Observer variation in the diagnosis of superficial oesophageal adenocarcinoma

BACKGROUND AND AIMS

When to perform oesophagectomy for neoplastic progression in Barrett's oesophagus is controversial. Some resect for high grade dysplasia whereas others defer treatment until intramucosal adenocarcinoma is diagnosed. Interobserver agreement for a diagnosis of high grade dysplasia or intramucosal adenocarcinoma remains unknown and may have therapeutic implications.

METHODS

Histological slides from 75 oesophagectomy specimens with high grade dysplasia or T(1) adenocarcinoma were blindly reviewed by two gastrointestinal pathologists and one general surgical pathologist, and classified as high grade dysplasia, intramucosal adenocarcinoma, or submucosal adenocarcinoma. A subsequent re-review of all 75 cases by the same observers following establishment of uniform histological criteria was undertaken. Interobserver agreement was determined by kappa statistics. Coefficients <0.21, 0.21-0.40, 0.41-0.60, 0.61-0.80, and >0.80 were considered poor, fair, moderate, good, and very good agreement, respectively.

RESULTS

Interobserver agreement among all pathologists and between gastrointestinal pathologists when comparing high grade dysplasia with intramucosal adenocarcinoma was only fair (k=0.42; 0.56, respectively) and did not substantially improve on subsequent re-evaluation following establishment of uniform histological criteria (K=0.50; 0.61, respectively).

CONCLUSIONS

When evaluating resection specimens and after implementation of uniform histological criteria, even experienced gastrointestinal pathologists frequently disagree on a diagnosis of high grade dysplasia versus intramucosal adenocarcinoma. Treatment strategies based on the histological distinction of high grade dysplasia from intramucosal adenocarcinoma using limited biopsy specimens should be re-evaluated.

Angiogenic polypoid proliferation adjacent to ileal carcinoid tumors: a nonspecific finding related to mucosal prolapse

Case reports have highlighted angiogenic polypoid proliferation in the mucosa adjacent to ileal carcinoid tumors, describing them as granulation tissue polyposis and florid angiogenesis. Some authors have proposed that the ileal carcinoid tumors themselves produce growth factors that cause the change. The purpose of this study was to determine the frequency of angiogenic polypoid proliferation in a large cohort of resected ileal carcinoid tumors compared with control groups. Search of the Cleveland Clinic and Summa Health System pathology files (1985 to 1999) yielded 65 resected ileal carcinoid tumors. Mucosal abnormalities adjacent to the ileal carcinoid tumors were graded 0 to 4+. Twenty ileal resection margins from colonic carcinoma cases served as normal controls. Ileal mucosa adjacent to 22 noncarcinoid neoplasms were also examined. The mucosa adjacent to 54/65 ileal carcinoid tumors (83%) showed mucosal abnormalities (vs. 3/20 normal controls), including mucosal edema, capillary ectasia, muscularis mucosae hypertrophy, fibrosis/smooth muscle proliferation within the lamina propria, club-shaped villi, and intramucosal capillary proliferation. Forty ileal carcinoid tumor cases (61%) showed some degree of angiogenic polypoid proliferation characterized by club-shaped villi and prominent intramucosal capillaries, with 17 (26%) graded as 3+ or 4+. Angiogenic polypoid proliferation was associated with hypertrophy of the muscularis mucosae, lamina proprial fibrosis/smooth muscle proliferation, and capillary ectasia similar to that described with gastrointestinal mucosal trauma/prolapse. This trauma/prolapse change was identified in 45 cases (69%) and was graded 3+ or 4+ in 23 (35%). Seventeen (77%) of the noncarcinoid neoplasms showed trauma/prolapse changes, with 7 (32%) graded as 3+ or 4+. Angiogenic polypoid proliferation also correlated with trauma/prolapse change in the noncarcinoid neoplasm controls. Neither APP (P =.24) nor the prolapse changes (P =.33) were found to be statistically different between the two tumor groups. Angiogenic polypoid proliferation of the adjacent ileal mucosa is common in patients with ileal carcinoid tumors and with noncarcinoid neoplasms. Angiogenic polypoid proliferation almost invariably coexists with fibromuscular change and capillary ectasia within the lamina propria, suggesting that mucosal trauma/prolapse plays a role in the histogenesis. The association of angiogenic polypoid proliferation with a variety of different neoplasms makes it unlikely that the tumors themselves secrete growth factors.

Endoscopic and histologic evaluation together with symptom assessment are required to diagnose pouchitis

BACKGROUND & AIMS

Pouchitis often is diagnosed based on symptoms alone. In this study, we evaluate whether symptoms correlate with endoscopic and histologic findings in patients with ulcerative colitis and an ileal pouch-anal anastomosis.

METHODS

Symptoms, endoscopy, and histology were assessed in 46 patients using Pouchitis Disease Activity Index (PDAI). Patients were classified as either having pouchitis (PDAI score > or =7; N = 22) or as not having pouchitis (PDAI score <7; N = 24).

RESULTS

Patients with pouchitis had significantly higher mean total PDAI scores, symptom scores, endoscopy scores, and histology scores. There was a similar magnitude of contribution of each component score to the total PDAI for the pouchitis group. Of note, 25% of patients with symptoms suggestive of pouchitis did not meet the PDAI diagnostic criteria for pouchitis. In both groups, the correlation coefficients between symptom, endoscopy, and histology scores were near zero (range, -0.26 to 0.20; P > 0.05).

CONCLUSIONS

The symptom, endoscopy, and histology scores each contribute to the PDAI and appear to be independent of each other. Symptoms alone do not reliably diagnose pouchitis.

Methylene blue staining of dysplastic and nondysplastic Barrett's esophagus: an in vivo and ex vivo study

BACKGROUND AND STUDY AIMS

Methylene blue selectively stains specialized columnar epithelium in Barrett's esophagus with high accuracy. We prospectively evaluated the methylene blue staining properties of dysplastic and nondysplastic Barrett's esophagus and the association of these properties with the risk for dysplasia and cancer.

PATIENTS AND METHODS

In a ex vivo study, we mapped, photographed, and sampled esophagectomy specimens with high grade dysplasia and/or early adenocarcinoma before and after methylene blue staining. In a concurrent in vivo study, we performed methylene blue staining and characterized methylene blue stain characteristics. Pathologists estimated the proportion of specialized columnar epithelium in each specimen and graded dysplasia.

RESULTS

We examined 551 biopsies from 47 patients with biopsy-proven Barrett's esophagus and 48 sections from five surgical specimens with Barrett's esophagus and dysplasia and early adenocarcinoma. The accuracy of ex vivo and in vivo methylene blue staining for specialized columnar epithelium was 87% and 90%, respectively. It was influenced by the length of Barrett's esophagus, biopsy location, and the presence of esophagitis and/or dysplasia. Light to absent staining (p = 0.01) and moderate to marked heterogeneity (p = 0.01) were significantly associated with high grade dysplasia or cancer in the univariate analysis and in a multivariate model that adjusted for the length of Barrett's esophagus and the presence of a lesion. These staining characteristics were present in all patients with severe dysplasia and/or adenocarcinoma.

CONCLUSIONS

Highly dysplastic or malignant Barrett's esophagus stains differently with methylene blue. Increased heterogeneity and decreased methylene blue stain intensity are significant independent predictors of high grade dysplasia and/or cancer. These features may help to direct biopsies in patients without a lesion.

Association of susceptibility locus for inflammatory bowel disease on chromosome 16 with both ulcerative colitis and Crohn's disease

A susceptibility locus for inflammatory bowel disease (IBD) on chromosome 16 (IBD1) has been linked to Crohn's disease in genome-wide linkage studies. We performed a case-control study with two markers for this locus using leukocyte DNA from 127 Crohn's patients, 83 ulcerative colitis patients, and 74 control patients. Allele, genotype, and haplotype frequencies of the polymerase chain reaction products were determined using autoradiography. Haplotype frequencies differed for ulcerative colitis and Crohn's disease, particularly for haplotype CC (22% ulcerative colitis vs 10% Crohn's disease, P = 0.002 Chi2 = 10.0) and haplotype CD (18% Crohn's disease vs 9% ulcerative colitis, P = 0.025 Chi2 = 5.02). These data demonstrate the association of the IBD1 locus with both ulcerative colitis and Crohn's disease in a group of unrelated IBD patients. The use of such microsatellite markers when combined with others, might help distinguish ulcerative colitis from Crohn's disease in patients with ambiguous clinical and histological features.

The diagnosis of low-grade dysplasia in Barrett's esophagus and its implications for disease progression

OBJECTIVE

The reported risk of progression from low-grade dysplasia (LGD) to high-grade dysplasia (HGD) or carcinoma (CA) in Barrett's esophagus varies. However, the validity of a diagnosis of LGD may be questioned because of interobserver variability.

