Prevalence and Predictors of Thyroid Dysfunction in Patients with HIV Infection and Acquired Immunodeficiency Syndrome: An Indian Perspective

Table 4

Regression analysis showing parameters that are predictors of subclinical hypothyroidism in patients with HIV infection.

Variable	⁢Model-1		⁢Model-2		⁢Model-3
	Standardized coefficient ()	value	Standardized coefficient ()	value	Standardized coefficient ()	value
Anti-TPO antibody	0.245	<0.001	0.247	<0.001	0.239	<0.001
CD4 count (baseline)	−0.178	0.012	−0.175	0.018	−0.185	0.013
CD4 cell count (6–12 months after diagnosis)	−0.099	0.216	−0.094	0.242	−0.125	0.130
CD4 cell count (at present)	−0.138	0.059	−0.138	0.062	−0.130	0.078
25OHD	−0.027	0.644	−0.025	0.679	−0.018	0.770
Tuberculosis	0.109	0.084	0.108	0.087	0.117	0.065
Lamivudine	−0.032	0.622	−0.030	0.653	−0.022	0.741
Stavudine	0.000	0.996	−0.010	0.903	−0.012	0.880
Tenofovir	−0.006	0.932	−0.001	0.991	0.009	0.904
Efavirinez	−0.126	0.053	−0.125	0.056	−0.125	0.056

Linear regression was initially performed with all parameters which are likely to influence serum testosterone levels age, weight, duration of HIV infection, antithyroid peroxidase (TPO) antibody titer, baseline CD4 count, CD4 count (6–12 months after diagnosis), CD4 count at the time of recruitment into the study, hemoglobin, erythrocytic sedimentation rate, history of tuberculosis, opportunistic fungal infections, viral infections (hepatitis-B and hepatitis-C), serum 25-hydroxy-vitamin-D (25OHD), and individual antiretroviral agents received by the patient to evaluate their role in the occurrence of subclinical hypothyroidism. Parameters with were included into the stepwise forward line regression analysis without adjustment for any variables (Model-1), after adjustment for age and duration of HIV infection (Model-2) and after adjustment for variables in Model-2 plus weight and history of opportunistic fungal infections and viral infections (Model-3). Standardized coefficient (β): change in odds ratio with 1 unit change in predictor variable; for the tuberculosis variable, patients without tuberculosis were taken as reference group.