Pathological evaluation in colorectal polyps endoscopic treatment

G Chir Vol. 27 - n. 11/12 - pp. 411-416 Novembre-Dicembre 2006

Pathological evaluation in colorectal polyps endoscopic treatment

G. BENFATTO, L. TENAGLIA, G. CATANIA, S. D’ANTONI, A. JIRYIS, D. CENTONZE,

S.M.R. GARUFI, F. MUGAVERO, A. GIOVINETTO

411

Introduction

Colorectal adenomas represent an anatomical and clinic entity rarely present before the age of 40, in absen-ce of familiar polyposis. The incidenabsen-ce’s peak is between 60 and 70 years of age; the distribution of the large inte-stine polyps is similar to that of the colorectal cancer.

The incidence of adenoma is much higher than that of cancer; however only few adenomas show ma-lignant potential and develop into cancer (1, 2). The identification and the removal of the colorectal polyps should reduce the incidence of cancer.

Out patient colonoscopy with endoscopic

polypec-tomy is the technique of choice to remove the majo-rity of polyps and their pathologic examinations shed light on the carcinogenesis of colorectal lesions (3-6).

We evaluate feasibility, safety, and effectiveness of endoscopic treatment of colorectal adenomas vs. surgi-cal treatment through the analysis of 1302 polypecto-mies in a period of 12 years long. The data considered by this study are: age, size of the lesion, location, hi-stological features.

Patients and methods

From January 1990 to December 2001, 1302 adenomatous polyps were removed, 1175 endoscopically, 127 surgically. The pa-tients, formerly surgically treated for colorectal carcinoma, those affected by any chronic bowel disease (Crohn’s disease, inflamma-tory bowel diseases), and those with familiar polyposis or with non adenomatous polyps weren’t included in this study.

All the endoscopic procedures were performed on unseated pa-tients at the Endoscopic Unit of our Department of Surgery, using Vi-deocolon Pentax EC-3840 or a colonoscope Olympus CF-30 I. SUMMARY: Pathological evaluation in colorectal polyps

endosco-pic treatment.

G. BENFATTO, L. TENAGLIA, G. CATANIA, S. D’ANTONI, A. JIRYIS,

D. CENTONZE, S.M.R. GARUFI, F. MUGAVERO, A. GIOVINETTO This retrospective study shows that endoscopic polypectomy is the tech-nique of choice to remove the majority of polyps; follow-up and pathologic examinations shed light on the carcinogenesis of colorectal lesions.

From January 1990 to December 2001, 1302 adenomatous polyps were removed, 1175 endoscopically, 127 with surgical procedures. The anatomical and morphologic conditions of the colon and some characteri-stics of the polyps represent limits to the feasibility and to the efficacy of polypectomy, and the most important variables for the correct management of the patients affected by colorectal adenomatous polyps.

RIASSUNTO: Valutazione patologica nel trattamento endoscopico

dei polipi colorettali.

G. BENFATTO, L. TENAGLIA, G. CATANIA, S. D’ANTONI, A. JIRYIS,

D. CENTONZE, S.M.R. GARUFI, F. MUGAVERO, A. GIOVINETTO Questo studio retrospettivo mostra che la polipectomia endoscopica rappresenta la tecnica da preferire per rimuovere la maggior parte dei po-lipi ed il follow-up con esami istologici ripetuti può chiarire la carcinoge-nesi delle lesioni colorettali.

Dal gennaio 1990 al dicembre 2001 sono state eseguite 1302 poli-pectomie, 1175 endoscopiche e 127 chirurgiche. Le condizioni anatomi-che del colon e le caratteristianatomi-che macro- e microscopianatomi-che dei polipi possono rappresentare un limite alla realizzazione ed all’efficacia di questo tipo di trattamento e rappresentano le più importanti variabili per un corretto trattamento terapeutico.

KEYWORDS: Colorectal adenomatous polyps - Endoscopic polypectomy - Dysplasia.

Polipi adenomatosi del colon-retto - Polipectomia endoscopica - Displasia.