METHODS

A search of the Cleveland Clinic Foundation surgical pathology files between 1986 and 1997 yielded biopsy specimens from 43 patients with Barrett's esophagus diagnosed and coded as LGD. Patients with concurrent or prior diagnoses of HGD or carcinoma were excluded. The LGD cases were randomized and blindly reviewed by three gastrointestinal (GI) pathologists along with cases originally diagnosed as Barrett's esophagus without dysplasia (ND; n = 28), indefinite for dysplasia (IND; n = 14), or HGD (n = 15). Each pathologist classified every biopsy specimen as ND, IND, LGD, or HGD, and interobserver agreements were determined by kappa statistics (K). Follow-up data were available on 25 patients originally diagnosed with LGD. Progression was defined as a subsequent diagnosis of HGD or CA on esophageal biopsy or resection specimens.

RESULTS

Agreement between two GI pathologists for a diagnosis of LGD was fair (K = 0.28) and poor (K = 0.21 and -0.04). Individual GI pathologists agreed with the original diagnosis of LGD in 70%, 56%, and 16% of cases. The 25 patients with follow-up included 21 men and four women (mean age, 67 yr) with a mean follow-up of 26 months (range: 2-84 months). Seven patients (28%) with follow-up developed HGD (five patients) or CA (two patients), 2-43 months (median: 11 months) after a diagnosis of LGD. The individual GI pathologists' diagnosis did not correlate with progression. However, when at least two GI pathologists agreed on LGD, there was a significant association with progression (seven of 17 patients, 41%, p = 0.04). When all three GI pathologists agreed on a diagnosis of LGD, four of five patients progressed (p = 0.012). In contrast, of the eight patients with follow-up and no agreement among GI pathologists for a diagnosis of LGD, none progressed.

CONCLUSIONS

A high degree of interobserver variability is seen in the histological diagnosis of Barrett's esophagus-related LGD. Although the number of observations is low, a consensus diagnosis of LGD among GI pathologists suggests an increased risk of progression from LGD to HGD or carcinoma.

Incidence and natural history of dysplasia of the anal transitional zone after ileal pouch-anal anastomosis: results of a five-year to ten-year follow-up

PURPOSE

Preservation of the anal transitional zone during ileal pouch-anal anastomosis is still controversial because of the risk of dysplasia and the theoretical risk of associated cancer. Without long-term follow-up data, the natural history and optimal treatment of anal transitional zone dysplasia are unknown. The aim of this study was to determine the long-term risk of dysplasia in the anal transitional zone and to evaluate the outcome of a conservative management policy for anal transitional zone dysplasia.

METHODS

Two hundred ten patients undergoing anal transitional zone-sparing ileal pouch-anal anastomosis for ulcerative or indeterminate colitis between 1987 and 1992 and who were studied with serial anal transitional zone biopsies for at least five years postoperatively were included. Median follow up was 77 (range, 60-124) months.

RESULTS

Anal transitional zone dysplasia developed in seven patients 4 to 51 (median, 11) months postoperatively. There was no association with gender, age, preoperative disease duration or extent of colitis, but the risk of anal transitional zone dysplasia was significantly increased in patients with prior cancer or dysplasia in the colon or rectum. Dysplasia was high grade in one and low grade in six. Two patients each with low-grade dysplasia detected on three separate occasions underwent mucosectomy 29 and 38 months after detection of low-grade dysplasia, but no cancer was found. The five other patients with dysplasia on one or two occasions were treated expectantly and were apparently dysplasia-free for a median of 72 (range, 48-100) months.

CONCLUSIONS

Anal transitional zone dysplasia after ileal pouch-anal anastomosis is infrequent, is most common in the first two to three years postoperatively and may apparently disappear on repeated biopsy. Anal transitional zone preservation did not lead to the development of cancer in the anal transitional zone after five to ten years of follow-up. Long-term surveillance is recommended to monitor dysplasia. If repeat biopsy confirms persistent dysplasia, anal transitional zone excision with neoileal pouch-anal anastomosis is recommended.

Incidence and natural history of dysplasia of the anal transitional zone after ileal pouch-anal anastomosis: results of a five-year to ten-year follow-up

PURPOSE

Preservation of the anal transitional zone during ileal pouch-anal anastomosis is still controversial because of the risk of dysplasia and the theoretical risk of associated cancer. Without long-term follow-up data, the natural history and optimal treatment of anal transitional zone dysplasia are unknown. The aim of this study was to determine the long-term risk of dysplasia in the anal transitional zone and to evaluate the outcome of a conservative management policy for anal transitional zone dysplasia.

METHODS

Two hundred ten patients undergoing anal transitional zone-sparing ileal pouch-anal anastomosis for ulcerative or indeterminate colitis between 1987 and 1992 and who were studied with serial anal transitional zone biopsies for at least five years postoperatively were included. Median follow up was 77 (range, 60-124) months.

RESULTS

Anal transitional zone dysplasia developed in seven patients 4 to 51 (median, 11) months postoperatively. There was no association with gender, age, preoperative disease duration or extent of colitis, but the risk of anal transitional zone dysplasia was significantly increased in patients with prior cancer or dysplasia in the colon or rectum. Dysplasia was high grade in one and low grade in six. Two patients each with low-grade dysplasia detected on three separate occasions underwent mucosectomy 29 and 38 months after detection of low-grade dysplasia, but no cancer was found. The five other patients with dysplasia on one or two occasions were treated expectantly and were apparently dysplasia-free for a median of 72 (range, 48-100) months.

CONCLUSIONS

Anal transitional zone dysplasia after ileal pouch-anal anastomosis is infrequent, is most common in the first two to three years postoperatively and may apparently disappear on repeated biopsy. Anal transitional zone preservation did not lead to the development of cancer in the anal transitional zone after five to ten years of follow-up. Long-term surveillance is recommended to monitor dysplasia. If repeat biopsy confirms persistent dysplasia, anal transitional zone excision with neoileal pouch-anal anastomosis is recommended.

The diagnosis of low-grade dysplasia in Barrett's esophagus and its implications for disease progression

OBJECTIVE

The reported risk of progression from low-grade dysplasia (LGD) to high-grade dysplasia (HGD) or carcinoma (CA) in Barrett's esophagus varies. However, the validity of a diagnosis of LGD may be questioned because of interobserver variability.

METHODS

A search of the Cleveland Clinic Foundation surgical pathology files between 1986 and 1997 yielded biopsy specimens from 43 patients with Barrett's esophagus diagnosed and coded as LGD. Patients with concurrent or prior diagnoses of HGD or carcinoma were excluded. The LGD cases were randomized and blindly reviewed by three gastrointestinal (GI) pathologists along with cases originally diagnosed as Barrett's esophagus without dysplasia (ND; n = 28), indefinite for dysplasia (IND; n = 14), or HGD (n = 15). Each pathologist classified every biopsy specimen as ND, IND, LGD, or HGD, and interobserver agreements were determined by kappa statistics (K). Follow-up data were available on 25 patients originally diagnosed with LGD. Progression was defined as a subsequent diagnosis of HGD or CA on esophageal biopsy or resection specimens.

RESULTS

Agreement between two GI pathologists for a diagnosis of LGD was fair (K = 0.28) and poor (K = 0.21 and -0.04). Individual GI pathologists agreed with the original diagnosis of LGD in 70%, 56%, and 16% of cases. The 25 patients with follow-up included 21 men and four women (mean age, 67 yr) with a mean follow-up of 26 months (range: 2-84 months). Seven patients (28%) with follow-up developed HGD (five patients) or CA (two patients), 2-43 months (median: 11 months) after a diagnosis of LGD. The individual GI pathologists' diagnosis did not correlate with progression. However, when at least two GI pathologists agreed on LGD, there was a significant association with progression (seven of 17 patients, 41%, p = 0.04). When all three GI pathologists agreed on a diagnosis of LGD, four of five patients progressed (p = 0.012). In contrast, of the eight patients with follow-up and no agreement among GI pathologists for a diagnosis of LGD, none progressed.

CONCLUSIONS

A high degree of interobserver variability is seen in the histological diagnosis of Barrett's esophagus-related LGD. Although the number of observations is low, a consensus diagnosis of LGD among GI pathologists suggests an increased risk of progression from LGD to HGD or carcinoma.

The importance of diagnostic accuracy in colonic inflammatory bowel disease

OBJECTIVE

Crohn's disease (CD) and ulcerative colitis (UC) may both affect the colon. However, in approximately 10-20% of these cases, it is impossible to distinguish between these two entities either clinically or histologically, and a diagnosis of indeterminate colitis (IC) is made. Correct diagnosis is important because surgical treatment and long-term prognosis differ for UC and CD. The purpose of this study was to determine the extent of interobserver agreement among board-certified pathologists and a specialist gastrointestinal (GI) pathologist regarding the histological diagnosis of colonic inflammatory bowel disease (IBD).

METHODS

A total of 24 university medical center pathologists from eight institutions evaluated 84 colectomy specimens and 35 sets of biopsy specimens from 119 consecutive patients with colonic IBD. A specialist GI pathologist subsequently reviewed all cases without knowledge of clinical data and prior diagnosis.

RESULTS

The GI pathologist's diagnoses differed from the initial diagnoses in 45% of surgical specimens and 54% of biopsy specimens. Of 70 cases initially diagnosed as UC, 30 (43%) were changed to CD or IC, whereas 4 of 23 cases (17%) initially diagnosed as CD were changed to UC or IC. The kappa coefficient for the overall agreement of initial diagnoses with the specialist GI pathologist's diagnoses was -0.01 (p = 0.98).

