BACKGROUND Systemic amyloidosis (SA) has a broad nonspecific clinical presentation. Its diagnosis depends on identifying amyloid in tissues. Abdominal fat pad fine needle aspiration (FPFNA) has been suggested as a sensitive and specific test for diagnosing SA. METHODS Thirty-nine FPFNA from 38 patients (16 women and 20 men, age range 40-88 years) during a 15-year period were reviewed. Smears and cell blocks were stained with Congo red (CR). A panel of antibodies (serum amyloid protein, serum amyloid A, albumin, transthyretin, kappa light chain and lambda light chain) was used on six cell blocks from five patients. The FNA findings were correlated with clinical and histological follow-up. RESULTS FPFNAs were positive, confirmed by CR in 5/39 (13%), suspicious in 1/39 (3%), negative in 28/39 (72%), and insufficient for diagnosis in 5/39 (13%) of cases. In all the positive cases, SA was confirmed within 2-16 weeks. Among the 28 negative cases, SA was diagnosed in 21, the rest were lost to follow-up. Among the insufficient cases, SA was diagnosed in four and one was lost to follow-up. Specificity was 100%, whereas sensitivity was 19%. SA typing using cell block sections was successful in three, un-interpretable in one, and negative in two cases. CONCLUSIONS FPFNA for SA is not as good as previously reported. This may be due to different practice setting, level of experience, diagnostic technique, or absence of abdominal soft tissue involvement. A negative result of FPFNA does not exclude SA. Immune phenotyping of amyloid is possible on cell block.