Duel in the sun: Two epi’ giants face off at APIC
ICPs may comment on CDC draft until Aug. 13
[Editor’s note: Two of the nation’s leading health care epidemiologists recently held a provocative debate on the controversial topic of active surveillance cultures for multidrug-resistant pathogens. As infection control professionals are well aware, the topic has become one of the hottest issues in the field. Accordingly, the debate was recently held in Phoenix at the annual conference of the Association for Professionals in Infection Control and Epidemiology (APIC). To capture the original flow of the exchange and rebuttal, we present the key points as they were made during the debate, with the results of the electronic voting by the audience at the end of this article.]
In keeping with a political season marked by polarization and acrimony, two of the nation’s leading health care epidemiologists recently squared off in a spirited debate about the use of active surveillance cultures to detect and isolate incoming patients with multidrug-resistant organisms (MDROs). The issue may seem mundane to those unversed in infection control, but the emergence of vancomycin-resistant staph strains has raised the stakes considerably on controlling MDROs.
In addition, the debate revealed the deep schism within the medical epidemiology community about how — and how aggressively — to attack the problem of emerging antibiotic resistance. It was telling that after two hours of point and counterpoint, an ICP in the audience seemed to sum up the mood of many in saying that her "feet were planted firmly in the air."
Here’s the background in a nutshell: The Society for Healthcare Epidemiology of America (SHEA) is urging active surveillance, which calls for culturing the nares of targeted patients on admission or periodically thereafter to detect and isolate the reservoir of resistant organisms.1
The primary pathogens are methicillin-resistant Staphylococcus aureus (MRSA) and vancomycin-resistant enterococci (VRE). The guidelines recommend the practice so colonized patients can be placed in contact isolation rather than serving as an undetected source to spread the pathogens to other patients.
More recently, new draft patient isolation guidelines were issued by the Centers for Disease Control and Prevention’s (CDC) Healthcare Infection Control Practices Advisory Committee (HICPAC).2 (See box, p. 104.) Rather than endorsing the SHEA recommendation on active surveillance culturing, the HICPAC guidelines establish a two-tiered approach that calls for more aggressive measures only in the face of ongoing transmission or if prevalence exceeds institutional goals.
Debating the respective positions at APIC were two epidemiologists with decades of experience in infection control.
Advocating the SHEA approach was William Jarvis, MD, a former top health care epidemiologist at the CDC who has left the agency to pursue private consulting.
Arguing in favor of the HICPAC approach was William Scheckler, MD, a member of the HICPAC panel that crafted the guidelines and, ironically, one of the founding members of SHEA.
The thousands of ICPs looking on at the APIC meeting included several hundred equipped with electronic voting devices to get a sample of audience opinion at the end of the exchange. Jarvis took the podium first and made it clear he would pull no punches with his old colleagues at the CDC.
ICPs urged to protest CDC inaction
"In the early 1970s, infection control programs were created to control antibiotic-resistant pathogens causing health care-associated infections," he said. "Between 1970 and 2004, the incidence and prevalence of antibiotic-resistant pathogens causing hospital-acquired infections has continued to rise. MRSA and VRE have become endemic in many, if not most, of your hospitals despite the recommendations of CDC and HICPAC for standard and contact isolation."
Indeed, the CDC’s own surveillance data chronicles the failure of the recommendations, with hospitals in its sentinel system showing a relentless onslaught of thousands of MRSA infections annually. "I hope that some of you have relatives or friends that have MRSA infections because we need to personalize this," Jarvis said. "These are infections that can be prevented and should be prevented. That’s why we are all in this room to begin with and why we went into infection control. . . . Why should we continue to do what we have done — which has failed — or even do less? We continue to ignore the source of the problem — the colonized patient."
Citing the well-known active surveillance program directed by Barry Farr, MD, hospital epidemiologist at the University of Virginia in Charlottesville, Jarvis said even "very conservative" cost estimates show the practice is cost-effective. "If you assume that there was a continuation of the 50% MRSA prevalence from the time of the [active surveillance] intervention for the 11 years after that; and if you assume . . . that the cost of any infection was $3,420, then the savings annually were $431,000," Jarvis said. "So Barry Farr has saved his hospital almost $5 million in 11 years."
In contrast, less aggressive approaches to MDROs can lead to a situation like that experienced by hospitals in the San Francisco Bay area in the early 1990s, he added. "In 1993, they had one [VRE] isolate at one hospital; and within five years, all the hospitals had VRE and they had over 100 bloodstream infections," Jarvis said. Taking a shot at the CDC’s increasing emphasis on patient safety, he added, "I do not call this protecting patients or making prevention primary."
Citing a wealth of research and references, Jarvis argued that the SHEA "seek-and-destroy" approach has been validated in Europe and in the United States centers that have tried it.
"This may all be by just chance alone, but I don’t think there is anybody in this room who can stand up and say that MRSA came into your facility, became endemic, and left," he said. "If there is, please stand up."
Colonized patients contaminate both health care workers and the environment just like infected patients, he argued. Yet routine clinical cultures do not detect most colonized patients, meaning that active surveillance cultures are necessary if the unrecognized reservoir is to be detected.
"The infection control community should demand that CDC and HICPAC take a more aggressive approach to controlling antibiotic-resistant pathogens," he told APIC attendees.
"They should revise the HICPAC isolation guideline to recommend the SHEA approach so you can go to administrators and get support to do it. . . . Write to [Secretary of Health] Tommy Thompson; write to CDC, because this is a political issue. It is not a scientific issue. I have shown you reams of evidence-based data. . . . Demand that the SHEA guideline approach — a proven and evidence-based approach — be adopted by CDC and HICPAC."