CONCLUSIONS

There is significant interobserver variation in the histological diagnosis of colonic IBD. This may have a profound effect on clinical patient care and, especially, on the choice of operation. More accurate diagnostic criteria are needed to facilitate patient care and to optimize treatment outcome.

The importance of diagnostic accuracy in colonic inflammatory bowel disease

OBJECTIVE

Crohn's disease (CD) and ulcerative colitis (UC) may both affect the colon. However, in approximately 10-20% of these cases, it is impossible to distinguish between these two entities either clinically or histologically, and a diagnosis of indeterminate colitis (IC) is made. Correct diagnosis is important because surgical treatment and long-term prognosis differ for UC and CD. The purpose of this study was to determine the extent of interobserver agreement among board-certified pathologists and a specialist gastrointestinal (GI) pathologist regarding the histological diagnosis of colonic inflammatory bowel disease (IBD).

METHODS

A total of 24 university medical center pathologists from eight institutions evaluated 84 colectomy specimens and 35 sets of biopsy specimens from 119 consecutive patients with colonic IBD. A specialist GI pathologist subsequently reviewed all cases without knowledge of clinical data and prior diagnosis.

RESULTS

The GI pathologist's diagnoses differed from the initial diagnoses in 45% of surgical specimens and 54% of biopsy specimens. Of 70 cases initially diagnosed as UC, 30 (43%) were changed to CD or IC, whereas 4 of 23 cases (17%) initially diagnosed as CD were changed to UC or IC. The kappa coefficient for the overall agreement of initial diagnoses with the specialist GI pathologist's diagnoses was -0.01 (p = 0.98).

CONCLUSIONS

There is significant interobserver variation in the histological diagnosis of colonic IBD. This may have a profound effect on clinical patient care and, especially, on the choice of operation. More accurate diagnostic criteria are needed to facilitate patient care and to optimize treatment outcome.

Immunohistochemistry in the differential diagnosis of acinar and endocrine pancreatic neoplasms

Histologic differential diagnosis of acinar cell carcinoma (ACC), mixed acinar-endocrine cell carcinoma (MAEC), and pancreatic endocrine tumors (PET) can be difficult but is important because of differences in their clinical behavior. This study investigates the utility of immunohistochemistry (IHC) in this differential diagnosis using immunohistochemical stains that are available in most laboratories. IHC was performed on paraffin-embedded tissue in ACC (n = 6), MAEC (n = 2), and PET (n = 13), using synaptophysin (SYN), chromogranin (CHR), chymotrypsin (CHY), and alpha-1-antitrypsin (AAT). Electron microscopy (EM) was performed in all cases to confirm the diagnosis. Long-term follow-up and death of disease (DOD) was known in all patients. The ACCs stained as follows: CHY (4/6), AAT (3/6), SYN (4/6); CHR was negative in all cases. Both cases of MAEC stained with CHY, AAT, and SYN (2/2); CHR was negative. PET stained as follows: SYN (13/13), CHR (8/13), CHY (4/13), AAT (5/13). In the ACC/ MAEC group, six of eight patients were DOD at mean follow-up of 11 months. Among the PET, two of 16 patients were DOD at mean follow-up of 37 months. Considerable immunophenotypic overlap exists between ACC, MAEC, and PET. Consequently, one can neither confirm nor rule out a diagnosis of ACC or MAEC using generally available immunohistochemical stains alone. These findings support a role for EM in the evaluation of exocrine and endocrine pancreatic neoplasms.

Short- and long-term follow-up of treatment for lymphocytic and collagenous colitis

Collagenous colitis and lymphocytic colitis cause chronic watery diarrhea. Multiple therapies have been found to improve symptoms but there have been few long-term follow-up studies. Our goal was to obtain long-term clinical follow-up on a cohort of patients with independently confirmed typical histopathologic changes. Pathology slides from 32 cases of collagenous or lymphocytic colitis patients from 1988-1992 were independently reviewed. Twenty-five cases were confirmed by both groups of pathologist as collagenous or lymphocytic colitis. For these 25 patients, charts were reviewed and telephone follow-up interviews were performed in 1992 and 1995. Seven of 32 (22%) of the original cases were not confirmed on independent pathologic interpretation. A 15.8% discordance rate was found between the different groups of pathologists. Patient demographics were similar to previously published reports except one-half of our patients had diarrhea of only 6 months or less. Eighty-one percent of patients receiving 5-ASA agents reported improvement as well as 100% of those receiving prednisone. At 23 month follow-up 86% of patients reported improvement in diarrhea and only 32% required routine medications. At 47 month follow-up all patients reported improved diarrhea and only 29% required routine medications. Collagenous and lymphocytic colitis can sometimes be identified in patients with relatively brief duration diarrhea. Clinical parameters and response to therapy are similar for collagenous or lymphocytic colitis. Most patients with lymphocytic and collagenous colitis improve with therapy such as 5-ASA preparations or steroids. Over a follow-up period of several years, most patients have improvement in diarrhea and generally do not require maintenance medications. Independent pathologic confirmation of the diagnosis should be obtained in patients not responding to therapy.

Colonic epithelial lymphocytosis without a thickened subepithelial collagen table: a clinicopathologic study of 40 cases supporting a heterogeneous entity

Lymphocytic colitis (LC) is classically described as a triad of chronic nonbloody, watery diarrhea, normal or nearly normal endoscopy findings, and colonic epithelial lymphocytosis without a thickened subepithelial collagen table (SECT). It is unknown how often patients with colonic epithelial lymphocytosis without a thickened SECT actually present with this classic triad. Cases diagnosed histologically as lymphocytic or microscopic colitis were reviewed. Criteria for inclusion were the presence of at least 15 surface lymphocytes per 100 epithelial cells and the absence of a thickened SECT (<12 microm). Clinical features and course were recorded by chart review and telephone follow-up. Forty patients met the inclusion criteria, including 25 women and 15 men with a mean age of 63.2 years (range, 25-83 years). Twenty-eight patients had the classic triad and were designated as having classic LC. The other 12 patients fulfilled the histologic criteria but not the clinical or endoscopic criteria for classic LC and were classified as having atypical LC (constipation, five patients; macroscopic colitis at endoscopy, five patients; hematochezia, one patient; and incidental finding, one patient). Clinically, patients with classic LC were predominantly women and had a higher incidence of autoimmune disease (p = 0.03) than did those with atypical LC. Histologically, surface eosinophilia was significantly greater in patients with classic LC (p = 0.04). Twenty patients were using nonsteroidal antiinflammatory drugs at the time of their colonic biopsy. Surface epithelial lymphocyte counts were higher in these patients, particularly in the distal sigmoid colon (p = 0.02). Fourteen patients had associated autoimmune disease, including three patients with sprue diagnosed by small bowel biopsy, all of whom responded to gluten withdrawal. Diarrhea present in 25 patients, without documented evidence of celiac sprue, was self-limited in five, resolved with treatment in three, required intermittent treatment in eight, daily treatment in five, and was refractory to treatment in four. All eight patients who experienced spontaneous or treatment-related symptom resolution had classic LC. No histologic feature correlated with clinical course. In conclusion, our study shows that colonic epithelial lymphocytosis without a thickened SECT is a histologic finding seen in a heterogeneous group of patients. Within this heterogeneous group is a distinct subset of patients who have the classic clinicopathologic triad of LC. This subset of patients has striking similarities to patients with collagenous colitis, lending further support to a close relationship between these two entities. Atypical LC comprises a heterogeneous group and includes patients with idiopathic constipation, coexisting LC and inflammatory bowel disease, and possibly infectious colitides. Because of the clinical heterogeneity among our study population, the descriptive term colonic epithelial lymphocytosis may be a more prudent diagnosis than lymphocytic colitis in the absence of adequate clinical information.

Intestinal adenocarcinoma in Crohn's disease: a report of 30 cases with a focus on coexisting dysplasia

There are relatively few reports that detail the types of intestinal adenocarcinoma complicating Crohn's disease and examine associated epithelial dysplasia. We determined the prevalence, grade, and type of dysplasia found adjacent to and distant from Crohn's-related adenocarcinomas. Thirty cases of resected Crohn's-related adenocarcinoma were reviewed, and histologic type, degree of differentiation, TNM stage, and the presence or absence, grade, and location of dysplasia were recorded. Most of the patients were male (70%). The median ages at diagnosis of Crohn's disease and adenocarcinoma were 34 and 49 years, respectively. The extent of Crohn's disease included ileocolitis in 21 patients, only colonic disease in six, and only small bowel disease in three. In most cases (67%), carcinoma was found incidentally at surgery. All carcinomas arose in areas involved by Crohn's disease. Eight (27%) adenocarcinomas arose in the small bowel, and 22 (73%) arose in the colon, including two in out-of-circuit rectums. Most carcinomas (63%) were poorly differentiated. Dysplasia was found adjacent to the carcinoma in 26 (87%) cases. Of the colorectal carcinomas, 19 (86%) had adjacent dysplasia, and nine (41%) had distant dysplasia. In conclusion, most cases of Crohn's-related intestinal adenocarcinoma have dysplasia in adjacent mucosa, and 41% of those arising in the colorectum have distant dysplasia, supporting a dysplasia-carcinoma sequence in Crohn's disease.

