International Journal of Antimicrobial Agents 11 (1999) 93 – 100

Review
Carbapenems in clinical practice: a guide to their use in
serious infection

J.S. Bradley a, J. Garau b, H. Lode c, K.V.I. Rolston d, S.E. Wilson e, J.P. Quinn f,*
a
Children’s Hospital, San Diego, USA
b
Department of Medicine, Hospital Mutua de Terrassa, Barcelona, Spain
c
Krankenhaus Zehlendorf, Berlin, Germany
d
MD Anderson Cancer Center, Houston, USA
e
Department of Surgery, Uni6ersity of California, Ir6ine, USA
f
Department of Medicine, Uni6ersity of Illinois at Chicago, 840 S. Wood Street, Chicago, IL 60612, USA

Received 18 September 1998; accepted 30 October 1998

Abstract

Meropenem and imipenem/cilastatin, currently the only available carbapenem agents in Europe and the United States, are
characterised by a broad spectrum of antimicrobial activity and stability to b-lactamase-mediated resistance mechanisms. A guide
to the use of carbapenems in clinical practice is presented; the role of carbapenems in the treatment of several types of serious
bacterial infection and an up-to-date account of their clinical efficacy and safety profiles are discussed. The good clinical efficacy
and favourable safety profiles of the carbapenems make them valuable as initial empirical therapy in the treatment of
ventilator-associated pneumonia, sepsis of unknown origin, post-operative peritonitis, paediatric meningitis, and febrile neutrope-
nia. However, to maintain superior efficacy, the carbapenems should be used appropriately for definitive therapy. © 1999 Elsevier
Science B.V. and International Society of Chemotherapy. All rights reserved.

Keywords: Carbapenem; Bacterial infection; b-Lactamase

1. Introduction lence of bacterial resistance for the pathogens under

consideration as there may be a recognised resistance
Patients hospitalised with serious bacterial infections problem at a particular centre.
require prompt empirical treatment with intravenous Carbapenems (meropenem and imipenem/cilastatin)
antibiotic therapy. However, the choice of which antibi- have a broad spectrum of antimicrobial activity which
otic, or combination of antibiotics, to use as first-line exceeds that of most other antimicrobial classes. Their
empirical therapy is a complex matter. Factors affecting potency extends to organisms resistant to many other
therapeutic choices include the type of infection to be types of antibiotics, including other b-lactam agents [1]
treated, the infection site, and the known or suspected (Table 1). For example, carbapenems are active against
organism(s). The spectrum of antimicrobial activity, antibiotic-resistant enteric bacteria that hyperproduce
clinical efficacy, and safety profile of the agent(s) class 1 (AmpC) b-lactamases, such as Enterobacter spp.,
should also be considered, as well as the local preva- or produce extended spectrum b-lactamases (ESBLs),
such as Klebsiella spp., the enzymes responsible for the
* Corresponding author. Tel.: +1-312-9967835; fax: + 1-312- growing problem of resistance to third-generation
4130342. cephalosporins.

0924-8579/99/$ - see front matter © 1999 Elsevier Science B.V. and International Society of Chemotherapy. All rights reserved.
PII: S 0 9 2 4 - 8 5 7 9 ( 9 8 ) 0 0 0 9 4 - 6
94 J.S. Bradley et al. / International Journal of Antimicrobial Agents 11 (1999) 93–100

Table 1 nia are considered. Other relevant practical issues, in-

Susceptibilities to the carbapenems of common bacterial pathogens
cluding bacterial resistance, infection management and
Carbapenem-susceptible Carbapenem-resistant pathogens cost-effectiveness of therapy, are also addressed.
pathogens

