INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

In 1992, 6.2% of the global population was >65 years of age; this proportion is projected to expand to 20% by the year 2050 (8). The elderly are at increased risk for serious infection that may result in death, for a variety of reasons.

The diagnosis of infection in elderly patients may be difficult due to the presence of fewer clinical signs and symptoms and the problems involved in microbiological confirmation (8). Therefore, empirical antibiotic therapy is often necessary, and since the range of pathogens implicated in infections in elderly patients is more diverse than that in younger patients, a broad-spectrum antimicrobial regimen is often required. The treatment of infections in elderly patients is potentially complicated by alterations in the pharmacokinetic handling of antibiotics in these patients, and this group may be subject to an increased risk of toxicity caused by some antibiotics (8).

Meropenem is a newer carbapenem which appears to have several advantages over imipenem. Meropenem is relatively stable to DHP-I (2, 26) and, consequently, does not have to be administered with a DHP-I such as cilastatin. Both carbapenems have a uniquely broad antimicrobial spectrum which covers most clinically important gram-negative and gram-positive aerobic and anaerobic cocci and bacilli. Meropenem is more active than imipenem against Enterobacteriaceae and Pseudomonas aeruginosa but slightly less active against certain gram-positive cocci (e.g., staphylococci) (9). The only bacterial species that are normally resistant to the carbapenems are Stenotrophomonas maltophilia, Burkholderia cepacia, Corynebacterium jeikeium, Enterococcus faecium, certain Enterococcus spp., and methicillin-resistant Staphylococcus aureus. Carbapenems are highly resistant to hydrolysis by almost all -lactamases, including the mutant extended-spectrum -lactamases produced by certain members of the family Enterobacteriaceae, e.g., Klebsiella pneumoniae (9).

Meropenem is well tolerated in the elderly population (27), but to the best of our knowledge only one European multicenter comparative study with meropenem has been performed. The study included elderly (>65 years) and nonelderly (<65 years) patients (22). Our study was designed to evaluate the efficacy of and tolerability for meropenem compared to those for a standard regimen in the Netherlands, cefuroxime plus gentamicin (with metronidazole if anaerobic pathogens were suspected) for the treatment of serious bacterial infections in patients 65 years of age.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Patients. This nonblinded, randomized, parallel-group study was conducted in five hospitals in The Netherlands. Patients were eligible to participate in the study if they were 65 years of age, able to provide informed consent, and had one or more (proven or suspected) of the following serious bacterial infections: sepsis syndrome, intra-abdominal infection, lower respiratory tract infection (LRTI), complicated urinary tract infection (UTI) (3), and/or bacteremia. Patients with known hypersensitivity to -lactam antibiotics were excluded, as were those with hepatic impairment (three times the upper reference limit of liver transaminases for each hospital), hepatic failure or hepatic coma, a granulocyte count of 500 cells/mm³, cystic fibrosis, or a life expectancy of <48 hours. Patients who had previously participated in the trial or received another investigational drug or antibiotic within 30 days or 3 days prior to randomization, respectively (unless the organism was resistant), were not eligible to participate in the study.

Classification of infections. Nosocomial infections were defined according to the criteria of the U.S. Centers for Disease Control (10). Intra-abdominal infection was defined as a suspected or proven complicated infection derived from the gastrointestinal or reproductive tract, with signs and symptoms of fever (>38.3°C), leukocytosis, abdominal wall rigidity, and/or ileus. The infectious process had to extend beyond the site of origin, causing peritonitis or abscess formation (such as perforation, acute cholangitis, and periappendicular abscess). Pneumonia or LRTI was defined according to signs and symptoms such as chest pain, cough, and/or auscultatory findings (rales and/or evidence of pulmonary consolidation) with or without fever (>38.3°C) or leukocytosis and radiographic or other laboratory evidence supporting the diagnosis. Sputum was cultured if a Gram stain showed 25 leukocytes and 10 epithelial cells per high-power field. UTIs were defined as pyelonephritis or as a complicated UTI in the presence of an indwelling catheter or the use of intermittent catheterization, >100 ml of residual urine after voiding, obstructive uropathy due to bladder outlet obstruction or a calculus, vesicoureteral reflux or urologic abnormalities, azotemia due to intrinsic renal disease or occurring after renal transplantation. Patients with a complicated UTI were included only when signs and symptoms of a systemic infection occurred.

