JAC Advance Access originally published online on September 22, 2006
Journal of Antimicrobial Chemotherapy 2006 58(5):916-929; doi:10.1093/jac/dkl354
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Review |
Two decades of imipenem therapy
1 Institut for Medical Microbiology and Epidemiology of Infectious Diseases, University of Leipzig D-04103 Leipzig, Germany 2 R M Alden Research Laboratory Santa Monica, CA 90404, USA 3 Institut Clínic de Pneumologia i Cirurgia Toràcica, Hospital Clínic de Barcelona Spain
*Corresponding author. Tel: +49-341-97-15-200; Fax +49-341-97-15-209; E-mail: acr{at}medizin.uni-leipzig.de
 |
Abstract |
|---|
Abstract
IntroductionImipenem, the first carbapenem discovered, was developed more than two decades ago in response to an unmet need for a highly potent, broad-spectrum antimicrobial agent with a strong safety profile. It has since been used to treat more than 26 million patients. In an era where antibiotic use has driven antibiotic resistance, choosing appropriate initial therapy for serious infection is critical. Appropriate antibiotic regimens must cover all likely pathogens, be administered promptly at the correct dosage and dosing interval, be well tolerated and prevent the emergence of resistance. While imipenem was initially reserved for use in intractable, serious infections, the benefits of early aggressive therapy are now known, making imipenem a core agent in de-escalation therapy due to proven efficacy and safety for indications such as nosocomial pneumonia, intra-abdominal infection, sepsis and febrile neutropenia. De-escalation therapy with an agent such as imipenem minimizes resistance development by initiating aggressive initial treatment and then tailoring therapy based on patient response and culture results, switching to a less expensive, narrower spectrum antibiotic regimen or shortening the duration of therapy. Imipenem has maintained sustained clinical efficacy, tolerability and in vitro activity against important bacterial pathogens for two decades. We review the factors that continue to make imipenem as appropriate an agent for de-escalation therapy now as it was 20 years ago.
Keywords: antibiotics , appropriate therapy , carbapenems , de-escalation , infections , pneumonia
|
Introduction |
|---|
ß-Lactams were among the first antimicrobial agents available for the therapy of infectious diseases. Over time, however, problems such as resistance development and selection of resistant organisms have become apparent. The medical need for compounds with broad-spectrum activity, rapid bactericidal action, limited resistance-promoting properties and good tolerability has been met with carbapenem compounds. Imipenem was the first carbapenem antibiotic selected for development more than two decades ago because it was a highly potent, broad-spectrum antimicrobial agent with a good safety profile.1,2 Since that time more than 26 million patients have been treated with imipenem (data on file, Merck & Co., Inc.), and imipenem continues to play an important role in the empirical and as well as the targeted treatment of severe and difficult to treat infections. This review will summarize important features of imipenem, discuss the accumulated treatment experience that has been established over 20 years of therapy with imipenem, and discuss its place in targeted and de-escalation therapy.
|
Chemical properties |
|---|
Imipenem (N-formimidoyl-thienamycin) is an amidine derivative of thienamycin that is 5–10 times more stable than the mother compound (Figure 1). Deliberate substitution of a methyl moiety in place of a sulphur was introduced to increase bactericidal activity and ß-lactamase stability in the hydroxyethyl side chain. Imipenem is rapidly degraded by kidney dehydropeptidase-1, thus it was combined with cilastatin, an inhibitor of this enzyme. Cilastatin not only prevents the degradation of imipenem but also protects the kidneys against potential toxic effects exerted by higher doses of imipenem. Imipenem and cilastatin are combined in a 1:1 ratio. Because cilastatin has no antibacterial activity of its own, only the amount of imipenem is given for dosing purposes.
|
|
Mechanism of action |
|---|
Like all other ß-lactams, imipenem inhibits bacterial cell wall synthesis by binding to and inactivating relevant transpeptidases, known as penicillin binding proteins (PBPs). In Escherichia coli, imipenem inhibits the transpeptidase activities of PBPs-1A, -1B and -2, and the D-alanine carboxypeptidase activities of PBP-4 and PBP-5. It also causes strong inhibition of the transglycosylase activity of PBP-1A while it inhibits the transpeptidase activity of PBP-3 only weakly, which is consistent with the finding that it has low binding affinity to PBP-3 and does not inhibit septum formation by the cells.3 This is in contrast to all other ß-lactams, including other carbapenems that preferentially bind to PBP-1 and PBP-3. Consequently, imipenem induces sphere formation with subsequent cell rupture but not the filamentous growth of bacteria observed for other ß-lactams. Therefore, imipenem therapy reduces the amount of lipopolysaccharide liberated during bacteriolysis.4 This effect was shown to be clinically relevant in a study with patients suffering from Gram-negative urosepsis.5 When compared with ceftazidime treatment, patients who received imipenem showed a faster defervescence, lower endotoxin levels and a tendency for faster normalization of cytokine levels.
