Incidence of immune reconstitution inflammatory syndrome among HIV patients infected with tuberculosis in a Dublin cohort

Purpose of the study

Globally, the incidence of HIV-associated tuberculosis (TB) infection is increasing. In this regard TB-associated immune reconstitution inflammatory syndrome (IRIS) after initiation of antiretroviral therapy is well recognised with rates as high as 45% being reported. The aim of this study was to review the clinical presentation of patients with TB/HIV co-infection and the incidence of IRIS in a large academic medical centre in Dublin.

Methods

A retrospective chart review was conducted on HIV-positive patients diagnosed with TB between December 2003 and May 2008. All patients were HIV positive and had a confirmed mycobacterial tuberculosis diagnosis, or a clinically presumptive diagnosis with response to anti-tuberculosis therapy. Data collected included patient demographics, clinical presentation, baseline CD4 counts and the development of a clinical syndrome consistent with IRIS.

Summary of results

Of the 30 charts reviewed, 18 (60%) were female and 12 (40%) were male. Seven patients (23%) were European (six Irish and one British) while the remainder of the patients were from sub-Saharan Africa or south-east Asia. Thirteen (43%) had pulmonary disease, the remainder (17 patients) presented with extra-pulmonary disease and of these, eight (27%) had disseminated disease. Five (17%) presented with lymphadenitis, three (10%) with cerebral TB or meningitis and one (3%) patient with peritoneal tuberculosis. One patient had multi-drug resistant tuberculosis and another had streptomycin resistance. At the time of their TB diagnosis, 18 (60%) had a CD4 count of less than 200 and 22 (73%) were antiretroviral therapy (ART) naïve. Eleven (37%) had a baseline HIV viral load greater than 100,000 copies/ml. Four (13%) patients had a clinical syndrome consistent with IRIS following initiation of their anti-TB therapy, all of whom had a CD4 count less than 100. Three of the patients presented with high grade temperatures and one presented with worsening lymphadenopathy. ART was initiated at a median duration of 8 weeks after TB treatment; one patient did not receive ART during their TB therapy. Two patients had disseminated disease, one extra-pulmonary disease and one pulmonary disease.

Conclusion

HIV/TB co-infection is more prevalent in the immigrant population. Our rate of IRIS is 13% in patients co-infected with HIV/TB, which is lower than previous series. The development of IRIS depends on the population studied, but extrapulmonary tuberculosis and low CD4 counts appear to be risk factors for IRIS.