Sepsis syndrome criteria obtained at time of entry. The diagnosis of sepsis syndrome was based on the definition of Bone et al. (5), i.e., clinical evidence of infection plus the presence of fever or hypothermia (rectal temperature of >38.3°C or <35.6°C), tachypnea (>20 spontaneous breaths/min), tachycardia (>90 beats/min), and at least one of the following manifestations of inadequate organ perfusion or function: oliguria (<30 ml/h), hypoxemia (arterial oxygen pressure of <75 mm Hg while room air was being breathed), elevated plasma lactate, and/or mental alteration compared to baseline.

Treatment. Eligible patients were randomly assigned (in blocks of four) to a study group by means of consecutive sealed envelopes. Meropenem (Zeneca Farma, Ridderkerk, The Netherlands) was administered at a dosage of 1 g (dissolved in 20 ml of sterile water-80 ml of sterile isotonic saline) every 8 h; in cases of renal impairment, the dosages were as follows: for a creatinine clearance rate of 26 to 50 ml/min, 1 g twice a day (BID); for a rate of 10 to 25 ml/min, 0.5 g BID; for a rate of <10 ml/min, 0.5 g once daily. Cefuroxime (Glaxo Wellcome, Zeist, The Netherlands) was given at a dosage of 1.5 g (dissolved in 100 ml of sterile isotonic saline) every 8 h, in cases of renal impairment, the dosages were as follows: for a creatinine clearance rate of 10 to 50 ml/min, 1.5 g BID and for a rate of <10 ml/min, 1.5 g once daily. Gentamicin (Schering-Plough, Amstelveen, The Netherlands) was administered at a dosage of 4 mg/kg of body weight (dissolved in 100 ml of sterile isotonic saline) once daily or in two or three divided doses; in cases of renal impairment, the dosages were as follows: for a creatinine clearance rate of 50 to 70 ml/min, 1.8 mg/kg once daily; for a rate of 10 to 50 ml/min, 1.5 mg/kg once daily; and for a rate of <10 ml/min, 1.5 mg/kg every 2 days. Metronidazole (Rhône-Poulenc Rorer, Amsterdam, The Netherlands) was given at a dosage of 0.5 g (dissolved in 100 ml of sterile isotonic saline) every 6 h. All drugs were administered intravenously over 20 to 30 min, with a controlled delivery system; in cases of renal impairment, the dosages were as follows: for a creatinine clearance rate of 10 to 50 ml/min, 0.5 g three times a day and for a rate of <10 ml/min, 0.5 g BID. The duration of treatment depended on the clinical and bacteriological response, but a duration of 5 to 10 days (maximum 28 days) was recommended.

Assessment. (i) Clinical. At the time of entry into the study, a medical history was taken and each patient was examined for signs and symptoms of infection. The pretreatment severity of the infection was assessed with the APACHE II scoring system (16). In addition, a chest X-ray was performed in patients with suspected LRTI. Hematological tests (hemoglobin, hematocrit, total leukocyte count with differentiation, platelet count) and serum biochemistry tests (aspartate aminotransferase, alanine aminotransferase, albumin, alkaline phosphatase, bilirubin, and creatinine) were also performed. All procedures were performed once a week, at the end of treatment, and at other times when necessary. Renal failure was defined as an increase in serum creatinine of 40 µmol/liter when the baseline value was <300 µmol/liter and an increase of 80 µmol/liter when the baseline value was 300 µmol/liter (13). A follow-up clinical examination was performed 4 to 6 weeks after the end of therapy.

