Small early tubular adenomas and mixed colonic polyps found on screening flexible sigmoidoscopy do not predict proximal neoplasia in males

CLINICAL GASTROENTEROLOGY AND HEPATOLOGY 2004;2:246 –251

Small Early Tubular Adenomas and Mixed Colonic Polyps Found on Screening Flexible Sigmoidoscopy Do Not Predict Proximal Neoplasia in Males NADEEM ULLAH,*,‡ KASHIF QURESHI,*,‡ JAMES HATFIELD,‡ PAULA SOCHACKI,*,‡ DOINA DAVID,‡ HASSAN ALBATAINEH,*,‡ LUIS MEJIA,*,‡ CHAYA KENKRE,‡ MICHAEL LAWSON,§ and MARTIN TOBI*,‡ *Wayne State University School of Medicine, Detroit; ‡John D. Dingell Veterans Affairs Medical Center, Detroit, Michigan; and §Kaiser Permanente Medical Group, Sacramento, California

Background & Aims: Distal colonic adenomas found on flexible sigmoidoscopy are associated with proximal neoplasias, thus requiring a complete colonoscopic examination, but data regarding the association of distal mixed polyps with proximal neoplasia are lacking. We conducted this study to elucidate the significance of distal mixed colonic polyps in predicting proximal neoplasia. Methods: We retrospectively analyzed data from patients who underwent a flexible sigmoidoscopic examination for colorectal cancer screening and a follow-up colonoscopic examination because of distal colonic polyps. Distal index polyps were classified by a pathologist as early tubular adenoma (ETA), serrated, or true mixed categories. Index polyps also were immunostained with a monoclonal antibody, Adnab-9, a marker for the colorectal adenoma carcinoma sequence. Results: In 636 patients with distal index polyps, 6% were malignant, 55% were adenomas, 13% were ETAs, 6% were serrated, 4% were true mixed, and 17% were hyperplastic. Compared with distal hyperplastic index polyps, distal malignant polyps (P ⴝ 0.0006) and adenomas (P ⴝ 0.001) were associated with a significantly increased number of synchronous proximal neoplasia, but the small distal mixed, serrated, or ETA did not predict the increased incidence of proximal neoplasia. Large distal polyps of each category were significantly associated with an increased number of synchronous proximal neoplasias. In support of these findings, immunostaining of distal polyps with Adnab-9 showed predictability for proximal neoplasia only in the adenoma category (P < 0.05). Conclusions: Small ETAs, serrated, or mixed polyps found on flexible sigmoidoscopic examination do not increase the probability of synchronous proximal neoplasia.

he incidence of colorectal cancer (CRC) is highest in the Western world, and the 5-year mortality rate remains at ⬃50%. Most CRCs arise from precursor adenomatous polyps, and there are abundant data to

T

support an adenoma carcinoma sequence.1,2 Colonoscopic polypectomy is recommended if an adenoma is found on screening flexible sigmoidoscopy because the risk for invasive CRC may be reduced by using this approach.3,4 Conversely, hyperplastic polyps are considered non-neoplastic, and their presence in the distal colon is believed not to predict an increased incidence of proximal adenomas.3– 8 Accordingly, a full colonoscopy usually is not recommended for these patients. Morphologically, adenomatous polyps are characterized by varying degrees of epithelial dysplasia, whereas hyperplastic polyps show a “saw-toothed,” or serrated, microscopic appearance without dysplasia. Another group of polyps termed “mixed” polyps show some features of dysplasia and a serrated appearance similar to hyperplastic polyps and are designated variously as mixed polyps, serrated adenomas, or hyperplastic polyps with adenomatous change. The natural history, cancer risk, and genetic features of this group are not yet completely defined, and there are no clear guidelines available on the follow-up of patients for whom a mixed polyp is found on screening sigmoidoscopy.9 –14 We conducted this study to elucidate whether distal mixed polyps and their histological features can predict the presence of proximal neoplasia. We stained a representative selection from various groups of these polyps with a monoclonal antibody, Adnab-9, a marker for the adenoma-carcinoma sequence. Adnab-9 binding has been shown to identify a subpopulation of cells in colonic adenomatous, but not carcinomatous, tissue, some of which resemble Paneth’s Abbreviations used in this paper: CI, confidence interval; CRC, colorectal cancer; ETA, early tubular adenoma; OR, odds ratio; VAMC, Veterans Affairs Medical Center. © 2004 by the American Gastroenterological Association 1542-3565/04/$30.00 PII: 10.1053/S1542-3565(04)00012-6