Gram-positi7e 2. Bacterial resistance

Staphylococcus aureus Methicillin-resistant
Staphylococcus aureus
Staphylococcus epidermidis Methicillin-resistant Antibiotic resistance is common among many bac-
Staphylococcus epidermidis terial pathogens. In Gram-negative bacteria, the pro-
Streptococcus pneumoniae duction of ESBLs confers resistance to most currently
(penicillin-susceptible) utilised b-lactam antibiotics, including third-generation
cephalosporins. In a recent study of the prevalence of
Streptococcus pneumoniae
(penicillin-resistant) ESBLs in Klebsiella spp. isolates from 35 European
Enterococci ICUs, these enzymes were found in 66% of the partici-
Gram-negati7e
pating hospitals [2]. There are distinct geographical
Serratia spp. Stenotrophomonas maltophilia differences in the prevalence of resistance. For example
Klebsiella spp. in Klebsiella, Livermore and Yuan [2] found the pro-
Haemophilus spp. portion of ESBL producers to be higher in isolates
Neisseria spp. from Portugal (49%) and Turkey (59%). Jarlier et al.
Pseudomonas aeruginosa
Acinetobacter spp.
found ceftazidime, tobramycin and ciprofloxacin resis-
Proteus spp. tance in Klebsiella to be considerably more frequent in
Citrobacter spp. France than in other European countries [3] (Table 2).
Enterobacter spp. Another b-lactamase-mediated resistance mechanism
Anaerobic growing in importance is the hyperproduction of chro-
Bacteroides spp. mosomal b-lactamase by derepressed mutants in species
Clostridia spp. which naturally produce inducible b-lactamase, such as
Enterobacter, Citrobacter and Serratia spp., and Pseu-
domonas aeruginosa [4]. In nosocomially-acquired infec-
This review, based on an official symposium held at tions, 23–37% Enterobacter spp., 7–53% Citrobacter
the Second European Congress of Chemotherapy spp. and 12–19% P. aeruginosa isolates were found to
(Hamburg, 1998), focuses on the role of carbapenems be stable derepressed mutants [5–8]; these organisms
and their appropriate use in the treatment of serious are not susceptible to initial empirical therapy by third-
infections. In particular, the treatment of intensive care generation cephalosporins.
unit (ICU), intra-abdominal and lower respiratory tract The carbapenems are stable to hydrolysis by serine-
infections, paediatric meningitis, and febrile neutrope- based b-lactamases including ESBLs and, unlike the

Table 2
Comparative rates of susceptibility of isolates from four European ICUs (bacteria commonly seen in an ICU)a

Organism Antibiotic susceptibility (%)

Country (number of strains) Ceftazidime Imipenem Tobramycin Ciprofloxacin

K. pneumoniae Belgium (148) 87 97 84 95

France (229) 64 100 55 65
Germany (57) 95 100 98 96
Netherlands (42) 88 93 98 95
E. cloacae Belgium (153) 61 97 84 97
France (158) 71 99 86 89
Germany (94) 59 100 96 95
Netherlands (87) 63 98 87 91
P. aeruginosa Belgium (326) 88 85 77 87
France (634) 81 78 64 68
Germany (154) 84 83 91 90
Netherlands (143) 81 90 92 80

a
Adapted from Intensive Care Medicine, Antibiotic susceptibility in aerobic gram-negative bacilli isolated in intensive care units in 39 French
teaching hospitals (ICU study), Jarlier V, Fosse T, Philippon A, 22, 1057–1065, Table 4, 1996, copyright notice of Springer-Verlag, with
permission.
 J.S. Bradley et al. / International Journal of Antimicrobial Agents 11 (1999) 93–100 95

Table 3
Clinical response rates from two prospective clinical trials of meropenem and imipenem/cilastatin in the ICU

Indication Satisfactory response (%)

Ref. [11] Ref. [4]

Meropenem (n= 90) Imipenem/cilastatin (n =87) Meropenem (n =66) Imipenem/cilastatin (n = 67)

LRTI (community) 80 80 n/aa n/aa

LRTI (hospital) 75 75 89 76
Intra-abdominal 82 81 73 73
Septicaemia 88 67 64 72

a
n/a, not assessed.