The mode of action of imipenem allows for activity against Gram-positive and Gram-negative bacteria, cocci and bacilli, aerobes and anaerobes. Imipenem may have activity against Mycobacteria spp., but Mycoplasma, Chlamydia, Legionella, Stenotrophomonas, Burkholderia, Clostridium difficile and methicillin-resistant Staphylococcus aureus (MRSA) are not within its antimicrobial spectrum.
|
Mechanisms of resistance |
|---|
Porins and efflux pumps
In order to reach their targets, carbapenems normally cross through protein channels in the outer membranes of Gram-negative bacteria called porins.6 Porin OprD deficiencies can precipitate resistance to carbapenems. The OprD mutation, in conjunction with AmpC production, is responsible for imipenem resistance in Pseudomonas aeruginosa.6,7 Porin defects in Klebsiella are also associated with carbapenem resistance.8,9 A recent study found ertapenem selected for OprD mutants of P. aeruginosa, albeit rarely.10
Efflux pumps are proteins that remove certain molecules from the bacterial cell (Figure 2). They play an integral role in the intrinsic and acquired resistance of P. aeruginosa to antibiotics.11,12 Imipenem is not a substrate of such efflux pumps, although meropenem is.12–14 The most prevalent efflux system, MexAB-OprM, has been found to cause resistance to meropenem, but not imipenem, in vitro.6,12–15 Overexpression of this system may emerge during antibiotic therapy and may result in treatment failure.13,16 The MexAB-OprM system induces parallel-resistance to other antimicrobials such as fluoroquinolones, penicillins, cephalosporins, macrolides and sulphonamides.6,13,14,17 This means that up-regulated efflux can quickly render fluoroquinolones and the majority of ß-lactams ineffective against Pseudomonas, leaving the activity of imipenem and aminoglycosides unaffected.
|
Extended-spectrum ß-lactamases
The most prevalent resistance mechanism against ß-lactams is the production of ß-lactamases. While many ß-lactamases are not capable of hydrolysing cephalosporins, the so called extended-spectrum ß-lactamases (ESBLs) have this ability with varying specificity for individual cephalosporins.18 Commonly found in Klebsiella pneumoniae, E. coli and other Enterobacteriaceae, ESBLs are plasmid-mediated and thus easily transferable. Some of the in vitro resistance to piperacillin/tazobactam among enteric pathogens may also be explained by the production of ESBL.19 Carbapenems are not affected by ESBLs and are the agents of choice in these cases.18,20–22 Chromosomally encoded type-1 ß-lactamases such as AmpC are produced by Enterobacter cloacae, for example. Cephalosporins may be hydrolysed by AmpC while carbapenems retain activity.23–25 Piperacillin/tazobactam appears to have limited in vitro activity against Enterobacter AmpC ß-lactamases.26
Carbapenemases
Among the ß-lactamases, metalloenzymes are distinguished by having a zinc ion required for enzymic activity.27 Sometimes referred to as carbapenemases, they are able to hydrolyse many ß-lactam antibiotics including carbapenems. Two types, IMP and VIM, have been identified, with IMP primarily isolated from the important pathogenic anaerobe Bacteroides fragilis28–31 and VIM from isolates of Pseudomonas and Enterobacteriaceae.32,33 The cfiA gene appears to be responsible for metallo-ß-lactamase production in imipenem-resistant strains, and its presence is a better marker than metalloenzyme activity, which may not be expressed when the gene is silent.30,34,35 Data do not point to plasmid-mediated transfer or spontaneous mutations causing resistance acquisition, although parallel-resistance arising from use of other classes of antibiotics such as quinolones may be an issue.30,36,37 Although few clinically relevant Bacteroides have been reported that either produce metalloenzymes or contain the cfiA gene, this is clearly an area of research activity affecting all ß-lactam antibiotics.29,34 Target modifications such as those caused by and encoded for by the mecA gene in MRSA confer resistance to all ß-lactams including the carbapenems.38,39 Otherwise imipenem remains active against staphylococci.
