Surgical audit under scrutiny--a prospective study.

Surgical audit must be shown to improve clinical practice and patient outcome if its widespread introduction is to be enthusiastically embraced by surgeons. Retrospective studies on hospital activity by their nature are often incomplete and unreliable. A 12-month prospective review (July 1990-June 1991) of the activity, morbidity and mortality that occurred within a district general surgical unit is analysed. During the study period, 3,927 patients were admitted to the unit, of whom 1,649 were elective and 2,278 (58%) were emergency cases. 48 patients (1.2%) were transferred to external specialist centres. 41% of the admissions did not require surgery. There were 2,335 in-patient and 765 out-patient operations performed. Using the BUPA classification (n = 3100), there were 388 major (12.5%), 802 intermediate (25.9%) and 1910 minor (61.6%) procedures. There were 15 perioperative and 38 non-operative (27 metastatic carcinoma) deaths. 80% of the perioperative deaths were high risk, elderly patients with acute abdominal pathology. 369 complications (39 in non-operative cases) were recorded among both in- and out-patients: 212 systemic, 133 local/wound and 24 major/life threatening. The perioperative mortality rate was 0.6%. The operative morbidity rate was 9.0% and the procedure-related morbidity 4.7%. The wound infection rate was 2%. In a non-specialist, general surgical unit with a broad case mix, it is possible to provide a standard of care and practice that produces very low mortality and an acceptable morbidity rate.