Desmoplastic melanoma (DM) is a rare subtype of melanoma characterized by malignant spindle cells associated with prominent fibrocollagenous stroma. Primary melanomas may be entirely desmoplastic ("pure" DM) or exhibit a desmoplastic component admixed with a nondesmoplastic component ("combined" DM). The cytologic features of only 5 cases of DM have been reported previously. Fine-needle biopsy (FNB) specimens from 20 recurrent or metastatic lesions in patients with cutaneous DM and 20 recurrent or metastatic lesions from patients with primary cutaneous non-DM were examined and compared. FNB specimens of patients with DM were less cellular (P = .009) and less often exhibited intranuclear cytoplasmic invaginations (P = .008) and mitotic figures (P = .006) than specimens from patients with non-DM. "Combined" DMs were more commonly composed of epithelioid cells (P = .017) and less often contained bizarre/giant tumor cells (P = .010) than did "pure" DMs. Recurrent and metastatic DM has a range of cytologic appearances. Awareness of the cytologic features and careful clinicopathologic correlation will assist in accurate FNB diagnosis.

In fluid specimens, reactive mesothelial cells and histiocytes may mimic epithelioid melanoma cells. Awareness of the morphological features and diagnostic pitfalls of melanoma in fluids is necessary to avoid the potentially serious consequences of misdiagnosis.

The cytologic findings in a case of mesenteric fibromatosis initially suggested the diagnosis of a benign spindle-cell soft-tissue tumor. Subsequent histology and electron microscopy were performed on the resected mass, and the definitive diagnosis was established. The patient had a history of previous abdominal surgery, but no features of Gardner's syndrome. The difficulties associated with the diagnosis of mesenteric fibromatosis and the cytologic diagnosis of benign and spindle-cell soft-tissue tumors and low-grade sarcomas in general are discussed.

Thirty‐one patients undergoing laparotomy for tumours in the region of the pancreas had both fine needle aspiration cytology (FNAC) and histological biopsy specimens taken to assess the accuracy of the FNAC technique. There were no false positive results but there were six false negative results following FNAC with only one false negative result of histological biopsy. However, there was sampling bias in favour of histology in each of the five patients with negative FNAC and positive histology; two had metastatic disease and three had histology repeated because the initial frozen section was negative. Two major complications may have resulted from the histological biopsy procedure. It is concluded that fine needle aspiration cytology is the ideal method of biopsying pancreatic lesions because of the inherent risk of complications following histological biopsy.

No abstract