Bacteremia Traced to Illicit Narcotic Use by Health Care Worker

Abstract & Commentary

Synopsis: A health care worker caused an outbreak of 26 cases of Serratia marcescens bacteremia in a surgical ICU to contamination of parenteral fentanyl infusions during illicit narcotic use.

Source: Ostrowsky BE, et al. Serratia marcescens bacteremia traced to an infused narcotic. N Engl J Med. 2002;346:1529-1537.

Over a 3-month period, 9 cases of Serratia marcescens bacteremia were detected in the surgical ICU of a 455-bed tertiary care facility. Because of the rarity of this infection and the similar antimicrobial-susceptibility patterns of the isolates, the latter were sent for genotyping. When the genotype could not be determined, the specimens were sent to the Centers for Disease Control and Prevention (CDC) for further analysis. Over the next 4 months, 17 more patients acquired S marcescens bacteremia and the CDC was invited to assist in identifying the source.

Surgical ICU patients with bacteremia during the epidemic period (case patients) were compared with randomly selected controls (patients who stayed in the surgical ICU for > 48 hours during the epidemic period without acquiring bacteremia). To assess health care worker exposure, physicians, nurses, and respiratory therapists who had exposure to these patients were identified. In univariate analysis, patients with S marcescens were more likely to have received fentanyl, particularly a continuous infusion, and the median days of infusion and median total dose were higher compared to controls. For the majority of cases (65%), the positive blood culture was collected while they were receiving fentanyl or within 24 hours of receiving a fentanyl infusion.

In multivariate analysis, only receipt of a continuous fentanyl infusion and receipt of care from 2 respiratory therapists were independent risk factors. Independent of this investigation, a report was filed by a nurse in the surgical ICU who witnessed one of these therapists manipulating IV infusions. The therapist agreed to hair sample testing and the test was positive for fentanyl. Employment was terminated and no additional cases were reported.

Comment by Leslie A. Hoffman, RN, PhD

Surveillance of nosocomial infections was recognized to be a major component of infection control in the late 1970s. Surveillance includes continuous monitoring of different infections or microorganisms to detect outbreaks that require specific and emergency measures. Using these principles, Ostrowsky and colleagues were able to trace the source of contamination to an unexpected source. Estimates based on the 1991 National Household Survey on Drug Abuse indicate that 4.2% of hospital workers acknowledge current use of illicit drugs and 8.9% report prior use of such drugs. Clues that might suggest drug use include mood swings, unexplained behavior or absences, repeated health problems, marital discord, bizarre working hours, medical errors, and arrests for driving under the influence of alcohol. Health care providers commonly hesitate to confront a colleague who appears to be in trouble with drugs or alcohol. By doing so, they facilitate the addict’s denial and enable the drug use to continue.

In this investigation, hair testing documented fentanyl abuse by the implicated health care worker. Hair testing offers the potential of detecting drug use over a longer period than a blood or urine sample. A 10-13 cm (4-5 inch) sample of hair can provide evidence of drug use for as long as 10 months previously since hair grows about 1 cm (0.4 inches) per month. Ostrowsky et al hypothesized that reuse of needles to remove fentanyl or replacement of fentanyl with contaminated liquid may have led to inadvertent contamination with S marcescens. A hospital official presented the case to the area district attorney who, implausibly, refused to pursue legal action because of insufficient evidence.

Dr. Hoffman is Professor, Medical-Surgical Nursing of Acute/Tertiary Care, University of Pittsburgh, School of Nursing.

Special Feature