Immune reconstitution inflammatory syndrome (IRIS) is not a significant cause of death in patients initiating antiretroviral therapy in Uganda, investigators report in the September 15th edition of Clinical Infectious Diseases, now available online. In a retrospective study, the investigators found that although 17% of patients died within the first three years of starting HIV treatment, only a small proportion of these deaths were attributable to IRIS.
"In our experience, the contribution of IRIS…to early mortality is limited," comment the investigators.
HIV treatment has been shown to be as effective in resource-limited settings as it is in developed countries. However, significant early mortality and loss to follow-up has been observed in patients starting antiretroviral therapy in a number of African countries. There has been little research into the conditions causing these deaths, nor has the contribution of IRIS to early mortality been well characterised.
After an individual starts HIV treatment, their immune system starts to improve and this can unmask sub-clinical infections, sometimes resulting in an inflammatory response that can cause unpleasant symptoms, illness, and – in rare cases – death.
Investigators from Kampala therefore undertook a retrospective study lasting three years that recorded mortality rates and the cause of death in 559 individuals who started HIV treatment for the first time between 2004 and 2005.
These patients had a median age of 38 years and 69% were women. At the time HIV treatment was started, the patients had advanced HIV disease, with 89% having an AIDS diagnosis, and the median CD4 cell count was 98 cells/mm3. A third of patients had a body mass index (BMI) of 18 kg/m2 or below and 36% had a haemoglobin level below 11 g/dl.
A total of 99 patients died during the three years of the study. As with other studies investigating deaths amongst patients starting HIV treatment in Africa, most deaths (80; 14% of the entire cohort) occurred during the year after HIV treatment was started (with 54, 73% of deaths, in the first three months). There was a significant fall in mortality thereafter (15 deaths in the second year and four in the third).
This provided a mortality rate of 17.9 per 100 person years in the first year after starting HIV treatment, 2.3 per 100 person years in the second year, and 1.2 per 100 person years in the third year.
Of the 80 deaths in the first year, 69 (86%) were HIV-related, with four being attributed to IRIS. All four cases were "unmasking events" including one case each of Cryptococcus meningitis, extra-pulmonary tuberculosis (TB), TB meningitis and a tumour in the brain.
Overall, TB and Cryptococcus were the most common causes of death, and almost two-third of the deaths attributed to TB occurred in individuals who had symptoms of the infection before initiating antiretroviral therapy.
In multivariate analysis that controlled for possible confounding factors, the baseline characteristics associated with both all-cause and HIV-related death were a CD4 cell count below 25 cells/mm3, a BMI below 18 kg/m2, and a haemoglobin level below 8 g/dl.
"We observed a high mortality rate of 14% during the first year of therapy, particularly during the first three months," write the investigators. They note that many of these deaths were due to infections and "may have been preventable if the infrastructure for opportunistic infection screening was routinely available and patients were given prophylaxis treatment prior to antiretroviral therapy initiation".
They emphasise that few deaths could be attributed to IRIS, supporting "the view that, in most cases, IRIS is a self-limited clinical entity".
Castelnuovo B et al. Cause-specific mortality and the contribution of immune reconstitution inflammatory syndrome in the first 3 years after antiretroviral therapy initiation in an urban African cohort. Clin Infect Dis 49 (online edition), 2009.