Abstract

Value of the tuberculin skin testing for isoniazid preventive therapy for HIV-infected patients

Presented by Jonathan E Golub (United States).

J.E. Golub¹, V. Saraceni², S. Cohn¹, A.G. Pacheco³, L.H. Moulton¹, S.C. Cavalcante², A. Efron¹, R.E. Chaisson¹, B. Durovni<sup>2

1</sup>Johns Hopkins University, Baltimore, United States, ²Rio de Janeiro City Health Secretariat, Rio de Janeiro, Brazil, ³Fiocruz, PROCC, Rio de Janeiro, Brazil

Background: Isoniazid preventive therapy (IPT) has been shown to be effective at reducing TB incidence among HIV-infected patients, particularly in conjunction with HAART. However, controversy continues regarding tuberculin skin testing (TST) for identifying HIV-infected patients most at risk for TB.
Methods: We compared TB risk by TST and IPT status in the THRio HIV cohort from Rio de Janeiro, Brazil. TB and IPT naïve patients were analyzed from HIV diagnosis until TB diagnosis, death, or censoring at last follow-up. Cox proportional hazards models compared TB risk across TST and IPT categories.
Results: Among 12,167 patients (36,286 person-years(PY) of follow-up) there were 1,644 incident TB cases (incidence rate (IR)=4.5/100PY);95%CI:4.3-4.8). TB incidences by IPT/TST category are in Table. Overall TB incidence was lower among those who received IPT (1.9/100PYs vs 4.8/100PYs). Patients with baseline CD4>200 (adjusted Hazard Ratio(aHR)=0.35 (95%CI:0.32-0.39) and those who received HAART (aHR=0.35;95%CI:0.30-0.37) had lower rates. Because of significant interaction between HAART exposure and baseline CD4 count, analyses were separated by HAART status(Table), and adjusted for baseline CD4, sex and age.

TST/IPT category	TB cases (n)	Person-years (pys)	Incidence rate per 100 pys (95% CI)	Incidence rate ratio (95% CI)	aHR (95% CI) with HAART	aHR (95% CI) no HAART
TST-negative/ no IPT	276	11,204	2.5 (2.2-2.8)	REF	REF	REF
TST-negative/ IPT	5	318	1.6 (0.6-3.8)	0.64 (0.21-1.51)	1.10 (0.40-2.97)	0.61 (0.08-4.44)
TST unknown/ no IPT	1,107	20,447	5.4 (5.1-5.7)	REF	REF	REF
TST unknown/ IPT	14	1,008	1.4 (0.8-2.3)	0.28 (0.14-0.43)	0.78 (0.42-1.42)	0.54 (0.17-1.69)
TST-positive/ no IPT	184	782	23.5 (20.4-27.2)	REF	REF	REF
TST-positive/ IPT	58	2,525	2.3 (1.8-3.0)	0.10 (0.07-0.13)	0.31 (0.20-0.48)	0.15 (0.07-0.30)

[TB Risk by TST and IPT category]


Conclusions: HIV-infected patients who are TST-positive greatly benefit from IPT regardless of HAART status. Patients with unknown or negative TST status had a greatly decreased risk of developing TB if they received IPT, but these associations were not statistically significant after adjustment for baseline CD4. Determination of TST status, where feasible, will help identify HIV+ patients most likely to benefit from IPT.