March 2004

Figure 1. Immunohistochemical staining of a true mixed polyp with Adnab-9 shows Adnab-9 –stained crypt lumen (open arrow) within the hyperplastic portion of the polyp. There is an abrupt transition (dark arrow) between the dysplastic or adenomatous glands and adjacent hyperplastic glands.

cells.15–17 The presence of Adnab-9 staining in nonneoplastic colonic epithelium has been associated with an increased risk for the development of CRC.15–18 In addition, Adnab-9 staining has a positive correlation with multiplicity and recurrence of colonic polyps and may be a predictor of the future development of dysplastic polyps with recognized malignant potential in these patients.19 Distal index polyps were stained with Adnab-9 to determine whether immunostaining characteristics of these polyps may help identify patients with an increased incidence of synchronous proximal neoplasia.

Materials and Methods Computerized medical records of patients who underwent flexible sigmoidoscopy for screening of CRC at the John D. Dingell VA Medical Center (VAMC; Detroit, MI) from 1992 to 2001 were reviewed, and those who had completed a follow-up colonoscopy because of polyp found on flexible sigmoidoscopy were included in this analysis. All patients at the VAMC have their medical records computerized, and the system is considered paperless. Site, size, and histological characteristics of each polyp found on sigmoidoscopic or colonic examination were noted. A distal colonic polyp with the most advanced pathological state was considered the index polyp, and patients were categorized into different groups based on the characteristics of the index polyp. Proximal colonic endoscopic findings of these different groups of patients, based on index polyps, were analyzed after follow-up colonoscopy was completed. Some of these index polyps were immunostained with Adnab-9 for possible correlation with synchronous proximal neoplasia. These polyps were reviewed by a pathologist blinded to the immunostaining results and reclassified based on H&E–stained slides according to the following definitions. Polyps were defined as true mixed if the polyp had both hyperplastic and

DISTAL MIXED POLYPS DO NOT PREDICT PROXIMAL NEOPLASIA

247

adenomatous components, with the adenomatous component ⬍20% of the polyp total volume,20 and dysplasia was adjacent to the hyperplastic component (Figure 1). Polyps were defined as serrated if the dysplasia was deep, with overlying hyperplastic glands and a ⬍20% adenomatous component (Figure 2). Polyps were defined as early tubular adenoma (ETA) if the dysplastic component was superficial and 20%– 40% of the polyp volume. A polyp was considered hyperplastic if it showed a serrated or saw-toothed appearance without dysplasia. Distal colonic polyps are those located in the rectum, sigmoid, or descending colon, and proximal polyps are from the right side of the colon, including the splenic flexure. Malignant polyps are defined as those with invasive cancer or high-grade dysplasia. Significant adenomas are defined as polyps with a villous component, adenomas with moderate dysplasia, multiple adenomas, or an adenoma ⬎9 mm in size. A polyp is defined as large if the endoscopist or pathologist described the size of the polyp as ⬎9 mm or a “large” polyp. Advanced neoplasia included malignant polyps in addition to significant adenomas, and both entities were evaluated separately.

Immunostaining Immunohistochemistry was performed in formaldehyde-fixed, paraffin-embedded, 5-␮m–thick representative tissue sections of colonic polyps. The avidin-biotin-peroxidase complex following the method of Hsu et al.21 was used for Adnab-9 staining using the Vectastin kit (Vector Laboratories, Burlingame, CA). Any discernable staining over background is considered positive.

Statistical Analysis Differences in each category were analyzed by means of comparison of proportions using the Fisher exact test and linear correlation between variables, using Spearman’s test, as

Figure 2. Photomicrograph of a serrated polyp. Note the saw-tooth– shaped hyperplastic glands near the surface of the polyp and dysplastic or adenomatous glands within the deep portion of the polyp. The dysplastic glands contain cells with stratified cigar-shaped nuclei, mitotic figures, and displacement of goblet cells to the base of some glands (H&E; original magnification 100⫻).