newer cephalosporins, they are not prone to inacti- patients, the main reason for change of therapy being
vation by mutant b-lactamases. Thus, b-lactamase ac- the isolation of organisms not susceptible to initial
tivity in these enzymes against the carbapenems is less therapy, such as Acinetobacter spp., P. aeruginosa, En-
likely to develop [9]. terobacter spp. and Serratia spp. In a prospective,
Emergence of resistant bacterial strains during car- observational study of 129 consecutive patients with
bapenem therapy occurs infrequently, with the notable Enterobacter bacteraemia, Chow and co-workers [13]
exception of serious P. aeruginosa infections. When concluded that avoiding third-generation
imipenem/cilastatin therapy is used in these infections, cephalosporins in the treatment of Enterobacter spp.
emergence of resistant strains occurs : 20% of the may lead to a decreased emergence of multiresistant
time. About half of these patients are merely colonised, strains and a lower associated mortality rate.
while the others have true infection. Typically, these The stability of the carbapenems to AmpC or ESBL
resistant strains have lost (by mutation) a specialised b-lactamase hydrolysis makes them an appropriate
porin protein, Opr D2, which allows selective penetra- monotherapy for nosocomial infections in the ICU.
tion of carbapenems across the outer membrane of P. Evidence supporting this notion comes from two recent,
aeruginosa [10]. prospective, European, multicentre clinical studies com-
In addition, the use of carbapenems has been compli- paring meropenem with imipenem/cilastatin (Table 3)
cated sporadically by emergence of unusual organisms which demonstrated a good clinical response in serious
that are intrinsically resistant, such as Stenotrophom- infections [4,11]. Since carbapenem resistance does not
onas maltophilia, Enterococcus faecium, and Candida emerge during treatment, except in some P. aeruginosa
spp. In clinical practice, the risk of superinfections can organisms, initial monotherapy is a desirable option at
be largely avoided by appropriate adjustments to the least until results of susceptibility tests are known [11]
choice of antimicrobial treatment once initial suscepti- and decisions on definitive therapy can be made.
bility results have become available [11]. Patient morbidity and mortality are the primary con-
To minimise the likelihood of the emergence of resis- siderations in the ICU. Drug costs pale into insignifi-
tance when using a carbapenem for high-risk organ- cance compared with personnel, running and
isms, such as P. aeruginosa or Acinetobacter spp., equipment expenses, so a clinically effective initial em-
clinicians commonly employ combination therapy with pirical therapy, such as that provided by the carbapen-
an aminoglycoside. However, there is no convincing ems, often represents the most cost-effective option.
evidence, to date, that this strategy is effective. Clear indications for empirical use of carbapenems in
the ICU include late ventilator-associated pneumonia
(discussed in a later section), severe sepsis, and post-
3. Intensive care unit infections operative peritonitis (discussed below). Carbapenems
are also extremely valuable where there is a history of
The prevalence of antimicrobial resistance in nosoco- previous antibiotic therapy which may have selected for
mial isolates, particularly those cultured from ICUs, is resistance, or a known problem in the unit due to
increasing. This is of major concern to the clinician as ESBLs or chromosomal b-lactamases, or in polymicro-
multiply resistant bacterial strains can be associated bial infections.
with high mortality rates. In a 1 year, prospective,
multicentre study of patients with pneumonia acquired
in the ICU [12], the reasons for modifying initial 4. Intra-abdominal infections
-empirical treatment were investigated. The results
showed that treatment had to be modified in 44% of The outcome of treatment of intra-abdominal infec-
96 J.S. Bradley et al. / International Journal of Antimicrobial Agents 11 (1999) 93–100

tions is influenced by a number of factors. For example, obstructive pulmonary disease (COPD), severe commu-
the physician’s technical expertise and clinical judgement nity-acquired pneumonia, and hospital-acquired pneu-
regarding the timing of surgery can have a major impact monia.
on outcome. Also, the anatomical site of infection [14,15] When patients with acute infective exacerbations of
severity of illness (as shown by a high APACHE II score) COPD are categorised by severity of disease (as mea-
[14,16], the nutritional and immune status of the patient sured by pulmonary function parameters such as
[14], and the presence of resistant organisms at the FEV1), it becomes apparent that Gram-negative bac-
infection site [17,18] may affect outcome. terial pathogens, such as Klebsiella spp., Proteus spp.,
Carbapenems have potential utility in the treatment of P. aeruginosa and Enterobacter spp., predominate in
intra-abdominal infections, which are often polymi- the more severe cases which are mainly older patients
crobial, since they exhibit activity against almost all of with long standing COPD [29].
the pathogens likely to be encountered. This contention Risk factors affecting outcome of pneumonia have
is supported by results of clinical trials which have shown been identified as age over 65 years, male gender,
equivalency or superiority of meropenem compared to difficult-to-treat pathogens, serum urea above
combinations of antibiotics. For example, for cefotaxime 7 mmol/l and serum albumin B 35 g/l [30]. Enter-
plus metronidazole [19], or tobramycin plus clindamycin obacteriaceae often play a major role in the aetiology
[20,21], clinical response rates of 89 – 100% were of nosocomial pneumonia, especially in cases of venti-
documented, compared to 91 – 92% for meropenem. lator-associated pneumonia. The so-called late onset
A chronological meta-analysis of ten randomised, pneumonia in ventilated-patients is often characterised
prospective trials of meropenem or imipenem/cilastatin by potentially multiply resistant Gram-negative patho-
monotherapy also demonstrated that these agents can be gens, such as Pseudomonas spp., Acinetobacter spp.
as effective as combinations of antimicrobials in the and Citrobacter spp.
treatment of intra-abdominal infection [22]. Carbapenems have demonstrated excellent clinical
Direct comparisons of the carbapenems have shown
efficacy in severe LRTI in several controlled clinical
little difference in outcome between these agents. For
trials. For example, in a comparison between
example, in three comparative studies, both meropenem
meropenem and imipenem/cilastatin in the treatment
and imipenem/cilastatin achieved clinical response rates
of acute exacerbations of COPD, clinical response at
of \94% [23–25].
treatment end was 98 and 96%, respectively [31]. In
On the basis of clinical data for serious, complicated
community-acquired pneumonia requiring hospitalisa-
intra-abdominal infections, carbapenem monotherapy
tion, satisfactory clinical response rates at the end of
should combat pre- and intra-operative bacteraemia,
therapy were observed for 91% of meropenem-treated
resolve the intraperitoneal inflammation, and minimise
and 90% of ceftazidime-treated patients [30]. In a ran-
post-operative intra-abdominal infection rates.
Bacteriology results from these and other studies [26,27] domised clinical trial of hospital-acquired LRTI [32],
indicate that the primary pathogens, E. coli and B. a satisfactory clinical response was obtained in 89%
fragilis, are well covered as were other intra-abdominal of patients treated with meropenem and 72% of pa-
pathogens including E. faecalis. tients treated with ceftazidime plus tobramycin. In
Stratification of b-lactam agents into peri-operative this study, the differences in both clinical and bacteri-
and therapeutic regimens offers a rational basis for ological efficacy response rates were statistically sig-
selection of antimicrobials. Selected second-generation nificant in favour of meropenem.
cephalosporins remain highly effective agents for At the City Hospital Zehlendorf (Heckeshorn,
prophylaxis in gastrointestinal surgery, as well as for Berlin, Germany), indications for carbapenem usage
treatment of uncomplicated appendicitis in adults [28]. are based on the local resistance situation and recom-
Patients with more serious intra-abdominal infection mendations from the American Thoracic Society
may be selected for initial carbapenem therapy based on (1996) consensus conference [33]. Carbapenems are
a high severity of illness score (such as APACHE II used to treat intensive care patients who have acute
calculation), the anatomical site of gastrointestinal exacerbations of severe long-standing chronic bronchi-
perforation, known predictors for a poor outcome (such tis. In nosocomial pneumonia, carbapenems are first-
as hypoalbuminaemia and immunosuppression) or line agents in late-onset ventilator-associated
post-operative infection and sepsis. pneumonia and/or in severe early-onset hospital-
acquired pneumonia where problematic pathogens,
such as P. aeruginosa and Acinetobacter spp., are sus-
5. Serious lower respiratory tract infections pected. Of currently isolated Gram-negative bacilli
responsible for hospital-acquired pneumonia, only S.
Serious lower respiratory tract infections (LRTIs) maltophilia is frequently resistant to carbapenems, due
include exacerbations of long established chronic to the presence of a unique metallo-b-lactamase.
 J.S. Bradley et al. / International Journal of Antimicrobial Agents 11 (1999) 93–100 97