Acquired increase in leucocyte binding by intestinal microvascular endothelium in inflammatory bowel disease

BACKGROUND

Endothelial cells that line microvascular blood vessels have an important role in inflammation through their ability to bind and recruit circulating leucocytes. Endothelial cells from the intestines of patients with chronically inflamed Crohn's disease and ulcerative colitis--the two forms of inflammatory bowel disease--display an increased leucocyte-binding capacity in vitro. We investigated whether this enhanced leucocyte binding is a primary or an acquired defect.

METHODS

We cultured human intestinal microvascular endothelial cells (HIMEC) from the uninvolved intestine and chronically inflamed bowel of three patients with inflammatory bowel disease (two Crohn's disease, one ulcerative colitis). We assessed HIMEC binding to polymorphonuclear leucocytes and U937 cells by means of an adhesion assay.

FINDINGS

After activation with interleukin-1beta or lipopolysaccharide, HIMEC from the chronically inflamed tissue in all three patients with inflammatory bowel disease bound twice as many polymorphonuclear leucocytes and U937 cells as endothelial cells from uninvolved tissue.

INTERPRETATION

Enhanced leucocyte binding by HIMEC from chronically inflamed tissue in patients with inflammatory bowel disease is an acquired defect since it is not found in the uninvolved intestinal segments from the same individuals. Because interaction between endothelial cells and leucocytes is a key regulatory step in the inflammatory process, this enhanced binding may contribute to the pathophysiology of chronic intestinal inflammation.

Solitary pulmonary papillomas in adults: a clinicopathologic and in situ hybridization study of 14 cases combined with 27 cases in the literature

Solitary endobronchial papillomas in adults are rare neoplasms. Only sporadic cases have been documented. The histologic classification of these tumors remains problematic, and little is known about their clinical behavior. The clinical and pathologic features of 13 endobronchial papillomas and a single endobronchiolar papilloma were reviewed. In situ hybridization for human papillomavirus (HPV) types 6/11, 16/18, and 31/33/51 was performed on seven cases. Twenty-seven additional well-documented cases were identified in a literature review. Human papillomavirus studies were performed in four of the previously reported cases. The 41 neoplasms combined from the Armed Forces Institute of Pathology and literature review were divided into three groups according to their histologic features. Thirty-one of 41 (76%) patients were men. The ages of the patients ranged from 26 to 74 years (median, 57 years). Three morphologically distinct histologic types were recognized; 27 squamous cell papillomas, 7 glandular papillomas, and 7 mixed squamous and glandular papillomas. Squamous papillomas: 23 of 27 (85%) patients were men, and the median age was 54 years. Six of eleven (55%) of these patients smoked. Twenty-six lesions were exophytic and a single lesion had an inverted pattern. Seven of 24 (29%) lesions featured cytologic atypia and 5 of 24 (14%) had viral cytopathic effect. Five of seven (71%) cases examined for HPV DNA were positive. Three of 18 (17%) recurred. Glandular papillomas: Four of seven (57%) patients were women. The mean age was 67 years. One of five (20%) patients smoked. Five lesions were central, and two were peripheral. Four lesions had columnar epithelium, and three had ciliated epithelium. One of six (17%) lesions recurred. Mixed papillomas: five of seven (71%) patients were men. The median age was 64 years. Three of five (60%) patients smoked. Three of seven (43%) lesions featured cytologic atypia. Four of five lesions were examined for HPV DNA and all were negative. No lesions recurred. This study demonstrates that solitary endobronchial papillomas can be separated into three distinct morphologic categories. Squamous cell and mixed papillomas are predominantly lesions of male smokers in their 6th decade. Although cytologic atypia is observed in many cases, the rarity of these tumors and difficulty in separating papillomas from endobronchial papillary squamous carcinomas make generalizations regarding the risk of progression to carcinoma tenuous at best. Human papillomavirus appears to play a pathogenetic role in some squamous cell papillomas, but not in mixed papillomas, yet its presence in the squamous lesions does not correlate with recurrence or malignancy. The first report of an inverted squamous cell papilloma indicates clinical features similar to the more common exophytic squamous cell papillomas. Glandular papillomas, the rarest of all endobronchial papillomas, are found in an older age group than squamous and mixed papillomas, and most-patients are nonsmokers. Based on these findings, all endobronchial papillomas should be completely excised.

Sarcoidosis presenting as granulomatous colitis

Sarcoidosis presenting solely as a granulomatous colitis is rare and appears identical to Crohn's disease. A 56-yr-old woman developed a Crohn's-like illness, which remitted after 5-ASA therapy. Two months later, she developed fever, adenopathy, muscle weakness, and peripheral neuropathy. A diagnosis of sarcoidosis was made after an extensive search for an infectious or rheumatological cause. This case illustrates the utility of serum angiotensin converting enzyme level in differentiating sarcoidosis from Crohn's disease.

Surveillance of patients with Barrett's esophagus for dysplasia and cancer with balloon cytology

BACKGROUND & AIMS

A less costly cancer surveillance method for Barrett's esophagus is desirable. The aim of this study was to compare nonendoscopic balloon cytology with biopsy and brush cytology for detecting dysplasia and carcinoma in patients with Barrett's esophagus.

METHODS

Patients in a surveillance program underwent balloon cytology before endoscopy with biopsy and brush cytology. Results of cytology were compared with those of histology.

RESULTS

Adequate columnar epithelium was obtained in 52 of 63 (83%) patients with balloon cytology and 59 of 61 (97%) with brush cytology. Balloon cytology obtained abnormal cells in 6 of 8 patients with adenocarcinoma, 2 of 2 patients with high-grade dysplasia, and 2 of 8 patients with low-grade dysplasia. Sensitivity of balloon cytology for high-grade dysplasia or carcinoma was 80% but only 25% for low-grade dysplasia. No patients without dysplasia or carcinoma had abnormal cells. Brush cytology was abnormal in all 11 patients with high-grade dysplasia or carcinoma but only 2 of 9 patients with low-grade dysplasia (sensitivity, 22%). Two of 39 patients without dysplasia had abnormal cells (specificity, 95%). Balloon cytology cost was sixfold less than endoscopy with biopsy.

CONCLUSIONS

Balloon cytology detected 80% of patients with high-grade dysplasia or carcinoma when sampling was adequate. Brush cytology data suggest that a more abrasive balloon may improve balloon cytology sensitivity. The potential cost savings of balloon cytology compared with endoscopic cancer surveillance in Barrett's esophagus support further studies of this technique.

Colonic epithelial lymphocytosis associated with an epidemic of chronic diarrhea

The term Brainerd diarrhea has been applied to outbreaks of chronic watery diarrhea of unknown etiology characterized by acute onset and prolonged duration. Our aim was to describe the histologic changes in gastrointestinal biopsy specimens from patients with Brainerd diarrhea. We examined 52 colonic and 12 small bowel biopsy specimens from 22 patients who were involved in an outbreak of Brainerd diarrhea that was linked to the water supply of a cruise ship visiting the Galapagos Islands. Small bowel biopsy specimens from seven patients were histologically normal. One patient had a duodenal biopsy specimen that resembled celiac sprue. Colonic biopsy specimens from 20 patients revealed surface epithelial lymphocytosis without distortion of mucosal architecture, surface degenerative changes, or thickened subepithelial collagen plates. The degree of surface epithelial lymphocytosis was greater than that seen in control groups of persons with normal colons, acute colitis, and ulcerative colitis (p < 0.001), similar to that seen with collagenous colitis, and less than that seen with lymphocytic colitis (p < 0.001). Three patients showed focal active colitis similar to that described in acute infectious-type colitis in addition to the epithelial lymphocytosis. Two patients had colonic biopsy specimens that were histologically normal. In summary, histologic abnormalities in the small bowel are generally absent in Brainerd diarrhea. Colonic biopsy specimens in Brainerd diarrhea frequently show epithelial lymphocytosis similar to that seen in collagenous and lymphocytic colitis. Although currently Brainerd diarrhea can be diagnosed only with epidemiologic data indicating an epidemic and a point source, the lack of surface degenerative changes and the relatively lower lymphocyte counts seen in our cases of Brainerd diarrhea may serve to distinguish it from lymphocytic colitis, and the lack of a thickened subepithelial collagen plate distinguishes it from collagenous colitis.

Detection of dysplasia at colonoscopy using laser-induced fluorescence: a blinded study

BACKGROUND

Laser-induced fluorescence spectroscopy has the potential to detect colonic dysplasia in vivo. However, previous studies have limited their analyses to multivariate regression techniques and unblinded retrospective evaluation. The purpose of this study was to develop a probability-based algorithm to detect colonic dysplasia using laser-induced fluorescence spectroscopy and to evaluate it in a blinded manner.

