Abstract
Biopsies of the liver, lung, and kidney are performed for many indications, including organ dysfunction, mass lesions, and allograft monitoring. The diagnosis depends on the sample, which may or may not be representative of the lesion or pathology in question. Further, biopsies are not without risk of complications. Autopsies are a resource for assessing the accuracy of biopsy diagnoses and evaluating possible complications. Herein, we aimed to compare liver, lung, and kidney biopsy diagnoses with those from autopsies conducted soon after the procedure and to assess the contribution of biopsy to mortality. A 28-year search of our database identified 147 patients who were autopsied after dying within 30 days of a liver, lung, or kidney biopsy. The concordance of the biopsy diagnosis with the autopsy findings was determined. Finally, medical records were reviewed to determine the likelihood that a biopsy contributed to the patient's death. The contribution of the biopsy to death was categorized as "unlikely," "possible," or "probable." Overall concordance between biopsy and autopsy diagnoses was 87% (128/147), including 95% (87/92), 71% (32/45), and 90% (9/10) for liver, lung, and kidney biopsies, respectively. Concordance was lower for biopsies of suspected neoplasms versus non-neoplastic diseases. Lung biopsy concordance was higher for wedge biopsy versus needle or forceps biopsy. A biopsy was determined to at least "possibly" contribute to death in 23 cases (16%). In conclusion, an autopsy is an important tool to validate liver, lung, or kidney biopsy diagnoses. Confirmation of biopsy diagnoses via post-mortem examination may be particularly valuable when patients die soon after the biopsy procedure. Furthermore, an autopsy is especially useful when patients die soon after a biopsy in order to determine what role, if any, the procedure played in their deaths. Though biopsy complications are uncommon, a biopsy may still contribute to or precipitate death in a small number of patients.