6. Paediatric meningitis Table 5

Bacterial pathogens isolated in febrile neutropenia
In the selection of appropriate therapy for paediatric Gram-positive Gram-negative
patients with meningitis, the microbiological and clini-
cal efficacy of the antibiotics are critical, as are their Coagulase-negative Escherichia coli
CNS safety profiles. staphylococci
To date, meropenem is the only carbapenem exten- Staphylococcus aureus Klebsiella spp.
Viridans streptococci Enterobacter spp.
sively evaluated in children. Worldwide, paediatric clin- Enterococcus spp. Pseudomonas aeruginosa
ical trials of meropenem have included over 1200 Bacillus spp. Stenotrophomonas maltophilia
children (Table 4) [34]. The results demonstrate the Corynebacterium spp. Acinetobacter spp.
efficacy of meropenem monotherapy in the treatment of
a variety of serious infections including meningitis, with
satisfactory responses being achieved in 90 – 100% of ensure effective levels of antibiotic reach the site of
children treated with meropenem. These data also infection by penetration across the meninges. Higher
showed meropenem was well tolerated. Meropenem is doses of meropenem are used in the treatment of
FDA-approved in the United States for use in infants meningitis, 2 g tds for adults, 40 mg/kg tds in children.
and children with meningitis and has demonstrated Imipenem/cilastatin has an upper limit of 4 g/day be-
good CNS tolerability at high doses. cause of its associated neurotoxicity and is not ap-
In a recent randomised, prospective study of over 200 proved for the treatment of meningitis. In an
children with meningitis, the clinical and microbiologi- uncontrolled trial of imipenem/cilastatin in children,
cal response rates of meropenem and cefotaxime in poor CNS tolerability was suggested [37].
children infected by Haemophilus influenzae, type b In summary, meropenem has a role in the initial
streptococci, Streptococcus pneumoniae, or Neisseria treatment of children with serious bacterial infections
meningitidis were comparable [35]. All strains of S. including meningitis. While cefotaxime and ceftriaxone
pneumoniae, including those documented to be non- are still considered to be the preferred therapeutic
susceptible to penicillin, were susceptible to agents for meningitis, in countries where the incidence
meropenem. Importantly, the incidence of seizures in of cefotaxime-resistance in S. pneumoniae is high (\
the meropenem-treated group was less than that seen in 5%), such as the USA, either addition of vancomycin to
the cefotaxime-treated group (11.6 and 19.2%, respec- cefotaxime or ceftriaxone, or monotherapy with
tively). This supports data from a previously published meropenem is recommended for initial therapy [38]. If
comparative trial in children which showed an inci- susceptibility data suggest that antibiotic therapy can
dence of seizures of 6/98 (6%) and 3/92 (3%), be effective with more narrow spectrum agents, treat-
respectively [36]. ment can be subsequently changed.
In meningitis, it is important to have high serum
concentrations of b-lactam antibiotics in order to
7. Febrile neutropenia

