Can unnecessary operations for abdominal stab would be safely reduced? A review of 255 patients.

Records of 255 patients who underwent laparotomies for stab wounds to the abdomen and lower chest were reviewed. Criteria for laparotomy were clinical and followed the conventional lines. Seventy-six patients (30%) were classified as having had an unnecessary laparotomy. Univariate analysis with the Chi-square test revealed 8 variables which differed significantly between the necessary and unnecessary laparotomy groups. Further stepwise discriminant analysis demonstrated 4 variables which had independent significance; they are the type of injury, generalized abdominal tenderness, haemoperitoneum and active intraabdominal bleeding. A mathematical model utilizing these 4 variables may be able to predict the need for laparotomy more accurately than utilizing any single variable. A prospective study is needed to test this hypothesis.