The class A carbapenemases (e.g. KPC-1, KPC-2, KPC-3) have also proven to be clinically important. In a study in 10 hospitals in Brooklyn, NY, all 96 Klebsiella with KPC carbapenemases were carbapenem-resistant, with only a few of them retaining susceptibility to cephalosporins or fluoroquinolones.40 Up to this time, strains producing KPC have primarily been reported in facilities in the United States.41–43
|
Resistance and infection control |
|---|
It has been argued that the broad use of carbapenems results in rapid resistance development particularly in P. aeruginosa and Acinetobacter spp. and thus should be avoided.44,45 However, these reports often have limitations such as lack of appropriate usage data (i.e. DDD per patient days), neglect of dosing issues, and lack of distinction between resistance development under therapy and spread of a resistant clone. Resistance due to clonal epidemiology is a public health issue that is best addressed by infection control measures. For example, an in vitro 1999 survey of 1599 clinical isolates of A. baumannii from 15 hospitals in Brooklyn, NY, USA found that more than 50% of these isolates were resistant to the carbapenems imipenem and meropenem.46 A single strain, Acinetobacter Type 1, accounted for 62% of all resistant isolates, and this strain was found in all 15 hospitals. Four strains accounted for 97% of all Acinetobacter isolates. It was concluded that the rate of carbapenem resistance correlated with the use of cephalosporins and aztreonam, suggesting the potential for parallel-resistance. The fact that a single clone accounted for the majority of a total of 419 unique patient isolates points to the rampant spread of that clone among the hospitals in the study and the need for infection control rather than antibiotic restriction.
An outbreak of KPC class A carbapenemase-positive Klebsiella was reported in Brooklyn, NY.40 All 96 isolates were carbapenem-resistant, and most were resistant to cephalosporins and fluoroquinolones. A citywide surveillance study reported rapid spread of this type of carbapenem resistance and parallel-resistance with other antibiotics, pointing to an immediate need for enhanced infection control measures.47
Finally, it has been argued that imipenem selects Stenotrophomonas maltophilia.48 Other studies and our own experience do not support this notion.49,50
|
Susceptibility surveys |
|---|
The potential for resistance has become as important a therapeutic consideration as efficacy or tolerability. It is a local problem with global implications. While resistance patterns must be determined at an institutional level to devise effective treatment strategies,51 there is a consistent trend towards increasing resistance and novel mechanisms of resistance. Worldwide surveys are helpful in assessing general aspects of resistance development and in anticipating future medical needs. Generally speaking, phenotypic tests that determine MIC values are being used for this purpose. In order to compare different studies and the impact of methodology it is necessary to see the original MIC distributions, because wild-type strains of individual species should cluster at specific MIC values regardless of the source of the isolate. Furthermore, breakpoints are set differently in various parts of the world, so that the same MIC could result in different S-I-R classifications. As an example of these differences Table 1 shows the current breakpoints for carbapenems as put forward by the CLSI and the EUCAST. Moreover, breakpoints may change over time as more information becomes available. It may then be difficult to judge trends in resistance development. Thus susceptibility data should be given, e.g. as shown in Figure 3 (a and b).
|
|
A number of worldwide studies have analysed the phenotypic susceptibility of imipenem and other antimicrobial agents over time. Of note is the SENTRY survey,52 the Nosocomial Prevalence and Resistance Survey (NPRS, Merck & Co., Inc.; Table 2) and the Meropenem Yearly Susceptibility Test Information Collection (MYSTIC) study (Table 3 and Figure 4).53
|
|
|
The SENTRY surveillance programme found ESBL-producing K. pneumoniae to be a major problem in Latin America (45.4% resistant), the Western Pacific (24.6%) and Europe (22.6%), with the vast majority of these isolates resistant to ceftazidime, ceftriaxone and aztreonam, as is characteristic of most ESBLs.54 Worldwide, 99–100% of these Klebsiella isolates were susceptible to imipenem, although susceptibility to other broad-spectrum agents was highly variable. For example, K. pneumoniae were susceptible to cefepime in Canada (94.4%) and the US (87.6%), but much less so in Europe (63.6%) and Latin America (49.6%). The situation is much the same with Acinetobacter and Pseudomonas. While 88.6–95.5% of Acinetobacter isolates from North America and Latin America were susceptible to imipenem, only 67–70% in North America and fewer than 30% in Latin America were susceptible to piperacillin/tazobactam, ceftazidime and ciprofloxacin.55 Between 1997 and 1999, susceptibility of P. aeruginosa to imipenem went from 90% to 86.2% in the Asia-Pacific region, from 83% to 92% in Canada, from 77% to 74.3% in Latin America, from 88% to 80.9% in the US and from 89.3% to 71.6% in Europe.55
NPRS has supported more than 500 susceptibility studies in 
50 countries. Similar to worldwide SENTRY,52 NPRS data show increasing worldwide cephalosporin resistance among Enterobacter, Klebsiella and E. coli, and sustained susceptibility to imipenem has been observed (Table 2).
