ALTANTA—Overcoming antibiotic resistance in the intensive care unit (ICU) requires multiple strategies, two of which were debated at the annual meeting of the American Thoracic Society. Marc Bonten, MD, a researcher at the Eijkman-Winkler Institute for Medical Microbiology, Infectious Diseases, and Inflammation in Utrecht, the Netherlands, advocated changing the duration and frequency of antibiotic use.[1] Presenting an argument in favor of antibiotic cycling was Marin Kollef, MD, Associate Professor of Pulmonary and Critical Care Medicine at the Washington University School of Medicine in St. Louis.[2]

HOW RESISTANCE DEVELOPS

Antibiotics lead to the selection of resistant organisms in various ways[3]:
• A patient has a susceptible strain of bacteria that undergoes a single-point mutation during treatment and becomes resistant. When resistance arises from a single-point mutation, inadequate treatment increases the likelihood that the resistant organism will spread.
• Multiple patients have susceptible organisms that undergo mutations. None of the mutations is sufficient on its own to produce resistance; however, antibiotic selection pressure raises the probability that the mutations will spread and ultimately combine to cause resistance. Ironically, successful treatment of susceptible infections increases the chances that the mutations will spread and combine.
• A patient with a susceptible strain is treated successfully but is surrounded by other patients who have resistant flora. The treated patient may have an increased chance of contracting a resistant strain once the susceptible flora have been eradicated because colonization resistance has been diminished.
• The density of resistant organisms is increased within a patient who is already colonized with such organisms. Elimination of susceptible organisms can allow the increase in density to occur.

The bottom line: Because bacteria are programmed to spread, treating and eradicating susceptible flora in one patient puts other patients at risk of acquiring resistance. Dr. Bonten noted that this happens with nearly every infection.

He acknowledged, however, that it is too simplistic to view these four mechanisms as independent processes; they can all interact. Dr. Bonten therefore concluded that the problem of resistance cannot be solved simply by looking at or measuring any one variable alone.

MANY FACTORS ARE INVOLVED

The time frame in which resistance develops depends on the number of mutations that a strain of bacteria must undergo. For example, Staphylococcus aureus became resistant to penicillin in five years, but it took Streptococcus pneumoniae more than 50 years to develop penicillin resistance.

The prevalence of resistance in an ICU may vary over time between 0% and 70% without any outside intervention, Dr. Bonten said. A number of factors influence the epidemiology of resistance in the ICU, he noted in an interview with PULMONARY REVIEWS; these include the admission rates of colonized patients and their length of stay. “It is … mainly related to patients admitted and discharged (with or without resistant bacteria) and the level of infection control procedures,” Dr. Bonten added. He also observed that if colonized patients are discharged as soon as they acquire resistance, its spread will be very unlikely. However, colonized patients with resistant bacteria are usually the ones who stay longest in the ICU and therefore contribute most to transmission.

Sometimes, isolating patients colonized with resistant bacteria may help prevent further spread. According to Dr. Bonten, “Patients most likely to be colonized can be identified … and put into isolation until proven to not be colonized. This has worked very well for methicillin-resistant S aureus in the Netherlands.”

REDUCING DRUG USE BY CHANGING PROCEDURES

Strategies that can help reduce antibiotic administration in the ICU include using more invasive diagnostic tests (eg, bronchoscopy) and, once results are obtained, either adjusting the antibiotic dose or stopping therapy. Another strategy is to use a very short course of antibiotics, reevaluate the patient after a few days, and stop treatment if there are no longer clinical signs of infection.

“My personal opinion is that more invasive and more frequent diagnostic tests should be performed,” Dr. Bonten said. “But this is a matter of controversy, especially for diagnosing ventilator-associated pneumonia.”

Dr. Bonten acknowledged that employing these methods would require an increase in the number of procedures and evaluations, but he thought that physicians would comply if they were convinced that the use of antibiotics and the spread of resistance would be reduced as a result. “Reduction in antibiotic use can never increase the prevalence of antibiotic resistance,” he concluded.

