Anorectal surgery in patients infected with human immunodeficiency virus: factors associated with delayed wound...

A review of all anorectal operations in patients infected with human immunodeficiency virus (HIV) was performed to assess the incidence, variety, and clinical course of anorectal disease in these patients and to identify factors influencing wound healing.

Anorectal disease is the most common indication for surgical intervention in patients infected with HIV. The cause and management of HIV-related anorectal conditions, which differ significantly from non-HIV-related diseases, are not clear. There also is considerable variation in the reported results of surgical procedures, including wound healing. St. Vincent's Hospital, Sydney, is situated in an area with the highest concentration of individuals infected with HIV in Australia.

The medical records of all identified patients infected with HIV who had an anorectal operation at St. Vincent's General Hospital between January 1, 1988, and January 31, 1995, were reviewed retrospectively. Logistic regression, Mann-Whitney U test, and Fisher's exact tests were used for analysis.

One thousand five hundred two patients with acquired immune deficiency syndrome (AIDS), equivalent to 26.8% of all known patients with AIDS in Australia at this time, were admitted to this hospital during the 7-year period. One hundred one patients infected with HIV underwent 161 anorectal operations. All patients were male homosexuals (98 patients, 97%) or bisexuals (3 patients, 3%), with intravenous drug use an additional risk factor in 5 patients (5%). Thirty-seven percent of patients had more than one operation. Seventy-two percent of patients were Centers for Disease Control (CDC) group 4 (AIDS) at operation, 27% were group 2, 1% was group 3, and none were group 1. Accurate information about wound healing was available for 74% of first operations, and univariate and multivariate logistic analyses of these showed that when the CD4+ T-lymphocyte count was <50 cells/ microL, healing was significantly retarded (p = 0.016). The Centers for Disease Control group, patient age, and serum albumin were not significant predictors of wound healing. The interval between HIV diagnosis and operation was not associated with impaired wound healing, but recognition of AIDS more than 1 year before operation was associated with significantly better wound healing compared with those in whom AIDS developed within the year before operation (p = 0.025). In the patients for whom accurate wound healing information was available, only 40% had healed their wounds by 3 months after operation. Wound healing was worst for patients with chronic fissures, only 16% of whom had healed their wounds at 3 months. The wound healing rate was worse for repeat operations than for first operations. Ten percent of patients had anorectal malignancies, none of which were diagnosed clinically before or during operation.

Wound healing is a significant problem after anorectal operations in patients infected with HIV, especially when the CD4 count is <50/microL. Although there seems to be little or no benefit from more invasive operations in some cases, thorough examination with adequate biopsies is required in all cases. The best management of anorectal disease in patients infected with HIV still is unclear.