TB Notes Newsletter
No. 3, 2013
CLINICAL RESEARCH BRANCH UPDATE
Results of TBESC Task Order 27: Health-System Benefits and Cost-Effectiveness of Using Mycobacterium Tuberculosis Direct Nucleic Acid Amplification Testing to Diagnose Tuberculosis Disease in the United States
Final results of the Tuberculosis Epidemiologic Studies Consortium’s Task Order 27 have been published in the journal Clinical Infectious Diseases.1 Investigators ascertained the health-system benefits and cost-effectiveness of using the Mycobacterium Tuberculosis Direct (MTD) nucleic acid amplification test (NAAT) (Hologic Gen-Probe Incorporated, San Diego, CA) to diagnose TB disease. Participating sites included metropolitan Atlanta, Georgia; the states of Hawaii and Massachusetts; and four areas of Maryland. The sites represented the diverse patterns of TB diagnosis in the United States, with nearly all patients being hospitalized in Atlanta, a small portion being hospitalized in Hawaii, and mixed reliance on inpatient and outpatient diagnosis in Maryland and Massachusetts. The study’s results will help improve CDC guidelines for the NAAT and its efficient use, and they also will serve as a baseline for newer NAAT assays such as the Xpert MTB/RIF assay (Cepheid, Sunnyvale, California).
In countries without uniform access to laboratories with capacity for mycobacteriology cultures, improvements in diagnosis of TB disease have been needed, because sputum-smear microscopy for acid-fast bacilli detects less than half of TB cases.2 NAATs can provide results that inform diagnosis of pulmonary TB disease within 24–48 hours of submission to a laboratory for analysis. CDC recommended in 1996 that NAAT be used on at least one respiratory specimen if the result of sputum-smear microscopy is positive, for each patient being examined for pulmonary TB.3 The enhanced version of the MTD NAAT was approved by the Food and Drug Administration (FDA) in 1999, and CDC recommended its use in 2009 on at least one (preferably the first) respiratory specimen (regardless of result from sputum-smear microscopy) of all patients suspected of pulmonary TB.4 However, anecdotal reports suggested that the use of NAATs was not widespread, for the presumed reason that it did not displace the need for any standard TB diagnostic tool and was technically demanding to perform. Moreover, individual providers and the directors of hospitals and laboratories determine whether to adopt NAAT, based on their own decision criteria. In addition, only one U.S. study has evaluated the cost-effectiveness of MTD NAAT, and only in a hospital setting. That study found the use of NAAT not cost-effective for exclusion of TB in smear-positive patients who have suspected TB disease.5 A more representative, comprehensive study of programmatic benefits and cost-effectiveness was needed to show that NAAT usage could be advantageous.
The purpose of Task Order 27 was to evaluate the MTD NAAT use, effectiveness, health-system benefits, and cost-effectiveness in the largest-ever multisite cohort of patients with suspected pulmonary TB disease. Methods included data analysis from provider initial suspicion of TB disease (which we defined as the earliest of the following events: respiratory isolation, laboratory result with positive results from sputum-smear microscopy, TB treatment start, report of a suspected case of TB, or TB diagnosis) through final TB determination (TB confirmation or definitive exclusion). The multisite study consisted of retrospective reviews of inpatient and outpatient medical records of a cohort of 2,140 patients suspected of having pulmonary TB reported to local jurisdictions during 2008–2010.
Of study patients, 40% had one or more specimens with positive results on sputum-smear microscopy and 60% had all specimens with negative results on microscopy. MTD NAAT was used for 43% of study patients, more for those with positive microscopy results (80%) than for those with negative results (19%). Thirty-seven percent of patients in the study had TB disease confirmed by culture isolation of M. tuberculosis, that is, 2.7 suspected cases per each confirmed case. Foreign-born patients were significantly more likely to have TB disease, but they were less likely to have had their specimens tested by NAAT when the sputum-smear microscopy results were negative. The study also found that public health laboratories (versus private laboratories) reported that they had protocols in place for use of MTD NAAT.
