Objective: To formulate a score system that would make the preoperative diagnosis of acute appendicitis more accurate.
Design: Retrospective then prospective study.
Setting: City University Hospital.
Subjects: 536 patients who had their appendixes removed between 1981 and 1986 (retrospective study), and 150 consecutive patients admitted with a presumptive diagnosis of appendicitis between 1987 and 1988 (prospective study).
Main outcome measures: Correlation between the histological diagnosis of appendicitis and variables representing history, clinical examination, and laboratory investigations.
Results: The rate of histologically proven negative appendicectomies in the retrospective series was 40% and in the prospective series 33%. The variables that were thought to be predictive were: male sex, white cell count of greater than 11 x 10(9)/l, history of less than 24 hours with no previous complaints, rebound tenderness, shift of pain from the epigastrium, and localised guarding, but all criteria had low specificities and sensitivities when applied prospectively, and combining the scores did not improve them.
Conclusion: The accurate diagnosis of appendicitis depends largely on the experience of the surgeon and is not improved by the application of a score system that includes the above variables.