METHODS

Fluorescence spectra were collected from normal mucosa and colonic polyps during colonoscopy using 370 nm excitation. Tissue was classified as normal, hyperplastic, or adenomatous by histologic examination. Preliminary data was used to devise an algorithm to differentiate tissue type based on probability distributions of the fluorescence intensity at 460 nm and the ratio of the intensity at 680 nm to that at 600 nm. The algorithm was then tested in a blinded fashion.

RESULTS

The algorithm correctly determined the tissue type in 88% of cases, equal to the agreement of independent pathologists. Sensitivity, specificity, and positive predictive value for the detection of dysplasia was 90%, 95%, and 90%, respectively.

CONCLUSIONS

Dysplasia was detected in vivo using fluorescence spectroscopy and a probability-based algorithm. This method may form the basis for a new surveillance technique for patients with increased risk for dysplastic transformation.

Methylene blue selectively stains intestinal metaplasia in Barrett's esophagus

BACKGROUND

Specialized columnar epithelium in Barrett's esophagus resembles gastric intestinal metaplasia, which selectively stains with methylene blue.

METHODS

We prospectively evaluated the safety, accuracy, reproducibility, cost, and diagnostic yield of methylene blue-directed biopsy in detecting specialized columnar epithelium and dysplasia in Barrett's esophagus. We performed upper endoscopy with methylene blue-directed biopsy and obtained 236 large cup biopsy specimens (145 stained, 91 unstained) from 14 patients with Barrett's esophagus of any length (Group 1) and 12 control patients. Biopsy specimens were independently examined by two pathologists unaware of the endoscopic results.

RESULTS

Methylene blue stained specialized columnar epithelium in 18 of the 26 patients, including those with intramucosal carcinoma (1), high-grade dysplasia (1), and indefinite/low-grade dysplasia (6). Methylene blue staining pattern, which was focal in 72% and diffuse in 28% of patients, was reproduced in 8 patients who had repeat staining within 4 weeks. The overall accuracy of methylene blue staining for detecting specialized columnar epithelium was 95%. The diagnostic yield of methylene blue staining for specialized columnar epithelium in "control" patients was 42%. The risk for dysplasia in stained biopsy specimens was greater than in unstained ones (odds ratio 17.7, p = .0004).

CONCLUSIONS

Methylene blue mucosal staining is a safe, inexpensive, reproducible, and highly accurate method of diagnosing specialized columnar epithelium in Barrett's esophagus.

A clinicopathologic study of 42 patients with granulomatous gastritis. Is there really an "idiopathic" granulomatous gastritis?

Idiopathic granulomatous gastritis (IGG) is a diagnosis made only by excluding other causes of granulomatous gastritis, such as infection, foreign bodies, and systemic granulomatous diseases. Recently, several investigators have questioned the existence of IGG. We reviewed the slides and clinical data of all cases of granulomatous gastritis seen at the Cleveland Clinic between 1975 and 1994. In addition to routine hematoxylin and eosin stains, slides from all cases were stained with Ziehl-Neelsen, Gomori's methenamine silver, and Giemsa stains. Clinical information and follow-up were available for 42 patients. The clinicopathologic diagnoses of the 42 patients with granulomatous gastritis were as follows: Crohn's disease (n = 23), three of whom had concomitant chronic active gastritis with Helicobacter pylori infection; sarcoidosis (n = 9), four of whom had concomitant chronic active gastritis with H. pylori infection; chronic active gastritis with H. pylori infection and no other systemic illness (n = 2); distal esophageal adenocarcinoma and chronic active gastritis (n = 2); mucosa-associated lymphoid tissue (MALT) lymphoma with chronic active gastritis and presumed H. pylori infection (n = 2); peptic ulcer complications (n = 2); hypertrophic gastropathy with chronic active gastritis (n = 1); and possible Crohn's disease (n = 1). We conclude that (a) in most cases of granulomatous gastritis, a diagnosis of Crohn's disease or sarcoidosis could be established; (b) the background inflammatory pattern was helpful in suggesting a diagnostic category for granulomatous gastritis; (c) granulomatous gastritis is not associated with H. pylori per se; however, if known cases of Crohn's disease and sarcoidosis are excluded, an association between H. pylori and granulomatous gastritis cannot be ruled out; and (d) IGG, if it exists, is extremely rare.

Diagnostic yield and therapeutic impact of flexible bronchoscopy in lung transplant recipients

BACKGROUND

Bronchoalveolar lavage and transbronchial biopsy are often used for definitive diagnosis of lung rejection and infection in lung transplant recipients. Although protected specimen brushing is of value in nosocomial bacterial pneumonia, its role in lung transplant recipients had not been widely reported. The aim of the study is to review the diagnostic yield and therapeutic impact of flexible bronchoscopy with the use of a combination of bronchoalveolar lavage, protected specimen brushing, and transbronchial biopsy in lung transplant recipients.

METHODS

We reviewed flexible bronchoscopy data in 83 transplant recipients between February 1990 and March 1995. Only those with bronchoalveolar lavage, protected specimen brushing, and transbronchial biopsy were included in the analysis. There were 282 bronchoscopies performed for clinically suspected lung rejection or infection (clinical bronchoscopy) and 38 bronchoscopies for follow-up of a previously detected histologic abnormality (follow-up bronchoscopy).

RESULTS

The total yields for rejection and infection for clinical and follow-up bronchoscopies were 67.4% and 58.9%, respectively. Acute rejection was detected with transbronchial biopsy in 26.2% and 34.2% of clinical and follow-up bronchoscopies, respectively. Cytomegalovirus pneumonitis was detected with transbronchial biopsy in 4.0% and 11.4% of clinical and follow-up bronchoscopies, respectively. Overall, bacteria was the most common cause of lower respiratory tract infection. When used together, protected specimen brushing and bronchoalveolar lavage were complementary techniques for detection of bacterial lower respiratory tract infection with a significantly higher proportion detected with protected specimen brushing ( > or = 10(3) colony forming units/ml) compared with bronchoalveolar lavage ( > or = 10(5) colony forming units/ml) (p < 0.001). Complications were hemorrhage (1.9%), pneumothorax (2.5%) and transient hypoxemia (10.5%). The results had an impact on management of rejection and infection in 57.8% of clinical and 39.5% of follow-up bronchoscopies.

CONCLUSIONS

We conclude that bronchoscopy, with the use of a combination of bronchoalveolar lavage, protected specimen brushing, and transbronchial biopsy, is safe with a high diagnostic yield and therapeutic impact for treating lung transplant recipients.

Management and outcome of patients with invasive carcinoma arising in colorectal polyps

BACKGROUND & AIMS

Treatment for invasive adenocarcinoma in colorectal polyps (malignant polyps) is controversial. The aim of this study was to evaluate our institutional treatment strategy for malignant polyps.

METHODS

Malignant polyps were designated as having favorable histology (grade I or II carcinoma with at least a 2-mm free margin) or unfavorable histology (grade III invasive adenocarcinoma, invasive adenocarcinoma with an unassessable margin, or a margin of < 2 mm). Malignant polyps with favorable histology were considered treated adequately by endoscopic polypectomy, whereas further therapy was recommended for malignant polyps with unfavorable histology. Recurrence, residual adenocarcinoma in a follow-up resection specimen, or metastasis during follow-up were considered adverse outcomes.

RESULTS

Of the 47 patients identified, 17 (36%) had favorable histology. Sixteen patients (94%) were treated with polypectomy alone. None had an adverse outcome (median follow-up, 70 months). Thirty patients (64%) had unfavorable histology, and 21 patients (70%) underwent colectomy. Five patients underwent radiation therapy alone. Four patients underwent no additional therapy. Ten of 30 patients with unfavorable histology had adverse outcomes that differed significantly from the favorable histology group (P = 0.03).

CONCLUSIONS

Endoscopic polypectomy alone is adequate therapy for malignant polyps with favorable histology.

Incidence, risk factors, and treatment of dysplasia in the anal transitional zone after ileal pouch-anal anastomosis

Preservation of the anal transitional zone (ATZ) after restorative proctocolectomy and stapled ileal pouch-anal anastomosis (IPAA) for ulcerative colitis is controversial.

PURPOSE

To evaluate the incidence, risk factors, and treatment options for dysplasia and/or cancer after restorative proctocolectomy and stapled IPAA.

METHODS

We reviewed the records of all 254 patients operated on for ulcerative colitis who had a restorative proctocolectomy, stapled IPAA, and annual postoperative biopsies of ATZ. Follow-up studies included an annual questionnaire and physical examination.

RESULTS

During a follow-up of 2.3 +/- 1.4 (mean +/- standard deviation) years, low-grade dysplasia was found in eight patients (3.1 percent), 16 (median: range, 6-56) months after surgery. Repeated biopsies revealed dysplasia in only two of eight patients, and completion mucosectomy was performed. Dysplasia in ATZ was associated with a preoperative (P = 0.02) or postoperative (P = 0.04) pathologic diagnosis of ulcerative colitis with concurrent dysplasia or cancer. No association (P > 0.05) was found between dysplasia and the following: age, sex, preoperative length of disease, use of a double-stapled versus single-stapled technique, or anastomotic distance from the dentate line.

