The relationship of pre mortem diagnoses and post mortem findings in a surgical intensive care unit

Crit Care Med. 1999 Feb;27(2):299-303. doi: 10.1097/00003246-199902000-00035.


Objective: To evaluate pre- and post mortem diagnoses and determine their relationship and the discrepancy rate.

Design: Retrospective, descriptive chart review.

Setting: A 36-bed surgical intensive care unit (ICU) of an academic, tertiary care center.

Patients: 149 adults who died in the ICU and had an post mortem examination.

Interventions: Review of the medical record for the ICU course, hospital discharge/death summary, major and minor clinical diagnoses, and the cause of death were directly compared with the major and minor diagnoses and cause(s) of death determined by post mortem examination.

Measurements and main results: Major and minor clinical diagnoses were categorized by the Goldman method and compared with post mortem findings to determine the discrepancy rate. Patients were categorized by the primary surgical service that provided medical and surgical care. Sixty-one (41%) patients had discrepancies uncovered at post mortem examination, of which 20 had two discrepancies. Twenty-three percent of the 149 patients had errors categorized as major and 18% as minor. Overall, 85% of the major errors were undiagnosed infectious processes. Complete agreement between the pre and post mortem diagnoses was present in 58% and varied with the surgical population: trauma group (86%) and cardiac surgery (69%) vs. the transplantation group (17%). Those with longer lengths of stay in the ICU were more likely to develop and, subsequently, have a major error discovered post mortem. Conversely, those who died early (<48 hrs), were less likely to have an undiagnosed disease at post mortem examination and, thus, more likely to have complete agreement between pre and post mortem findings.

Conclusions: The overall discrepancy rate as well as the infectious discrepancy rate between pre mortem clinical diagnoses and post mortem findings were substantially higher in a surgical ICU compared with a hospital-wide population. The majority of these discrepancies were undiagnosed infections. The length of time spent in the ICU before death appeared to influence the rate of errors uncovered at the post mortem examination, suggesting that a longer ICU course, as well as the particular type of surgical patient population, may increase the chance of developing an infectious process, only to be uncovered at post mortem examination.

Publication types

  • Comparative Study

MeSH terms

  • Adult
  • Autopsy* / statistics & numerical data
  • Cause of Death
  • Classification
  • Connecticut
  • Cross Infection / mortality
  • Diagnosis*
  • Diagnostic Errors / statistics & numerical data
  • Hospital Mortality*
  • Hospitals, University
  • Humans
  • Intensive Care Units / standards
  • Intensive Care Units / statistics & numerical data*
  • Length of Stay / statistics & numerical data
  • Retrospective Studies