Table 4 Patients with persistent (] 14 days) and profound

Efficacy data from paediatric clinical trials of meropenem versus
comparatora
(5100 PMN/mm3) neutropenia often develop bacterial
infections [39]. In many cases, however, the patient’s
Indication Satisfactory response/number evaluable fever is unexplained since there is no focus of infection
(%) and no positive culture. It is vital nonetheless to effec-
tively treat febrile neutropenia from day one with
Meropenem Comparator
prompt administration of broad-spectrum, parenteral,
Meningitis 165/184 (90%) 162/176 (92%) high-dose antibiotic therapy, as infections can dissemi-
Lower respiratory tract 142/144 (99%) 118/123 (96%) nate rapidly due to impairment in normal host
Urinary tract 85/86 (99%) 43/44 (98%) defences.
Septicaemia 39/39 (100%) 26/27 (96%)
Options for the treatment of febrile neutropenia,
Skin/skin structure 47/48 (98%) 38/41 (93%)
Intra-abdominal 20/22 (91%) 12/13 (92%) following recommendations by the Infectious Disease
Society of America [40], include duotherapy without
a
Comparator was cefotaxime with/without additional agents, in- vancomycin (such as b-lactam plus aminoglycoside or
cluding metronidazole, amikacin, clindamycin, or tobramycin, except double b-lactam regimen), duotherapy of b-lactam plus
for meningitis when monotherapy was employed of either cefotaxime
vancomycin, or monotherapy with either extended
or ceftriaxone. Adapted from Diagnostic Microbiological Infectious
Diseases, 31, Bradley JS, Selecting therapy for serious infections in
spectrum cephalosporin or carbapenem.
children: maximizing safety and efficacy. Copyright (1998), with A wide spectrum of both Gram-positive and Gram-
permission from Elsevier Science. negative bacterial pathogens may be responsible for
98 J.S. Bradley et al. / International Journal of Antimicrobial Agents 11 (1999) 93–100

