Article Text

PDF
229 TUBERCULOSIS DIAGNOSIS AT DEATH AT AN URBAN MEDICAL CENTER: MISSED OPPORTUNITIES.
  1. M. K. Leonard1,
  2. M. Barragán1,
  3. A. Patel1,
  4. H. M. Blumberg1
  1. 1Emory University/Grady Memorial Hospital, Atlanta, GA

Abstract

Background Tuberculosis (TB) cases in the United States are concentrated among underserved populations in the inner city. Grady Memorial Hospital (GMH) cares for 80% of patients with TB in Atlanta and 20% of all TB cases in Georgia. An unfortunate consequence of late diagnosis is that some patients are initially diagnosed with TB very late in their disease process, resulting in an unacceptably high mortality rate shortly thereafter. Regretfully, TB is sometimes diagnosed postmortem. Delayed diagnosis has major public health implications with regard to the persistent transmission of TB in the community and in hospital settings.

Methods As part of a larger TB mortality study, we identified patients seen at Grady Memorial Hospital, a 1,000-bed inner-city university-affiliated public hospital in Atlanta, GA, who were diagnosed with TB postmortem by performing a retrospective review of medical records from June 1993 to June 2004. Death at diagnosis was confirmed by postmortem culture results, autopsy findings, and/or dying within 72 hours of presentation.

Results In a preliminary study, over an 11-year period, 20 cases of missed diagnoses of TB were identified. All of the patients were African American, 13 (65%) were male, and the mean age at death was 47.8 (range 24-81). Eighteen (90%) were culture confirmed, whereas 2 (10%) were clinical cases. Of these 18, 13 (72%) had pulmonary TB, whereas 5 had extrapulmonary TB; among these 5, 4 had disseminated TB, whereas 1 had pleural. Five (25%) cases were identified at autopsy. Eleven (55%) were HIV positive with a mean CD4 of 89.8, median 51.5 (range 8-290). Eight (40%) patients were in the ICU, and 6 of the 8 (75%) were on ventilators. Radiologic findings at admission (n = 18) were as follows: miliary (2, 11%), interstitial (5, 28%), cavitary (1, 0.06%), lobar infiltrate (1, 0.06%), pleural effusion (8, 44%), and normal (1, 0.06%). Despite radiologic findings and admission diagnoses of AIDS, pneumonia, weight loss, or dyspnea, TB was considered for 8 patients and only 6 received presumptive therapy within 72 hours prior to death.

Conclusions In areas of high prevalence, TB is frequently underdiagnosed. Furthermore, given the protean clinical features of TB and its ability to mimic other diseases, its diagnosis may be missed or overlooked. Our results demonstrate that in many cases, TB is diagnosed late or even at autopsy. Among these cases, there were many missed opportunities to diagnose and treat TB. Clinicians treating patients with TB risk factors should have a low threshold of suspicion in diagnosing TB and should consider treating empirically while awaiting the results of smears, cultures, and molecular tests in some cases.

Statistics from Altmetric.com

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.