248

ULLAH ET AL.

CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 2, No. 3

Table 1. Probability of Synchronous Proximal Neoplasia Based on Distal Index Polyp Type

Distal polyp

Proximal cancer

Proximal significant adenoma

Proximal advanced neoplasia

Proximal neoplasia

Significance for proximal neoplasia (P)

Malignant polyp (n ⫽ 35) Adenoma (n ⫽ 348) ETA (n ⫽ 80) Serrated (n ⫽ 37) Mixed (n ⫽ 25) Hyperplastic (n ⫽ 111)

2 8 0 1 0 0

15 72 9 2 2 11

17 (49%) 80 (23%) 9 (11%) 3 (8%) 2 (8%) 11 (10%)

19 (54%) 135 (39%) 18 (23%) 8 (22%) 3 (12%) 25 (23%)

0.0006a 0.001b Not significant Not significant Not significant Compare with all above

ETA, early tubular adenoma; OR, odds ratio; CI, confidence interval. aOR, 4.09; 95% CI, 1.83–9.17. bOR, 2.18; 95% CI, 1.33–3.58.

described previously.15,17,18 Significance is defined as P ⬍ 0.05.

Results A total of 4814 patients underwent a screening flexible sigmoidoscopy from 1992 to 2001. Six hundred thirty-six patients who had a polyp on flexible sigmoidoscopy and underwent a follow-up colonoscopy were included in this analysis. Average time between initial flexible sigmoidoscopy and follow-up colonoscopy was 4 months. The majority of patients were men (⬎99%), with a mean age 63.84 ⫾ 7.84 years. Distal index polyps among these 636 patients were 55% adenoma (n ⫽ 348), 13% ETA (n ⫽ 80), 4% true mixed (n ⫽ 25), 6% serrated (n ⫽ 37), 17% hyperplastic (n ⫽ 111), and 6% malignant (n ⫽ 35). Forty-three percent of adenomas (148 of 348 adenomas) were small or diminutive, whereas the majority of mixed polyps (138 of 142 polyps; 97%) were small or diminutive. Proportions of the various polyp categories used for Adnab-9 staining were almost identical for the average age of patients in that category (Spearman’s ⫽ 1; P ⫽ 0.017), showing that the staining sample was representative. The distribution of proximal polyps based on index polyp category is listed in Table 1. Of the various groups of distal polyps, proximal neoplasia was found in 54% of the malignant polyp group, 39% of the adenoma group, 23% of the ETA group, 22% of the serrated group, 12% of the true mixed group, and 23% of the hyperplastic polyp group. Patients with distal malignant (odds ratio [OR], 4.09; 95% confidence interval [CI ], 1.83–9.17; P ⫽ 0.0006) or adenomatous polyps (OR, 2.18; 95% CI, 1.33–3.58; P ⫽ 0.001) had a significantly increased number of synchronous proximal neoplasia compared with those with distal hyperplastic polyps. Patients with distal polyps of the mixed variety, including true mixed, serrated, or ETA, did not show a significantly increased incidence of proximal neoplasia compared with those with distal hyperplastic polyps (Table 1).

When a comparison was made between subgroups of index polyps based on significant polyp size or advanced pathological state, large index polyps in all categories of mixed polyps (OR, 9.23; 95% CI, 1.68 –50.76; P ⫽ 0.009) and advanced adenomas (OR, 2.96; 95% CI, 1.729 –5.078; P ⬍ 0.0001) were associated with a significantly increased incidence of synchronous proximal neoplasia (Table 2). Small distal mixed polyps (true mixed, serrated, or ETA) were not associated with an increased number of proximal neoplasia compared with distal hyperplastic polyps (19% vs. 23%; P ⫽ 0.775). The presence of small distal adenomas predicted a trend to an increase in proximal neoplasia, but this was not statistically significant (OR, 1.51; 95% CI, 0.846 – 2.708; P ⫽ 0.193). The distribution of proximal polyps among patients with index polyps stained with Adnab-9 is listed in Table 3. Adnab-9 staining was predominantly associated with cytoplasmic distribution in nonneoplastic regions of polyps. In adenomas, focal staining of dysplastic areas also was observed (data not shown). Although there was a trend toward an increased incidence of proximal neoplasia with Adnab-9 staining, statistical significance was observed only in the case of adenomas.

