|
August 2003 Just over three years ago, following the 4th Decennial International Conference on Nosocomial Infections, I wrote an editorial in this space that questioned the cost-benefit equation in using contact precautions for the control of vancomycin-resistant enterococci (VRE) in the hospital setting. I wondered, literally, whether the juice was worth the squeeze.
A few people, at least, expressed support that the question needed to at least be raised. In the last three years, however, not only has the question not been addressed, but the proponents of maximum effort to control VRE have seemingly become more strident than ever. In the May 2003 issue of Infection Control and Hospital Epidemiology (ICHE) there appeared a special report entitled “SHEA Guideline for Preventing Nosocomial Transmission of Multidrug-Resistant Strains of Staphylococcus aureus and Enterococcus.” Like previous Society of Healthcare and Epidemiology of America (SHEA) guidelines, and similar to the Healthcare Infection Control Practices Advisory Committee CDC Guidelines, this report consists of a focused discussion of the relevant issues, followed by a series of specific recommendations, each ranked according to the strength of the supporting literature. The first subheading in the recommendations deals with active surveillance cultures to identify the reservoir for spread of such organisms. The first nine of the specific recommendations deal with VRE, the remaining seven deal with methicillin-resistant Staphylococcus aureus (MRSA). Almost all of them are ranked as category IA or IB, that is, based on solid science (IA) or semi-solid science, but with a strong theoretical rationale (IB). For purposes of this discussion, I will focus my comments exclusively on VRE, and ignore the MRSA recommendations. The VRE recommendations deal with the need to establish an institutional program for active surveillance, the timing, frequency and breadth of the program within the institution, techniques of culturing, and the like. Again, all of them are highly ranked as being supported by strong evidence. My concern is simply this — is the evidence in support of the recommendations dealing with “active” surveillance really as solidly based in science as the authors of the SHEA document evidently believe? I am frankly skeptical. I do not profess to have read every article on the subject cited in their extensive bibliography, but I do claim to have read every article dealing with the issue that has appeared in the Journal of the American Medical Association, The New England Journal of Medicine, Annals of Internal Medicine, Clinical Infectious Diseases and SHEA’s own journal, Infection Control and Hospital Epidemiology. Most of it is descriptive, experiential, and largely uncontrolled. Those that have attempted to control their data have used historical controls, or used other hospitals with little or no evidence of comparability. My own personal evaluation of the evidence is that it is inconclusive, and therefore an unresolved issue. The first and most obvious problem with active surveillance is the sensitivity, or rather the insensitivity of the surveillance instrument, the rectal culture. Its sensitivity is at about the 60% level. It could be argued, of course, that this insensitive instrument still manages to identify those patients excreting the largest burden of VRE, and therefore the most likely to be a source of nosocomial spread, and that might indeed be true. It is not yet, however, documented scientifically. Much of the rest of the country is seemingly unconvinced as well. In recent SHEA meetings (other than 2003), the proportion of hospitals not carrying out “active” surveillance for VRE has seemed to be in the range of 60% to 70%.
The principal proponents are clearly within the Charlottesville, Va., Baltimore and CDC axis. They appear to have succeeded, however, in creating a bandwagon for active surveillance, even though that bandwagon, in the view of at least a few people like me, is not yet as scientifically based as we would wish. At what proved to be the First Decennial International Conference on Nosocomial Infections, held in 1970, Professor R. E. O. Williams, acting as rapporteur, used the term “bandwagon” to describe what many participants felt was a major need, that is, surveillance itself. The need for this was similarly not believed to be scientifically established at that time. Subsequently, of course, students of Alexander Langmuir pointed out that surveillance was the key tool necessary to develop epidemiologic data. All this would be quite arcane, and of little interest to anyone other than some hospital epidemiologists, were it not for the fact that there are some dangers in premature declaration of scientific truth before the scientific community actually agrees and supports it. The principal danger is that the error, if it should be an error, becomes locked in as policy. This has already started to happen, since it is now a SHEA policy. Will it then become a HICPAC/CDC policy? The answer is very likely “yes.” And how many years after that might it become part of a Joint Commission on Accreditation of Healthcare Organizations policy and therefore a requirement for accreditation? The second danger is as potentially damaging as the first, and that is potential damage to the academic respectability of hospital epidemiology. This has been painstakingly established over the last three decades, starting out as a wholly new discipline, and possessing today a respectable level of academic currency. This is now placed at potential risk. We do not operate in a vacuum, and the rest of the scientific community will evaluate what hospital epidemiologists claim to be scientific truth. Finally, attention to the cost-effectiveness of active surveillance for VRE has continued to be woefully lacking. The pressures for careful resource management in hospitals have increased sharply just in the last three years, and many hospital epidemiologists would have a difficult time justifying the additional expense of active surveillance for VRE, even if they believed strongly that it was the appropriate course of action. Reasonable people may interpret data differently, as lawyers are fond of pointing out, and I therefore do not believe I am being disloyal to SHEA — an organization to which I belong and support — in criticizing one segment of its report. I recognize, further, that not everyone on the Editorial Advisory Board of Infectious Disease News is likely to agree with these comments, including the current President of SHEA, Dr. Mike Tapper. He and others will be given every opportunity to respond. |