infection in febrile neutropenia (Table 5), including febrile neutropenia [50]. This method identifies patient
staphylococci, viridans streptococci, Klebsiella spp., E. groups with high-, moderate-, and low-risk infections,
coli, and P. aeruginosa [41]. Gram-positive organisms thereby enabling those not at high risk to have their
are isolated in : 70% of documented infections [42], anti-infective therapy ‘stepped-down’. Carbapenem
are generally less aggressive than Gram-negative or- monotherapy may provide the possibility of patients
ganisms, and most are associated with a low overall requiring parenteral therapy being treated at home.
mortality. However, treatment is complicated by the
fact that some Gram-positive isolates are resistant to
initial b-lactam empirical therapy. Furthermore, 8. Summary
a-haemolytic (or viridans) streptococci have become
more important as they sometimes cause an aggressive Carbapenems exhibit one of the broadest antibacte-
rial spectra of any currently available class of antimi-
syndrome, the so-called ‘toxic strep syndrome’, and
crobial agents. The stability of the carbapenems to
there is wide-spread penicillin resistance (15 –50%)
hydrolysis by ESBL and AmpC chromosomal b-lacta-
among these organisms.
mases establishes them as a viable therapeutic alterna-
Gram-negative infections can lead to substantial
tive to third-generation cephalosporins in the context of
morbidity, particularly if they are unrecognised or in- increasing b-lactamase-mediated resistance. Evidence
appropriately treated. Any delay in instituting potent, shows that ESBL-producing pathogens are causing in-
Gram-negative coverage in the febrile neutropenic creasing problems in ICUs; however, they remain sus-
patient can result in increased mortality [43]. In con- ceptible to carbapenems.
trast, there is more time available to make treatment In ICUs, carbapenems are especially valuable in units
modifications when dealing with Gram-positive infec- with known third-generation cephalosporin resistance
tions [44]. problems, in patients with infection who have received
Thus, although the approach to empirical coverage previous antibiotic courses, and in polymicrobial infec-
for Gram-positive and Gram-negative infections is dif- tions. Carbapenems are appropriate for use as first-line
ferent, potent Gram-negative coverage is a necessary empirical therapy in severe sepsis, post-operative peri-
part of the initial empirical regimen. Specific Gram-pos- tonitis and late onset ventilator-associated pneumonia.
itive agents, such as vancomycin, can be added later as Carbapenems also have a major role to play in the
and when appropriate. Limiting the use of vancomycin treatment of nosocomially acquired pneumonias, severe
is becoming of increased importance to prevent further community acquired pneumonias and in acute infective
dissemination of resistance to this antibiotic (the last episodes in patients with long-standing COPD. There
resort for some Gram-positive pathogens). is, also, a definite role for meropenem in the treatment
The emergence of resistance in Gram-negative organ- of paediatric meningitis. In patients with febrile neu-
isms due to ESBL production is of great concern in tropenia early empirical use of carbapenems is recom-
neutropenic patients. As carbapenems have a broad mended as using them as reserve agents may result in
antimicrobial spectrum and stability to most commonly effective treatment being given too late, resulting in
encountered b-lactamases, they are suitable candidates increased morbidity.
for initial empiric monotherapy. Furthermore, agents There are potential advantages of carbapenem
with a relatively low potential for induction of seizures monotherapy over multiple drug therapy, such as re-
duced nursing time and possibly reduced side-effects.
(important in cancer patients with an underlying CNS
Also, with meropenem rapid bolus administration is
disorder, such as metastatic disease to the brain) and a
possible (but not with imipenem/cilastatin due to solu-
low incidence of nausea and vomiting (important to
bility problems) which makes home administration a
patients receiving chemotherapy) are preferred. Both
viable option and may permit small volume administra-
carbapenems have been studied in this patient group. tion (this can be beneficial in small children or patients
Imipenem/cilastatin has proven efficacy although evi- with a fluid overload).
dence suggests seizures at 3 – 4 g may be a problem [45] Carbapenem agents, thus, have a definite role as
and some studies have reported incidences of nausea initial empirical therapy as well as definitive therapy in
and vomiting irrespective of dose used [45,46]. a range of serious infections. Treatment should be
Meropenem has been shown to be effective [47–49] individualised based on patient factors, known/sus-
with no seizures and low incidences of nausea and pected pathogen(s) and known resistance problems in
vomiting reported in these studies. Thus, saving car- the treatment unit. While b-lactamase-mediated resis-
bapenems for use as ‘last resort’ drugs will lessen their tance to the carbapenems may be less likely to emerge
overall impact in this special patient population. than with other b-lactams, the carbapenems should
Risk-based therapy may become an important new always be used appropriately so that their superior
way of providing appropriate care for patients with breadth of antimicrobial efficacy is maintained.
 J.S. Bradley et al. / International Journal of Antimicrobial Agents 11 (1999) 93–100 99

