Context.—

Autopsies can reveal clinically undiagnosed diseases. However, the frequency of first diagnoses at autopsy and their association with clinically known risk factors are not well understood because of lack of systematic analyses addressing this topic.

Objective.—

To perform a large retrospective cohort analysis on the frequency of clinically undiagnosed postmortem findings and correlate these with patients’ risk factors.

Design.—

Six hundred forty-eight consecutive and complete autopsies of adults (age >18 years), performed in the University Hospital Zurich, Switzerland, during a 3-year time period were retrospectively analyzed. Clinical diagnoses and postmortem findings were compared in order to identify clinically undiagnosed lesions and clarify their correlation with common risk factors.

Results.—

In 633 of 648 patients (98%), at least one clinically undiagnosed finding was identified at autopsy. The most common nonneoplastic entities were bronchopneumonia (198; 31%), coronary artery disease (155; 24%) and acute or subacute myocardial infarction (94; 15%), and the most common malignancies were prostate cancer in men (14; 2.2%), followed by kidney cancer (10; 1.5%), gastrointestinal stromal tumor (10; 1.5%), and lung carcinoma (9; 1.4%) in both genders. Clinically undiagnosed cardiac amyloidosis was demonstrated in 8% (52 of 648) of patients and was significantly associated with age, hypertension, chronic kidney disease, coronary artery disease, and hypertensive cardiomyopathy.

Conclusions.—

Autopsy is a useful investigation for the detection of clinically undiagnosed entities. In our cohort, cardiac amyloidosis showed the highest number of underlying risk factors, but was clinically underdiagnosed. Our findings underline the necessity of improved clinical detection of cardiac amyloidosis, especially in light of emerging therapeutic options. Moreover, we characterize the most common entities prone to clinical underdiagnosis.

This content is only available as a PDF.

Competing Interests

The authors have no relevant financial interest in the products or companies described in this article.

Supplementary data