MYSTIC data reported that 99.6% of all Enterobacteriaceae isolated between 1997 and 2003 from 130 centres in Europe, North America and Latin America remained susceptible to imipenem, compared with 85–86% that remained susceptible to ceftazidime or piperacillin/tazobactam (Figure 4).56 These susceptibility patterns have been noted worldwide (Table 3).57
|
Dosing, pharmacokinetics and pharmacodynamics |
|---|
Dosing
Imipenem dosing is well established in adults, children, patients with impaired renal function and geriatric patients. It should be noted that the often used 4x 500 mg and 3x 1 g dosing result in approximately the same time above the MIC (t > MIC).58 Patients with creatinine clearance of 
70 mL/min/1.73 m2 and/or body weight <70 kg require dosage reductions. For paediatric patients 
3 months of age, the recommended dose for non-CNS infections is 15–25 mg/kg/dose every 6 h.
Pharmacokinetics and pharmacodynamics
The pharmacokinetics and pharmacodynamics of imipenem are well established and will be reviewed only briefly. Intravenous infusion of imipenem over 20 min results in peak plasma levels of imipenem ranging from 14 to 24 mg/L (250 mg dose), 21 to 58 mg/L (500 mg dose) and 41 to 83 mg/L (1000 mg dose). At these doses, plasma levels of imipenem decline to <1 mg/L after 4–6 h. The plasma half-life is 1 h. The median concentration of imipenem 1 h after dosing is 5.6 mg/kg in lung tissue, 11.1 mg/kg in endometrial tissue, 22 mg/L in pleural fluid, 2.6 mg/L in cerebrospinal fluid (2 h post-dose) and 16.4 mg/L in interstitial fluid.
Co-administration with probenecid extends the half-life and increases the serum concentration. Approximately 10–20% of imipenem binds to human serum proteins. Imipenem is excreted renally, with 70% of imipenem recovered in the urine within 10 h and no detectable urinary excretion after that time. Accumulation is not observed in plasma or urine even with regimens administered as frequently as every 6 h in patients. Imipenem is distributed extensively in tissues and fluids.59,60
Carbapenem efficacy depends on the dosing interval duration during which free drug concentration exceeds the MIC. t > MIC is the best pharmacodynamic predictor of carbapenem efficacy, with optimal cell kill achieved when 40% of the dosing interval has drug concentrations higher than the MIC.58,61 A recent study compared the pharmacokinetic profiles of imipenem (1g, intravenous, 30min infusion) and meropenem (1g, intravenous, 30min infusion) in 20patients who were critically ill with sepsis.62 Peak serum concentration (Cmax) was significantly higher with imipenem than meropenem (90.1 ± 50.9 versus 46.6 ± 14.6 mg/L, P < 0.01) as was the area under the curve (216.5 ± 86.3 versus 99.5 ± 23.9 mg · h/L, P < 0.01). The mean volume of distribution and mean total plasma clearance were significantly higher with meropenem than imipenem (25 ± 4.1 versus 17.4 ± 4.5 L, P < 0.01 and 191 ± 52.2 versus 116.4 ± 42.3 mL/min, P < 0.01). The study estimated that imipenem would maintain t > MIC for 8 h after a single infusion, whereas meropenem would only maintain t > MIC over this 8 h period for pathogens with an MIC 2 mg/L. The authors concluded that the two carbapenems were not dose equivalent in patients with sepsis, with more research needed to clarify whether the more favourable pharmacokinetic profile of imipenem balances any greater in vitro potency of meropenem against Gram-negative pathogens.