CONCLUSIONS

Incidence of low-grade dysplasia in ATZ was low. Restorative proctocolectomy with total mucosectomy of the anal canal and handsewn IPAA is recommended for patients with preoperative diagnosis of ulcerative colitis and concurrent cancer or dysplasia. Frequent follow-up with biopsies is recommended for patients with incidental finding of cancer or high-grade dysplasia after restorative proctocolectomy and stapled IPAA with preservation of ATZ. For persistent or recurrent low-grade dysplasia, we recommend a completion mucosectomy.

Pseudomyxoma peritonei. A clinicopathologic study of 19 cases with emphasis on site of origin and nature of associated ovarian tumors

Nineteen patients (10 men, 9 women) with pseudomyxoma peritonei were studied to determine the site of origin of the disease and the nature of associated ovarian tumors. A primary appendiceal mucinous neoplasm was found in 16 of the 17 patients with an evaluable appendix. One woman's appendix had intramural and serosal mucinous deposits without neoplastic epithelial cells. The appendiceal tumors were classified as mucin-producing adenoma ("mucinous cystadenoma") in 13 patients (seven women, six men), intramucosal adenocarcinoma associated with mucin-producing adenoma in one man, and invasive adenocarcinoma associated with mucin-producing adenoma in one man and one woman. Five of the nine women had cystic mucinous ovarian neoplasms; each also had an appendiceal mucinous neoplasm (adenoma in four, invasive adenocarcinoma with adenoma in one). The ovarian neoplasms had histologic features resembling a borderline (low malignant potential) mucinous tumor in four and a mucinous cystadenoma in one; all five ovarian tumors also had features of pseudomyxoma ovarii. Mucinous implants were also on the ovarian surface of the contralateral ovary in four of the five women with ovarian tumors and in the other four women without ovarian tumors. The intraperitoneal mucus deposits contained neoplastic mucinous epithelial cells in 16 patients (eight men, eight women) and were acellular in three (two women, one man). Of 17 patients with more than 6 months of follow-up, 12 (seven women, five men) were alive after postoperative intervals of 7 to 147 months, including three with known residual disease. Five (three men, two women) died of disease 16 to 60 months after initial operation, including two patients with appendiceal carcinoma. Acellular intraperitoneal mucus appeared to be a favorable prognostic feature. We conclude that (a) the appendix is the primary site of origin of pseudomyxoma in the vast majority of cases in both men and women, and (b) the associated mucinous ovarian tumors are most likely secondary neoplasms resulting from incorporation of implanted mucus and neoplastic mucinous epithelial cells of the pseudomyxoma peritonei.

Surgical management of high-grade dysplasia in Barrett's esophagus

The role of surgery in patients with Barrett's esophagus and high-grade dysplasia is controversial. The aims of this study were to determine the prevalence of unsuspected early cancer and to evaluate surgical outcome in a cohort of patients with high-grade dysplasia. Records of all 16 patients who underwent esophagectomy for high-grade dysplasia from 1986 to 1991 were reviewed. All had preoperative endoscopy with no gross evidence of carcinoma, and none had a preoperative diagnosis of intramucosal or invasive carcinoma. Intramucosal carcinoma was found in six (38%) resection specimens. There were no cases of invasive carcinoma or lymph node metastases. One patient (6%) died 3 months postoperatively. The remaining patients are alive without evidence of recurrent cancer (range of follow-up, 2-68 months). Early postoperative complications occurred in seven patients (44%). Late complications occurred in 11 patients (73%). Anastomotic strictures accounted for seven of the 11 (64%) late complications. Complications were successfully managed conservatively in all but two patients. One required laryngectomy for chronic aspiration and another required a gastrojejunostomy for gastric outlet obstruction. Intramucosal carcinoma that had been unsuspected is frequently found in patients with Barrett's esophagus and high-grade dysplasia. Mortality associated with esophagectomy is low, and perioperative complications can usually be managed conservatively. Esophageal resection is indicated in appropriately selected patients with Barrett's esophagus and high-grade dysplasia.

Leiomyosarcoma of the rectum and anal canal

Leiomyosarcoma of the rectum and anal canal is rare and the optimal treatment is not clear. Eight patients with isolated anorectal leiomyosarcoma treated surgically were reviewed. The age ranged from 44 to 76 years (median 63 years) and the follow up ranged from 6 months to 4.5 years (median 2 years). All patients were symptomatic at presentation. All tumours involved the muscularis propria of the low and/or mid-rectum with three tumours also involving the anal sphincters. The tumour size ranged from 1.2 to 10 cm (median 4 cm). Mucosal involvement occurred in only three patients and there was no lymph node involvement. All showed microscopic infiltration at the advancing border, despite macroscopic circumscription. Only one patient was thought to have a tumour sufficiently small (3 cm) and localized on clinical and intrarectal ultrasound examinations (UST2N0) to be suitable for wide local excision. That patient remained tumour-free after 2 years. The remaining patients (88%) were treated by abdomino-perineal resection. The disease free interval in this latter group ranged from 3 months to 4.5 years. All recurrences were detected within 15 months of surgery and the mean interval from detection of recurrence to death was 9 months. Using a histological grading system that included tumour differentiation, mitotic count and amount of necrosis, high grade sarcomas were associated with a worse prognosis. Other factors associated with a poor outcome included large tumour size (> 6-7 cm) and prior incomplete local excision.

Anorectal malignant melanoma has a poor prognosis

Clinicopathologic features and surgical treatment of 15 patients with primary anorectal malignant melanoma were studied retrospectively. There was a female preponderance (2:1). The median age was 66 years. Common initial symptoms were rectal bleeding (87%) and/or anal pain (33%); 25% of the melanomas were amelanotic. The maximum tumor size ranged between 0.8 and 8.4 cm (median 3.0 cm). Of the tumors evaluated histologically (n = 12), tumor thickness ranged from 0.9 to 11.3 mm (median 6.1 mm). All melanomas invaded at least into the subepithelial tissue (n = 8) and/or the submucosa of the distal rectum (n = 4), with extension into the internal anal sphincter (n = 5) and lamina propria (n = 3). Endoluminal ultrasound accurately demonstrated depth of invasion in 3 of 3 patients. Three (20%) patients with distant metastases at initial presentation had a mean survival of 8 mo; one of these primary melanomas measured 0.8 cm. Of 12 patients undergoing "curative" treatments--4 by abdominoperineal resection (APR) and 8 by local excision (LE), the incidence of loco-regional recurrence was similar (2/4 and 5/8). All these 7 patients with loco-regional recurrence developed distant metastases within 3 months. The mean survival was similar between APR and LE in the total group (25 mo vs 20 mo), in the decreased (27 mo vs 24 mo) and in those treated with a curative intent (29 mo vs 22 mo). There was no long-term survivor but four patients remained tumor-free up to 19 mo after APR (n = 1) or LE (n = 3).(ABSTRACT TRUNCATED AT 250 WORDS)

Clinical and pathologic factors associated with delayed diagnosis in solitary rectal ulcer syndrome

Solitary rectal ulcer syndrome (SRUS) often goes unrecognized or is misdiagnosed. Of 98 patients with a final clinicopathologic diagnosis of SRUS, an initially incorrect diagnosis had been made in 25 patients (26 percent). In these 25 patients with a misdiagnosis, the median age was 43 years and the female-to-male ratio was 3.2:1. The median duration of incorrect diagnosis was five years (range, three months to 30 years), and seven patients received prednisone (> 30 mg/day) for a mistaken diagnosis of inflammatory bowel disease. The main clinical symptoms were rectal bleeding (84 percent) and a disturbance of bowel function (56 percent). Rectal prolapse was present in 13 patients. Original rectal biopsy specimens from 23 patients were reviewed; inadequate specimens and failure to recognize diagnostic features of SRUS contributed to delayed diagnosis in 13 and 10 patients, respectively. The most common clinicopathologic misdiagnoses in SRUS patients with rectal ulcers or mucosal hyperemia were Crohn's disease and mucosal ulcerative colitis. In patients with "polypoid" SRUS, diagnostic confusion was usually with a neoplastic polyp. Persistent bowel symptoms and rectal lesions led to review of the presentations and repeat biopsy directed toward the edge of the rectal ulcers or from within the polypoid or hyperemic rectal lesions, finally establishing the diagnosis of SRUS. Intractable symptoms led to surgery in 15 patients (60 percent), with symptomatic improvement in over two-thirds.

Ki-67-determined growth fraction versus standard staging and grading parameters in colorectal carcinoma. A multivariate analysis

BACKGROUND

The antibody Ki-67 binds to nuclei in all cell cycle phases except GO and can be used to measure growth fraction. Because proliferative activity has been linked to prognosis in neoplasia, the authors analyzed 100 cases of colorectal carcinoma, each with 3 or more years of follow-up, using Ki-67 immunostaining.

METHODS

The Ki-67-positive nuclear area and total nuclear area of carcinoma cells in 20 microscopic fields were measured by computed morphometry. A Ki-67 score (percent positive nuclear area x 100) was calculated. The following characteristics also were recorded for each case: patient age and sex, tumor site and size, modified Dukes' stage, spread beyond bowel wall, lymph node status, tumor grade, histologic type, extramural venous spread, tumor growth pattern, fibrosis, lymphocytic infiltration, and mitotic rate.