Università degli Studi di Catania Ospedale Vittorio Emanuele di Catania Cattedra di Chirurgia Generale (Direttore: Prof. F. Basile)

The patients submitted to a complete endoscopic polypectomy w er e 1021 (96 of them had 2 or mor e polyps). These patients, 726 men and 295 women, ranged in age fr om 23 to 92 years (mean 64,2 years, 467 ov er sixty). The patients under w ent surgical pr oce -dur es w er e 127. These patients, 86 men and 41 women, ranged in age fr om 49 to 86 years (mean 67,4; 73 ov er sixty). In all patients pancolonoscopy has been per formed. Bo w el pr e-paration befor e colonoscopy was achiev ed using 4000 ml of a poly ethilen gly col electr olytic lav age solution. In 127 patients it ha -sn ’t been possible to per form an endoscopical polypectomy , in 66 cases because of the siz e of the polyps, in 42 cases because of the lo -cation, in 19 cases because of the risk of per foration related to the anatomical and morphological conditions of the intestine (div er ti -culitis, adhesions). All tissues w er e fix ed in 10% formalin and stai -ned with hemato xylin and eosin for the pathologic examination. The location, the siz e and the histologic featur es ar e the data that w e hav e ev aluated also on the repor t of a recent study of Ike -da et al. (7). W e also hav e consider ed as an impor tant variant the age, in or der to distribute the patients in two gr oups: under sixty (554) and ov er sixty (467). O n the basis of the anatomical distribution of color ectal polyps w e classified right sided polyps (RSP), left sided polyps (LSP), rectum polyps (RP). R egar ding to the siz e, a ver y small adenoma (VSA) was defined as lesser than 5 mm in diameter ,a small adenoma (SA) as lar -ger than 5 mm but lesser than 10 mm in diameter , a large adenoma (L A) was as larger than 10 mm in diameter . The pathological featur es includes: the morphology of the le -sion (pedunculated polyp , shor t stalked polyp , sessile polyp), the glandular ar chitectur e (tubular adenoma, villous adenoma, tubulo -villous adenoma), and the grade of dysplasia. In accor ding to the criteria of the W orld H ealth O rganization w e recogniz e two grades of dysplasia: high and lo w . A denomas with high grade dysplasia (AHGD) includes the car cinoma in situ. If the histologic examina -tion sho ws the inv asion of the submucosa, w e consider the lesion as a real car cinoma of the colon and w e always per form a surgical operation ev en if it has been possible to per form an endoscopic polypectomy .

Results

The distribution of all the polyps was: 321 polyps (24,65%) localiz ed in the right colon, 574 (44,10%) in the left colon and 407 (31,25%) in the rectum. Among 1175 endoscopic polypectomies, 508 w er e le -sions less than 5 mm of diameter , 411 betw een 5 and 10 mm, and 383 mor e than 10 mm. The polyps remo -ved b y surger y w er e mor e than 30 mm (giant polyps). M acr oscopically sessile shapes w er e pr ev alent (843/ 64,75%), the pedunculated ones w er e 297/22,80%, and the par tially pedunculated 162/12,45%). H isto -logy sho w ed 920 tubular adenomas (70,65%), 204 vil -lous adenomas (15,65%), and 178 tubulo-villous ade -nomas (13,70%). A t the histologic examination of the endoscopically remo ved polyps 956 polyps w er e ade -nomas lo w grade dysplasia (AL GD), 150 adenomas high grade dysplasia (AHGD), 69 malignant (MP); in this cases it is not right to define them as adenomas. The histologic examination of the surgically remo ved polyps included 41 AL GD, 54 AHGD and 32 MP . The 82% of villous adenomas and the 74% of the tu -bulo-villous ones sho w ed a high grade of dysplasia or w er e inv asiv e car cinomas; these histologic ar chitectu -res hav e been found in 67% of the cases aged ov er 60 years. T ables 1-9 sho w the distribution of the lesions, the siz e and the grade of dysplasia in the patients ov er and under 60 years. The mor e remar kable data ar e referr ed to the grade of dysplasia, that incr eases in both the gr oups of age as the siz e’ s incr easing. In the patients ov er 60 fur thermo -re the per centage of AHGD is major than the AL GD ev en if the siz e of the polyp is similar . In the patients ov er 60, the per centage of AHGD is major as regar ds to the patients under 60 for the pr esence of whatev er siz e of polyp . M or eo ver , the per centage of malignant polyps, endoscopically remo ved, was respectiv ely of 53,5% in the patients under 60 and of 56,5% in the patients ov er 60. In the patients ov er 60, the pr esence of AHGD is mor e incr eased in the right colon than in the patients under 60. Surgical operation to remo ve polyp was necessar y in 73 patients ov er 60 and in 63 patients under 60. All the patients with malignant polyps hav e been successiv ely subjected to surgical resection. F our of the patients who had been endoscopically polypectomiz ed hav e been fur ther subjected to surgical operation in urgency because of complications (3 iatr ogenic per fo -rations and one massiv e bleeding not endoscopically tr eatable). B ecause of the siz e of the lesions and/or of the lo -cation, in 97 patients it hasn ’t been possible to remo ve the polyp thr ough endoscopy (“ pr oblematic polyps ”). So, in these patients it has been per formed surger y with a differ ent pr ocedur e, accor ding to the site and to the information taken fr om the biopsies. The practices adopted for the follo w-up hav e been giv en fr om the grade of dysplasia of the lesion. The patients with AL GD should undergo colonosco -pic sur veillance only at two years after complete re -mo ving of all polyps found during initial examina -tion; in the patients with a ver y danger ous grade of malignancy adenomas at six months, one and thr ee years. F inally , the patients with malignant polyps ar e follo w ed with a mor e car eful and longer not only en -doscopic follo w-up (tumoral mar kers, hepatic sono -graphy , CT scan).

