BACKGROUND: Brush cytology and histology have been found to be complementary in the evaluation of Barrett's esophagus at a referral medical center. This study evaluated the usefulness of brush cytology and histology in a community hospital setting. METHODS: One hundred consecutive patients with Barrett's esophagus underwent esophagogastroscopy performed by four staff gastroenterologists. Four quadrant biopsy specimens for histopathology at 3-cm intervals throughout the Barrett's segment and one brushing for cytology were obtained. All specimens were interpreted by four board-certified staff pathologists in a blinded fashion. RESULTS: Histologic diagnosis included three adenocarcinomas, one high-grade dysplasia, six low-grade dysplasias, and one indeterminate dysplasia. Cytology diagnosed the same three adenocarcinomas, no high-grade dysplasia, three low-grade dysplasia, and two indeterminate dysplasias. The case of high-grade dysplasia on histology was diagnosed as normal by cytology. The six patients found to have low-grade dysplasia by histology were found to have low-grade dysplasia (3), indeterminate dysplasia (2), and no abnormality (1) by cytology. In no case was a higher grade of dysplasia diagnosed by cytology than by histology. CONCLUSION: Adding brush cytology to histology increased the cost but not the diagnostic yield in the evaluation of Barrett's esophagus in a community hospital setting.
BACKGROUND: Brush cytology and histology have been found to be complementary in the evaluation of Barrett's esophagus at a referral medical center. This study evaluated the usefulness of brush cytology and histology in a community hospital setting. METHODS: One hundred consecutive patients with Barrett's esophagus underwent esophagogastroscopy performed by four staff gastroenterologists. Four quadrant biopsy specimens for histopathology at 3-cm intervals throughout the Barrett's segment and one brushing for cytology were obtained. All specimens were interpreted by four board-certified staff pathologists in a blinded fashion. RESULTS: Histologic diagnosis included three adenocarcinomas, one high-grade dysplasia, six low-grade dysplasias, and one indeterminate dysplasia. Cytology diagnosed the same three adenocarcinomas, no high-grade dysplasia, three low-grade dysplasia, and two indeterminate dysplasias. The case of high-grade dysplasia on histology was diagnosed as normal by cytology. The six patients found to have low-grade dysplasia by histology were found to have low-grade dysplasia (3), indeterminate dysplasia (2), and no abnormality (1) by cytology. In no case was a higher grade of dysplasia diagnosed by cytology than by histology. CONCLUSION: Adding brush cytology to histology increased the cost but not the diagnostic yield in the evaluation of Barrett's esophagus in a community hospital setting.