PI Perspective #47

December 2008     View PDF     En español

Risk of a hospital stay high after starting HAART

by Paul Dalton

A poster at the joint 2008 ICAAC / IDSA meeting in Washington, DC showed a high rate of hospitalizations within the first three months following the start HIV therapy (HAART). Further, the rate of hospital stays within the first 45 days of being on HAART for those who responded well to therapy were surprisingly comparable to those who didn’t respond well.

Although taking HAART reduces HIV-related illnesses over time, the study sought to find the rate of hospital stays within the first year after starting therapy. People who start HIV therapy are at risk for a variety of illnesses. These include adverse drug reactions, opportunistic infections, and immune reconstitution inflammatory syndrome (IRIS).

In this retrospective study (1997–2005) by the John Hopkins HIV Clinical Cohort, data on the hospital stays of 1,329 people were collected and examined. The study included from 6 months before starting HAART up to a year afterwards.

Two groups were followed based upon their ability to reach a 1 log (90%) reduction in their viral loads within six months. Responders (965) were those who achieved the 1 log decrease, and non-responders (362) were those who didn’t. Women comprised about 35% of the individuals, while nearly 4 in 5 were Black and nearly 1 in 5 was White in this cohort. More than 2 in 5 were injection drug users (IDUs).

The results show a high rate of hospital stays for both groups within the first 45 days after starting HAART — just above 80 stays per 100 person-years on average. For responders, their hospital stays fell significantly between 45 and 90 days while non-responders showed no significant change over one year of follow-up. Also, both 45-day rates were comparable to the rates seen before starting HAART for each group.

Risk factors for hospitalization after starting HAART included lower CD4 count (<100), being female and IDU. Other factors did not contribute to these rates, which include age, type of HIV drugs used, and various time periods during the study.

Several factors may limit this study’s findings. First, it included a high number of injection drug users, which is significant for that community but may not translate well to the general population. Second, the study was conducted at a single health care center, which may limit how it applies to demographic differences seen in parts of the US.

From these data, keeping a close eye for symptoms of illness during the first three months on HAART appears to be in order for both health providers and patients to help prevent possible hospital stays. Prospective studies that look at the reasons for these stays could provide more specific information on the types of clinical and lifestyle factors that better predict an individual’s risk for a hospital stay.