Role of brush cytology in the diagnosis of Barrett's esophagus: an analysis of eight cases.

Although debatable, role of cytology in diagnosing Barrett's esophagus has been stressed by some authors. Our brief report analyses the role of brush cytology (BC) in its diagnosis. Eight patients who presented with upper gastrointestinal (GI) manifestations and subsequently diagnosed to have Barrett's esophagus with or without adenocarcinoma on brush cytology (BC) or, endoscopic biopsy (EB) or, a combination of both the techniques were included in the study. In all the cases routine cytologic smears and histologic sections with relevant special stains (when essential) were studied. On cytology, the diagnosis of "Barrett's esophagus" was made when the esophageal brushings obtained from a region beyond 3 cm from the gastroesophageal junction showed closely intermingled clusters of squamous and columnar cells, or when there was evidence of intestinal metaplasia. Adenocarcinoma in Barrett's esophagus was diagnosed by the usual criteria for malignancy with features of adenocarcinoma. Cytologic diagnoses were correlated with the histologic diagnoses. Of the eight cases, seven had histopathologic correlation of which six had evidence of BE on BC; four had cytohistologic concordance, of which three cases revealed adenocarcinoma arising in BE. Two endoscopic biopsies were nonrepresentative. In one case, cytology missed Barrett's esophagus. For two cases in which cytology detected Barrett's mucosa, biopsies were nonrepresentative. Our study showed brush cytology (BC) to be a fairly reliable test for detecting Barrett's esophagus. Owing to its inherent advantage of sampling a wider and circumferential area, a technically well performed brushing procedure is likely to be more representative and superior than multiple endoscopic biopsies.