Inflammation of the female reproductive tract increases susceptibility to HIV-1 and other viral infections and, thus, it becomes a serious liability for vaginal products. Excessive release of proinflammatory cytokines may alter the mucosal balance between tissue destruction and repair and be linked to enhanced penetration and replication of viral pathogens upon chemical insult. The present study evaluates four surface-active microbicide candidates, nonoxynol-9 (N-9), benzalkonium chloride (BZK), sodium dodecyl sulfate, and sodium monolaurate for their activity against human sperm and HIV, and their capacity to induce an inflammatory response on human vaginal epithelial cells and by the rabbit vaginal mucosa. Spermicidal and virucidal evaluations ranked N-9 as the most potent compound but were unable to predict the impact of the compounds on vaginal cell viability. Interleukin (IL)-1 release in vitro reflected their cytotoxicity profiles more accurately. Furthermore, IL-1 concentrations in vaginal washings correlated with cumulative mucosal irritation scores after single and multiple applications (P < 0.01), showing BZK as the most damaging agent for the vaginal mucosa. BZK induced rapid cell death, IL-1 release, and IL-6 secretion. The other compounds required either more prolonged or repeated contact with the vaginal epithelium to induce a significant inflammatory reaction. Increased IL-8 levels after multiple applications in vivo identified compounds with the highest cumulative mucosal toxicity (P < 0.01). In conclusion, IL-1, IL-6, and IL-8 in the vaginal secretions are sensitive indicators of compound-induced mucosal toxicity. The described evaluation system is a valuable tool in identifying novel vaginal contraceptive microbicides, selecting out candidates that may enhance, rather than decrease, HIV transmission.