On the Contagious Nature of Tuberculosis (Continued)

Kent A. Sepkowitz, M.D.

Memorial Sloan-Kettering Cancer Center, New York, New York

	ARTICLE

TOP ARTICLE REFERENCES

In this issue of the AJRCCM (pp. 927-933), the admirable investigators of the Canadian Collaborative Group in Nosocomial Transmission of Tuberculosis present the latest installment in their story of how and why and when TB is transmitted in hospitals (1). They previously have furthered our understanding in many related areas: the role of ventilation in TB control, the fate of a positive tuberculin reactor, and the association between job duties and risk of occupational TB (2, 3). This time around, they focus on the consequences of delayed TB diagnosis on both patient outcome and on the risk of purified protein derivative (PPD) conversion among workers caring for such patients.

Their report on the causes ofand deleterious results from diagnostic delay aligns with most previous reports. Early diagnosis of an infectious disease is always a good idea. In addition, the authors find an assuring association between clinical experience managing a disease and favorable patient outcome. Similar to articles demonstrating comparable benefit in cancer (4) and myocardial infarction (5), outcome was improved for those cared for in hospitals with high rates of TB. At play here, however, is the uneasy tension between the need for all services to be available on demand to any and every patient versus the advantages of a specialist in a specialty hospital.

The more remarkable, and unique, aspect of this thorough study is the authors' ability to finally establish a connection between the missed diagnosis of tuberculosis and subsequent nosocomial spread to health care workers. For years, experts have stated that the overlooked case, rather than the one already diagnosed, posed the biggest risk for nosocomial spread. This old saw has existed for decades as found truth, but with little objective backing. Certainly the assumption is easy enough to swallow: in a hospital where TB is but one of countless potential diagnoses (less than 0.05% of all admissions in these hospitals), the likelihood of prompt diagnosis on every last case is low. It is a perverse testament to the wily acid fast bacillus that in the era of molecular diagnosis, real-time PCR, and all the rest, TB continues to escape routine diagnosis with alarming regularity. So given that diagnostic delays predictably occur, it makes sense that undiagnosed patients would, as they careen around the hospital to X-ray, to the pulmonary functions lab, to the patient lounge, spread TB as they went.

The proof for this phenomenon, however, has been scant (6). The authors have provided us with a simple and great fact: yes, Virginia, the undiagnosed case of tuberculosis does indeed represent a substantial risk to the tuberculin-negative hospital worker. With this building block in place after all of these years, an even more familiar and important "unproved fact" of TB mythology can be taken on. This one pertains to the relative contribution of community versus nosocomial transmission of TB. To dateremarkablyoutside the outbreak setting and a few isolated reports, the prevailing wisdom has been that community exposure accounts for the bulk of acute TB infections in hospital workers. Reporting on a 10-hospital survey of PPD results among health care workers in 1984, CDC authors could find "no correlation between patient exposure or job classification and the prevalence of significant [PPD] reactions" (7).

Yet the evidence supporting a predominance of community transmission too rests on shaky ground. As such, it resembles what these authors describe as "trying to see the wind": trying to document the non-event, the unseen transmission. Very few individual cases of TB result in conversions, leading many to conclude that risk simply is not there. Hospitals, for example, test hundreds and thousands of workers annually and seldom see a cluster or a worrisome trail of acute infection.

This article would argue otherwise. Looking not at individual exposures, but rather the aggregate of hundreds of workers caring for patients for many years at 17 different hospitals, they are able to clearly demonstrate a "dose-response" association between overall institutional rates of unsuspected, preliminarily undiagnosed TB and the rate of tuberculin conversion among the workers. By broadening their net across time zones, years, and buildings, the authors have overcome the problem inherent in smaller inquiries. And, given that none of the hospitals had a known outbreak during the study period and, furthermore, that the hospitals, under the steady leadership of the authors, could be expected to have a thoughtful approach to TB control, the only explanation can be that nosocomial transmission occurred during these years in these hospitals. And finally, we must conclude that nosocomial transmission, not community exposure, still represents a substantial and inexplicably underappreciated risk for hospital workers throughout North America.

Finally, and perhaps most impressive and important of all, this article and the work that the group has done before it show the potent benefit of collaborative studies. It is somewhat embarrassing that these Canadian investigators, miles and miles away from the multidrug-resistant TB outbreaks that nearly paralyzed several urban areas in the U.S. a decade ago, were able to calmly and soberly develop a program to answer the fundamental questions that the U.S. outbreaks urgently posed: How does TB spread in hospitals? What is the role of ventilation? What is the contribution of a childhood BCG vaccination? Does clinical experience matter? It is a puzzle why no group in the U.S. stepped forward to confront these same questions. One suspects that many good ideas were left stranded at the planning stage, victims of posturing and egos and the entire nasty sideshow that characterizes so many initially honorably academic pursuits. In the meantime, our Canadian colleagues have executed a program that is a model of simplicity, flexibility, and productivity. Perhaps in addition to learning our lessons about the consequences of delayed TB diagnosis, we would be well-served to learn the benefits of real collaborative work from the Canadian Collaborative Group in Nosocomial Transmission of TB.

	References

TOP ARTICLE REFERENCES

1. Greenaway C, Menzies D, Fanning A, Grewal R, FitzGerald M, and The Canadian Collaborative Group in Nosocomial Transmission of Tuberculosis. Delay in diagnosis among hospitalized patients with active tuberculosis predictors and outcomes. Am J Respir Crit Care Med 2002;165:927-933.

2. Menzies D, Fanning A, Yuan L, FitzGerald JM. Tuberculosis in health care workers: a multicentre Canadian prevalence survey: preliminary results. Canadian Collaborative Group in Nosocomial Transmission of Tuberculosis. Int J Tuberc Lung Dis 1998;2(9 Suppl 1):S98-S102.

3. Menzies D, Fanning A, Yuan L, FitzGerald JM. Hospital ventilation and risk for tuberculous infection in Canadian health care workers. Canadian Collaborative Group in Nosocomial Transmission of TB. Ann Intern Med 2000; 133: 779-789 [Abstract/Free Full Text].

4. Begg CB, Cramer LD, Hoskins WJ, Brennan MF. Impact of hospital volume on operative mortality for major cancer surgery. JAMA 1998; 280: 1747-1751 [Abstract/Free Full Text].

5. Thiemann DR, Coresh J, Oetgen WJ, Powe NR. N Engl J Med 1999; 340: 1640-1648 [Abstract/Free Full Text].

6. Blumberg HW, Watkins DL, Berschling JD, Antle A, Moore P, White N, Hunter M, Green B, Ray SM, McGowan JE Jr.. Preventing the nosocomial transmission of tuberculosis. Ann Intern Med 1995; 122: 658-663 [Abstract/Free Full Text].

7. Snider DE Jr,, Cauthen GM. Tuberculin skin testing of hospital employees: infection, "boosting," and two-step testing. Am J Infect Control 1984; 12: 305-311 [Medline].