Preoperative prediction of mortality following surgery for esophageal cancer

Ninety-five patients with thoracic esophageal cancer who had undergone radical esophagectomies through right thoracotomies from 1986 to 1989 were statistically analyzed semi-quantitatively to identify the risk factors predicting "operative (within 45 days of operation) or hospital death." Age, pulmonary function (%VC or %FEV1.0), cardiac function (EKG or Master test), renal function (Ccr), hepatic function (R15'ICG), diabetes mellitus (75 OGTT), extent of tumor invasion to the adventitia, and the type of operative procedure were each scored according to severity; 0 (no risk), 1, 2, or 3 (high risk). Patients with no severe postoperative complications had an accumulated score of less than 8 and comprised Group I, while those suffering an an "operative death" had a total score of 8 or more, and comprised Group II. Group III included those suffering a "hospital death." There was a significant difference between Group I and Group II (p < 0.005), but not between Group I and Group III. "Operative deaths" could be preoperatively predicted from the risk factors. However, "hospital deaths" could not be predicted, since they resulted mainly from recurrences of the cancer. In cases with scores of 8 or more the operative procedure should be changed to a simpler one for an improved prognosis.