Why couldn't an accurate diagnosis be made? An analysis of 1044 consecutive autopsy cases

Pathol Int. 1999 May;49(5):408-10. doi: 10.1046/j.1440-1827.1999.00879.x.


The protocols of 1044 consecutive patients autopsied between 1983 and 1997 at Sumitomo Hospital (Osaka, Japan) were retrospectively analyzed and the findings were compared with clinical diagnoses. In 73 cases, the clinical diagnosis apparently differed from the autopsy findings, and in six cases the origin of a malignant neoplasm remained unsolved even at autopsy. Of the 73 discrepant cases, 24 were a result of clinician misjudgment and a neglect to conduct further examinations. Missed diagnosis due to an erroneous pathological report, technical error of endoscopy, and misleading results obtained by new non-invasive technologies accounted for seven, nine, and 11 cases, respectively. Twenty-two cases were missed because the clinician could not carry out precise examination. It is concluded that advances in diagnostic technology and medical knowledge have not reduced the value of an autopsy.

MeSH terms

  • Autopsy
  • Diagnosis, Differential
  • Diagnostic Errors / classification*
  • Diagnostic Errors / statistics & numerical data*
  • Evaluation Studies as Topic
  • Humans
  • Neoplasms, Unknown Primary / diagnosis
  • Retrospective Studies