How Many Sputum Specimens are Necessary to Rule Out TB?
Abstract & Commentary
David J. Pierson, MD, Pulmonary and Critical Care Medicine, Harborview Medical Center, University of Washington, is Editor for Critical Care Alert.
Synopsis: Among patients placed in respiratory isolation because of suspected tuberculosis, the diagnosis could be made as accurately with 2 sputum specimens as with the currently recommended 3.
Source: Leonard MK, et al. Am J Infect Control. 2005; 33:58-61.
Leonard and colleagues reviewed all the sputum specimens obtained to test for tuberculosis for nearly 4 years in a large public teaching hospital. Sensitivities of first, second, and third acid-fast bacilli (AFB) smears were calculated overall and also separately for HIV-negative and HIV-positive patients. Patients from whom less than 3 sputum specimens were collected were excluded from calculation of sensitivity of second and third specimens. All sputum cultures that eventually proved to be positive for Mycobacterium tuberculosis were identified and used in determining sensitivities.
Between January 1, 1997, and October 1, 2000, 425 patients were diagnosed with culture-proven pulmonary tuberculosis. AFB smears and cultures were performed on 951 sputum specimens from these 425 patients; 61% of all respiratory samples were positive on smear for AFB. Leonard et al compared the demographics (age, sex, race) of patients with at least 1 positive smear to those of patients with 3 negative smears, and found no significant differences.
The sputum specimens from 147 patients who had at least 3 specimens obtained were used in calculating the sensitivity of the second and third specimens. Overall, the sensitivity of a positive AFB sputum smear was 67% with 1 sputum specimen, 71% with 2, and 72% with 3 specimens.
Also, 239 of the 425 patients were HIV-negative, and 142 were HIV-positive. The sensitivities of the AFB smear for HIV-negative patients were 75%, 79%, and 80% with 1, 2, and 3 specimens, respectively, collected for analysis. This compared with 57%, 61%, and 62%, respectively, for HIV-infected patients. Leonard et al conclude that 2 specimens were adequate for establishing the diagnosis of tuberculosis, and that the third specimen added little diagnostic value.
Comment by David J. Pierson, MD
At Grady Memorial Hospital in Atlanta, where this study was carried out, about 1500 patients are isolated each year for suspected pulmonary tuberculosis, and approximately 150 cases of tuberculosis are diagnosed. Tuberculosis is much more prevalent at that institution than in most US hospitals, and vary large numbers of patients are isolated because of concern about this diagnosis. Placing patients in respiratory isolation is cumbersome and expensive, and it would be advantageous if the duration of isolation for patients who prove not to have tuberculosis could be minimized.
This study failed to demonstrate any substantial increase in the sensitivity of sputum smears for tuberculosis with sequential specimens, with essentially no difference between 2 and 3 specimens. These findings are consistent with those of a study at Harborview Medical Center in Seattle a few years ago.1 In that study, also a retrospective review of all culture-positive sputum specimens for tuberculosis, AFB staining of a single sputum specimen was sufficient to establish the diagnosis in all HIV-infected patients with pulmonary tuberculosis. Even in HIV-negative patients, a single negative sputum smear made the diagnosis of tuberculosis much less likely, although a minimum of 2 sputum specimens was necessary to achieve an acceptable early diagnostic yield. Partly as a result of this study, infection control policy was changed in that institution with respect to when patients can be taken out of respiratory isolation. Now, 2 negative sputum smears are sufficient.
It is important to point out that, in both of these studies, a significant number of smear-negative sputum cultures eventually turned out to be positive for tuberculosis. While a negative-AFB smear does not rule out tuberculosis as the diagnosis, it suggests that the number of infectious organisms being shed via the respiratory tract is relatively small, and that the risk of spreading the infection is relatively low. Thus, 2 negative-sputum smears means that the likelihood that health care workers or other patients will become infected with tuberculosis from that patient during the hospitalization is low, and that the patient may be taken out of respiratory isolation, although it does not completely rule out that diagnosis in the patient.
Reference
- Finch D, Beaty CD. Chest. 1997;111:1174-1179.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content