(ii) Bacteriological. At least two blood specimens for cultures (both aerobic and anaerobic) were drawn when each subject entered the study. Also, urine samples were collected and, when possible, specimens from the site of infection were obtained for culture. If prior antimicrobial therapy had been administered, samples were taken after the previous antibacterials were stopped and before the therapy in the present study was started. Repeat cultures were taken from the same relevant sites from all patients except those with UTIs during and preferably immediately posttreatment. For patients with UTIs, a urine specimen was cultured before therapy, after 48 to 72 h, 5 to 9 days posttherapy, and 4 to 6 weeks after the end of therapy to identify relapses and superinfections. A blood culture set consisted of two bottles, Bactec (Becton Dickinson) and BacT Alert (Organon Technika, Oss, The Netherlands); one was aerobic, and the other was anaerobic. Each bottle was filled with 10 ml of blood. All procedures were followed in accordance with the manufacturer's recommendations, and isolates were identified by standard methods. Primary bacteremia was defined as isolation of a pathogen from the circulating blood without a known site of infection. Contamination was defined as isolation of a common (skin) contaminant (e.g., Bacillus spp., coagulase-negative staphylococci, and Corynebacterium spp.) from one of at least two blood cultures (10).

Statistical analysis. A power calculation was performed after recruitment; the expected midpoint of a 95% confidence interval (CI) for a response rate of approximately 70% is 20%. The primary endpoint was the clinical response to therapy at the end of treatment. The secondary endpoints were bacteriological response and tolerability. Statistical analyses of dichotomous variables were done by the two-sided Fisher exact test at a 5% level of significance. Standard approximate 95% CIs for differences in proportions are given. Results were analyzed by the intention-to-treat principle as well as by evaluability.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Patients. During an 11-month period, a total of 79 patients participated the study (University Hospital Utrecht, Utrecht, The Netherlands, n = 20; St. Antonius Hospital, Nieuwegein, The Netherlands, n = 9; Elkerliek Hospital, Helmond, The Netherlands, n = 7; St. Elisabeth Hospital, Tilburg, The Netherlands, n = 19; and Sophia Hospital, Zwolle, The Netherlands n = 24). Of these patients, 39 were randomized to receive meropenem and 40 were randomized to receive cefuroxime-gentamicin. Gentamicin was administered once daily, BID, or in three divided doses to 30, 7, and 2 of the latter patients, respectively (1 patient did not receive gentamicin due to severe preexisting renal failure). Metronidazole was administered to 15 patients receiving the combination regimen. The mean (range) duration of treatment was 7.5 days (3 to 21 days) in the meropenem group and 7.4 days (3 to 17 days) in the combination arm (Table 1). The means of the total numbers of doses of meropenem, cefuroxime, gentamicin, and metronidazole administered were 19, 19, 5, and 26, respectively.

Clinical efficacy. On an intention-to-treat basis, the clinical response at the end of therapy was satisfactory in 69% (27 of 39) of patients treated with meropenem and in 63% (25 of 40; P = 0.64; 95% CI, 14 to 23%) of those treated with cefuroxime-gentamicin (±metronidazol). All patients that died were classified as failures. The clinical response in evaluable patients at the end of therapy was satisfactory in 26 of 37 (70%) patients treated with meropenem and 24 of 33 (73%; P = 1.00; 95% CI, 24 to 19%) of those treated with cefuroxime-gentamicin (±metronidazole) (Table 2) (P = 1.00). In patients with infections other than UTIs, a satisfactory response occurred in 23 of 31 (74%) of meropenem patients compared with 21 of 28 (75%) of combination therapy patients.

Bacteriological efficacy. Microbiologically documented infections were present in 41 of 70 (59%) clinically evaluable patients (Table 4). The majority of infections were caused by gram-negative organisms (n = 57 [26 patients receiving meropenem, 31 patients receiving combination therapy]). There were 26 gram-positive infections (19 meropenem, 7 combination therapy) and 6 anaerobic infections (4 meropenem, 2 combination therapy) (Table 4). There were more gram-positive infections in the meropenem group and more gram-negative infections in the combination therapy group.

Tolerability. All patients were included in the evaluation of tolerability. Adverse events were reported in 19 of 39 (48.7%) meropenem-treated patients compared to 18 of 40 (45.0%; P = 0.82; 95% CI, 18 to 26%) patients treated with combination therapy. The most common adverse events were impaired liver function (meropenem, n = 9; combination therapy, n = 5), diarrhea (meropenem, n = 4; combination therapy, n = 1), and renal failure (meropenem, n = 2; combination therapy, n = 5; P = 0.43; 95% CI, 20 to 5%). All adverse events, except renal failure with electrolyte disturbance, were mild. Three patients (two receiving meropenem, one receiving combination therapy) required hemodialysis for a short period.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

In elderly patients with serious infections, a delay in the selection of antibiotics and/or the incorrect choice of antibiotics are likely to have a significant impact on treatment outcome. Combination antibiotic therapy is widely used but, compared to monotherapy, may require more personnel time and may increase treatment costs, the risk of toxicity, and patient inconvenience (4, 13). Effective empirical therapy with a single-drug regimen would help to overcome these problems.