Discussion Flexible sigmoidoscopy is used as a tool for CRC screening, mainly for the average-risk patient population. If a large polyp is found on sigmoidoscopic examination, a full colonoscopy and polypectomy are recommended.3,4 Conversely, a small polyp is biopsied, and a colonoscopy is recommended only if morphological examination shows it to be an adenoma. Hyperplastic polyps found distally do not predict an increased risk for proximal neoplasia.6 – 8 Patients at our institution are offered flexible sigmoidoscopy starting at the age of 50 years if they fall into the category of average risk for colon cancer. The majority of these patients tended to have a full colonoscopic examination if a polyp was seen

March 2004

DISTAL MIXED POLYPS DO NOT PREDICT PROXIMAL NEOPLASIA

249

Table 2. Probability of Proximal Neoplasia Based on Index Polyp Morphological Characteristics and Size Proximal cancer

Proximal significant adenoma

Proximal advanced neoplasia

Proximal neoplasia

Significance for proximal neoplasia

Advanced adenoma (n ⫽ 200)

5

52

57 (29%)

89 (45%)

Small adenoma (n ⫽ 148)

3

20

23 (16%)

43 (29%)

Nonadenoma ⬎ 9 mm (n ⫽ 4)

0

3

3 (75%)

3 (75%)

Nonadenoma ⬍ 1 cm (n ⫽ 138)

1

10

11 (8%)

26 (19%)

Hyperplastic (n ⫽ 111)

OR, 2.96; 95% CI, 1.729–5.078; P ⬍ 0.0001a OR, 1.51; 95% CI, 0.846–2.708; P ⫽ 0.193b OR, 9.23; 95% CI, 1.68–50.76; P ⫽ 0.009c OR, 0.9; 95% CI, 0.526–1.604; P ⫽ 0.775d

0

11

11 (10%)

25 (23%)

Distal polyp

OR, odds ratio; CI, confidence interval. aAdvanced adenoma vs. hyperplastic. bSmall adenoma vs. hyperplastic. cLarge nonadenoma vs. hyperplastic. dSmall nonadenoma vs. hyperplastic.

on sigmoidoscopic examination, regardless of the pathological characteristics of the polyp. Therefore, a large population of our patients with hyperplastic or mixed variety of rectosigmoid polyps had undergone a colonoscopic examination. This provided an opportunity to compare proximal colonic findings of these patients with

those of patients who had more advanced lesions distally, such as adenomas or malignant polyps. Our study suggests that the presence of a mixed polyp found on sigmoidoscopy is not associated with an increased incidence of proximal neoplasia. Although there are some reports describing an association between ser-

Table 3. Incidence of Proximal Neoplasia in Subgroups of Distal Colonic Polyps Based on Adnab-9 Staining Proximal cancer

Proximal significant adenoma

Proximal nonsignificant adenoma

Proximal neoplasia

Significance for proximal neoplasia

All Adnab positive (n ⫽ 39)

0

9

7

16 (78%)

All Adnab negative (n ⫽ 51) Adenoma Adnab positive (n ⫽ 10)

OR, 2.26; 95% CI, 0.91–5.61; P ⫽ 0.107a

0 0

5 6

7 2

12 (24%) 8 (80%)

Adenoma Adnab negative (n ⫽ 15) ETA Adnab positive (n ⫽ 17)

0 0

3 1

2 3

5 (33%) 4 (24%)

ETA Adnab negative (n ⫽ 17) Serrated Adnab positive (n ⫽ 6)

0 0

2 1

2 1

4 (24%) 2 (66%)

Serrated Adnab negative (n ⫽ 7) Mixed Adnab positive (n ⫽ 2)

0 0

0 0

1 0

1 (14%) 0

Mixed Adnab negative (n ⫽ 9)

0

0

1

1 (11%)

Distal polyp

ETA, early tubular adenoma; OR, odds ratio; CI, confidence interval. aAdnab positive vs. negatively stained polyp. bAdnab positive vs. negatively stained adenoma. cAdnab positive vs. negatively stained ETA. dAdnab positive vs. negatively stained serrated polyp. eAdnab positive vs. negatively stained mixed polyp.

