Volume 12, Number 5—May 2006
THEME ISSUE
Tuberculosis Special Section
Research
Isoniazid Preventive Therapy and Risk for Resistant Tuberculosis
Maria Elvira Balcells*1, Sara L. Thomas*, Peter Godfrey-Faussett*, and Alison D. Grant*
Table 3
Studies comparing isoniazid treatment with no treatment in HIV-infected populations*
| Author, country, dates | Population | Intervention/ comparison; blinding | Enrolled (n) INH/control | Follow up; loss to follow-up; overall or INH vs. control | TB cases: culture positive/total (%) |
Definition of INH resistance | Resistant cases/total tested (% culture positive tested) |
Risk for resistant TB/1,000 |
||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| INH | Controls | INH | Controls | INH | Controls | RR (95% CI) | ||||||
| Randomized controlled trials | ||||||||||||
| Gordin, USA, 1991–1996 (30) | Clinic attendees; med. CD4 233/247 | 6 mo INH 300 mg daily vs. placebo; double blind | 260/257 | 34 mo/33 mo; 6.2% vs. 7% | NS/3 | NS/6 | NS | 0/3 (NS) | 0/5 (NS) | 1.92† | 1.94† | 0.99 (0.06–6,298.19) |
| Hawken, Kenya, 1992–1997 (31) | Clinic or VCT attendees; med. CD4 321.5/346 | 6 mo INH 300 mg daily/placebo; double blind | 342/342 | Med. 1.83 y (range 0–3.41); 32% vs. 27.3% not seen in final 6 m | 19/25 (76) | 22/23 (95.7) | Growth on 0.2 μg/mL INH >1% growth on control medium | 2/17 (90) | 0/21 (96) | 10.05† | 1.46† | 6.88 (0.01–3,882.85) |
| Mwinga, Zambia, 1992–1996 (33) | VCT attendees | 6 mo INH 900 mg twice weekly/placebo; double blind | 350/352 | Med. 1.8 y; 32.4% vs. 30.3% not seen in final 6 m | NS/27 | NS/44 | NS | 0/3 (NS) | 1/5 (NS) | 1.43† | 26.38† | 0.05 (0.00–30.47) |
| Johnson, Uganda, 1993–NS (34) | Clinic or counseling attendees | 6 mo INH 300 mg daily/placebo; partially double blind‡ | 931/787 | Mean 2 y/1.6 y (PPD+/anergic); 16.1% vs. 30.6% | 36/51 (70.6) | 46/64 (71.9) | Growth on 0.1 μg/mL INH (BACTEC radiometric method) >1% growth on control medium | 5/20 (56) | 1/24 (52) | 13.69 | 3.39 | 4.04 (0.50–32.80) |
| Rivero, Spain, 1994–2000 (35) |
Clinic attendees; med. CD4 193/215 |
6 mo INH 300 mg daily/no treatment; not blind |
82/77 |
24 mo; 26.8% vs. 7.8% |
3/3 (100) |
4/4 (100) |
NS |
3/3 (100) |
4/4 (100) |
36.59 |
51.95 |
0.70 (0.16–3.05) |
| Cohort study | ||||||||||||
| Moreno, Spain, 1985–1994 (32) | Clinic attendees; med. CD4 689/648 | 9–12 mo INH (dose NS)/no treatment; not blind | 29/92 | 89 mo vs. 60 mo; NS | 3/3 (100) | 39/43 (90.7) | Growth on 0.2 μg/mL INH >1% growth on control medium | 2/2 (67) | 0/12 (31) | 118.64† | 5.41† | 21.95 (0.04–11,582.31) |
*INH, isoniazid; TB, tuberculosis; RR, relative risk; CI, confidence interval; med., median; NS, not stated; Rx, treatment; VCT, voluntary counseling and testing; PPD, purified protein derivative.
†Calculated by adding 0.5 to numerator and denominator of both groups.
‡Unclear whether isoniazid and placebo group received the same number of tablets.
1Current affiliation: Department of Medicine, Pontificia Universidad Católica de Chile, Santiago, Chile