Acknowledgements [17] Hopkins JA, Lee JCH, Wilson SE. Susceptibility of intra-ab-
dominal isolates at operation: a predictor of postoperative infec-
tion. Am Surg 1993;59:791 – 6.
This review is based on presentations from an official [18] Christou NV, Turgeon P, Wassef R, Rotstein O, Bohnen J,
symposium held at the Second European Congress of Potvin M, The Canadian Intra-abdominal Infection Study
Chemotherapy (Hamburg, 1998) which was sponsored Group. Management of intra-abdominal infections. The case for
by Zeneca Pharmaceuticals. intraoperative cultures and comprehensive broad-spectrum an-
tibiotic coverage. Arch Surg 1996;131:1193– 1201.
[19] Huizinga WK, Warren BL, Baker LW et al. Antibiotic
monotherapy with meropenem in the surgical management of
References intra-abdominal infections. J Antimicrob Chemother 1995;36
Suppl A:179 – 189.
[1] Edwards JR. Meropenem: a microbiological overview. J Antimi- [20] Condon RE, Walker AP, Sirinek KR, et al. Meropenem versus
crob Chemother 1995;36 Suppl A:1–17. tobramycin plus clindamycin for treatment of intraabdominal
[2] Livermore DM, Yuan M. Antibiotic resistance and production infections: results of a prospective, randomized, double-blind
of extended-spectrum b-lactamases amongst Klebsiella spp. from clinical trial. Clin Infect Dis 1995;21:544 – 50.
intensive care units in Europe. J Antimicrob Chemother [21] Berne TV, Yellin AE, Appleman MD, Heseltine PN, Gill MA.
Meropenem versus tobramycin with clindamycin in the antibiotic
1996;38:409 – 24.
management of patients with advanced appendicitis. J Am Coll
[3] Jarlier V, Fosse T, Philippon A. Antibiotic susceptibility in
Surg 1996;182:403 – 7.
aerobic gram-negative bacilli isolated in intensive care units in 39
[22] Chang DC, Wilson SE. Meta-analysis of the clinical outcome of
French teaching hospitals (ICU study). Intensive Care Med
carbapenem monotherapy in the adjunctive treatment of intra-
1996;22:1057 – 65.
abdominal infections. Am J Surg 1997;174:284 – 90.
[4] Garau J, Blanquer J, Cobo L, et al. Prospective, randomised,
[23] Kanellakopoulou K, Giamarellou H, Papadothomakos et al.
multicentre study of meropenem versus imipenem/cilastatin as
Meropenem versus imipenem/cilastatin in the treatment of in-
empiric monotherapy in severe nosocomial infections. Eur J Clin
traabdominal infections requiring surgery. Eur J Clin Microbiol
Microbiol Infect Dis 1997;16:789–96.
Infect Dis 1993;12:449 – 453.
[5] Buirma RJ, Horrevorts AM, Wagenvoort JH. Incidence of
[24] Brismar B, Malmborg AS, Tunevall G, et al. Meropenem versus
multi-resistant gram-negative isolates in eight Dutch hospitals:
imipenem/cilastatin in the treatment of intra-abdominal infec-
The 1990 Dutch Surveillance Study. Scand J Infect Dis Suppl
tions. J Antimicrob Chemother 1995;35:139 – 48.
1991;78:35 – 44.
[25] Geroulanos SJ. Meropenem versus imipenem/cilastatin in intra-
[6] Verbist L. Incidence of multi-resistance in gram-negative bacte-
abdominal infections requiring surgery: Meropenem Study
rial isolates from intensive care units in Belgium: a surveillance
Group. J Antimicrob Chemother 1995;36:191 – 205.
study. Scand J Infect Dis Suppl 1991;78:45–53.
[26] Hemsell DL, Martens MG, Faro S, Gall S, McGregor JA. A
[7] Snydman DR. Clinical implications of multi-drug resistance in multicenter study comparing intravenous meropenem with clin-
the intensive care unit. Scand J Infect Dis Suppl 1991;78:54 – 63. damycin plus gentamicin for the treatment of acute gynecologic
[8] Lopez-Yeste M, Xercavins M, Lite J, Cuchi E, Garau J. Fluo- and obstetric pelvic infections in hospitalized women. Clin Infect
roquinolone and aminoglycoside resistance in chromosomal Dis 1997;24 Suppl 2:S222 – S230.
cephalosporinase-overproducing Gram-negative bacilli strains [27] Nichols RL, Smith JW, Geckler RW, Wilson SE. Meropenem
with inducible b-lactamase. Enferm Infec Microbiol Clin versus imipenem/cilastatin in the treatment of hospitalized
1996;14:211 – 4. patients with skin and soft tissue infections. South Med J
[9] Livermore DM. b-lactamase-mediated resistance and opportuni- 1995;88:397 – 404.
ties for its control. J Antimicrob Chemother 1998;41 Suppl [28] Hopkins JA, Wilson SE, Bobey DG. Adjunctive antimicrobial
D:25 – 41. therapy for complicated appendicitis:bacterial overkill by combi-
[10] Quinn JP, Studemeister AE, DiVincenzo CA, Lerner SA. Resis- nation therapy. World J Surg 1994;18:933 – 8.
tance to imipenem in Pseudomonas aeruginosa: clinical experi- [29] Eller J, Ede A, Schaberg T, Niederman MS, Mauch H, Lode H.
ence and biochemical mechanisms. Rev Infect Dis Infective exacerbations of chronic bronchitis - relation between
1988;10:892 – 8. bacteriologic etiology and lung function. Chest 1998;113:1542–8.
[11] Colardyn F, Faulkner KL, The meropenem serious infection [30] Finch RG, Pemberton K, Gildon KM. Pneumonia: the impact
study group: intravenous meropenem versus imipenem/cilastatin of risk factors on the outcome of treatment with meropenem and
in the treatment of serious bacterial infections in hospitalised ceftazidime. J Chemother 1998;10:35 – 46.
patients. J Antimicrob Chemother 1996;38:523–537. [31] Hamacher J, Vogell F, Lichey J, et al. Treatment of acute
[12] Alvarez-Lerma F. Modification of empiric antibiotic treatment bacterial exacerbations of chronic obstructive pulmonary disease
in patients with pneumonia acquired in the intensive care unit: in hospitalised patients – a comparison of meropenem and
ICU-Acquired Pneumonia Study Group. Intensive Care Med imipenem/cilastatin: COPD Study Group. J Antimicrob
1996;22:387 – 94. Chemother 1995;36 (Suppl A):121 – 133.
[13] Chow JW, Fine MJ, Shlaes DM, et al. Enterobacter bacteremia: [32] Sieger B, Berman SJ, Geckler RW, Farkas SA. Empiric treat-
clinical features and emergence of antibiotic resistance during ment of hospital-acquired lower respiratory tract infections with
therapy. Ann Intern Med 1991;115:585–90. meropenem or ceftazidime with tobramycin:a randomized study:
[14] Pacelli F, Doglietto GB, Alfieri S, et al. Prognosis in intra-ab- Meropenem Lower Respiratory Infection Group. Crit Care Med
dominal infections: multivariate analysis on 604 patients. Arch 1997;25:1663 – 70.
Surg 1996;131:641 – 5. [33] American Thoracic Society. Hospital-acquired pneumonia in
[15] Wilson SE, Faulkner K. Impact of anatomical site on bacterio- adults:diagnosis, assessment of severity, initial antimicrobial
logical and clinical outcome in the management of intra-abdom- therapy, and preventive strategies. Am J Respir Crit Care
inal infections. Am Surg 1998;64:402–7. 1996;153:1711– 1725.
[16] Bohnen JM, Mustard RA, Oxholm SE, Schouten BD. APACHE [34] Bradley JS. Selecting therapy for serious infections in children:
II score and abdominal sepsis: a prospective study. Arch Surg maximizing safety and efficacy. Diagn Microbiol Infect Dis
1988;123:225 – 9. 1998;31:405 – 10.
100 J.S. Bradley et al. / International Journal of Antimicrobial Agents 11 (1999) 93–100

