Peter N.R. Heseltine

Treatment of Disseminated Mycobacterium avium Complex Infection in AIDS with Amikacin, Ethambutol, Rifampin, and Ciprofloxacin

OBJECTIVE To determine the efficacy of combination drug therapy for disseminated Mycobacterium avium complex infection in patients with the acquired immunodeficiency syndrome (AIDS). DESIGN Prospective, nonrandomized, before-after comparison. SETTING Outpatient clinics at three university medical centers. PATIENTS Seventeen patients with at least two consecutive blood cultures positive for M. avium complex who had not been previously treated with antituberculous medications. Fifteen of the seventeen patients completed at least 4 weeks of treatment. INTERVENTION Patients received daily intravenous amikacin (7.5 mg/kg body weight) for the first 4 weeks plus the following oral medications for at least 12 weeks: ciprofloxacin, 750 mg twice daily; ethambutol, 1000 mg daily; and rifampin, 600 mg daily. MEASUREMENTS AND MAIN RESULTS The baseline geometric mean colony count from blood cultures decreased from 537/mL to 14/mL (P less than 0.001) after 4 weeks of therapy. The microbiologic suppression was sustained while on treatment and was associated with a decrease in systemic symptoms related to M. avium complex infection. Premature withdrawal from treatment (less than 12 weeks) occurred in 7 of 17 patients. The commonest reasons for early withdrawal were gastrointestinal intolerance and hepatic toxicity. CONCLUSIONS Mycobacterial load and systemic symptoms in patients with AIDS and disseminated M. avium complex infection can be effectively reduced by a regimen containing amikacin, ethambutol, rifampin, and ciprofloxacin.

Antibiotic management of surgically treated gangrenous or perforated appendicitis: Comparison of gentamicin and clindamycin versus cefamandole versus cefoperazone☆☆☆

A study of 130 adult patients with surgically treated gangrenous or perforated appendicitis was undertaken to evaluate the efficacy of three antibiotic regimens. Forty-eight patients received cefamandole, 40 were given the combination of clindamycin and gentamicin, and 42 were treated with cefoperazone. Side effects from these antibiotics were infrequent and mild. When all cases were compared for infectious failure, clindamycin-gentamicin showed a clear advantage over cefamandole. Because of the heterogeneity of the total study population, patients with perforation and peritonitis were compared separately. This analysis confirmed the advantage of clindamycin-gentamicin over cefamandole. In addition, it appears that clindamycin-gentamicin is more efficacious than cefoperazone.

In Vitro Activities of Nontraditional Antimicrobials against Multiresistant Acinetobacter baumannii Strains Isolated in an Intensive Care Unit Outbreak

ABSTRACT Fifteen multiresistant Acinetobacter baumannii isolates from patients in intensive care units and 14 nonoutbreak strains were tested to determine in vitro activities of nontraditional antimicrobials, including cefepime, meropenem, netilmicin, azithromycin, doxycycline, rifampin, sulbactam, and trovafloxacin. The latter five drugs were further tested against four of the strains for bactericidal or bacteriostatic activity by performing kill-curve studies at 0.5, 1, 2, and 4 times their MICs. In addition, novel combinations of drugs with sulbactam were examined for synergistic interactions by using a checkerboard configuration. MICs at which 90% of the isolates tested were inhibited for antimicrobials showing activity against the multiresistant A. baumannii strains were as follows (in parentheses): doxycycline (1 μg/ml), azithromycin (4 μg/ml), netilmicin (1 μg/ml), rifampin (8 μg/ml), polymyxin (0.8 U/ml), meropenem (4 μg/ml), trovafloxacin (4 μg/ml), and sulbactam (8 μg/ml). In the kill-curve studies, azithromycin and rifampin were rapidly bactericidal while sulbactam was more slowly bactericidal. Trovafloxacin and doxycycline were bacteriostatic. None of the antimicrobials tested were bactericidal against all strains tested. The synergy studies demonstrated that the combinations of sulbactam with azithromycin, rifampin, doxycycline, or trovafloxacin were generally additive or indifferent.

Pharmacokinetics of meropenem in patients with intra-abdominal infections.