A recent study simulated target attainment rates of imipenem and meropenem using the same dose of each drug (250, 500 and 1000 mg every 6 h and every 8 h) and 30 min infusions.63 Imipenem achieved the pharmacodynamic goal more often than meropenem (58.3–99.2% for imipenem versus 46.9–99% for meropenem), although no statistically significant differences were noted. Studies have used 3 h infusions to optimize meropenem efficacy by increasing the t > MIC, while keeping the drug at room temperature for <4 h because drug potency decreases after this point.64,65 A study examining a 3 h infusion of imipenem and meropenem at a variety of doses found no significant differences in ability to achieve the pharmacodynamic goal between imipenem and meropenem.66
Although these models suggest that imipenem and meropenem should have equal efficacy in a 3 h infusion, they may explain the higher dosing necessary for meropenem. If a 3 h infusion is necessary for meropenem to achieve the same efficacy as is possible with imipenem in a 1 h infusion, this would need to be taken into consideration. In other words, in order to be effective, it may be necessary to provide a higher dose of meropenem, or a longer infusion time. This was outlined in a literature review that suggested that administering meropenem 500 mg every 8 h infused over 30 min would have a much lower probability of attaining 40% t > MIC against P. aeruginosa (72.5%) than would 500 mg every 8 h infused over 3 h (87.9%) or 1000 mg every 8 h infused over 30 min (93.4%).67 The pharmacokinetics of meropenem may need to be optimized to achieve the critical t > MIC.61,67
|
Indications and clinical studies |
|---|
Imipenem is active against a broad spectrum of pathogens, making it particularly useful in the treatment of serious polymicrobial and mixed aerobic/anaerobic infections, as well as for initial empirical treatment. Guideline documents developed worldwide for serious infection recommend imipenem aseffective initial targeted or empirical therapy in ventilator-associated pneumonia (VAP) hospital- and healthcare-acquired pneumonia (HAP and HCAP), intra-abdominal infection and febrile neutropenia. Imipenem is indicated for hospitalized patients with intra-abdominal, lower respiratory tract, gynaecological and genitourinary tract and skin and soft tissue infections, as well as for those with sepsis or endocarditis.68,69 Extensive study has shown imipenem to be effective in these disease areas as well as in nosocomial and ventilator-associated pneumonia, febrile neutropenia and for the empirical treatment of serious infection.70–84 Imipenem is also considered to be appropriate empirical therapy for serious infection when there is a high likelihood of infection with resistant organisms or multiple organisms that might otherwise require multidrug regimens.85 In addition, using imipenem in preference to other antibiotics during the 5 day period after onset of bacteraemia due to an ESBL-producing organism was independently associated with lower mortality in one study.86 Imipenem has become a standard comparator in clinical studies for several indications.76,87,88
|
Efficacy in special situations |
|---|
Mycobacterium spp.
There are multiple reports of imipenem efficacy against Mycobacterium spp. Three patients with M. tuberculosis infections recalcitrant to most standard antituberculosis agents had their infections cleared with imipenem and amikacin, with no recurrence in 12 months of follow-up.89 However, an MIC of >100 mg/L for imipenem against M. tuberculosis strain H37Rv was described.90
A review of non-tuberculous mycobacteria listed imipenem as a parenteral drug of choice in the treatment of infections due to Mycobacterium fortuitum, Mycobacterium chelonae, Mycobacterium smegmatis, M. chelonae-like organism, Mycobacterium peregrinum, and sorbitol-positive and -negative biovariants of M. fortuitum.91 Case reports have been published on the use of imipenem in the treatment of various infections due to M. chelonae, M. fortuitum and Mycobacterium abscessus.92–98 While imipenem-containing regimens were successful in some of these cases, several involving M. chelonae showed little or no improvement during treatment that included imipenem. This may be due to variations in the in vitro activity of imipenem against different Mycobacteria isolates. One study reported that imipenem had good activity against three M. fortuitum biovariants but only slight activity against two M. chelonae subspecies.99 A mouse model of disseminated Mycobacterium avium complex infections reported that ciprofloxacin in combination with amikacin and imipenem reduced both bacteraemia and mortality.100
Transplant patients
Enterobacteriaceae, including E. coli, Klebsiella spp. and Enterobacter spp., are common causes of intra-abdominal, respiratory tract and bloodstream infections in transplant patients. These organisms are in many parts of the world commonly cephalosporin-resistant, due to ESBLs and inducible group 1 cephalosporinases. Because of this, third-generation cephalosporins should generally be avoided for treatment of serious infections in transplant recipients.101
Nocardia infections, although infrequent among heart, lung or heart–lung transplant recipients, are of concern. In a retrospective review of 540 transplant patients, 10 were found to have Nocardia infection; all isolates were susceptible to imipenem, trimethoprim/sulfamethoxazole and amikacin.102 A single-centre study found that 3 of 233 heart-transplant recipients had Nocardia infections.103 The authors suggested that a ß-lactam/ß-lactamase inhibitor with ciprofloxacin or amikacin followed by a short course of trimethoprim/sulfamethoxazole could be effective. A retrospective survey of nocardiosis in 9 hospitals in Italy found 30 patients with documented nocardiosis from 1982 and 1992.104 Most strains tested were susceptible to imipenem and amikacin. In a rare case of post-cardiac transplantation mediastinal infection due to Nocardia, surgical debridement, dressing, sugaring and imipenem with ciprofloxacin were used for 4 weeks followed by oral ciprofloxacin for 1 year. Treatment was successful and well tolerated.105 In addition, multiple case reports have described successful treatment of HIV-infected patients with concurrent pulmonary or cerebral Nocardia infection.106–111
