Cheryl L. Wright

Histology of the stomach and duodenum in Crohn's disease.

Crohns disease (CD) not uncommonly affects the stomach and duodenum, but its histologic appearance is not well described beyond the identification of granulomas. We retrospectively identified 209 upper gastrointestinal biopsy samples from 80 sets of biopsies from 49 patients with CD. Age- and sex-matched control biopsies were selected from recent cases of Helicobacter pylori gastritis (73 biopsy samples from 34 patients), from patients with a known history of nonsteroidal antiinflammatory drug use (18 biopsy samples from 12 patients), and from three patients with ulcerative colitis. Architectural and inflammatory changes were evaluated and compared. Over three fourths of the patients with CD had abnormal biopsy results. Fifty-six percent of patients with CD had acute inflammation, but only 10% of the patients were infected with H pylori. Focal acute inflammation was a characteristic of H pylori-negative CD (stomach, 31%; duodenum, 40%), which was much less common in the non-CD group (stomach, 2%; duodenum, 8%). Surface intraepithelial neutrophils of the duodenum were more common in H pylori-negative patients with CD (25%) than in those who did not have CD (4%), and deep acute inflammation of the duodenum was more likely in H pylori-negative patients with CD (19% vs. 0%). Granulomas were found in only 9% of the CD group. Focal acute inflammation of the gastroduodenum, especially in a background of noninflamed mucosa, is strong evidence for CD in the appropriate clinical context, but the stomach and duodenum must be properly sampled and carefully examined for any evidence of H pylori.

Histopathology and mismatch repair status of 458 consecutive colorectal carcinomas

Defects in the mismatch repair (MMR) genes hMLH1 and hMSH2 have been found in 10% to 20% of sporadic colorectal carcinomas and also many cases of hereditary nonpolyposis colorectal cancer syndrome. Patients with these tumors have an improved prognosis and may show greater sensitivity to chemotherapy. We examined 458 resected colorectal carcinomas from 430 consecutive patients and used immunohistochemistry to determine which tumors lacked expression of these genes (MMR-d). We correlated the status of MMR-d or “intact” expression with stage, site, and histology. Eighty-nine of 458 tumors (19.4%) were MMR-d, including 80 hMLH1 and 9 hMSH2 tumors. A total of 6% of patients had synchronous tumors, and 37.7% of these were MMR-d (P = 0.0008). A high proportion of patients with previous breast cancer (4 of 6 patients) had hMLH1-defective colorectal carcinomas. MMR-d tumors presented at an earlier stage than intact tumors, and the node-positive MMR-d tumors were less likely than intact tumors to have pericolonic extranodal tumor deposits (18.2% vs. 44%). The proportion of tumors at each site that were MMR-d increased progressively from cecum (32%) to ascending (35%) to transverse colon, where 41% of all tumors were defective. The proportions then rapidly decreased, reaching the lowest rate (4.7%) in the rectum. Both types of MMR-d tumors more often had expansive borders, intraepithelial lymphocytosis, peritumoral lymphoid, and Crohns-like lymphoid responses than the intact tumors; the frequencies of these features diminished with advancing stage. Tumor budding was less common in stage II and III MMR-d tumors than in intact tumors. Keloid and myxoid type stromas correlated with stage and vascular invasion and were not related to mismatch repair status. Significant differences existed between the hMLH1 and hMSH2 tumors. The reported right-sided preponderance of MMR-d tumors is due to most hMLH1, but not hMSH2, tumors being found there (87.5% vs. 44.4%). hMSH2 tumors were most common in the rectum (55.6%). Mucinous tumors were common in hMLH1 tumors (36.3%) but not in hMSH2 tumors (11.1%). hMLH1 tumors were most likely to be poorly differentiated (70%), which was uncommon with hMSH2 tumors (22.2%). hMSH2 tumors were more likely to be confined to the wall (66.7%) than hMLH1 (20%) or intact tumors (23%). We conclude that hMLH1 and hMSH2-defective tumors have distinctly differing histologic features from each other.

