Characteristics of Primary Thrombotic Microangiopathy in Patients with and without Human Immunodeficiency Virus: Thirteen Year Experience at the University of Maryland

Background: Thrombotic microangiopathy (TMA) is a syndrome characterized by thrombocytopenia and microangiopathic hemolytic anemia. TMAs can be primary or secondary to variety of clinical conditions: pregnancy, malignancy, autoimmune diseases and medications. Human immunodeficiency virus (HIV) infection is associated with thrombotic thrombocytopenic purpura (TTP), although the pathophysiology and clinical presentation has not been well-defined.

Methods: In an urban center with a 3% prevalence of HIV infection, we retrospectively reviewed the charts of patients who underwent therapeutic plasma exchange (TPE) for TTP from January 2000 to December 2013 after IRB approval. The aim of our study is to describe and analyze the clinical differences seen in primary TMA with or without HIV infection. Differences of laboratory data between the 2 groups were assessed using the t -test.

Results: Of 188 patients requiring TPE over 13 years at our urban tertiary medical center, 31 (17%) were HIV-positive and 154 (83%) were HIV negative. HIV-positive patients had median age of 43 years (range, 26-65), 30 (96%) black, 23 (74%) female, 23 (74%) with primary TMA. HIV-negative patients had a median age of 46 years (range, 16-82) 92 (59%) black and 104 (66%) female, 54 (35%) with primary TMA. Fifteen of 31 HIV-positive patients (48%) presented with comorbid infections caused by bacterial, viral, fungal and opportunistic organisms, whereas only 29 of the 157 (18%) HIV-negative patients presented with concomitant infection.

The following comparisons were made between HIV-positive (n=23) and negative (n=54) patients with primary TMA. Laboratory data revealed that hemoglobin was lower (mean Hb: 7.0 vs 8.1 g/dL, P=0.03) and renal failure was more severe (mean creatinine 2.3 vs 1.6, P=0.04) in HIV-positive patients compared with HIV-negative. There was no statistically difference between 2 groups in LDH and platelet counts. ADAMTS13 level was available in 6/23 HIV positive and 18/54 HIV-negative primary TMA patients. The level <10% was seen in 11 HIV-negative patients. ADAMTS13 level of HIV-positive patients ranged from 14 to 74 (Sample collection was delayed in 3/6 patients). All HIV-positive cases showed immunosuppression, with a mean CD4 count of 124 cells/ µL, ranging from 2 to 444 cells/ µL. Regarding treatment, HIV-negative patients required an average of 23% more TPE sessions (18 vs. 15) than HIV-positive patients. Steroids, intravenous immune gamma-globulin and rituximab were used in both groups. Mortality and complications were similar in both groups.

Discussion: The frequency of 1 in 5 patients with TMA also having HIV at the University of Maryland is one of the highest frequencies reported in the literature outside of South Africa. Our study suggests that HIV-positive patients might have a thrombotic microangiopathy triggered by an infection more commonly than patients without HIV. The absence of ADAMTS13 deficiency was notable in our HIV-positive patients. However, delayed sample collection in 3/6 patients might limit the significance of this data. The etiology of TMA in HIV infection remains unclear. Prior studies have suggested that a HIV-specific mechanism such as B cell dysregulation as a result of T cell depletion; cytokine production and HIV infection of endothelial cells play a role. The limited number of patients seen individual centers precludes controlled clinical trials to study the efficacy of treatment for this serious disease. Future prospective collaborative research studies are needed to expand our knowledge of TMA in HIV-positive populations.