OR, 8.0; 95% CI, 1.21–52.71; P ⫽ 0.041b OR, 1.0; 95% CI, 0.204–4.880; P ⫽ 1.0c OR, 3.0; 95% CI, 0.198–45.27; P ⫽ 0.559d OR, 1.133; 95% CI, 0.34–37.46; P ⫽ 1.0e

250

ULLAH ET AL.

rated adenoma and adenocarcinoma, and some have even suggested a serrated pathway to CRC,2,10,11 to our knowledge, there are no reports available about the significance of distal serrated/mixed polyp in predicting proximal neoplasia. The lack of data regarding the clinical significance of finding a mixed polyp on distal colonic examination occurs because there is no universally agreed-on definition of a mixed polyp. Longacre and Fenoglio-Preiser10 suggested there are at least 2 types of mixed polyps; 1 type in which adenomatous and hyperplastic glands are admixed, but clearly identifiable as such, and another type in which a spectrum of changes occur such that the combined architectural and cytological features do not completely meet the standard histological criteria for either hyperplastic or adenomatous polyp. These polyps have architectural features of a hyperplastic polyp but cytological features of an adenoma. They determined the nomenclature for this latter type as a “serrated polyp.”10 Conversely, Sommers20 described serrated adenoma in which adenomatous epithelium, usually at the base, composes up to 20% of the polyp. We classified these mixed polyps into true mixed, serrated, or ETA based on percentage and location of adenomatous tissue to see whether this can elucidate which mixed polyp subtype is associated more often with synchronous proximal neoplasia, necessitating a full colonoscopic examination. Most of our patients with mixed polyps probably belong to the category described by Longacre and FenoglioPreiser10 as “admixed,” rather than “serrated.” Our study shows that the probability of finding a neoplastic lesion proximally on colonoscopic follow-up is high for patients with distal malignant polyps (54%; P ⫽ 0.0006) and adenomas (39%; P ⫽ 0.001) compared with those with distal hyperplastic polyps (23%). The presence of ETA, serrated, or mixed polyps found distally did not increase the risk for proximal neoplasia compared with those with distal hyperplastic polyps. Regarding the significance of the size of the index polyp, we found that large distal polyps were associated with statistically significant proximal neoplasia regardless of histological characteristics. The incidence of proximal neoplasia for patients with distal small index polyps of mixed variety was not greater than that for patients with distal hyperplastic index polyps (19% vs. 23%; OR, 0.9; 95% CI, 0.52–1.60; P ⫽ 0.77). A single small or diminutive adenoma ⬍9 mm found distally was associated with a modest trend of finding a proximal neoplasia compared with a distal hyperplastic polyp. Although there are conflicting reports about the significance of finding a small or diminutive adenoma on flexible sig-

CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 2, No. 3

moidoscopy, most agree that larger size and advanced pathological state of a distal polyp is associated with a greater incidence of synchronous advanced proximal neoplasia compared with a diminutive distal adenoma.2,22–29 In addition, we could further stratify patients with distal colonic polyps by staining some of these index polyps with Adnab-9 to determine which patient would have needed a full colonoscopic examination for potential proximal neoplasia. Our results show that overall, distal polyp staining with Adnab-9 tended to have synchronous proximal neoplasia compared with those with no staining, but the difference was not significant. When different polyp categories were analyzed based on morphological characteristics, distal adenomas staining for Adnab-9 were associated with more frequent proximal neoplasia than Adnab-9 –negative adenomas (80% vs. 33%; OR, 8.0; P ⫽ 0.041). This is in agreement with morphological correlation and previous reports that Adnab-9 staining showed a positive correlation with multiplicity and recurrence of colonic adenomas.19 In support of our clinical observation, Adnab-9 staining of distal ETAs, serrated, mixed, or hyperplastic polyps did not predict proximal neoplasia, although the numbers in these subgroups were small. It is important to note that being a VAMC, our patient population consists mainly of men; thus, these results should be applied cautiously to the general population. We conclude that a small ETA, serrated, or mixed polyp found on sigmoidoscopic examination does not increase the probability of synchronous proximal neoplasia compared with distal hyperplastic polyps. Thus, patients with these lesions may be followed up as for those who have distal hyperplastic polyps. Conversely, a large polyp (ⱖ1 cm), regardless of pathological state, or a polyp with advanced pathological characteristics found on distal colonic examination is associated with an increased possibility of proximal neoplasia, necessitating a full colonoscopic examination. Small distal adenomas also are associated with a trend toward the presence of proximal neoplasia compared with distal hyperplastic polyps. In adenomas, Adnab-9 staining may be used to predict the presence of proximal synchronous neoplasia.