Omeprazole produces parietal cell hypertrophy and hyperplasia in humans

While there is evidence that omeprazole may induce changes in parietal cells, the effect of acid suppression on parietal cells in humans is poorly documented. This study was undertaken to evaluate the effects of omeprazole in human parietal cells over time. The light microscopic morphology of parietal cells in gastric biopsies from 17 patients on omeprazole were compared with those from 13 patients on ranitidine and 20 patients on no acid-lowering medication. Light microscopic and ultrastructural morphology of parietal cells was also evaluated in an additional 14 patients before and after omeprazole administration. Objective measurements of parietal cell height, mass and number were analyzed using analyses of variance. Electron microscopy was used to evaluate parietal cell enlargement. Twenty-five of 31 biopsies from patients on omeprazole, 1 of 13 from patients on ranitidine, and 0 of 20 from patients on neither drug showed parietal cell enlargement. Parietal cell height, mass, and number were increased in omeprazole-treated patients compared with ranitidine-treated patients and those on neither drug, and with the group also evaluated prior to beginning omeprazole treatment. Parietal cell height and mass were increased in patients on omeprazole longer than 12 months compared with biopsies from patients on the drug for less than 12 months. Resin-embedded sections and electron microscopy showed enlarged parietal cells with prominence of cytoplasmic tubulovesicles with sparse secretory canaliculi. Parietal cell hypertrophy and hyperplasia develops in patients on chronic omeprazole therapy; this can be recognized on routine examination of histologic sections. These morphologic changes increase with duration of therapy.While there is evidence that omeprazole may induce changes in parietal cells, the effect of acid suppression on parietal cells in humans is poorly documented. This study was undertaken to evaluate the effects of omeprazole in human parietal cells over time. The light microscopic morphology of parietal cells in gastric biopsies from 17 patients on omeprazole were compared with those from 13 patients on ranitidine and 20 patients on no acid-lowering medication. Light microscopic and ultrastructural morphology of parietal cells was also evaluated in an additional 14 patients before and after omeprazole administration. Objective measurements of parietal cell height, mass and number were analyzed using analyses of variance. Electron microscopy was used to evaluate parietal cell enlargement. Twenty-five of 31 biopsies from patients on omeprazole, 1 of 13 from patients on ranitidine, and 0 of 20 from patients on neither drug showed parietal cell enlargement. Parietal cell height, mass, and number were increased in omeprazole-treated patients compared with ranitidine-treated patients and those on neither drug, and with the group also evaluated prior to beginning omeprazole treatment. Parietal cell height and mass were increased in patients on omeprazole longer than 12 months compared with biopsies from patients on the drug for less than 12 months. Resin-embedded sections and electron microscopy showed enlarged parietal cells with prominence of cytoplasmic tubulovesicles with sparse secretory canaliculi. Parietal cell hypertrophy and hyperplasia develops in patients on chronic omeprazole therapy; this can be recognized on routine examination of histologic sections. These morphologic changes increase with duration of therapy.

Enterocolic lymphocytic phlebitis with lymphocytic colitis, lymphocytic appendicitis, and lymphocytic enteritis

We describe a 53-year-old man with a history of diarrhea temporally related to the use of flutamide. He developed an acute abdomen, and presented with an ileocecal intussusception due to an edematous ischemic cecum. The ischemia was due to enterocolic lymphocytic phlebitis (ELP), with numerous associated thrombi. The phlebitis involved not only the ischemic area but also the grossly unaffected areas, including the entire right colon, terminal ileum, and appendix. All layers of the bowel wall were involved. Mesenteric veins were also prominently affected, but the arteries were spared. This rare form of vasculitis was associated with a marked lymphocytic infiltrate involving the epithelium of the entire right colon, ileum, and appendix. This is the first reported case of ELP occurring in conjunction with lymphocytic colitis, lymphocytic enteritis, and lymphocytic appendicitis. The temporal association of the patients symptoms with flutamide use suggests that this peculiar form of lymphocytic inflammation of the veins and mucosa likely represents a drug reaction. We suggest that some cases of lymphocytic colitis may also be associated with ELP but are unlikely to be recognized unless affected submucosal vessels happen to be included in the biopsy.

Terminal ileal imaging with ileoscopy versus small-bowel meal with pneumocolon

Disease of the terminal ileum can be diagnosed by ileocolonoscopy or barium radiography. We compared the diagnostic accuracy in the terminal ileum of ileocolonoscopy and small-bowel meal with pneumocolon. Consecutive patients during an 18-month period who had both ileocolonoscopy and small bowel meal with pneumocolon were identified and their colonoscopy reports, radiographs, and ileal biopsies reviewed blindly by paired gastroenterologists, radiologists, and pathologists, respectively. A gold-standard diagnosis was determined for each patient by consensus. Of 48 study subjects, 14 (29.2%) had Crohns disease, 5 (10.4%) had lymphoid nodular hyperplasia, and 29 (60.4%) were normal. The sensitivity for a diagnosis of Crohns ileitis was 92.9% for ileocolonoscopy and 100% for small-bowel meal, while their specificities were 100% and 97.1% respectively. The gold standard diagnosis confirmed ileocolonoscopic findings in 45 patients (93.8%) and radiographic findings in 42 patients (87.5%). Agreement between ileocolonoscopy and small bowel meal occurred in 39 cases (81.2%). By combining histology with ileocolonoscopy, the sensitivity and specificity could be increased to 100% for all diagnoses. Ileocolonoscopy and small-bowel meal with pneumocolon are complementary techniques for imaging the terminal ileum. A prospective comparative trial is now needed to more objectively assess their accuracy, cost effectiveness, and adverse effects.

Nonsteroidal anti-inflammatory drug-induced strictures of the colon

The phenomenon of strictures of the colon induced by nonsteroidal anti-inflammatory drugs is a newly recognized pathologic entity that has gained little exposure in the surgical literature to date. A further case is reported and the clinical features of this entity are discussed. Most patients present with symptoms suggestive of malignancy, namely anemia, obstructive symptoms, or weight loss. Pathologic changes are characterized by diaphragm-like strictures with submucosal fibrosis. Surgical resection to exclude malignancy and treat symptoms along with cessation of the nonsteroidal anti-inflammatory drug is the treatment of choice.

Sclerosing sweat-duct carcinoma (malignant syringoma) of the upper eyelid: a patient report with immunohistochemical and ultrastructural analysis.

Summary: Sclerosing sweat-duct carcinoma or malignant syringoma is a rare eyelid tumor. The authors report the first patient with sclerosing sweat-duct carcinoma involving the upper eyelid and present its immunohistochemical and ultrastructural features. The clinical features and the histopathologic and ultrastructural findings that distinguish this tumor from its benign counterparts are discussed. Sclerosing sweat-duct carcinomas are slow growing but locally invasive neoplasms that have a propensity to recur. Wide and deep surgical excision with frozen section evaluation of margins is recommended.