RESULTS

Ki-67 scores ranged from 1 to 90 (mean, 34.6). Ki-67 scores were higher in Stage A disease (versus Stage B, C, and D disease) but were not associated with survival. Survival curves differed by stage, lymph node metastases, infiltrative growth pattern, lymphocytic infiltration, fibrosis, extramural venous spread, and tumor grade in a univariate analysis. The infiltrative growth pattern (P = 0.04) and lymphocytic infiltration (P = 0.003) were features associated independently with survival after adjusting for modified Dukes' stage. Furthermore, the lack of a significant lymphocytic infiltrate was associated with a death rate 3.4 times greater than that occurring in patients with Stage B disease with a significant infiltrate.

CONCLUSIONS

The authors conclude that proliferative activity in colorectal carcinoma as measured by Ki-67 immunostaining was not associated with prognosis.

Colonic vasoactive intestinal peptide nerves in inflammatory bowel disease

Vasoactive intestinal peptide is a neuropeptide with potent modulatory activity on intestinal immunity and may be implicated in the pathogenesis of inflammatory bowel disease (IBD). Previous studies have reported abnormal morphology of vasoactive intestinal peptide-stained enteric nerves, in addition to increased, normal or decreased levels of extractable peptide in Crohn's disease (CD) and ulcerative colitis (UC) tissues. These observations have not been correlated with the amount of enteric nerve fibers or the degree of mucosal inflammation. The investigation was intended to determine whether abnormalities of vasoactive intestinal peptide in IBD are related to quantitative changes of enteric nerve fibers or mucosal inflammation, and whether they are specific for CD or UC. To do this, digitized morphometric analysis was applied to a large number of IBD and control colonic surgical specimens that were immunostained for vasoactive intestinal peptide and S100 protein and scored for severity of inflammation. The results showed that, as compared with controls, there is a marked decrease of vasoactive intestinal peptide-immunoreactive nerve fibers in the lamina propria and submucosa (P less than 0.0001), and of S100-immunoreactive nerve fibers in the lamina propria (P less than 0.0001) of patients with IBD. In the lamina propria but not the submucosa, the variation of decrease is significantly associated with the severity (P less than 0.0001) but not the type (P greater than 0.9) of IBD because it is detected in both CD and UC. We conclude that in IBD there is loss of mucosal neuropeptidic innervation that is intimately associated with inflammation. This loss probably represents a nonspecific event subsequent to damage to enteric nerve fibers but may contribute to disruption of local immunoregulation.

Do microscopic abnormalities at resection margins correlate with increased anastomotic recurrence in Crohn's disease? Retrospective analysis of 100 cases

The relationship between histologic changes at resection margins and anastomotic recurrence was evaluated in patients with Crohn's disease. Pathology and medical records from 1960 to 1977 identified 100 patients who met the following criteria: 1) no prior surgery for Crohn's disease, 2) small bowel or small bowel and colonic resection with anastomosis done for Crohn's disease at the Cleveland Clinic, and 3) resection margins available for microscopic analysis. The following histologic features of the margins were evaluated: edema, inflammation, lymphoid aggregates, pyloric metaplasia, fibrosis, cryptitis and crypt abscesses, ulcers, granulomas, villous shortening, mucin depletion, neuronal hyperplasia, and transmural inflammation. Additionally, margins were categorized as histologically normal, showing nonspecific changes, showing changes suggestive of Crohn's disease, and showing changes diagnostic for Crohn's disease. Anastomotic recurrence occurred in 50 patients after an average follow-up period of 11.5 years. Cumulative recurrence-free rates for the four margin categories were not significantly different. Anastomotic recurrence was not associated with any clinical or histologic feature or combination of features.

The spectrum of eosinophilic gastroenteritis. Report of six pediatric cases and review of the literature

Eosinophilic gastroenteritis is an inflammatory disease of unknown etiology characterized by infiltration of the gastrointestinal tract with eosinophilic leukocytes, accompanied by varying abdominal symptoms and usually by peripheral blood eosinophilia. We report our experience with six pediatric cases presenting to the Cleveland Clinic Foundation over the past eight years. Unusual findings in our patients included ascitic fluid without eosinophilia and eosinophilic pericarditis (one patient), and eosinophilic cholecystitis (one patient). Endoscopic examination and biopsy helped to establish the diagnosis in all patients. Bone marrow aspiration supported the diagnosis by demonstrating eosinophilia and identifying reactivation of the disease, even in cases without peripheral eosinophilia. All six patients responded promptly to prednisone. Diagnosis is challenging and eosinophilic gastroenteritis may be more common than is recognized. This series of cases significantly expands the spectrum of the disease in children, and documents the usefulness of diagnostic endoscopy in this condition.

Spectroscopic diagnosis of colonic dysplasia

We have developed a method for defining diagnostic algorithms for pathologic conditions based on fluorescence spectroscopy. We apply this method to human colon tissue and show that fluorescence can be used to diagnose the presence or absence of colonic adenoma. This method uses fluorescence excitation-emission matrices (EEM) to identify optimal excitation regions for obtaining fluorescence emission spectra which can be used to differentiate normal and pathologic tissues. In the case of normal and adenomatous colon tissue, these were found to be: 330, 370, and 430 nm +/- 10 nm. At these excitation wavelengths, emission wavelengths for use in diagnostic algorithms are identified from average difference and ratio of the spectra from normal and pathologic tissues. In colon tissue, at 370 nm excitation, 404, 480, and 680 nm were found to be useful emission wavelengths for diagnosing the presence of adenoma in vitro. The basis of colon tissue autofluorescence was investigated using EEM of pure molecules and relevant excitation-emission maxima in the literature.

Escherichia coli O157:H7-associated colitis. A clinical and histological study of 11 cases

Hemorrhagic colitis is characterized by abdominal cramps, bloody diarrhea, and no or low-grade fever. Most cases are caused by the Shiga-like toxin-producing bacteria, Escherichia coli O157:H7. Nineteen colonic biopsy specimens and one resection specimen were reviewed from 11 patients with E. coli O157:H7-associated colitis to determine whether histologic features could be useful in diagnosis or in suggesting pathogenesis. All specimens showed hemorrhage and edema in the lamina propria. Specimens from nine patients were focally necrotic and showed hemorrhage and acute inflammation in the superficial mucosa with preservation of the deep crypts, similar to the pattern of injury associated with acute ischemic colitis. Specimens from five patients showed neutrophils focally infiltrating the lamina propria and crypts, resembling the pattern of injury seen in infectious colitis. One or both of these histologic patterns were observed in specimens from all but one patient. Specimens from four patients had poorly formed inflammatory pseudomembranes. It is concluded that the histologic features of E. coli O157:H7-associated colitis resemble a combination of ischemic and infectious injuries similar to those described in toxin-mediated Clostridium difficile-associated colitis. This suggests that the toxin(s) produced by these E. coliplay a role in the colonic injury. Infection with E. coli O157:H7 should be considered in the differential diagnosis of ischemic and infectious colitis.

Deoxyribonucleic acid ploidy and cell cycle analysis of colorectal carcinoma by flow cytometry. A prospective study of 137 cases using fresh whole cell suspensions

Several retrospective studies suggest that abnormal deoxyribonucleic acid (DNA) content in colorectal carcinoma correlates with adverse clinical outcome. Many of these studies have used naked nuclei retrieved from formalin-fixed, paraffin-embedded tissues for flow cytometry. The purpose of this study was to prospectively analyze 137 colorectal carcinomas using fresh whole-cell suspensions for flow cytometry and to determine whether abnormal DNA content (DNA aneuploidy or tumors with high proliferative activity) correlates with Dukes' stage, histologic grade, lymphocytic infiltration of the tumor, tumor fibrosis, extramural venous spread, or tumor size. Cell suspensions for flow cytometry were prepared by enzyme disaggregation with collagenase XI, DNase, and trypsin. Satisfactory DNA histograms were obtained from 132 of the 137 samples. The mean coefficients of variance for the G1/G0 of the external 2C control, internal 2C populations, and aneuploid populations were 2.5, 3.5, and 3.5, respectively. The mean percentage of viable cells was 97%. Of 132 cases, 102 (77%) demonstrated abnormal DNA histograms, of which 77 (58%) showed DNA aneuploidy. Abnormal DNA histograms of DNA aneuploidy did not correlate with Dukes' stage. Tumors of higher histologic grade were more likely to demonstrate DNA aneuploidy, however, these differences did not reach statistical significance. The authors conclude that (1) satisfactory DNA histograms can be obtained with the use of a fresh, whole-cell technique; (2) abnormal DNA histograms did not statistically correlate with standard clinical, grading, or staging parameters; and (3) carcinomas of high histologic grade showed an increased proportion of aneuploid DNA histograms, but this trend did not reach statistical significance.