References 1. Tobi M. Polyps as biomarkers for colorectal neoplasia. Frontiers in Bioscience 1999;4:1– 8. 2. Burger LJ. Colorectal polyps and other precursor lesions: need for an expanded view. Gastroenterol Clin North Am 2002;31:959 – 970. 3. Bond JH. Polyp guideline: diagnosis, treatment, and surveillance for patients with colorectal polyps. Am J Gastroenterol 2000;95: 3053–3063. 4. Winawer S, Fletcher R, Rex D, Bond J, Burt R, Ferrucci J, Ganiats

March 2004

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

T, Levin T, Woolf S, Johnson D, Kirk L, Litin S, Simmang C, for the U.S. Multisociety Task Force on Colorectal Cancer. Colorectal cancer screening and surveillance: clinical guidelines and rationale— update based on new evidence. Gastroenterology 2003; 124:544 –560. Brady PG, Straker RJ, McClave SA, Nord HJ, Pinkas M, Robinson BE. Are hyperplastic rectosigmoid polyps associated with an increased risk of proximal colonic neoplasms? Gastrointest Endosc 1993;39:481– 485. Provenzale D, Garrett JW, Condon SE, Sandler RS. Risk for colon adenomas in patients with rectosigmoid hyperplastic polyps. Ann Intern Med 1990;13:760 –763. Sciallero S, Costantini M, Bertinelli E, Castiglione G, Onofri P, Aste H, Casetti T, Mantellini P, Bucchi L, Parri R, Boni L, Bonelli L, Gatteschi B, Lanzanova G, Rinaldi P, Giannini A, Naldoni C, Bruzzi P. Distal hyperplastic polyps do not predict proximal adenomas: results from a multicentric study of colorectal adenomas. Gastrointest Endosc 1997;46:124 –130. Rex DK, Smith JJ, Ulbright TM, Lehman GA. Distal colonic hyperplastic polyps do not predict proximal adenomas in asymptomatic average-risk subjects. Gastroenterology 1992;102:317–319. Lino H, Jass JR, Simms LA, Young J, Leggett B, Ajioka Y, Watanabe H. DNA microsatellite instability in hyperplastic polyps, serrated adenomas, and mixed polyps: a mild mutator pathway for colorectal cancer? J Clin Pathol 1999;52:5–9. Longacre TA, Fenoglio-Preiser CM. Mixed hyperplastic adenomatous polyps/serrated adenomas. Am J Surg Pathol 1990;14: 524 –537. Makinen MJ, George SMC, Jernvall P, Makela J, Vihko P, Karttunen TJ. Colorectal carcinoma associated with serrated adenoma—prevalence, histological features, and prognosis. J Pathol 2001;193:286 –294. Weston AP, Campbell DR. Diminutive colonic polyps: histopathology, spatial distribution, concomitant significant lesions, and treatment complications. Am J Gastroenterol 1995;90:24 –28. Uchida H, Ando H, Maruyama K, Kobayashi H, Toda H, Ogawa H, Ozawa T, Matsuda Y, Sugimura H, Kanno T, Baba S. Genetic alterations of mixed hyperplastic adenomatous polyps in the colon and rectum. Jpn J Cancer Res 1998;89:299 –306. Kang M, Mitomi H, Sada M, Tokumitsu Y, Takahashi Y, Igarashi M, Katsumata T, Okayasu L. Ki-67, p53, and Bcl-2 expression of serrated adenomas of the colon. Am J Surg Pathol 1997;21: 417– 423. Tobi M, Maliakkal B, Zitron I, Alousi M, Goo R, Nochomouitz L, Luk G. Adenoma-derived antibody, Adnab-9 recognizes a membrane-bound glycoprotein in colonic tissue and effluent material from patients with colorectal neoplasia. Cancer Lett 1992;67: 61– 69. Tobi M, Maliakkal BJ, Alousi MA, Voruganti V, Shafiuddin M, Yang S, Gesell MS, An T, Hatfield JS, Fligiel S. Cellular distribution of a colonic adenoma-associated antigen as defined by monoclonal antibody Adnab-9. Scand J Gastroenterol 1992;27:737–742. Tobi M, Elitsur Y, Moyer MP, Halline A, Deutsch M, Nochomouitz