Febrile neutropenia
The treatment of infectious complications in cancer patients has evolved as a consequence of the developments in the chemotherapy of cancer patients. An early non-comparative study of imipenem in 71 cancer patients with 79 febrile episodes reported a response rate of 67%.112 A prospective, open label, randomized, multicentre study comparing cefepime and imipenem/cilastatin in patients with haematological malignancies and febrile neutropenia (n = 180) reported monotherapy success rates of 40% and 51% in the cefepime and imipenem groups, respectively (P = NS). The 4 week overall mortality rate was 5%.87 Another study comparing imipenem and cefepime in the treatment of patients with cancer and febrile neutropenia found a 68% response rate to the imipenem regimen, compared with a 75% response rate to the cefepime regimen (P = 0.2).77 Imipenem and ceftazidime plus tobramycin were found to have similar efficacy in cancer patients with febrile neutropenia, with successful outcomes in 78% (35 of 45) of the imipenem group patients and 71% (29 of 41) of the ceftazidime group patients.113 In a prospective, randomized study, imipenem and sulbactam/cefoperazone plus amikacin were compared for empirical therapy of febrile neutropenia in 30 evaluable episodes in cancer patients. The clinical response rate was 60% (P > 0.05) for both regimens.114 A prospective study of 83 febrile neutropenic cancer patients randomized patients to imipenem or piperacillin plus amikacin. The overall response rate for clinically or microbiologically documented infections was 90% with imipenem versus 76% with the piperacillin regimen (P = NS).115 Sixty-six patients undergoing bone marrow transplant were randomized to receive netilmicin plus either imipenem or ceftazidime as empirical antimicrobial therapy for febrile neutropenia.116 Positive clinical response was observed in 80% of receiving the imipenem regimen and 73% of those receiving the ceftazidime regimen.
Paediatric patients
The use of imipenem is well established in children with non-CNS infections who were from 3 months to 16 years of age, as indicated below. The dose of imipenem used in most studies was 100 mg/kg/day for patients 3 years of age or 60 mg/kg/day for those >3 years of age, divided in four equal doses. A multicentre, open, non-comparative trial involving 178 infants and children with bacterial infections evaluated the efficacy of imipenem.117 A favourable clinical response was achieved in 98 of 100 patients. Adverse reactions, generally mild and reversible, included diarrhoea/vomiting (5.1%), irritation of intravenous infusion site (3.3%) and rash (2.2%). Changes in laboratory test values reported most frequently were thrombocytosis (8.9%), elevations in aspartate aminotransferase (7.9%) and alanine aminotransferase (5.6%) and eosinophilia (8.4%). Infections were caused by a broad spectrum of pathogens including Haemophilus influenzae, S. aureus, P. aeruginosa and anaerobes.
In another study, children (n = 144, 22 days to 15 years old) hospitalized for non-CNS bacterial infections received imipenem (100 mg/kg/day for patients 3 years of age; 60 mg/kg for those >3 years of age) for 9.4 days (range 3–28 days).118 Diagnoses in the 74 evaluable children included bronchopneumonia with or without empyema (20%), peritonitis complicating appendicitis (16%), skin/soft tissue abscesses (14%), septicaemia (11%) and miscellaneous other infections (39%). Among evaluable patients, 95% were clinically cured or improved. One child, a marasmic child with Pseudomonas pneumonitis, died during therapy; 12% of children had non-serious adverse experiences possibly related to imipenem.
A multicentre study evaluated imipenem in severe infections in children with granulocytopenia and haematological diseases and cancers.119 Children who had received prior antibiotics had an efficacy rate of 79.2%, while those who had not received previous treatment had an efficacy rate of 80.6%. Three children experienced nausea, vomiting and/or diarrhoea; two children had abnormal liver function test parameters that recovered after cessation of the drug treatment.
Twenty-five children (5 months to 11.3 years) with acute osteomyelitis (n = 7), suppurative arthritis (n = 11) or both (n = 7) were treated with imipenem.120 Bacterial pathogens identified included S. aureus, H. influenzae, P. aeruginosa, Streptococcus pneumoniae, group A Streptococcus, Kingella kingae and Citrobacter amalonaticus. All isolates were susceptible to imipenem in vitro. All but one child with P. aeruginosa osteomyelitis responded favourably to imipenem. The median duration to resolution of symptoms was 6 days. Adverse reactions included maculopapular rash (n = 1), diarrhoea (n = 1) and mild transient elevation of alanine aminotransferase levels (n = 3).
|
Safety and tolerability |
|---|
The good safety profile of imipenem is well established after 29 years of research and 20 years of clinical experience in 26 million patients from more than 100 countries. Adverse effects seldom require discontinuation and serious side effects are rare. The most common drug-related adverse events reported in at least 1% of patients in clinical trials included phlebitis/thrombophlebitis (3.1%), nausea (2.0%), diarrhoea (1.8%) and vomiting (1.5%).
