HIV-positive people with higher viral load have increased risk of bacterial pneumonia

The incidence of bacterial pneumonia amongst HIV-positive individuals treated with an antiretroviral regimen containing a protease inhibitor is similar to that seen in the general, HIV-negative population, according to a French study published in the May edition of HIV Medicine. However, the investigators, from the APROCO cohort, found that some HIV-positive individuals had a higher risk of bacterial pneumonia than others, and that bacterial pneumonia was much more common amongst patients with poor control of HIV and suboptimal adherence.

Before effective antiretroviral therapy became available, bacterial pneumonia occurred much more frequently in HIV-positive patients than in the general population, and caused a significant amount of illness and death in people with HIV. Unlike most opportunistic infections which occur in individuals with HIV, bacterial pneumonia can occur even when a patient has a good CD4 cell count. Although several studies have reported a decline in the incidence of bacterial pneumonia since potent HIV treatment became available, certain groups of HIV-positive patients still appear to be more vulnerable to the illness.

The French doctors therefore analysed the incidence and risk factors for bacterial pneumonia in a cohort of 1203 patients who started a protease inhibitor-containing antiretroviral treatment regimen between 1997 and 1999. They write, “such an analysis could provide insights into the pathophysiology of bacterial pneumonia in HIV-infected patients and could help better target specific preventative measures to patients at higher risk.”

To be included in the investigators’ anaylsis, bacterial pneumonia required hospitalisation, and was diagnosed using a chest x-ray or sputum culture, alternatively a presumptive diagnosis could be made if the patient recovered after antimicrobial therapy.

A total of 29 cases of bacterial pneumonia were diagnosed, giving an incidence of 0.8 per 100 patient years, similar to the incidence seen in the general population in the US and Europe in the 1990s. Eleven cases of bacterial pneumonia had a confirmed cause, with Streptococcus pneumoniae accounting for nine of these. No patient died because of bacterial pneumonia.

In multivariate analysis the investigators identified several non-HIV-related factors associated with an increased risk of bacterial pneumonia. Each ten-year increase in age increased the risk of bacterial pneumonia by 108% (p = 0.0004), and injecting drug use increased the risk by 176% (p = 0.02). Not smoking, however, reduced the risk of bacterial pneumonia by 70% (p = 0.05).

Surprisingly, the investigators found that individuals who started HIV therapy with a CD4 cell count above 500 copies/mm³ had a higher risk of bacterial pneumonia than patients who started antiretroviral therapy with a weaker immune system (p = 0.0005). They were at a loss to explain this finding.

Individuals who took unboosted saquinavir, a component of an antiretroviral regimen which would now be considered suboptimal, also had an increased risk of bacterial pneumonia (p = 0.002).

Viral load was also connected with the risk of bacterial pneumonia, with the illness occurring with much greater frequency amongst individuals whose viral load was above 10,000 copies/ml than amongst patients with better virological control (p

“Our data suggest that some HIV-infected patients have a much higher risk of bacterial pneumonia”, write the investigators, adding, “from a cost-benefit point of view, it may be more efficient to…target antipneumococcal vaccination at such patients, i.e. older patients, intravenous drug users, smokers and patients who have poor adherence or a poor virological response to [HIV therapy].”