Background:  Portal hypertension without liver cirrhosis is a newly described life-threatening complication of unknown cause in HIV-infected persons without hepatitis virus co-infections. Postulated pathogenesis includes prolonged exposure to ART, particularly didanosine (ddI).

Methods:  We performed a nested case control study including 15 patients with cryptogenic non-cirrhotic portal hypertension (NCPH) and 75 matched controls of the Swiss HIV Cohort Study to investigate risk factors for the development of NCPH. Matching criteria were absence of hepatitis virus infection, similar date of HIV infection, and follow-up to at least the date of diagnosis of NCPH in the respective case.

Results:  All 15 cases (13 male; 11 of whom were homosexual and 4 heterosexual) had endoscopically documented esophageal varices and absence of liver cirrhosis on biopsies; 15 patients had splenomegaly; 7 variceal bleeding; 8 ascites, 5 portal thrombosis; 2 hepatic encephalopathy; 4 died of hepatic complications. At time of diagnosis of NCPH, no differences in characteristics of cases vs. controls were found in:  median HIV infection date, April 1990 vs January 1990; sex; ethnicity; duration of follow-up (12.0 vs 11.9 years); CDC (Centers for Disease Control and Prevention) disease stage; peak HIV RNA; HIV RNA on ART; plasma lipids, fat loss, fat accumulation, and blood pressure. Differences were found in:  median age, 52 vs 43 years (conditional logistic regression OR /10 years older 2.9 [95%CI 1.4 to 6.1], p = 0.004); men who have sex with men (MSM; OR 4.5 [1.2 to 17], p = 0.03); CD4 count, 197 vs 522 (OR for CD4 <200, 29.4 [3.6 to 242], p = 0.002); CD4 nadir, 103 vs 164 (ns); body mass index, 20.8 vs 23.7 (ns); diabetes mellitus, 27% vs 4% (OR 8.8 [1.6 to 49], p = 0.01); median alanine aminotransferase, 39 vs 26 IU/L (OR for above normal, 6.3 [1.2 to 34], p = 0.03); alkaline phosphatase, 171 vs 85 U/L (ns); platelets, 182 vs 242x10³/L (ns), current smoking, 7% vs 41% (OR 0.11 [0.01 to 0.88], p = 0.04). Median cumulative exposure to ART (8.5 [IQR 5.1 to 11.4] vs 6.8 [3.3 to 8.7] years, OR per year exposure, 1.3 [1.0 to 1.6] p = 0.02), NRTI (OR 1.3 [1.1 to 1.7], p = 0.01), ddI (3.4 [1.5 to 8.1], p = 0.005), ritonavir (1.4 [1.0 to 1.9], p = 0.03), and nelfinavir (1.4 [1.0 to 1.90], p = 0.03) were longer in cases. Exposure to NNRTI and other PI were not different between groups. In bi-variable models, only the association of NCPH with ddI exposure was robust; other co-variables, particularly low CD4 cell count were not independent risk factors.

Conclusions:  We found a strong association with prolonged exposure to ddI and the development of NCPH.