Seizures
Imipenem is not indicated for CNS infections because of its proconvulsive activity.121–124 Dosage monitoring systems have resulted in decreased seizure incidence in patients treated with imipenem, making it important to examine the current, real-life incidence of seizure with carbapenems. A meta-analysis of 37 papers published between 1984 and 1999 reporting the use of imipenem was done to ascertain the true incidence of seizures when variables known to increase the risk or seizure are factored in.125 Among 5761 adult patients treated with imipenem 81 seizures were reported, corresponding to a 1.4% incidence of seizures. Yearly data from US product labelling indicate that the seizure rate for imipenem has consistently been 0.4%, compared with 0.5–0.7% for meropenem. No seizures were reported in two comparative studies of imipenem and meropenem (n = 200 and n = 232, respectively).74,126 Recognition of the factors that predispose patients to seizure, such as kidney dysfunction, prior history of seizure, metabolic derangement, anoxia, and phenytoin discontinuation, and use of appropriate carbapenem doses have substantially reduced concerns about seizure risk with carbapenems.127
Overdose
The main dosage concern with imipenem is correct dosage adjustment in patients with renal impairment. The risk of producing a seizure is highly associated with improper dose adjustment in relation to kidney function.128 If appropriate care is taken, seizures occur in <1% of patients treated. It is possible that concomitant administration of drugs with neurotoxic profiles, such as theophylline given in overdose, may increase the risk of seizures.128 In patients with normal kidney function, the maximum total daily dosage should not exceed 50 mg/kg/day or 4.0 g/day, whichever is lower.
Drug interactions
In general, ganciclovir and imipenem should not be used concomitantly because seizures have been reported in patients who received these drugs together. Concomitant administration of imipenem and probenecid results in only minimal increases in plasma levels of imipenem and plasma half-life, thus it is not recommended that imipenem and probenecid be co-administered. Imipenem should not be mixed into solution with other antibiotics, although it may be administered concomitantly with other antibiotics, such as aminoglycosides.
|
Carbapenems as appropriate therapy |
|---|
With the introduction of new carbapenems with different spectra of activity, it has become necessary to differentiate among the available agents so that the most appropriate agent can be selected. The place of carbapenems in appropriate therapy has been addressed in several recent publications. A carbapenem classification scheme was proposed to delineate the optimal use of these agents (Table 4).69 Ertapenem, a Group 1 carbapenem, is best suited for use in infections where non-fermenting Gram-negative bacilli such as Pseudomonas and Acinetobacter are not usually involved. Imipenem and meropenem, classified as Group 2, are considered to be among the most reliable agents for nosocomial infections. Brink et al.129 agreed that Group 2 agents are appropriate empirical treatment for patients with severe or nosocomial infection when used in accordance with local surveillance data. Nevertheless, it is still difficult to give clear recommendations for the selection of a single carbapenem compound in a particular situation. Whenever imipenem and meropenem were tested against each other in clinical studies, significant differences were not observed. This does not come as a surprise since clinical trials are usually designed to show non-inferiority rather than superiority of a compound. This is a clear limitation of the evidence base that is generated with such studies. Ertapenem was not tested against imipenem or meropenem in clinical trials. However, despite the fact that ertapenem has no significant activity against enterococci and only 5% of the drug is unbound in the serum, it has shown non-inferiority to the established broad-spectrum therapy with piperacillin/tazobactam in patients with intra-abdominal infections.130 In vitro advantages of one compound in terms of MIC values may be contrasted by less favourable pharmacokinetic properties. Because of differences in local epidemiologic situations, all recent guidelines point to this fact and the necessity to adjust treatment recommendations accordingly. Finally, pharmaco-economic considerations may vary with country, healthcare system, re-embursement forms, contracts and medical practice.
|
Imipenem has an important role in the treatment of HAP and VAP.131–134 These infections are burdened with treatment failure, often in the range of 30–40%, and these are clearly related to an increased mortality.135,136 If adequate therapy is used, treatment failure decreases.135–138 Ibrahim et al.139 demonstrated a decrease in inadequate initial therapy when initiating a strategy that included imipenem plus vancomycin and a quinolone or ß-lactam. Recent American Thoracic Society (ATS) guidelines recommend among other compounds ertapenem for early onset HAP or VAP in patients with no known risk factors for multidrug-resistant pathogens.140 Conversely, imipenem and meropenem are among those that in combination with an antipseudomonal fluoroquinolone or an aminoglycoside are recommended for patients with late onset disease or risk factors for multidrug-resistant pathogens. Moreover, the addition of linezolid or vancomycin is recommended if there is a likelihood for MRSA.
