Human Immunodeficiency Virus (HIV) infection and venous thromboembolism

Abstract

Introduction: Venous thromboembolism (VTE) consists of two related conditions: deep vein thrombosis (DVT) and pulmonary embolism (PE). Several risk factors for VTE have been identified and although each of them does not carry a similar risk, the number of risk factors present greatly increases the risk of VTE and mortality. These risk factors include; being bedridden/immobility, hormonal contraceptive use, malignancies, age, pregnancy and pueperium, surgery and thrombophilias. Recently some studies have suggested that HIV infection may be a risk factor for VTE. Several studies have been done and have shown that patients infected with HIV have a hypercoagable state and there has been some evidence suggesting that HIV may be associated with VTE. There has however been no study done in Mulago hospital looking at the association of HIV and VTE though an estimated 12.8% of the patients attending the hospital are infected with HIV. Objectives: To determine the prevalence of HIV infection among patients with VTE and to assess the association between HIV infection and VTE among patients attending medical inpatient wards and medical outpatient clinics in Mulago hospital. Methods: A case-control study was conducted in the medical wards and medical outpatient clinics of Mulago hospital from February 2010 to May 2010. The cases were patients with VTE that was confirmed by compression/doppler ultrasound. The controls on the other hand were patients who presented in the emergency medical ward in the accident and emergency unit (ward 3BEM) or medical outpatient with no symptoms of VTE. Controls were matched with the cases on age, sex, inpatient or outpatient status and level of function (Kanorfsky). The cases and controls were all assessed for history or presence of HIV infection and other risk factors for VTE. Characteristics of the study population were then described and the test of significance done using independent t-test for continuous variables and chi square or fisher’s exact test for categorical variables. The prevalence of HIV in the VTE patients (cases) was determined and a 95% confidence was computed. The univariate analysis (unadjusted odds ratio) for HIV infection and other risk factors for VTE were done and so was the multivariate analysis (adjusted odds ratios). Using conditional logistic regression the effect of the other covariates on VTE was also assessed. Results: Among the 30 patients with VTE 11(36.7%) were seropositive for HIV whereas among the 60 controls there were 32(53.3%) who were seropositive for HIV. On univariate analysis the unadjusted odds ratio (OR) of HIV infection was 0.507 (p=0.136) which was not significant. Only the cases had past history of DVT (p = 0.013), history of surgery (p = <0.005), active cancer (p=0.043) and immobility (p = 0.013) and their association with VTE was statistically significant. On multivariate analysis after adjusting for other covariates the OR of HIV was 0.293 (p=0.058) and not statistically significant. After adjusting for other covariates the past history of DVT (p=0.008) and history of immobility (p=<0.005) respectively were statistically significant. Conclusion: The prevalence of HIV among patients with VTE presenting at Mulago hospital medical wards and outpatient clinics was 36.7%. This was lower than what was seen in a case control study done in South Africa but was not comparable to most of the studies done outside Africa due to the different study designs. More than two thirds of the patients presenting with DVT in the lower limbs are likely to have left sided lower limb DVT. The reasons for this finding are however not yet clear. There was no association between HIV infection and VTE which is contrary to what studies done elsewhere have shown. This may be because of the high prevalence of HIV among the patients presenting to the medical wards and outpatient clinics of Mulago hospital. Recommendations: Patients diagnosed with VTE should be counseled and tested for HIV given that more than one third of them are likely to be infected with HIV. A bigger follow up study may be needed to compare the VTE prevalence among the cohorts in the HIV care centers and to that of the general population so as to confirm whether HIV is or is not associated with VTE in our setting.