1987 Gallie lecture. Quality control in surgical research: importance to the patient.

One of the deficiencies in surgical scholarship is the rarity of timely, well-controlled research to evaluate new operative procedures. The invasive aspect of surgery makes randomized studies more difficult to initiate, and in studies with large numbers of participating surgeons, the impact of unsatisfactory surgical expertise (craftsmanship) is a major variable. In an effort to improve surgical research, the use of quality control data for early assessment of the effectiveness of the operative procedure is strongly recommended. Early postoperative studies of stimulated gastric acid secretion have led to improved surgical performance during trials of operations using vagotomy for control of peptic ulcer disease. Similar, early, quality control studies have been effective when investigating selective shunts for control of variceal bleeding; necessary data include shunt patency, portal perfusion of the liver and obliteration of major portoazygous collaterals. Failure to incorporate objective tests for surgical quality control can lead to erroneous assessment of operations. Carefully planned prospective studies of well-performed operations are urgently needed.