To confirm the presence of an outbreak of postoperative infections following laminectomy and to determine the infection rate after interventions were instituted.

Retrospective cohort study. Medical records were reviewed, personnel interviewed, and premises examined.

Surgical unit of hospital A in Pakistan.

Patients who had surgical laminectomy between January 1993 and July 1994.

Instructive program for nursing and medical staff in December 1993.

From January to December 1993, 6 (15%) of 41 laminectomy patients developed postoperative discitis. The risk of discitis varied significantly by surgeon (P=.016); patients who had one particular surgeon, surgeon A, were nine times more likely to develop postoperative infections than patients who did not have surgeon A. Patients were not consistently cleaned or shaved before coming to the operating room, and personnel moved back and forth between the operation theater and other parts of the hospital without changing their gowns or slippers. After the instructional intervention, between January and July 1994, 2 (6%) of 31 laminectomy patients developed postoperative discitis, a rate not significantly lower than in the preceding 12 months (P=.45). Overall, from January 1993 through July 1994, female patients were more likely to develop discitis than males (31% vs 7%; relative risk, 4.4; 95% confidence interval, 1.3-15.6; P<.032).

Endemic conditions require that laminectomy at hospital A be limited to those situations where the benefits of the surgery exceed the considerable risk of postoperative discitis.