Characteristics of discrepancies between clinical and autopsy diagnoses in the intensive care unit: a 5-year review

Intensive Care Med. 2003 Feb;29(2):321-4. doi: 10.1007/s00134-002-1576-z. Epub 2002 Nov 30.


Objectives: To characterise discrepancies between clinical and autopsy diagnoses in patients who die in the intensive care unit.

Design: Retrospective chart review.

Setting: Ten-bed closed mixed adult intensive care unit in a tertiary referral teaching hospital.

Participants: All the clinical notes and autopsy reports of 346 patients who died in the intensive care unit in 1996-2000.

Interventions: Discrepancies between clinical and autopsy diagnoses were reviewed by two intensivists, a specialist in infectious diseases, a pathologist and an anaesthesiologist. New findings which would have changed current therapy in the intensive care unit were categorised as a Class I discrepancy, and those related to death but which would not have altered therapy as a Class II discrepancy.

Results: Of 2370 patients admitted, 388 (16.4%) died. An autopsy was performed in 346 (89%) of the deceased patients. A Class I discrepancy was found in 8 patients (2.3%) and a Class II discrepancy in 11 patients (3.2%). Five of the eight (62%) Class I discrepancies were infections which occurred in patients already treated for another infections.

Conclusion: Despite the availability of advanced diagnostic facilities, especially infectious complications seem to remain undiagnosed. Autopsy is a valuable tool with which to monitor diagnostic accuracy in these patients.

Publication types

  • Comparative Study

MeSH terms

  • Aged
  • Autopsy / economics
  • Autopsy / standards*
  • Cause of Death*
  • Clinical Competence / standards
  • Diagnostic Errors / economics
  • Diagnostic Errors / statistics & numerical data*
  • Finland / epidemiology
  • Hospital Mortality*
  • Hospitals, University
  • Humans
  • Intensive Care Units / economics
  • Intensive Care Units / standards*
  • Intensive Care Units / statistics & numerical data
  • Length of Stay / statistics & numerical data
  • Medical Audit
  • Middle Aged
  • Retrospective Studies