If I had a hammer
As the wave of applause for Jarvis gradually faded, Scheckler rose to the HICPAC defense, using a quote by Mark Twain to set up an opening salvo. "To a man with a new hammer, everything looks like a nail. Active surveillance is their hammer," he said.
While SHEA focused on MRSA and VRE, the HICPAC guidelines have to address multiple drug-resistant gram-negative bacilli as well, covering a broad variety of organisms including Pseudomonas, he noted. In addition, the guiding HICPAC principle is to make the recommendations applicable to all settings, not just hospitals. But even if applied only to hospitals, a blanket recommendation for active surveillance would not be valid because of the differences among those institutions, he added.
Noting that he has been a epidemiologist at a community hospital for 34 years — St. Mary’s in Madison, WI — Scheckler said the lion’s share of the active surveillance data is from teaching hospitals and major academic centers. "To take data from academic centers and assume that they are extrapolative to all hospitals, I think is questionable," he said. By the same token, the efficacy of data cannot necessarily be extrapolated across medical units, just as data collected in controlling an outbreak are not necessarily relevant to nonoutbreak, endemic problems. Typically, in addition to active surveillance cultures, the studies cited by its proponents include seven or eight other simultaneously administered infection control interventions.
"Active surveillance cultures of selected populations has rarely, if ever, been done as a single intervention," Scheckler explained. "So to try to tease out the one that made the difference is difficult. . . . There is total absence of randomized controlled trials and a paucity of comparative studies."
In addition, putting more patients in isolation should not be undertaken lightly, as some studies are now suggesting that these patients receive compromised medical care, he added. Tackling the cost issue, Scheckler said there is a difference between cost and charges, so the money spent on surveillance cultures is not necessarily recouped in prevented infections. "None of these costs can be charged to the patient; they certainly can’t be charged to Medicare," he said, noting that he estimated that it would cost him $200,000 to $300,000 annually to undertake the practice at his hospital.
Moreover, Scheckler showed data that show his hospital is preventing and controlling MDROs in the absence of active surveillance cultures.
"We have no evidence of MRSA spread [or] VRE spread, and we do look for it, Dr. Jarvis," he said. "Targeted surveillance for MDRO prevalence is a reasonable thing to do if you have a problem. We don’t all have a problem."
Thus a flexible approach using a hierarchy of controls is warranted. "We have to have an understanding of the diversity of health care settings," he said. "The flexible approach does require some expertise, but you already have that. Most importantly, I would hate to lose [staff] positions because I am spending all of my money on cultures. I think you will find that the HICPAC guideline — when you look through it — meets all the requirements of evidence and is the one place where we ought to go."
Jarvis opened his 10-minute rebuttal by chastising Scheckler for using his own hospital’s "anecdotal experience" to defend a HICPAC guideline that is supposedly evidence-based. He rejected the inflexible charge, saying the SHEA guideline allows ICPs to select the patient populations they want to culture. "There actually is much more flexibility in the SHEA guideline than there is in the HICPAC guideline," he said.
While questioning culturing cost estimates as too high, Jarvis says the practice is cost-effective because it reduces or eliminates costs linked to infections and extended lengths of stay. "The fact is those surveillance cultures are much less expensive than one MRSA infection that is costing you $27,000," he said.
Arguing that there are plenty of data justifying the practice at various types of hospitals, Jarvis rejected the demand for controlled randomized trials with a hilarious bit on parachutes. "There are no randomized controlled trials of parachute interventions," he said in mock seriousness to much audience laughter. "The basis for parachute use is purely observational. Individuals who insist that all interventions need to be validated by a randomized controlled trial need to come down to earth — with a thud."
Scheckler took a conciliatory tact in his rebuttal, reminding the audience that the two guidelines aren’t all that different. "If you have a problem — as defined in a variety of ways in the HICPAC guidelines — then you do need to do the intensified approach, which is basically the SHEA approach," he said. "We don’t really disagree; we just say we need the maximum amount of flexibility to do these things."
Scheckler went for the jugular shortly thereafter, however, saying he was aware that SHEA members were split on their own guideline, which was approved by "a Florida margin" by a split SHEA task force. "I think you need to know that to understand why we are having a debate today at all," he said. "Because there is a disagreement. I will say it is honest disagreement, but there was substantial disagreement within the [SHEA] organization of hospital epidemiologists as well."
The verdict: A virtual draw
At the end of the debate, the audience members with electronic voting devices gave these results to the following three statements:
1. I agree with the SHEA guideline for control of MRSA and VRE and plan to implement it or have implemented it.
A. Agree: 26%
B. Disagree: 28%
C. Undecided: 46%
2. I agree with the new HICPAC guideline concerning MDROs as presented today and plan to implement it as soon as it is available.
A. Yes: 28 %
B. No: 17%
C Undecided: 55%
3. Both guidelines are valid, and either could be used.
A. Agree: 51%
B. Disagree: 22%
C. No opinion: 27%
1. Muto CA, Jernigan JA, Ostrowsky BE, et al. Special report: SHEA guideline for preventing nosocomial transmission of multidrug-resistant strains of Staphylococcus aureus and Enterococcus. Infect Control Hosp Epidemiol 2003; 24:362-386.
2. Centers for Disease Control and Prevention. Healthcare Practices Infection Control Advisory Committee (HICPAC). Draft Guideline for Isolation Precautions: Preventing Transmission of Infectious Agents in Healthcare Settings 2004. Atlanta: 2004. Web: www.cdc.gov/ncidod/hip/isoguide.htm.