The safety of intraosseous infusions: risks of fat and bone marrow emboli to the lungs

The technique of intraosseous infusion is a life-saving emergency alternative when IV access is impossible or will be critically delayed. Concerns about its safety remain, especially concerning the risk of bone marrow and fat emboli to the lungs. We examined autopsy pulmonary specimens on two children who had received intraosseous infusions during resuscitation attempts and found an average of 0.23 to 0.71 bone marrow and fat emboli per mm2 of lung. We studied normotensive dogs with intraosseous infusions of emergency drugs and solutions into the distal femur. Three dogs were studied with each of the following emergency drugs or solutions: controls with normal saline (0.9% NaCl), epinephrine 0.01 mg/kg, NaHCO3 1 mEq/kg, CaCl 10 mg/kg, atropine 0.01 mg/kg, hydroxyethyl starch 6% in normal saline 10 mL/kg, 50% dextrose in water 0.25 g/kg, and lidocaine 1 mg/kg. Four hours after infusion, the animals were killed, and representative sections of the lung were examined with oil red-0 and hematoxylin and eosin stains for the presence of fat and bone marrow emboli. Fat and bone marrow emboli were found in all lung sections, varying from 0.11 to 4.48 emboli/mm2 lung (mean, 0.91 emboli/mm2 lung) for the emergency drugs and solutions and 0.06 to 0.53 emboli/mm2 (mean, 0.29 emboli/mm2 lung) for the controls. Analysis of variance revealed no significant difference (P = .07) in mean number of fat and bone marrow emboli per square millimeter of lung among the emergency drugs and compared with controls.(ABSTRACT TRUNCATED AT 250 WORDS)

Fluorescein endoscopy. A technique for noninvasive assessment of intestinal ischemia

Late diagnosis contributes significantly to the mortality and morbidity of mesenteric ischemia. Although flexible endoscopy permits noninvasive assessment of the colon, ischemic colitis is often difficult to differentiate from nonspecific proctitis/colitis or mucosal changes seen in inflammatory bowel diseases. Intravenously administered fluorescein sodium (FSC) has long been used intraoperatively to assess bowel viability because its uptake is inversely proportional to the degree of intestinal ischemia. The authors wished to determine if FSC could be used to augment conventional endoscopy in an attempt to identify and monitor ischemic areas of the bowel within reach of the endoscope. Segmental mesenteric ligation of canine rectum was performed, and serial transanal biopsies obtained. Endoscopy was performed after intravenous FSC administration, using a rigid proctoscope, with a long-wave ultraviolet light source inserted into the lumen of the scope. Areas of nonspecific-appearing proctitis did not take up FSC and were easily differentiated from well-perfused areas by their failure to fluoresce. Ischemic areas were monitored in this manner over a 28-day period. Complete resolution of ischemic mucosal changes occurred with late stricture formation. In humans, the sigmoid and left colon are the areas most frequently affected by ischemic colitis. Because long-wave ultraviolet light is not transmitted through glass fibers, a different light source was needed for potential clinical application. To this end, a single fiber of an argon laser was inserted through the instrument channel of a colonoscope, resulting in greater fluorescence than seen with the ultraviolet light. FSC endoscopy appears to be a sensitive and accurate method of detecting and monitoring ischemic colitis and has many potential clinical applications.

Epithelial downgrowth

We present a surgically treated case of epithelial downgrowth with clinicopathologic correlation and special emphasis on the ultrastructural findings. Long-term follow-up revealed a salvaged eye, but with only peripheral vision due to glaucomatous optic atrophy.

Crohn's disease and intestinal carcinoma. A report of 11 cases with emphasis on associated epithelial dysplasia

Eleven patients with intestinal carcinoma complicating Crohn's disease are reported. In the 4 patients with small intestinal carcinoma, carcinomas were poorly differentiated or signet-ring cell type in 3 cases and mucinous type in 1 case. All 4 patients had high-grade dysplasia in the mucosa immediately adjacent to the carcinoma only. In the 7 patients with colonic carcinoma, tumors were mucinous type in 3, signet-ring cell type in 1, and well or moderately differentiated in 3. Dysplasia could be assessed in 6 patients. In 4 patients it was both adjacent to and distant from the invasive carcinoma. This distant dysplasia was focal in 2 cases. One patient showed dysplasia adjacent to the carcinoma only and in 1 other patient no dysplasia was seen. In a comparison group, all 7 cases of carcinoma associated with ulcerative colitis showed dysplasia adjacent to and distant from the invasive carcinomas. The authors conclude that (a) associated dysplasia is prevalent enough in Crohn's disease-associated carcinomas to support a dysplasia-carcinoma sequence and (b) dysplasia is more limited with carcinoma in Crohn's disease than ulcerative colitis, and this may make dysplasia less useful as a cancer surveillance marker in Crohn's disease.

Gastric antral vascular ectasia. A histologic and morphometric study of "the watermelon stomach"

Gastric antral vascular ectasia (GAVE) is an uncommon but important cause of gastrointestinal blood loss and iron deficiency anemia that is characterized by a distinctive endoscopic appearance consisting of parallel erythematous folds traversing the gastric antrum. In order to clarify the histologic features of this lesion, nine antral biopsy specimens from seven patients with the clinical and endoscopic diagnosis of GAVE were reviewed and compared with specimens from normal controls and patients with other common antral lesions. Specimens obtained using standard endoscopic biopsy forceps were evaluated for mucosal vascularity, presence of intravascular fibrin thrombi, and the following histologic changes: mucosal inflammation, fibromuscular hyperplasia of lamina propria, epithelial regeneration, and mucosal architectural distortion. Mucosal vascularity was determined by counting the absolute number of vascular lumina per slide, measuring the mean cross-sectional area of each vessel lumen, and determining the percentage of each specimen occupied by vessels. Histologic changes were graded as absent to minimal, moderate, or marked. Significant differences (p less than 0.05) between GAVE and control groups were mean vessel cross-sectional area, percentage of area occupied by vessels, presence of intravascular fibrin thrombi, and fibromuscular hyperplasia. GAVE appears to demonstrate sufficiently distinctive histopathologic features to allow its recognition in antral biopsy specimens obtained by endoscopy.

Strumal carcinoids of the ovary. An immunohistologic and ultrastructural study

A series of six ovarian strumal carcinoids was examined. The presence of thyroid follicular epithelium was conclusively proved by the presence of follicular cells immunohistochemically positive for thyroglobulin, and by the ultrastructural identification of non-neuroendocrine cells with features of thyroid epithelium lining the follicles. Progressive replacement of thyroid epithelial cells by carcinoid cells accounts for the predominance of neuroendocrine granule-containing cells and the scarcity of thyroid epithelial cells lining many of the follicles. A variety of neuroendocrine hormones and other immunoreactive substances was demonstrated within carcinoid cells, including somatostatin (five cases), chromogranin (five cases), serotonin (five cases), glucagon (four cases), insulin (two cases), and gastrin (one case). Only one case contained calcitonin-positive cells. None were carcinoembryonic antigen-positive or had amyloid deposits. The carcinoid element in five cases stained positively for prostatic acid phosphatase. While strumal carcinoid shares some features with medullary carcinoma of the thyroid gland, it has sufficient differences to warrant a separate designation.

Gastric and duodenal polyps in familial adenomatous polyposis: a prospective study of the nature and prevalence of upper gastrointestinal polyps

One hundred patients with familial adenomatous polyposis have prospectively undergone gastroduodenoscopy to identify and characterise polyps found. Forty six patients had polyps in the stomach or duodenum. Thirty five patients had adenomas (33 in duodenum, two in stomach) and 26 patients had fundic gland polyps. Some of these patients had polyps in the stomach and the duodenum. Adenomas in the duodenum were present in 33% of patients studied with Gardner's syndrome variant (p = 0.04). Adenomas were also more common in older patients. As adenomas may be a precursor of adenocarcinoma, routine surveillance of the stomach and duodenum with gastroduodenoscopy is recommended in patients affected with familial adenomatous polyposis.

Villous and tubulovillous adenomas of the colon and rectum. A retrospective review, 1964-1985

The records of 1,040 patients with 1,440 villous and tubulovillous adenomas of the colon and rectum treated at the Cleveland Clinic over a 21 year period were reviewed. The incidence of invasive adenocarcinoma was higher in patients with multiple adenomas at the time of presentation, in patients with a family history of colorectal cancer, and in those with adenomas greater than 4 cm in diameter. Eighty percent of the adenocarcinomas were Dukes' A or B, 15 percent were Dukes' C, and 5 percent presented with distant metastases. Clinical impression on examination was more accurate than biopsy in assessing the presence of malignancy in these tumors. Recurrence after local treatment of both benign and malignant tumors was high. Recurrence was significantly related to the size of the adenoma (more common when the tumor was more than 4 cm in diameter), location (more common when it was located in the rectum), and the type of treatment. Only 4 percent of the patients had died from colorectal cancer at the time of follow-up. The overall 5 year actuarial survival rate was 85 percent.

Papillary cystic epithelial neoplasms of the pancreas. A clinical study of four cases

We treated four cases of papillary cystic epithelial neoplasm of the pancreas. This neoplasm typically presents in young female patients as a slowly growing mass that may attain large size. It is of low malignant potential, and surgical resection will be curative in almost all cases. One patient presented with a liver metastasis from the pancreatic tumor; both lesions were successfully resected. Papillary cystic epithelial neoplasm should be considered in the differential diagnosis of any cystic pancreatic neoplasm.