The use of minimally invasive procedures in autopsies such as image-guided fine-needle aspiration may facilitate family acceptance and, therefore, increase the number of postmortem examinations. The objective of the current study was to validate the use of cytology in a prospective set of conventional autopsies.
All lesions and organs sampled for histological examination were concomitantly evaluated by scrape cytology of exactly the same location. The cytopathologist and the surgical pathologist were blinded to each other's microscopic findings. Final cytological and histological diagnoses were divided into 6 main diagnostic groups: normal, neoplasms (benign and malignant), inflammatory conditions, adaptive processes, degeneration, and cardiovascular disorders. Cytohistological agreement was assessed with Cohen's κ coefficient. The simple percentage agreement was also reported for each diagnostic group and for all different organs sampled in the postmortem examinations.
Two hundred eleven samples were studied from 36 consecutive autopsies (21 males and 15 females; median age, 58 years). Complete cytohistological agreement was achieved for 151 samples (71.6%) with a κ coefficient of 0.43 (moderate correlation). When samples were divided by diagnostic group, the best results were seen in normal specimens (93.8%) and neoplasms (82.3%). Organs with the best performance included the thyroid, cerebellum, lymph nodes, and adrenal glands (all with 100% agreement), the brain (90.5%), and the pancreas (84.6%).
Cytology showed a good correlation with histology, particularly for neoplastic cases and specimens with minimal pathological alterations, and could be used as an alternative diagnostic method in partial or restricted postmortem examinations. Cancer Cytopathol 2016;124:785-90. © 2016 American Cancer Society.