Portal de Periódicos da CAPES

Clinical Characteristics of Iris Syndrome in Patients with HIV and Tuberculosis

2005; SAGE Publishing; Volume: 10; Issue: 3 Linguagem: Inglês

10.1177/135965350501000303

2040-2058

Christos Michailidis, Anton Pozniak, Sundhiya Mandalia, Sheena Basnayake, Mark Nelson, Brian Gazzard,

Streptococcal Infections and Treatments

Background Some patients with HIV/tuberculosis (TB) coinfection who are on anti-TB treatment and highly active antiretroviral therapy (HAART) will develop an exacerbation of symptoms, signs or radiological manifestations of TB that are not due to relapse or recurrence of their TB. The aetiology of these immune reconstitution inflammatory syndrome (IRIS) reactions is unknown but it is presumed that they occur, at least in part, as a consequence of HAART-related reconstitution of immunity. Methods Patients who were diagnosed with their first episode of definitive or presumed TB between January 2001 and July 2003 were identified from the Chelsea and Westminster TB/HIV database. The patients were classified into those who developed IRIS and those who did not using a set definition of the syndrome. Demographic, clinical and laboratory data relating to both HIV and TB were compared between the two groups. Results A total of 55 cases of TB were identified, of which 45 cases were confirmed on culture or gene probe and 10 were presumed cases. Fourteen cases (25.5%) developed IRIS with a median (range) duration of 2.53 (0.53–14.97) months. The median baseline CD4 [interquartile range (IQR)] for the IRIS group was significantly lower at 80 (33–117) cells/mm 3 ( P=0.05) than the non-IRIS group at 139 (77–284) cells/mm 3 . A significantly greater proportion of patients in the IRIS group [11/14 (78.6%), P=0.011] had baseline CD4 <100 cells/mm 3 compared with the non-IRIS group [16/41 (39.0%)]. There was no significant difference between the two groups when comparing the log 10 baseline viral load (VL). Eight (57.0%) patients in the IRIS group had disseminated TB at baseline compared with seven (17.0%) in the non-IRIS group ( P=0.006). In those who had a detectable VL at baseline, the median fold change (IQR) in CD4 from baseline to 3 months was significantly higher in the IRIS group patients, 1.5 (0.6–5.6), compared with 0.7 (-0.2 to 1.0) for those in the non-IRIS group ( P=0.046). Conclusions Patients who develop IRIS are more likely to present with disseminated TB, have a CD4 count <100 cells/mm 3 and have a prompt rise in CD4 count in the initial 3 months of HAART.