A study conducted amongst urban women in the United States during the early years of potent antiretroviral therapy has found one of the highest incidence rates of bacterial pneumonia ever recorded in well-resourced countries. Nevertheless, the study, reported in the July 1st edition of Clinical Infectious Diseases found that both potent antiretroviral therapy and use of cotrimoxazole prophylaxis reduced the risk at all CD4 counts, although the use of both together reduced the risk even further.
Bacterial pneumonia is a major cause of illness and death among HIV-positive individuals, occurring much more frequently than in the general population. Since the advent of powerful antiretroviral therapy combinations, the incidence of bacterial pneumonia is decreasing but the risk for people living with HIV remains significantly higher than the general population at all CD4 counts.
The use of cotrimoxazole (TMP-SMX, Bactrim, Septrin) in order to prevent Pneumocystis pneumonia (PCP) , in addition to antiretroviral therapy, may also have reduced the incidence of bacterial pneumonia, although previous studies have produced conflicting results. In addition, most previous studies of bacterial pneumonia have focused on HIV-positive men, even though HIV-positive women may be at higher risk for bacterial pneumonia.
Consequently, investigators from the HIV Epidemiologic Research (HER) study in New York, Baltimore, Detroit, and Providence, examined the rate of, and risk factors for, bacterial pneumonia, in this prospective study of the natural history of HIV disease in women.
The HER study recruited 885 HIV-positive and 425 at-risk women aged 16-55 between April 1993 and January 1995. The mean age of the HIV-positive women was 35.5 years; and 60.8% were African American, 20.7% were white, and 17.2% were Latina. Baseline CD4 counts were above 500 cells/mm3 in 32.6%; 49.3% had between 201-500 cells/mm3; and 17.5% had less than 200 cells/ mm3. Women with clinical AIDS were excluded from the study.
A total of 316 cases (in 195 HIV-positive women) of community-acquired bacterial pneumonia were seen during a mean of 4.21 years of follow-up. In comparison, only thirteen cases in ten HIV-uninfected women were seen during a mean of 4.15 years of follow-up. This produced an incidence rate of bacterial pneumonia of 8.5 episodes per 100 person-years among the HIV-positive women, compared with 0.7 episodes/100 person-years among the HIV-uninfected women (p
Bacterial pneumonia appeared at all CD4 counts, although the rate was highest in women with baseline CD4 counts below 200 cells/ mm3: 17.9 episodes per 100 person-years. This compares with 8.7 and 4.9 episodes per 100 person-years in women with baseline CD4 counts between 201-500 cells/mm3 and above 500 cells/ mm3, respectively (p
Demographic and lifestyle factors associated with a higher risk of bacterial pneumonia included being African American, having less than twelve years of education, injection drug use, and cigarette smoking. Notably, current smoking more than doubled the risk of bacterial pneumonia (Adjusted Hazard Ratio, 2.12; 95% CI, 1.26-3.55).
Factors associated with a lower risk of pneumonia included duration of potent antiretroviral therapy and/or TMP-SMX use. Although TMP-SMX and potent antiretroviral therapy use each decreased the risk of bacterial pneumonia, the combination of both further decreased the risk. This was most pronounced in women with CD4 counts below 200 cells/mm3, but was seen across all CD4 counts.
Of the 658 HIV-positive women in the HER study after January 1996, those who had taken TMP-SMX for a year (which itself decreased the risk of bacterial pneumonia by 3% for each month of use; p = 0.002), decreased their risk of bacterial pneumonia by a further 8% for each month of potent antiretroviral therapy (p
After four years of follow-up, 94% of women who used potent antiretroviral therapy for more than a year and 81% who used it for less than a year were free of bacterial pneumonia, compared with 69% of women who had never taken antiretroviral therapy (p
There was no association seen between multiple episodes of bacterial pneumonia and prior PCP, injection drug use, alcohol use, cigarette smoking, race/ethnicity, or living with children.
A total of fifteen out of the 68 deaths that were observed in the HER study occurred during a hospitalisation for bacterial pneumonia. This resulted in a case-fatality rate of 7.7% for the 195 HIV-positive women who developed bacterial pneumonia. After adjusting for CD4 count, potent antiretroviral therapy use, and TMP-SMX use, HIV-positive women with bacterial pneumonia had a fivefold increased risk of death compared to the HIV-positive women without bacterial pneumonia (p
The investigators note that the high incidence rate of bacterial pneumonia (8.5 episodes per 100 person-years) observed in their study is "among the highest in the literature reporting on populations in developed countries with access to [potent antiretroviral therapy]." This high rate is partially attributed to the fact that "less than one-half of eligible women reported [potent antiretroviral therapy] use."
Nevertheless, they point out that the women in their study "are not atypical of populations in the United States with HIV, supporting the generalizability of our data to women and drug users with HIV infection. Furthermore, given that we excluded women with clinical AIDS, our rates of bacterial pneumonia are likely an underestimation."
This compares starkly with a recent study from France, reported on aidsmap in April, which found that rates of bacterial pneumonia were similar to HIV-negative populations in patients with low viral loads thanks to widespread use of potent antiretroviral therapy.
The HER investigators add that although potent antiretroviral therapy and TMP-SMX use "decreased the rate, bacterial pneumonia was associated with an accelerated progression to death. These findings support the need for additional methods to prevent the substantial morbidity and mortality associated with bacterial pneumonia.
"Interventions aimed at timely initiation of [potent antiretroviral therapy], TMP-SMX therapy, and smoking cessation programs are warranted," they conclude.
Kohli R et al. Bacterial pneumonia, HIV therapy, and disease progression among HIV-infected women in the HIV Epidemiologic Research (HER) study. Clinical Infectious Diseases, 43:90-98, 2006.