Amyloidosis in surgically resected atrial appendages: a study of 345 consecutive cases with clinical implications.

Histomorphologic parameters of atrial appendages removed during the Cox-Maze procedure have been shown to correlate with recurrence of atrial fibrillation. While amyloid deposition has been noted within atrial appendages, the incidence and significance remains incompletely understood. More accurate amyloid typing methodologies and targeted pharmacotherapeutics have recently been developed, prompting pathologists to provide more detailed information about the type of amyloid identified in such samples. This study sought to fully characterize the morphologic characteristics of atrial amyloid as well as its incidence and clinical significance. Tissue archives were queried for atrial appendages removed during the cardiac surgeries (2010-2014). Patient demographics, imaging features, and salient clinical findings were recorded. Pattern and extent of amyloid deposition were recorded. Typing of the amyloid protein, when present, was performed on a subset of cases by laser capture microdissection with mass spectrometry-based proteomic analysis. A total of 383 atrial appendages from 345 consecutive patients were included in the study (mean age, 69 years; range, 26-92 years). Amyloid was present in 46% of patients. A linear relationship was observed between age and presence of atrial amyloidosis. Women were more likely to have atrial amyloidosis. Two distinct morphologies of amyloid were observed: filamentous and nonfilamentous, and correlated perfectly with amyloid type (filamentous = AANF-type amyloid; nonfilamentous = ATTR-type amyloid). Filamentous deposits were observed in 91% of those with amyloid. Amyloid was more likely to be found in the left atrial appendage than the right. Patients with atrial amyloid, irrespective of type, were more likely to have experienced stroke or TIA and more likely to have atrial arrhythmia preoperatively. Postoperatively, those with atrial amyloid are more likely to experience recurrence of arrhythmia than those who did not have atrial amyloid. Understanding the morphologic characteristics of AANF-type amyloid will allow for identification by the light microscopy and obviates the need for expensive ancillary typing techniques. The finding of nonfilamentous amyloid, should still prompt confirmation of amyloid type so that targeted therapy may be employed.