The effects of HIV on the host that predispose to bacterial infections include impairment of neutrophil and mononuclear cell function, cytokine production, and B-cell function in addition to T-cell depletion. Disruption of skin and mucous membrane barriers, indwelling intravascular catheters, and the use of broad-spectrum antibiotics all increase the risk of bacterial infections. The individual contribution of each of these factors has not been quantified.

Recently several studies have investigated the specific defects in neutrophil function that occur during the course of HIV infection. Bandres and colleagues evaluated phagocytic function of neutrophils and monocytes from HIV-infected men with CD4 counts of >400/mm³.⁽⁹⁾ In this study, phagocytic or reactive oxygen-generating functions were preserved and in some patients were enhanced compared to age and sex-matched controls. Direct infection of neutrophils by HIV has also been reported.⁽¹⁰⁾ Using the polymerase chain reaction (PCR), Cabrilovich and colleagues identified HIV DNA in 11 of 37 (29.7%) HIV-infected patients' neutrophils. Detection of HIV DNA was more common in patients with symptomatic disease than in asymptomatic patients. Neutrophil superoxide production was found to be depressed in patients with HIV infection compared to controls.⁽¹¹⁾ The degree of impairment was increased at lower CD4 counts.

Polymorphonuclear neutrophil chemiluminescence (a measure of oxidative burst) was found to be depressed at all stages of infection, particularly in patients with AIDS.⁽¹²⁾ The most profound impairment in neutrophil chemiluminescence in this study was found in a group of HIV-infected individuals with severe, recurrent bacterial pneumonia. Neutrophil oxidative burst was examined in a study of 34 HIV infected adults and 12 seronegative controls.⁽¹³⁾ The percentage of stimulated neutrophils was similar in control subjects and HIV-infected subjects with CD4 counts of >200/mm³, however it was significantly reduced in those with CD4 counts of <200/mm³ and further reduced in the group with CD4 counts <50/mm³. Finally, Vanacore and colleagues looked at the expression of CD16, a low-affinity receptor for IgG, on neutrophils in 65 patients with different stages of HIV infection using flow cytometry.⁽¹⁴⁾ Expression of CD16 was not reduced in asymptomatic patients compared to controls; however, significant reductions were seen in symptomatic patients and those with AIDS. Taken together, these studies demonstrate a number of different functional defects, which may predispose to bacterial infections. Further investigation is warranted to determine whether specific alterations in neutrophil function can predict the development of AIDS.

Abnormalities in B-cell function include the spontaneous secretion of antibody and lack of, a response to signals for activation.⁽¹⁵⁾ It has been postulated that the chronic stimulation of B cells precludes a response to new antigens. Recently, a reduction in the density of complement receptor 2 (CR2) on B cells has been demonstrated in HIV-infected subjects.⁽¹⁶⁾ The role of this receptor in B-cell activation and the implications of diminished levels of CR2 on B cells in HIV infection remain to be determined.

Several defects in the HIV-infected host's defense against bacterial infections in the respiratory tract have recently been reviewed.⁽¹⁷⁾ In the upper respiratory tract, a reduction in salivary immunoglobulin A may lead to colonization with pathogenic bacteria. In the lower respiratory tract, alveolar macrophages may show impaired function due to direct effects of HIV. In addition, impaired neutrophil function and reduced production of opsonizing antibody by B lymphocytes have been implicated as important factors predisposing to bacterial pneumonia.

In addition to the impact of HIV on bacterial infections, recent attention has been given to the possible role of mycoplasma in the pathogenesis of HIV infection.^(18-25) Contamination of cell culture led to the initial observation that certain species of mycoplasma increase the cytopathic effect of HIV in vitro. These results were later confirmed by experimental infection with several species of mycoplasma.⁽¹⁸⁾ Most recently the presence of antibody to a new species, Mycoplasma penetrans, has been shown to correlate with HIV seropositivity.^{(26, 27)} Using an ELISA and Western blot assay, the seroprevalence of antibody to M. penetrans was found to be 31.4% among HIV-infected adults compared to 1.0% among uninfected controls (p<0.05).⁽²⁶⁾ No differences in rates of seroprevalence were noted by stage of disease or risk group. In another study done in the Congo by the same group of investigators, the seroprevalence for M. penetrans was 15% among HIV-infected adults compared to 4% among uninfected controls; this difference did not reach statistical significance. Further prospective longitudinal investigation conducted in a diverse population of HIV-infected individuals is needed to determine whether co-infection with M. penetrans alters the natural history of HIV infection.