Noncompartmental and compartmental analyses of meropenem disposition in patients receiving 1-g intravenous intermittent infusions every 8 h were performed. Twelve patients (one woman and 11 men) participated in the meropenem pharmacokinetic analysis. Operative findings included perforated appendicitis (five patients), gangrenous appendicitis (five patients), peri-appendical abscess (one patient), and gunshot wound to the abdomen (one patient). The most common associated adverse drug reactions to meropenem were diarrhea and increased liver enzymes. The estimated noncompartmental pharmacokinetic parameters, mean +/- standard deviation, are as follows: maximum drug concentration in plasma, 47.58 +/- 17.59 micrograms/ml; half-life, 1.04 +/- 0.19 h; elimination rate constant, 0.68 +/- 0.12 h-1; area under the concentration-time curve from 0 h to infinity, 57.5 +/- 20.12 micrograms x ml/h; total plasma clearance, 315.40 +/- 71.94 ml/min; renal clearance, 136.7 +/- 89.20 ml/min; volume of distribution at steady state, 26.68 +/- 6.88 liters; and mean residence time, 1.47 +/- 0.28 h. The two-compartment model best described meropenem disposition in our patients. Our findings differed from estimates for healthy volunteers possibly because of the physiologic changes as a result of surgery. Our findings suggest that meropenem (1,000 mg) administered intravenously every 8 h provides adequate concentrations for most intra-abdominal infections.

Surgically treated gangrenous or perforated appendicitis. A comparison of aztreonam and clindamycin versus gentamicin and clindamycin.

A randomized, double-blinded, controlled clinical study of 84 patients with surgically treated gangrenous or perforated appendicitis was done to compare the efficacy of the combination of aztreonam, the first monobactam antibiotic, with gentamicin when either was combined with clindamycin. Fifty-six patients who were treated with aztreonam and clindamycin (A/C) and 28 patients who were treated with gentamicin and clindamycin (G/C) fulfilled criteria for evaluation. A matched historic control group of 56 G/C patients was also included for comparison. All measures of outcome, including days of fever, hospitalization, antibiotic therapy, and the incidence of antibiotic failures, were similar. It was concluded that aztreonam was as effective as gentamicin in this study and may offer some advantages with regard to toxicity and serum drug level monitoring.

Readers:' Forum: Nosocomial Rubella—Consequences of an Outbreak and Efficacy of a Mandatory Immunization Program

An outbreak of rubella in a large metropolitan hospital is described. Nineteen cases among employees and three secondary cases in family members occurred. Nosocomial cases occurred among the 3,900 employees of an adult medical-surgical unit where a voluntary program of rubella immunization was in effect. No cases occurred among the 1,400 employees of the womens and pediatric units with mandatory policies, despite interfacility and community exposure. Ten pregnant women were among the 377 contacts of the cases. Five were sero-negative to rubella. Two who developed clinical rubella, one asymptomatic sero-conversion and one other, all elected to terminate their pregnancies. The remaining woman, exposed in her third trimester delivered a normal infant. We conclude a policy requiring new employees to be rubella immune is more effective in preventing nosocomial rubella than a voluntary program and is desirable in view of the potential consequences of an outbreak to pregnant employees.

Why don't doctors and nurses wash their hands?

For almost 150 years healthcare workers have been taught that cross-infections are transmissible but not contagious and that the most effective way to prevent these cross-infections is to wash their hands before and after every patient contact.1,2 But, they don’t do it. They don’t merely not do it every time, they don’t do it most of the time and sometimes not even when it might be most expected, as when caring for an intensive care unit (ICU) or emergency room patient.3,4 A brief observation in almost any clinic or hospital will confirm these lapses. Why don’t physicians and nurses follow this most basic care practice? Some years ago, while visiting a foreign military hospital, my colleague asked, half-jokingly, the purpose of the armed soldier stationed outside the ICU. We were told quite seriously that the sergeant had instructions to arrest anyone entering the unit who did not don a gown and gloves after washing their hands. We were impressed less by the show of force than by the example of commitment to a tenet of infection control we professed to be essential. But, we asked ourselves, are the lives of our ICU patients less important than these, that we cannot muster the same enthusiasm at home? Why don’t physicians and nurses wash their hands? Is it that we really don’t believe that hand washing works? If we didn’t believe in the spread of disease on our hands, why did we adopt the use of gloves so quickly in the acquired immunodeficiency syndrome era? And why do we not change those gloves—always—between patient contacts? The reasons that come to mind do not seem very compatible with our professional ethics or sworn oaths. Maybe we don’t believe it’s necessary. For a few years after the introduction of penicillin, it seemed possible that all infections might be vanquished by antibiotics, but, in less than a blink of evolution’s eye, the promise was gone, replaced by organisms resistant to all known antimicrobials, such as vancomycin-resistant enterococci, a leading cause of mortality in liver transplant patients.5 Perhaps it has to do with our expectation of successful outcomes for our patients. My pediatric colleagues are both enthusiastic about their patients and, usually, enthusiastic about the outcomes. Although not perfect by any measure, in general they practice better infection control. Maybe the daunting task of believing a tiny 800-g baby will survive to become a fulfilled adult requires a different perspective on the contribution of each small act of care. The authors of an article in this issue give us some measure of how much healthcare workers do care when challenged and how science is best applied.6 The article is about the scientific examination of an electronic faucet. The authors found one that didn’t work as advertised and that may have enhanced the growth of pathogenic bacteria. They warn us not to rely on technology as an answer to our problems. What they don’t tell you is that, during this time, they were in the aftermath of a terrible flood and were rebuilding their town—from scratch and memories of what was good. They used technology to measure how safe their hospital was for their patients. They did not use the technology as a substitute for care. That they carried out this study at a time of great hardship is extraordinary. There are healthcare workers throughout the developing world who work under the same or worse conditions. And in the United States there are some who also work under these conditions. They work in crowded public and community hospitals and clinics with few resources and antiquated systems. It’s hard to believe in—or to achieve— excellent medical outcomes when you don’t have the space, the supplies, or the trained people to succeed. They are told that new technology will make it all work; but, it is not about the technology, it’s about how much excellence you expect. Healthcare workers’ expectations of excellence are perhaps not sufficiently part of modern medicine. A recent report by the Institute of Medicine tells us more than we wanted to know about our performance.7 As healthcare costs more and the budgets contract, there has emerged a philosophy of the reversed Field of Dreams. If we don’t build it, they won’t come. Maybe they are right, but it numbs the soul and loses sight of excel-

