Fine needle aspiration cytology (FNAC) of mammary granular cell tumour: a report of three cases.

This report describes the FNAC findings in three cases of granular cell tumour of the breast. The patients comprised two females aged 59 and 62 years and one male aged 28 years. All patients presented with a breast lump which was clinically and radiologically suspicious of malignancy. FNAs yielded moderately cellular specimens which on cytologic examipation consisted of groups of cells and single cells with small regular nuclei and abundant granular cytoplasm. Bare nuclei were also present but these did not have the characteristic bipolar appearance of myoepithelial cells. In two cases there was a granularity to the background. The aspirates were reported as equivocal or atypical, probably benign, and surgical biopsy was performed. Histological examination showed typical benign granular cell tumours with strong positive staining for S-100 protein. Pathologists should be aware that granular cell tumour may occur in or around the breast and should consider this diagnosis in aspirates containing a population of cells with regular nuclei and abundant granular cytoplasm. The main cytologic differential diagnoses are likely to be apocrine cells and histiocytes. The suspicion of a granular cell tumour should be heightened when these features are present in an aspirate from a clinically and radiologically suspicious mass. These cases highlight the role of the triple approach encompassing clinical, radiological and cytological features in the assessment of a breast lesion.

Gastric as well as duodenal biopsies may be useful in the investigation of iron deficiency anaemia.

BACKGROUND: Iron absorption is known to be impaired in the setting of gastric achlorhydria, yet gastric atrophy is not usually considered an aetiological factor for iron deficiency anaemia. We aimed to determine the prevalence of achlorhydric gastric atrophy in patients with iron deficiency and no identifiable source of gastrointestinal blood loss, and to assess whether gastric, as well as duodenal, biopsies should be routinely performed in these patients. PATIENTS: Forty-one consecutive patients with iron deficiency anaemia and no specific gastrointestinal symptoms or evidence of a bleeding lesion on faecal occult blood testing or upper gastrointestinal or colonic endoscopy. METHODS: As well as routine duodenal biopsies, samples were taken from gastric corpus and antrum for evidence of gastric atrophy. Achlorhydria was considered to be present if plasma gastrin measured on a sample obtained with the patient fasting was over 200 ng/l. Serum was tested for intrinsic factor and gastric parietal cell antibodies. RESULTS: Haemoglobin concentrations ranged from 4.1 to 10.9 g/dl. Eight (20%) of the 41 patients had corpus-predominant or generalized atrophy and high plasma gastrin levels, of whom six had serum intrinsic factor and/or gastric parietal cell antibodies: two also had Giardia lamblia organisms in duodenal biopsies. Four other patients (10%) had villous atrophy of the duodenum. CONCLUSIONS: As well as confirming the importance of seeking coeliac disease in patients with iron deficiency anaemia, our results suggest that achlorhydric gastric atrophy is also a common association. Gastric biopsies should be taken in patients with no other explanation for anaemia. The finding of Giardia organisms in two achlorhydric patients, with a possible contributory role, suggests that duodenal biopsies should be obtained even if serum coeliac-related antibodies are absent.

Effect of proton pump inhibitors on the detection of Helicobacter pylori in gastric biopsies.

BACKGROUND: Proton pump inhibitors are known to decrease the activity of Helicobacter pylori organisms within the stomach and to shift their distribution proximally. This effect may reduce the sensitivity of histological examination and rapid urease testing for H. pylori on biopsies taken from recommended sites. It is of particular relevance if a proton pump inhibitor has been prescribed before the patient has undergone diagnostic endoscopy. METHODS: We studied patients referred to our open-access upper gastrointestinal endoscopy service who had either been on no medication (controls) or were already taking proton pump inhibitors. Biopsies taken from the gastric antrum and corpus were used for rapid urease testing and for histological examination. Sera, taken from patients who had no evidence of H. pylori in biopsies, were tested for IgG H. pylori antibodies as an alternative indicator of infection. RESULTS: H. pylori organisms were detected by histological examination in 27 of 40 controls (68%) and in 13 of 25 patients taking proton pump inhibitors (52%). Among patients with positive histology (organisms detected in either antral or corpus biopsies, or both), only the sensitivity of the antral urease test read at 1 h was significantly lower in patients taking proton pump inhibitors than in controls, with no significant difference in sensitivities of the antral urease test at 24 h, of the corpus urease test at 1 or 24 h, or of histology from the antrum or corpus. Of patients with negative histology, none of 13 controls compared with six of 12 patients taking proton pump inhibitors (50%) had positive serology (P = 0.005). Five (83%) of the six histology-negative, seropositive patients taking proton pump inhibitors had histological changes consistent with H. pylori gastritis even though no organisms were detected. CONCLUSIONS: Treatment with a proton pump inhibitor before endoscopy reduces the sensitivity of antral and corpus biopsies for H. pylori detection, both by urease testing and histological examination. If proton pump inhibitors already prescribed cannot be discontinued for an adequate period before endoscopy, patients should have biopsies taken from the corpus as well as from the antrum, and serum should be tested for H. pylori.

Unusual extraskeletal myxoid chondrosarcoma.

We describe an unusual soft tissue tumor occurring within the rectus femoris muscle of a 64-year-old woman. The site, size, macroscopic, and histological appearances were all consistent with an extraskeletal myxoid chondrosarcoma. However, the present case differs significantly from previous reports of this uncommon tumor in that electron microscopy did not show any evidence of chondroblastic differentiation. Furthermore, the cells failed to stain for vimentin while labeling intensely for neuron-specific enolase, contained large numbers of cytolysosomes having a multivesicular appearance, and focally produced an external lamina. Based on the typical histological appearances we conclude that this is an unusual variant of extraskeletal myxoid chondrosarcoma in which there is evidence of nerve sheath differentiation.

The role of morphometry in predicting prognosis in pancreatic islet cell tumors.

Morphometry of 31 pancreatic islet cell tumors was examined to determine the value of this technique in assessing tumor behavior. Patients were followed for a mean period of 5.1 years (range, 1 month-14 years) after diagnosis. Initially 17 localized and nine metastatic tumors were studied. Discriminant analysis was carried out on these cases and identified nuclear/cytoplasmic ratio and number of nuclei/mm2 as the significant discriminatory features. These were combined to derive a classification rule which was capable of correctly identifying localized and metastatic tumors in 92% of cases. The classification rule was applied subsequently to an additional five test cases, all of which were classified successfully. The failure of increased nuclear size and pleomorphism to correlate with malignancy in these tumors was confirmed. Tumors which metastasized had significantly greater gross diameters than localized lesions, but overlap existed. Mitotic counts were not a helpful discriminatory feature. Morphometry may be useful in improving histologic assessment of pancreatic islet cell tumor behavior.

The role of intraluminal tension and pH in the development of necrotizing enterocolitis: an animal model.

Presented is an in vivo animal model of necrotizing enterocolitis (NEC) in which we used an isolated loop of rabbit colon. Changes in mucosal perfusion and macroscopic and microscopic appearances were studied after exposure of the mucosa to different intraluminal tensions and pHs. Mucosal perfusion was reduced with increasing intraluminal tension and decreasing pH. Histologic changes consistent with NEC were identified after five hours when the mucosa had been exposed to both increased intraluminal tension and decreased pH. Factors affecting intraluminal tension and pH could influence the development of NEC in the neonate.