Daniel M. Keller, PhD

September 25, 2012

September 25, 2012 (San Francisco, California) — A French hospital has reduced the duration of carbapenems use by an average of 4 days with an antibiotic management team (AMT) that controls the daily dispensing of these antibiotics on the basis of real-time microbiological results.

Lead investigator Carine Couzigou, MD, from the infection control unit and committee at Groupe Hospitalier Paris Saint-Joseph in France, who is a member of the AMT, presented the study findings in a poster session here at the 52nd Interscience Conference on Antimicrobial Agents and Chemotherapy.

Dr. Couzigou said their system could help prevent the emergence of carbapenemase-producing enterobacteria in hospitals.

On the AMT is an infectious diseases specialist and an intern with infectious diseases training. A team member makes daily rounds in all hospital units, providing real-time monitoring of antimicrobial use. For each case, the infectious diseases specialist assesses whether or not the carbapenem should be continued and transmits that information to the dispensing pharmacy. In addition, members of the AMT provide intervention and education when recommending changes in treatment, which can lead to the development of local therapeutic guidelines.

During a 6-month study period last year, the investigators evaluated the effect of the AMT on the duration of carbapenem prescription at their 534-bed acute care hospital. For every prescription of a carbapenem, the prescribing physician must provide written justification and specify the expected duration of use on an order form.

The investigators compared the duration of therapy expected by the prescribing physician and the actual duration of therapy after the 58 prescriptions (36 prescriptions for imipenem, 15 for ertapenem, and 7 for meropenem) were assessed by the ATM. The 3 patients who were transferred or died were excluded from the analysis.

The most common reasons for a carbapenem prescription were urinary tract infection (n = 18), intraabdominal infection (n = 9), and pneumonia (n = 9). The most common colonizing or infecting organisms (33 cases, 57%) were extended-spectrum beta-lactamase-producing enterobacteria. However, in 34% of cases, the infection was not documented microbiologically.

There was a mean delay of 1.6 days before evaluation by the AMT. The team agreed with 62% of the carbapenem prescriptions and disagreed with 38%. In cases of disagreement, the carbapenem was stopped because there was no evidence of infection or because the patient was switched to a narrower-spectrum antibiotic. Prescribing physicians complied with the AMT recommendations 100% of the time.

With the AMT, carbapenem administration was reduced by a median of 4 days. The mean duration originally planned by the prescribing physicians was 11 days (median, 7 days). After assessment, the actual mean duration of use was 7 days (median, 4 days) (P = .004).

Dr. Couzigou told Medscape Medical News that using an AMT is feasible only if there is an adequate number of infectious disease specialists. The system "needs a little money from the hospital to have an infectious disease specialist who will check for carbapenem monitoring," she advised.

Pharmacists might be able to oversee the use of carbapenem for uncomplicated infections, such as urinary tract infections, "but for more complicated ones, an infectious disease specialist would still be required," she said.

Poster session moderator Stephan Harbarth, MD, MS, associate professor of medicine, attending physician in infectious diseases, and associate hospital epidemiologist at Geneva University Hospitals in Switzerland, who was not involved in the study, told Medscape Medical News that this small study is interesting because the investigators "tried to really monitor and supervise the correct use of carbapenems."

"That's what we do in Geneva.... We often see that carbapenem use is not appropriate and tell the colleagues to go for more narrow-spectrum antibiotics. But we don't do specific carbapenem audits. It's too time consuming," he said.

In his opinion, the AMT system used in this study is not be feasible in the long term for most hospitals because it requires too much time. "Rather, the way to go is a computerized decision-support system with warnings; if somebody prescribes more than 3 days' worth of carbapenems, an alert pops up so physicians can adjust or at least consider decreasing the duration of carbapenem use," he explained.

Dr. Couzigou agrees that computerized systems can help identify patients reaching or exceeding a set duration of carbapenem and alert the AMT. But she expects her hospital will not have such a system for 2 or 3 more years.

Aside from deciding whether to continue carbapenem or not, Dr. Harbarth said it is important to check dosing. He cited data showing that for certain patients, especially in the intensive care unit, carbapenems are underdosed, particularly in young septic patients.

"I think that this is a much more interesting question at the moment. How can we better tailor the dosing schemes to the needs of individual patients?" he asked. "This kind of therapeutic drug monitoring is expensive. You need the infrastructure, and many hospitals don't do it on a regular basis. This is certainly something we have to work on for the future," he noted.

The study received no commercial funding. Dr. Couzigou and Dr. Harbarth have disclosed no relevant financial relationships.

52nd Interscience Conference on Antimicrobial Agents and Chemotherapy (ICAAC): Abstract K-234. Presented September 9, 2012.