The results of this small multicenter study indicate that meropenem (1 g three times per day) is an effective empirical monotherapy for infections in severely ill, elderly patients with or without bacteremia. The patient population was characterized by relatively high mean APACHE II scores and the presence of underlying diseases. There was no significant difference between the rates of satisfactory clinical response achieved with meropenem and those achieved with cefuroxime-gentamicin (±metronidazole), a standard treatment regimen in The Netherlands (70 versus 73%). Another randomized study reported a higher overall rate of clinical success with meropenem (93 versus 79% with ceftazidime-amikacin) in patients with serious bacterial infections (22). Currently, a randomized study with a low dosage of meropenem (500 mg three times per day) for serious infections is under way.

A wide range of pathogens were isolated from patients in this study. The observed rates of satisfactory microbiological response with meropenem are lower than those from certain other trials with meropenem in patients with community-acquired LRTIs (22, 29), intra-abdominal infections (6, 11, 14), complicated UTI (7), and septicemia (30). A lower rate of microbiological eradication would be expected in our study because it involved only elderly patients, in whom bacterial eradication may be compromised by age-related immune deficiencies. The risk of resistance emerging during therapy is always a concern when empirical broad-spectrum monotherapy is used. However, there was no evidence of the development of meropenem resistance in this study within the follow-up period, but the size of the study does not allow conclusive evidence for single pathogens or for clinical conditions other than pneumonia.

Both drug regimens employed were well tolerated. Most adverse events reported were mild (e.g., diarrhea, phlebitis, and elevated liver transaminases) and occurred with similar frequency in both groups. We observed a higher incidence of diarrhea in the meropenem group (10%) than that reported from the international clinical trials program with meropenem (4.3% overall, 1.9% drug related) (27). However, since the latter analysis (which involved a total of 3,220 patient exposures) indicated that the adverse event profile of meropenem in elderly patients does not differ from that in younger patients, this difference may simply be due to the relatively small number of patients in our trial.

Since renal function declines in the elderly, the pharmacokinetics of a drug in the presence of renal impairment is an important concern for elderly patients. Meropenem is primarily excreted by the kidneys, and the elimination half-life in healthy young volunteers is approximately 1 h (2). The elimination half-life is slightly prolonged in the elderly; a value of 1.27 h was observed in one study with healthy elderly men (18, 26). Therefore, dosage adjustments, made according to creatinine clearance, may be required in elderly patients (21).

The development of renal failure during therapy was reported in two patients (5%) treated with meropenem in the present trial and in 5 (15%) of those treated with combination therapy. Data from both animal and clinical studies have shown that meropenem is well tolerated by the kidneys (27, 31). An overall incidence of 0 to 25% for nephrotoxicity caused by aminoglycosides is reported in the literature (13, 19, 20, 23, 28), and the risk is increased in patients with renal impairment. In recent years, once-daily administration of aminoglycosides has been shown to be at least as effective as multiple daily administration in the treatment of certain types of infections (1, 12, 15). Once-daily administration of gentamicin was associated with a lower incidence of nephrotoxicity in two studies (1, 12) but not in another meta-analysis (23). In elderly patients, however, pharmacokinetic monitoring may be better (17). Thus, any drugs for elderly patients must be selected carefully and may require monitoring. The routine use of aminoglycosides in elderly patients should be avoided when possible (8, 17). In The Netherlands, a 1-day dosage of meropenem costs approximately $120 versus $79 in a control group, not including costs for monitoring gentamicin levels in the blood and for extra personnel for the administration of antibiotics.

In conclusion, in this small study, meropenem monotherapy (1 g every 8 h) was as efficacious as combination therapy for the empirical treatment of serious infections in the elderly, producing high rates of clinical and microbiological success without serious adverse events. Meropenem was well tolerated and offers greater flexibility of administration than cefuroxime plus gentamicin.