Imipenem is regarded as an agent of choice in the treatment of severe nosocomial infections.69,129,141–143 The patients most likely to benefit from imipenem would include those who are suspected of being infected with P. aeruginosa, an ESBL-producing E. coli or Klebsiella, Acinetobacter or any multidrug-resistant pathogen, and those with polymicrobic or mixed aerobic/anaerobic infections. The utility of this recommendation was illustrated by data suggesting that when a carbapenem was administered during the first 5 days in patients with ESBL-producing K. pneumoniae bacteraemia, the mortality rate was 5% compared with a mortality rate of 43% when any other antibiotic was used (P = 0.01).86 Similarly, a bowel colonization study suggested that piperacillin/tazobactam may not retain activity against AmpC producers in some cases.144,145
|
Mono versus combination therapy |
|---|
There are three reasons given for combining antimicrobial agents: (i) to broaden the spectrum of activity; (ii) to achieve synergistic effects; and (iii) to prevent resistance development. In the case of carbapenem therapy these reasons might not be convincing since these agents have an inherent broad spectrum of activity, synergy even in vitro has only inconsistently been demonstrated and prevention of resistance development has not been well documented. Hence, a number of studies have addressed the question of whether imipenem can be used as monotherapy in severe nosocomial infections. The results were conflicting.146,147 However, two meta-analyses clearly showed that there was no advantage for combination therapy employing aminoglycosides, not even in Pseudomonas infections.125,147 However, combination therapy was associated with a significantly higher rate of nephrotoxicity.
|
De-escalation therapy |
|---|
For decades escalation therapy was advocated in the treatment of infectious disease. Antibiotic therapy was initiated with a basic agent and only if this approach failed after 72 h, more potent compounds were used. Rising resistance rates and better understanding of the inflammatory process prompted some experts to advocate initial therapy with broad-spectrum, highly active compounds at least in severe infections. This concept was initially referred to as ‘intervention therapy’. Although not the only suitable agents for this concept, carbapenems always played an essential role in this approach. The concerns about engendering resistance through overuse of potent antibiotics were met by expanding the concept to the ‘de-escalation therapy’, whereby the initial therapy was tailored once microbiological culture results and susceptibility tests were available. The two-stage approach of using aggressive initial therapy followed by de-escalation allows serious infection to be treated immediately and effectively while avoiding antibiotic overuse, potential resistance and excessive cost. Meanwhile this concept has been proven by a number of studies that showed that the appropriate initial therapy is a crucial factor in outcome of the patient.148–150
Cost is a practical consideration in selecting treatment. Imipenem is a cost-effective treatment for serious infection,151–154 particularly if de-escalation to an oral regimen is possible.152 A formulary feasibility study found that imipenem and piperacillin/tazobactam were both effective and suitable for intra-abdominal infection, pneumonia, febrile neutropenia, and skin and soft tissue infection, but that imipenem should be retained due to the prevalence of multidrug-resistant Gram-negative pathogens.75 A more recent study of febrile neutropenic patients reported that overall treatment costs were 189.55 euros less with imipenem than piperacillin/tazobactam (P < 0.001).155 Imipenem monotherapy has been recommended in polymicrobial infections where combination therapy would be more costly, although imipenem combination therapy was recommended if Pseudomonas was present.85 This sustained susceptibility, efficacy, tolerability and cost-effectiveness are the hallmarks of appropriate initial therapy.
|
Conclusions |
|---|
Imipenem, the first in the carbapenem class, has pioneered broad-spectrum antibiotic usage. It has been used in 26 million patients over 20 years and still demonstrates potent antibiotic activity against life-threatening pathogens, including those that are multidrug resistant. Imipenem is still used as the benchmark for comparison with other agents and remains an antibiotic of choice for the treatment of serious infection. Early aggressive therapy with a broad-spectrum agent is essential to reduce mortality and minimize resistance. Imipenem has remained a key asset in the setting of appropriate antibiotic therapy for 20 years due to its efficacy, tolerability and sustained pathogen susceptibilities. The option to de-escalate allows physicians to achieve the goal of treating critically ill patients aggressively while maintaining the option to narrow treatment as needed to avoid resistance.
|
Transparency declarations |
|---|
 Top |
|---|