DISTAL MIXED POLYPS DO NOT PREDICT PROXIMAL NEOPLASIA

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

251

L, Luk GD. Mucosal origin and shedding of an early colonic tumor marker defined by Adnab-9 monoclonal antibody. Scand J Gastroenterol 1993;28:1025–1034. Tobi M, Prabhu S, Gage RE, Orr T, Lawson MJ. Colorectal cancer risk. The impact of evidence of a field effect of carcinogenesis on blinded diagnosis using an anti-adenoma antibody test performed on colonoscopic effluent. Dig Dis Sci 2002;47:317–321. Sheikh R, Min B, Yasmeen S, Teplitz R, Tesluk H, Ruebner BH, Tobi M, Hatfield J, Fligiel S, Lawson M. Comparison of Ki-67, Adnab-9, p53 immunohistochemical staining, and ploidy with clinical and histological features of severely dysplastic colorectal adenomas. Dig Dis Sci 2003;48:223–229. Sommers SC. In: Rotterdam H, Sheahan DG, Sommers SC, eds. Biopsy diagnosis of the digestive tract. Volume 2. 2nd ed. Biopsy interpretation series. New York: Raven, 1993:684. Hsu SM, Raine L, Fanger V. Use of avidin-biotin-peroxidase complex (ABC) in immunoperoxidase techniques: a comparison between ABC and unlabeled antibody (PA) procedures. J Histochem 1981;21:577–580. Ryan ME, Norfleet RG, Kirchner JP, Parent K, Nanez JF, Rhodes RA, Wyman JB. The significance of diminutive colonic polyps found at flexible sigmoidoscopy. Gastrointest Endosc 1989;35: 85– 89. Read TE, Read JD, Butterly LF. Importance of adenomas 5 mm or less in diameter that are detected by sigmoidoscopy. N Engl J Med 1997;336:8 –12. Collett JA, Platell C, Fletcher DR, Aquilia S, Olynyk JK. Distal colonic neoplasms predict proximal neoplasia in average-risk asymptomatic subjects. J Gastroenterol Hepatol 1999;14:67– 71. Spencer RJ, Melton LJ III, Ready RL, et al. Treatment of small colorectal polyps: a population-based study of risk of subsequent carcinoma. Mayo Clin Proc 1984;59:305–310. Zarchy TM, Ershoff D. Do characteristics of adenomas on flexible sigmoidoscopy predict advanced lesions on baseline colonoscopy? Gastroenterology 1994;105:1501–1504. Wallace MB, Kemp JA, Trnka YM, Donovan JM, Farraye FA. Is colonoscopy indicated for small adenomas found by screening flexible sigmoidoscopy. Ann Intern Med 1998;129:273–278. Levin TR, Palitz A, Grossman S, Connell C, Finkler L, Ackerson L, Rumore G, Selby JV. Predicting advanced proximal colonic neoplasia with screening sigmoidoscopy. JAMA 1999;281:1611– 1617. Schoen RE, Corle D, Cranston L, Weissfield JC, Lance P, Burt R, Iber F, Shike M, Kidendall JW, Hasson M, Lewin KJ, Appelman HD, Paskett E, Selby JV, Lanza E, Schatzkin A. Is colonoscopy needed for the nonadvanced adenoma found on sigmoidoscopy? Gastroenterology 1998;115:533–541.

Address requests for reprints to: Nadeem Ullah, M.D., Division of Gastroenterology, 11-M/GI, John D. Dingell VA Medical Center, 4646 John R, Detroit, Michigan, 48201.