A prospective study of the risk of tuberculosis among HIV-infected patients

• Published in: AIDS (London) Vol. 7; no. 10; p. 1345
• Main Authors: Guelar, A, Gatell, J M, Verdejo, J, Podzamczer, D, Lozano, L, Aznar, E, Miró, J M, Mallolas, J, Zamora, L, González, J
• Format: Journal Article
• Language: English
• Published: England 01-10-1993
• Subjects: Adult; Aged; AIDS-Related Opportunistic Infections - epidemiology; Cohort Studies; Female; Follow-Up Studies; Humans; Isoniazid - therapeutic use; Longitudinal Studies; Male; Middle Aged; Prospective Studies; Risk Factors; Tuberculin Test; Tuberculosis - complications; Tuberculosis - epidemiology; Tuberculosis - prevention & control; Drugs; Prospective Studies; Laboratory Examinations And Diagnoses; Risk Factors; Cohort Analysis; Viral Diseases; Biology; Infections; Hiv Infections; Research Methodology; Examinations And Diagnoses; Diseases; Studies; Treatment; Tuberculosis; Data Analysis
• ISSN: 0269-9370
• DOI: 10.1097/00002030-199310000-00007

To evaluate the risk of developing active tuberculosis (TB) in a cohort of HIV-1-infected patients.
Prospective longitudinal follow-up of 839 HIV-infected patients, of whom 505 (60%) were parenteral drug users and 269 (32%) homosexual men. Tuberculin skin tests were performed at baseline and annually thereafter. Prophylaxis with isoniazid (300 mg daily for 9 months) was offered to those with a positive tuberculin test (induration > or = 5 mm). Diagnosis of TB was accepted if it could be confirmed microbiologically (acid-fast bacilli seen in Ziehl-Neelsen stains or grown in Lowenstein-Jensen cultures) or pathologically (presence of caseating granulomas) and patients had consistent clinical manifestations.
Active TB developed in 23 out of the 733 (3.1%) patients with a negative tuberculin skin test after a mean follow-up of 16 +/- 11 months (range, 2-52 months), with an estimated cumulative probability of 1.5 and 7% after 1 and 3 years, respectively (or 2.4 per 100 patient-years). None of the 87 patients with a negative tuberculin test but a positive Multitest developed TB. Conversely, 106 patients had a positive tuberculin skin test (97 at baseline and nine who converted during follow-up). Active TB developed in seven out of the 26 not receiving prophylaxis or in whom prophylaxis had to be discontinued (16.2 per 100 patient-years), in four out of 61 patients 3-27 months after having completed 9 months of prophylaxis with isoniazid (8.9 per 100 patient-years) and in none of the 19 still receiving isoniazid. When TB was diagnosed, the mean CD4 lymphocyte count of the 34 patients who developed it during follow-up was 77 +/- 103 x 10(6)/l (range, 1-400 x 10(6)/l).
Among HIV-infected patients in whom the tuberculin skin test is negative, the risk of developing active TB is sufficient to consider prophylaxis if the CD4 count falls below 400 x 10(6)/l, at least in those patients with skin anergy living in high-risk geographical areas such as Spain. When the tuberculin skin test was positive, isoniazid (9 months) provided a 45% protection beyond the period of its administration.