[35] Rodriguez W, Bradley JS, Odio C. Meropenem vs cefotaxime in [43] Bodey GP, Jadeja L, Elting L. Pseudomonas bacteremia: retro-
the treatment of pediatric meningitis. In: Program and Abstracts spective analysis of 410 episodes. Arch Intern Med
of the 37th Interscience Conference on Antimicrobial Agents and 1985;145:1621– 9.
Chemotherapy (Toronto, Canada) 1997. Washington DC: Amer- [44] Rubin M, Hathorn JW, Marshall D, Gress J, Steinberg SM,
ican Society for Microbiology (abstract LM-002). Pizzo PA. Gram-positive infections and the use of vancomycin in
[36] Klugman KP, Dagan R, The Meropenem Meningitis Study 550 episodes of fever and neutropenia. Ann Intern Med
Group. Randomized comparison of meropenem with cefotaxime 1988;108:30 – 5.
for treatment of bacterial meningitis. Antimicrob Agents [45] Winston DJ, Ho WG, Bruckner DA, Champlin E. Beta-lactam
Chemother 1995;39:1140–1146. antibiotic therapy in febrile granulocytopenic patients. Ann In-
[37] Wong VK, Wright HT Jr, Ross LA, et al. Imipenem/cilastatin tern Med 1991;115:849 – 59.
treatment of bacterial meningitis in children. Pediatr Infect Dis J [46] Freifeld AG, Walsh T, Marshall D, et al. Monotherapy for fever
1991;10:122 – 5.
and neutropenia in cancer patients: a randomized comparison of
[38] Bradley JS, Scheld WM. The challenge of penicillin-resistant
ceftazidime versus imipenem. J Clin Oncol 1995;13:165–76.
Streptococcus pneumoniae meningitis: current antibiotic therapy
[47] Cometta A, Calandra T, Gaya H, et al. Monotherapy with
in the 1990s. Clin Infect Dis 1997;24 Suppl 2:S213-S221.
meropenem versus combination therapy with ceftazidime plus
[39] Bodey GP, Buckley M, Sathe YS, Freireich EJ. Quantitative
amikacin as empiric therapy for fever in granulocytopenic
relationships between circulating leukocytes and infection in
patients with cancer. Antimicrob Agents Chemother
patients with acute leukemia. Ann Intern Med 1966;64:328 – 40.
[40] Hughes WT, Armstrong D, Bodey GP, et al. 1997 guidelines for 1996;40:1108 – 15.
the use of antimicrobial agents in neutropenic patients with [48] Meropenem Study Group of Leuven, London and Nijmegen.
unexplained fever. Infectious Diseases Society of America. Clin Equivalent efficacies of meropenem and ceftazidime as empirical
Infect Dis 1997;25:551–573. monotherapy of febrile neutropenic patients. J Antimicrob
[41] Rolston KVI, Bodey GP. Infections in patients with cancer. In: Chemother 1995;36:185 – 200.
Holland JF, Frei E III , Bast RC Jr, Kufe DW, Morton DL, [49] Mason B, Gildon KM, Bell JM, Megson GM. Meropenem in
Weichfelbaum RR, ed. Cancer Medicine, 3rd ed. Baltimore: the empirical management of febrile neutropenia. Abstract no.
Williams and Wilkins, 1997:2416–2441. 40 presented at the Third International Symposium on Febrile
[42] Cometta A, Glauser MP. Empiric antibiotic monotherapy with Neutropenia, December 10 – 13 1997.
carbapenems in febrile neutropenia: a review. J Chemother [50] Rolston KV. Expanding the options for risk-based therapy in
1996;8:375 – 81. febrile neutropenia. Diagn Microbiol Infect Dis 1998;31:411–6.