Too Many or Too Few Hands

In 1975, Haley investigated a series of neonatal infections at a large municipal hospital and determined that understaffing and overcrowding had contributed significantly to the outbreak. Almost 25 years later, though intuitive to infection control practitioners, surprisingly little controlled evidence exists to support this concept. The paper by Harbath et al in this issue identifies and measures the effects of understaffing, overcrowding, and handwashing on an outbreak of Enterobacter cloacae in a neonatal intensive care unit (ICU). The authors conclude that these factors had a primary role in sustaining the outbreak, which was only brought under control when they were changed.

Analysis of Cefepime Tissue Penetration Into Human Appendix

Cefepime is a new extended‐spectrum cephalosporin with gram‐positive and gram‐negative coverage including Staphylococcus aureus and Pseudomonas aeruginosa. We evaluated the drugs plasma, peritoneal fluid, and appendix tissue concentrations in patients with a postoperative diagnosis of perforated or gangrenous appendicitis. Patients 18 years of age or older were randomly assigned to receive either cefepime 2 g every 12 hours plus metronidazole 500 mg every 6 hours intravenously, or gentamicin 1.5 mg/kg plus clindamycin 900 mg every 8 hours intravenously. During surgery, appendix tissue, plasma, and peritoneal fluid samples were obtained, and frozen at −70°C for high‐pressure liquid chromatographic analysis. Thirty‐five patients with perforated (26) or gangrenous (9) appendicitis had concentrations acceptable for analysis. The mean time between the administration of cefepime and the time of sampling (referred to as A time) was 5.99 ± 3.75 hours (mean ± SD). The values for plasma (n=34), tissue (n=33), and peritoneal fluid (n=25) concentrations were 16.27 ± 21.87 μg/ml, 4.84 ± 6.15 μg/g, and 14.4 ± 22.84 μg/ml, respectively. The appendix tissue:plasma ratio was 0.66 ± 0.52 and the peritoneal fluid:plasma ratio was 0.66 ± 0.51. Spearman rank correlations indicated statistically significant correlations between plasma concentration (r=‐0.889; p<0.0001), peritoneal fluid concentration (r=‐0.783; p=0.0002), and appendix tissue concentration (r=‐0.704; p=0.0016) versus Δ time. There was a significant correlation between peritoneal fluid and plasma concentration (r=0.853; p<0.0001), and appendix tissue and plasma concentration (r=0.815; p=0.0001). These results indicate that tissue and peritoneal fluid concentrations are approximately 51% and 74%, respectively, of a simultaneous plasma sampling after approximately 5.8 hours.

Pharmacoeconomics of Piperacillin/Tazobactam and Imipenem/Cilastatin in the treatment of patients with intra-abdominal infections

Costs involved in using piperacillin 4 g/tazobactam 500 mg, given as intermittent intravenous infusions every 8 hours, were compared with those for imipenem/cilastatin 500 mg, given as intermittent intravenous infusions every 6 hours, for the treatment of patients with gangrenous or perforated appendicitis. A total of 88 patients were included in our cost analyses: 42 patients in the piperacillin/tazobactam group and 46 patients in the imipenem/cilastatin group. Durations (mean +/- SD) of antibiotic therapies were 7.8 +/- 3.3 days and 7.1 +/- 2.6 days for the piperacillin/tazobactam and imipenem/cilastatin groups, respectively. No statistical significance was found for the difference in duration of therapy (P = 0.376). Total drug treatment costs were