COUNTERPOINT: THE CASE FOR ANTIBIOTIC CYCLING

Another method of modifying antibiotic use is antibiotic cycling, which rests on the premise that using more than one class of drugs increases the chance of controlling resistance. In his presentation, Dr. Kollef explained how to limit the use of less effective antibiotics while also administering more effective ones, thereby increasing heterogeneity.[2]

WORKHORSES DON’T WORK

Citing a study by Rahal et al,[4] Dr. Kollef illustrated the ineffectiveness of homogeneous antibiotic use. The study hospital, which had been using cephalosporins as their “workhorse” drug, had developed a problem with resistant Klebsiella species. In 1996, the investigators implemented a guideline that limited the use of cephalosporins for one year. During that time, a carbapenem became their new workhorse.

The result was that the hospital decreased its cephalosporin use by 80%, and the incidence of resistant Klebsiella dropped by 44%. However, a 140% rise in carbapenem use was accompanied by the emergence of imipenem-resistant Pseudomonas, which increased in incidence by 57%.

“Homogeneous patterns of use aren’t going to be effective,” Dr. Kollef said. “Limiting the use of one class of compounds may be counteracted by corresponding changes in prescribing and drug resistance.”

HETEROGENEITY

According to Dr. Kollef, there are several ways of achieving heterogeneity: The LDS Hospital in Salt Lake City has an automated antibiotic consultant, which allows physicians to get rapid information on culture results, antibiotics the patient is currently receiving, and antibiotics the patient has been exposed to. In a study by Evans et al,[5] the automated antibiotic consultant was used exclusively from 1994 to 1995 in the LDS Hospital’s ICU. During that period, excessive dosing occurred 87 times with the automated consultant, compared with 405 times in the preceding two years, when the computerized system had not been used. Patients received an average of 4.7 fewer doses of antibiotics when the automated consultant was used, with an average decrease in antibiotic cost of $81 per patient.

After reviewing the LDS data, Dr. Kollef commented, “Not only does a system like that encourage heterogeneous use of antibiotics, but it tends to modify the use of drugs overall.”

CYCLING REMOVES THE PRESSURE

Between 1997 and 1999, Raymond et al[6] conducted a study of a surgical ICU in which antibiotics were administered conventionally for one year; in the following year, antibiotics were cycled on a quarterly basis. Cycling was found to reduce mortality associated with infection, as well as the incidence of gram-positive coccal and gram-negative bacillary infections; it was also an independent predictor of survival. “When you withdraw the antibiotic from the environment for a period of time, you reduce the pressure for those resistance genes and they will be deselected,” said Dr. Kollef. “By having more effective drugs in the environment, Raymond et al were seeing an improvement in outcomes.”

EDUCATION AND COOPERATION

To start a cycling protocol in an ICU requires a multidisciplinary effort. “[There must be] cooperation between the ICU team, microbiologists working in the hospital, and hospital epidemiologists,” Dr. Kollef said. “And, it requires having a feedback of information, for example, … quarterly unit-specific antibiogram data to track changes in organisms … as antibiotics are being modified.

“The ICU is just one part of the problem,” Dr. Kollef stated in his closing remarks, noting that resistance is an issue in the general population as well. “[Physicians] need to focus on the broad implications [of resistance], which include its impact on the community.”

—Gale Jurasek

References
1. Bonten MJM. Shortening the duration of therapy is the best approach to reduce the occurrence of antibiotic resistance. Presented at: Annual Meeting of the American Thoracic Society; May 20, 2002; Atlanta, Ga.
2. Kollef MH. Antibiotic cycling is the best approach to reduce the occurrence of antibiotic resistance. Presented at: Annual Meeting of the American Thoracic Society; May 20, 2002; Atlanta, Ga.
3. Lipsitch M, Samore MH. Antimicrobial use and antimicrobial resistance: a population perspective. Emerg Infect Dis. 2002;8:347-354.
4. Rahal JJ, Urban C, Horn D, et al. Class restriction of cephalosporin use to control total cephalosporin resistance in nosocomial Klebsiella. JAMA. 1998;280:1233-1237.
5. Evans RS, Pestotnik SL, Classen DC, et al. A computer-assisted management program for antibiotics and other antiinfective agents. N Engl J Med. 1998;338:232-238.
6. Raymond DP, Pelletier SJ, Crabtree TD, et al. Impact of a rotating empiric antibiotic schedule on infectious mortality in an intensive care unit. Crit Care Med. 2001;29:1101-1108.

Return to table of contents