Autopsy analysis in surgical patients: a basis for clinical audit.

An autopsy study was performed to quantify diagnostic fallibility in clinical surgery. Autopsy results in 312 surgical patients were compared with clinical findings. The primary clinical diagnosis was correct in 93 per cent of patients; complications had been correctly diagnosed in 60 per cent and error in treatment was found in 16 per cent. Error in treatment had an adverse impact on the course of disease in 11 per cent of patients. Infective complications such as abdominal sepsis and bronchopneumonia were encountered most often. Sensitivity was low for the clinical diagnosis of pulmonary embolism, bronchopneumonia, myocardial infarction and terminal haemorrhage. Statistical analysis showed that sudden unexpected death is the most obvious condition in which a high yield is expected from a post-mortem examination. Autopsy remains a valuable means of quality control in clinical surgery and could be a basis for surgical audit.