Vancomycin-resistant enterococci (VRE) are a still growing concern in hospital units tending to seriously ill patients. VRE are easily transmitted via healthcare workers’ hands or clothes, or via contaminated instruments or environment [1]. The risk of VRE colonization resulting in systemic infection is substantially increased in immunosuppressed subjects, such as cancer patients or transplant recipients [2], and these infections are difficult to treat. Strategies to prevent VRE transmission include the use of barrier precaution, the isolation of colonized patients and active surveillance cultures to identify asymptomatically colonized patient. The effectiveness of active surveillance cultures that are cumulative expensive and resource intensive, in reducing VRE transmission remains controversial [3]. However, the non-compliance with active surveillance programs may outweigh the added expense of the program itself.

In this setting, this work evaluated the cost of a failure in the active surveillance of VRE that had resulted in an outbreak in a French University Hospital.

Overall, the VRE colonized 13 patients. In one patient, VRE was also recovered from biliary drainage fluid. Two of them could be discharged directly from the HBC, the remaining eleven were treated in the IDU. No new colonization occurred in the HBC after 21 October — less than two months after discovery of the index case — and no case was observed among VRE-negative patients from the IDU seven weeks after the index case admission. The isolation sector was cancelled out on 1st December, after 60 days functioning.

The median age of the colonized patients was 59 years and most of the patients (83%) were male. Their underlying conditions were the following: hepatocellular carcinoma (n=7), uncompensated liver cirrhosis (n=9), liver transplantation (n=3), chronic viral hepatitis (n=4), HIV infection (n=1), cholangiocarcinoma (n=1) and gastric carcinoma (n=1). Six patients died from their underlying disease or from other infections. None of these deaths was related to the VRE. All the patients were successfully decontaminated as shown by three consecutive negative weekly cultures.

The median length of stay in hospital was 59 days (range, 22–77 days) whereas it was 29 days (range, 4–68 days) for the six patients that were in intensive care unit. In most patients, the length of stay was determined by the underlying condition or complications thereof. Yet, for one patient, transfer to an aftercare facility was delayed because of VRE colonization.

Liver transplantation activity, in the HBC was not impaired by the outbreak: 138 transplantations were performed in 2007, 130 in 2008 and 116 in 2009.

This report indicates that eradicating even a limited outbreak requires substantial efforts and resources. Here, the outbreak was triggered by a failure in the systematic VRE screening. Indeed, the index patient was a candidate for liver transplantation. Owing to the increasing risk of VRE infection in transplant recipients, this patient should have been screened for VRE even if he was admitted in a non-hepatobiliary ward. Furthermore, the index patient came from Portugal where the prevalence of VRE among clinical Enterococcus faecium is approximately 25% [4]. Thus, this patient belonged to a population “at risk” for VRE colonization. Altogether, this suggests that the active surveillance program has failed. Such failure has been previously reported in liver transplant ICUs where active surveillance cultures were performed only on approximately 50% of the patients within 24h of admission [5]. In this study [5], the failure has been imputed to staff workload. In our study, same reason may have caused the nonfulfillment of systematic VRE screening. Altogether, this underlines that the implementation of the active surveillance program could have some limitations in the real life setting.

Despite this failure, the time taken to eradicate the VRE outbreak was rapid as it was less than two months. Indeed, experiences suggest than any delay in appropriate measures strengthens the threat of endemicity [6]. In this study, the involvement of the infection disease unit and hospital management provided the necessary leadership to enforce strict control measures. Consistently, no new colonization occurred once colonized patients were identified and transferred to infectious disease unit. This highlight that the isolation of colonized patient should not be delayed.

The cost of these measures that include the staffing, the barrier precautions and contact isolation far exceeded the cost of active surveillance cultures. However, one might wonder whether resort to such expensive measure was mandatory. Indeed, nosocomial pathogen outbreaks often trigger a debate between advocates of strict control measures and those who favor a less resource-intensive attitude, preserving usual clinical activity [7]. The Western Australian (WA) experience has convincingly shown that enhanced infection control practices are able to prevent transition from a large hospital outbreak to endemicity [8]. An often-debated measure is isolation, as opposed to in situ barrier precautions. In many hospitals, isolation sectors have to be set up by subtracting beds to routine clinical activity, which often results in a loss of income for the hospital since beds are then reserved for newly colonized or infected patients [6, 7]. The case for isolation is not fully substantiated [9]. Simple cohorting may be sufficient where adequate architectural conditions are met (single-bed rooms, separate nurse station) and trained staff is present. Such conditions are often met in infectious disease units. Finally, the appeal of a resource-sparing infection control policy should be balanced against the likely and lasting cost of endemicity [10]. Estimating the cost-benefit of the eradication of a hospital outbreak is difficult as it depends on several factors, notably the delay before the next outbreak. However, Montecalvo et al.[10] showed that even in an endemic setting strict infection control measures may yield back up to 2.70 dollars for each dollar spent.

Our report both underlines the burdensome and costly consequence of the non-compliance with active surveillance programs and the benefice of this program as the outbreak had been halted owing to the implementation of the VRE screening. This further suggests that beliefs related to the potential cost of VRE active surveillance programs should be balanced with the cost of the eradication of VRE outbreak. Thus, in the view of reducing health care cost, effort has to be made to strictly adhere to active surveillance program.

We acknowledge that the data presented here are somewhat anecdotal. They bear only on a limited outbreak, in a single clinical setting, within the framework of the French healthcare system. However, the failure resulting in this outbreak and caused by reduced resource in the summertime may likely be happened in the future. Indeed, the healthcare resource may be reduced due to the economic crisis. In this setting, our study warns that the improper implementation of active surveillance cultures related to lack of resource might increase VRE transmission and the associated expenses.

A partial report of these data has been presented as a poster at “Dixièmes Journées Nationales d’Infectiologie”, 10-12 June 2009, Lyon, France.