Compared with no MTD NAAT use, study investigators found that use of MTD NAAT
- Improved diagnostic accuracy (i.e., detected a higher proportion of true positives among all positives, true negatives among all negatives, and true positives among all those with TB) among all subpopulations of patients. It was also more specific (i.e., detected a higher proportion of true negatives among all those without TB), in all subpopulations except patients experiencing homelessness.
- Reduced time to TB diagnosis for patients who had positive results from both sputum-smear microscopy and MTD.
- Reduced the number of medical procedures (biopsies, bronchoscopies, computed tomographies) and the usage of respiratory isolation for patients with positive smears and negative MTD results but without TB.
- Resulted in an average 1.5 fewer months of unnecessary and potentially toxic TB medications in MTD-negative patients without TB.
- Resulted in fewer contact investigations initiated for patients who had positive smears and negative MTD results, and who ultimately did not have TB.
- Showed health system cost savings per additional patient who is HIV infected or homeless with TB diagnosed or excluded regardless of sputum-smear microscopy. Cost savings also occurred, per additional patient having a history of substance abuse with TB excluded and negative results on sputum-smear microscopy.
MTD NAAT improved diagnostic accuracy and shortened turn-around time for an initial result, and it reduced unnecessary respiratory isolation, treatment, and contact investigations. In multivariable analysis, providers used MTD NAAT results to significantly reduce time to confirm TB disease.
In light of the experience from Task Order 27 and many other studies, using MTD NAAT is clearly more accurate6 than using sputum-smear microscopy as the determining factor in rapid TB disease diagnosis. In addition, Task Order 27 provides data showing cost savings if NAAT is targeted to persons living with HIV, homelessness, or substance abuse (patients who are more likely to be hospitalized during TB evaluation.)
The study serves as an incentive for providers and/or laboratories to adopt NAAT where it is not currently being used. The National Institutes of Health has recommended use of the NAAT for TB diagnosis in HIV-infected adults and adolescents with advanced immunodeficiency, because of their risk for rapid clinical progression to TB and risk of early death from TB, because NAAT has a high positive predictive value (proportion of true positives among all positives) for sputum specimens with positive smear microscopy results and acceptable sensitivity (proportion of true positives among al those with TB) for specimens with negative microscopy results.7 The Xpert MTB/RIF NAAT (which, compared to MTD NAAT, is less technically demanding, less prone to error, can concurrently diagnose potential multidrug-resistant TB, and is less expensive internationally) received FDA approval in July 2013. This will significantly contribute to prompt, accurate TB diagnosis, and it will possibly save lives.
Div of TB Elimination
Epidemiologist, Principal Investigator for TBESC Task Order 27
- Marks SM, Cronin W, Venkatappa T, Maltas G, Chon S, Sharnprapai S, Gaeddert M, Tapia J, Dorman SE, Etkind S, Crosby C, Blumberg HM, Bernardo J. The Health-System Benefits and Cost-effectiveness of Using Mycobacterium Tuberculosis Direct Nucleic Acid Amplification Testing to Diagnose Tuberculosis Disease in the United States. Clinical Infectious Diseases. 2013;57(4);532-42.American Thoracic Society. CDC, Council of the Infectious Diseases Society of America. Diagnostic standards and classification of tuberculosis in adults and children. Am J Respir Crit Care Med 2000; 161:1376-95.
- CDC. Nucleic acid amplification tests for tuberculosis. MMWR 1996;45:950-1.
- CDC. Updated guidelines for the use of nucleic acid amplification tests in the diagnosis of tuberculosis. MMWR 2009; 58 (01); 7-10.
- Dowdy DW, Maters A, Parrish N, Beyer C, Dorman SE. Cost-effectiveness of the Gen-Probe amplified Mycobacterium tuberculosis direct test as used routinely on smear-positive respiratory specimens. J Clin Microbiol 2003; 41: 948-53.
- Greco S, Girardi E, Navarra A, Saltini C. The current evidence on diagnostic accuracy of commercial based nucleic acid amplification tests for the diagnosis of pulmonary tuberculosis. Thorax 2006; 61: 783-90.
- National Institutes of Health. Guidelines for Prevention and Treatment of Opportunistic Infections in HIV-infected Adults and Adolescents. 2013. http://aidsinfo.nih.gov/guidelines.