Hospital admissions remain high amongst those with HIV

Hospital admissions remain high amongst patient with HIV, even in the era of modern antiretroviral therapy, investigators from the US military report in the online edition of the Journal of Acquired Immune Deficiency Syndromes.

Between 1999 and 2006, a third of patients in the study were hospitalised, and the rate of admission to hospital did not change over time.

However, the investigators found that there were important changes in the causes of hospitalisation, with admissions for liver disease and skin infections increasing. Maintaining a CD4 cell count above 350 cells/mm³ was associated with a significant reduction in the risk of admission to hospital.

Soon after effective combination antiretroviral therapy became available in the late 1990s, studies recorded a decline in rates of hospitalisation and mortality in patients with HIV.

Less is known about hospitalisation trends in the era of modern HIV treatment. However, some evidence suggests that rates have stablised or even increased, possibly as a result of ageing, co-infections, and treatment side-effects.

An understanding of trends in the hospitalisation of patients with HIV can assist in the design of interventions to further improve the help of this population and in health service planning.

Therefore, investigators from the US military’s HIV Natural History Study monitored trends and risk factors for hospitalisations for 2429 patients between 1999 and 2007.

This period was divided into three eras that corresponded with changing standards in HIV treatment and care: 1999-2001; 2002-2004; 2005-2007. Hospitalisation was defined as an admission to hospital that lasted for at least 24 hours.

Overall, the patients had a mean age of 37 and had been living with HIV for a mean of seven years. Antiretroviral therapy was taken by 62% of individuals, the mean duration of treatment being four years. Mean CD4 cell count at the time such therapy was started was 350 cells/mm³, and mean CD4 cell count through the study period was 554 cells/mm³, with 52% of patients having a viral load below 400 copies/ml.

A total of 822 patients, one-third of the cohort were hospitalised at least once, the total number of hospital admissions being 1770.

The mean duration of hospitalisation was six days, and this did not change significantly over the study.

Similarly, there was no change in the overall rate of hospitalisations over the eight years of the study (137 per 1000 person-years).

Gastrointestinal illnesses were the most common cause of hospitalisation (24 per 1000 person-years), followed by bacterial infections (18 per 1000 person years), respiratory disease (16 per 1000 person years), and cardiovascular disease (12 per 1000 person years).

There was a 50% increase in rates of hospitalisation for cancers, which was of borderline significance (p = 0.06). Hospitalisations for cardiovascular disease increased by 24% (p = 0.06), but admissions for neurological disease fell by 25% (p = 0.05).

Admissions for AIDS-defining conditions occurred at a rate of 10 per 1000 person-years, and this did not alter over the course of the research.

Infections were the single biggest cause of admissions (49 per 1000 person-years), and once again this rate was unaltered during the study.

The risk of admission because of hepatitis co-infections, cirrhosis or other forms of liver disease increased by 71%. Although the total number of infections was small, the investigators noted that admissions because of MRSA increased by 300%, and that there hospitalisations due to skin infections went up by 50%.

Surgery also became an increasingly important reason for admission, and accounted for 22 hospitalisations per 1000 person years.

Next the investigators looked at the proportion of hospitalisations that were due to specific illnesses and conditions.

AIDS-related illnesses account for 8% of admissions and this did not change over time.

The proportion of hospitalisations related to MRSA increased from 0.3% to 3%, and an increase was also observed in the percentage of admissions for surgery from 13% to 20%.

An examination of the characteristics of patients being admitted to hospital showed that they had a mean age of 41 years, which increased steadily over the study period (39 – 43 years, p < 0.001).

The percentage of hospitalised patients with hepatitis C increased from 8% to 14% (p < 0.01).

An increase was also observed in CD4 cell count at the time of hospitalisation from a mean of 409 to 466 cells/mm³.

Use of HIV treatment by hospitalised patients remained steady at approximately 70%.

Finally, the investigators analysed the factors associated with hospitalisation.

Each 50 cells/mm³ increase in lowest ever CD4 cell count was associated with an 8% reduction in the risk of hospitalisation (p < 0.01). Patients with a current CD4 cell count above 350 cells/mm³ also had a significantly reduced risk of admission to hospital (p < 0.01).

For individuals with a current CD4 cell count below 350 cells/mm³, use of antiretroviral therapy reduced the risk of hospitalisation (p = 0.02).

Co-infection with hepatitis C increased the risk of hospitalisation by 46% (p = 0.02), and increasing duration of infection with HIV was also a factor associated with an increased risk of hospitalisation (p = 0.05).

Separate analysis was then undertaken looking at the factors associated with hospitalisation for infectious causes. Once again a higher nadir CD4 cell count (p < 0.01), a current CD4 cell count above 350 cells/mm³ (p < 0.01), and use of HIV treatment at CD4 cell counts below 350 cells/mm³ (p < 0.01) were protective.

Increasing age was an important risk factor for admission for surgery (p < 0.01).

“Hospitalisations continue to occur at high rates among HIV-infected persons and… these rates have not changed in the late HAART era”, comment the investigators. They add, the causes of hospitalisations have diversified with non-AIDS-related co-morbidities currently being the most common cause of admissions.”

CD4 cell counts above 350 cells/mm³ were shown to reduce the risk of hospitalisation, and the investigators write, “these data suggest that HAART used by treatment guidelines seems to be protective of hospitalisations due to non-AIDS-related causes.”