BACKGROUND: Severe thrombocytopenia (TCP, platelets <50 × 10⁹/L) is relatively frequent during HIV infection and is associated with bleeding risk and disease progression. We investigated the changes in the incidence of severe TCP and its predisposing conditions in a cohort of HIV-positive subjects. METHODS: The incidence and predictors of platelet counts <50 × 10⁹/L were investigated in all patients enrolled at our institution between 1985 and 2012. Three different periods were considered on the basis of the available antiretroviral regimens: P1 (1985-1989), P2 (1990-1996), and P3 (1997-2012). Incidence rates were assessed using Poisson regression models and the predictors by means of Cox regression. RESULTS: The study involved 5137 patients with platelet counts >50 × 10⁹/L at enrollment. Severe TCP occurred in 597 subjects, and its incidence decreased over time. The incidence decreased in patients with opportunistic diseases and malignancies but increased in patients with chronic liver disease; TCP unrelated to any cause other than HIV infection remained stable. Multivariate analysis showed that injected drug use, a diagnosis of AIDS, low CD4⁺ cell counts, increased serum alanine aminotransferase levels, and an earlier year of enrollment were predictors of an increased risk of severe TCP, whereas the use of highly active antiretroviral therapy (HAART) was associated with a reduced risk. CONCLUSIONS: A considerable reduction in the incidence of severe TCP after the introduction of HAART was found, probably because of its ability to limit bone marrow damage induced by uncontrolled HIV replication and opportunistic infections. On the contrary, HAART may have a reduced impact on TCP related to chronic liver disease.
BACKGROUND: Severe thrombocytopenia (TCP, platelets <50 × 10⁹/L) is relatively frequent during HIV infection and is associated with bleeding risk and disease progression. We investigated the changes in the incidence of severe TCP and its predisposing conditions in a cohort of HIV-positive subjects. METHODS: The incidence and predictors of platelet counts <50 × 10⁹/L were investigated in all patients enrolled at our institution between 1985 and 2012. Three different periods were considered on the basis of the available antiretroviral regimens: P1 (1985-1989), P2 (1990-1996), and P3 (1997-2012). Incidence rates were assessed using Poisson regression models and the predictors by means of Cox regression. RESULTS: The study involved 5137 patients with platelet counts >50 × 10⁹/L at enrollment. Severe TCP occurred in 597 subjects, and its incidence decreased over time. The incidence decreased in patients with opportunistic diseases and malignancies but increased in patients with chronic liver disease; TCP unrelated to any cause other than HIV infection remained stable. Multivariate analysis showed that injected drug use, a diagnosis of AIDS, low CD4⁺ cell counts, increased serum alanine aminotransferase levels, and an earlier year of enrollment were predictors of an increased risk of severe TCP, whereas the use of highly active antiretroviral therapy (HAART) was associated with a reduced risk. CONCLUSIONS: A considerable reduction in the incidence of severe TCP after the introduction of HAART was found, probably because of its ability to limit bone marrow damage induced by uncontrolled HIV replication and opportunistic infections. On the contrary, HAART may have a reduced impact on TCP related to chronic liver disease.