, McReynolds DB. A prospective
random study of a single agent versus combination antibiotics as therapy in penetrating injuries of the abdomen. Surg Gynecol Obstet
1983; 156:28994.
Warren JW, Miller EH Jr, Fitzpatrick B, et al. A randomized, controlled
trial of cefoperazone vs. cefamandole-tobramycin in the treatment of
putative, severe infections with gram-negative bacilli. Rev Infect Dis
1983; 5(Suppl 1):S17380.
Coppens L, Hanson B, Klastersky J. Therapy of staphylococcal infections with cefamandole or vancomycin alone or with a combination
of cefamandole and tobramycin. Antimicrob Agents Chemother
1983; 23:3641.
Korzeniowski O, Sande MA. Combination antimicrobial therapy for
Staphylococcus aureus endocarditis in patients addicted to parenteral
drugs and in nonaddicts: a prospective study. Ann Intern Med 1982;
97:496503.
Hanson B, Coppens L, Klastersky J. Comparative studies of ticarcillin
and mezlocillin plus sisomicin in gram-negative bacillary bacteraemia
and bronchopneumonia. J Antimicrob Chemother 1982; 10:33541.
Duff P, Keiser JF. A comparative study of two antibiotic regimens for
the treatment of operative site infections. Am J Obstet Gynecol
1982; 142:9961003.
Thompson SE III, Hager WD, Wong KH, et al. The microbiology and
therapy of acute pelvic inflammatory disease in hospitalized patients.
Am J Obstet Gynecol 1980; 136:17986.
Sculier JP, Coppens L, Klastersky J. Effectiveness of mezlocillin and
endotracheally administered sisomicin with or without parenteral sisomicin in the treatment of gram-negative bronchopneumonia. J Antimicrob Chemother 1982; 9:638.
Gentry LO, Wood BA, Martin MD, Smythe J. Cefamandole alone and
combined with gentamicin or tobramycin in the treatment of acute
pyelonephritis. Scand J Infect Dis Suppl 1980; (Suppl 25):96100.
Gibbs RS. A trial of spectinomycin hydrochloride compared with aqueous penicillin G plus kanamycin for treatment of severe pelvic inflammatory disease. Sex Transm Dis 1980; 7:213.
Gentry LO. Treatment of skin, skin structure, bone, and joint infections
with ceftazidime. Am J Med 1985; 79:6774.
Badaro R, Molinar F, Seas C, et al. A multicenter comparative study
of cefepime versus broad-spectrum antibacterial therapy in moderate
and severe bacterial infections. Braz J Infect Dis 2002; 6:20618.
Dupont H, Carbon C, Carlet J. Monotherapy with a broad-spectrum
beta-lactam is as effective as its combination with an aminoglycoside
in treatment of severe generalized peritonitis: a multicenter randomized
controlled trial. The Severe Generalized Peritonitis Study Group. Antimicrob Agents Chemother 2000; 44:202833.
Sexton DJ, Tenenbaum MJ, Wilson WR, et al. Ceftriaxone once daily
for four weeks compared with ceftriaxone plus gentamicin once daily
for two weeks for treatment of endocarditis due to penicillin-susceptible streptococci. Endocarditis Treatment Consortium Group. Clin
Infect Dis 1998; 27:14704.

70.

71.

72.

73.

74.

75.

77.

158 CID 2005:41 (15 July) Bliziotis et al.

78.

79.

80.

81.

82.

83.

84.

85.

cephalosporins in patients with Enterobacter bacteremia. Eur J Clin


Microbiol Infect Dis 2002; 21:57781.
Kaye KS, Cosgrove S, Harris A, Eliopoulos GM, Carmeli Y. Risk factors
for emergence of resistance to broad-spectrum cephalosporins among
Enterobacter spp. Antimicrob Agents Chemother 2001; 45:262830.
Kosmidis J, Koratzanis G. Emergence of resistant bacterial strains during treatment of infections in the respiratory tract. Scand J Infect Dis
Suppl 1986; 49:1359.
Alvarez-Lerma F. Modification of empiric antibiotic treatment in patients with pneumonia acquired in the intensive care unit. ICU
Acquired Pneumonia Study Group. Intensive Care Med 1996; 22:
38794.
Furno P, Bucaneve G, Del Favero A. Monotherapy or aminoglycosidecontaining combinations for empirical antibiotic treatment of febrile
neutropenic patients: a meta-analysis. Lancet Infect Dis 2002; 2:23142.
Paul M, Soares-Weiser K, Leibovici L. Beta lactam monotherapy versus
beta lactamaminoglycoside combination therapy for fever with neutropenia: systematic review and meta-analysis. BMJ 2003; 326:1111.
Carmeli Y, Troillet N, Eliopoulos GM, Samore MH. Emergence of
antibiotic-resistant Pseudomonas aeruginosa: comparison of risks associated with different antipseudomonal agents. Antimicrob Agents
Chemother 1999; 43:137982.
Safdar N, Handelsman J, Maki DG. Does combination antimicrobial
therapy reduce mortality in gram-negative bacteraemia? A metaanalysis. Lancet Infect Dis 2004; 4:51927.
Korvick JA, Bryan CS, Farber B, et al. Prospective observational study
of Klebsiella bacteremia in 230 patients: outcome for antibiotic combinations versus monotherapy. Antimicrob Agents Chemother 1992;
36:263944.

Downloaded from http://cid.oxfordjournals.org/ by guest on April 27, 2016

76.

of ceftriaxone versus ceftriaxone plus amikacin for the empirical treatment of infections in patients with altered host defense: microbiological
and clinical evaluation. Chemotherapy 1992; 38:4207.
Melekos MD, Skoutelis A, Chryssanthopoulos C, Bassaris HP. A comparative study on aztreonam, ceftazidime and amikacin in the treatment of complicated urinary tract infections. J Chemother 1991; 3:
37682.
Krumpe PE, Cohn S, Garreltes J, et al. Intravenous and oral monoor combination-therapy in the treatment of severe infections: ciprofloxacin versus standard antibiotic therapy. Ciprofloxacin Study Group.
J Antimicrob Chemother 1999; 43(Suppl A):11728.
Mouton YJ, Beuscart C. Empirical monotherapy with meropenem in
serious bacterial infections. Meropenem Study Group. J Antimicrob
Chemother 1995; 36(Suppl A):14556.
Biglino A, Bonasso M, Gioannini P. Imipenem/cilastatin as empirical
treatment of severe infections in compromised patients. J Chemother
1991; 3(Suppl 1):20812.
Croce MA, Fabian TC, Stewart RM, et al. Empiric monotherapy versus
combination therapy of nosocomial pneumonia in trauma patients.
J Trauma 1993; 35:3039.
Wolff M. Comparison of strategies using cefpirome and ceftazidime
for empiric treatment of pneumonia in intensive care patients. The
Cefpirome Pneumonia Study Group. Antimicrob Agents Chemother
1998; 42:2836.
Chow JW, Fine MJ, Shlaes DM, et al. Enterobacter bacteremia: clinical
features and emergence of antibiotic resistance during therapy. Ann
Intern Med 1991; 115:58590.
Lee SO, Kim YS, Kim BN, Kim MN, Woo JH, Ryu J. Impact of previous
use of antibiotics on development of resistance to extended-spectrum