Natural killer (NK) cells are an important defense against intravascular tumor dissemination. Tumor embolization can occur at surgery, so we tested whether surgical stress decreased perioperative NK cell cytotoxicity, and examined the underlying mechanism of suppression. Patients with solid tumors underwent NK cell cytotoxicity assay just before and 24 hours after surgery in a 3-hour chromium 51 release assay. The NK cell cytotoxicity was significantly decreased postoperatively. We considered that surgical NK cell impairment might be due to (1) NK cell redistribution, (2) presence of suppressor cells, or (3) direct "toxic" effects on NK cells. Impaired NK cell cytotoxicity was not due to NK cell redistribution, because differential counts showed no significant changes in the percentage of large granular lymphocyte NK morphology. To isolate possible suppressor cells, postoperative cells from patients were selectively depleted of NK cells using anti-Leu-11b monoclonal antibody plus complement; these cells were then mixed with autologous preoperative cells. Postoperative NK cell cytotoxicity was markedly impaired, but the postoperative NK depleted cells did not suppress preoperative NK cells. We conclude that NK cell functional impairment from surgical stress is due to direct "toxic" effects on NK cells rather than either NK cell redistribution or the generation of NK-directed suppressor cells.