Discussion

The term “polyp ” generally refers to any pr otube -rant lesion of the intestinal mucosa. T wo thir d of all colonic polyps ar e adenomas. Age is the most impor -tant risk factor for the dev elopment of colonic adeno -mas. Thanks to colonoscopic scr eening, it is suggested that the incidence of adenomas in asymptomatic pa -412 G. B enfatto e Coll.

P athological ev aluation in color ectal polyps endoscopic tr eatment T ABLE 1 -DISTRIB UTION OF VER Y SMALL SIZE (< 5 mm ) COL ORECT AL AL GD AND AHGD ENDOSCOP ICALL Y REMO -VED IN P A TIENT S UNDER 60 YEARS. _Right colon Left colon R ectum T otal AL GD (%) 76 (97,45%) 144 (96,65%) 81 (95,3%) 301 (96,5%) AHGD (%) 2 (2,55%) 5 (3,35%) 4 (4,7%) 11 (3,5%) T otal 78 149 85 312 T ABLE 2 -DISTRIB UTION OF VER Y SMALL SIZE (< 5 mm ) COL ORECT AL AL GD AND AHGD ENDOSCOP ICALL Y REMO -VED IN P A TIENT S O VER 60 YEARS. Right colon Left colon R ectum T otal AL GD (%) 45 (93,75%) 73 (92,4%) 64 (92,75%) 182 (92,85%) AHGD (%) 3 (6,25%) 6 (7,6%) 5 (7,25%) 14 (7,15%) T otal 48 79 69 196 T ABLE 3 -DISTRIB UTION OF SMALL SIZE (5-10 mm ) COL ORECT AL AL GD AND AHGD ENDOSCOP ICALL Y REMO VED IN P A TIENT S UNDER 60 YEARS. Right colon Left color R ectum T otal AL GD (%) 43 (91,5%) 106 (93,8%) 72 (92,3%) 222 (93,25%) AHGD (%) 4 (8,5%) 7 (6,2%) 6 (7,7%) 16 (6,75%) T otal 47 113 78 238 T ABLE 4 -DISTRIB UTION OF SMALL SIZE (5-10 mm ) COL ORECT AL AL GD AND AHGD ENDOSCOP ICALL Y REMO VED IN P A TIENT S O VER 60 YEARS. Right colon Left colon R ectum T otal AL GD (%) 34 (87,15%) 61 (87,15%) 56 (87,5%) 151 (87,3%) AHGD (%) 5 (12,85%) 9 (12,85%) 8 (12,5%) 22 (12,7%) T otal 39 70 64 173 T ABLE 5 -DISTRIB UTION OF L AR GER SIZE (> 10 mm ) COL ORECT AL AL GD AND AHGD ENDOSCOP ICALL Y REMO VED IN P A TIENT S UNDER 60 YEARS. Right colon Left colon R ectum T otal AL GD (%) 19 (65,5%) 36 (63,15%) 11 (45,85%) 66 (60%) AHGD (%) 10 (34,5%) 21 (36,85%) 13 (54,15%) 44 (40%) T otal 29 57 24 110 T ABLE 6 -DISTRIB UTION OF L AR GER SIZE (> 10 mm ) COL ORECT AL AL GD AND AHGD ENDOSCOP ICALL Y REMO VED IN P A TIENT S O VER 60 YEARS. Right colon Left colon R ectum T otal AL GD (%) 9 (40,9%) 14 (45,15%) 11 (45,85%) 34 (44,15%) AHGD (%) 13 (59,1%) 17 (54,85%) 13 (54,15%) 43 (55,85%) T otal 22 31 24 77

tients is about 25-30% at age 50, mor eo ver they ar e mor e common in men (8-10). The distribution of adenomatous polyps in the co -lon has impor tant implications for scr eening pr o-grams. A dv ancing age is a risk factor for right-sided polyps and cancers (11). This hypothesis is assur ed fr om the results of our study wher e it is demonstrated an incr ease of the per centage in the subjects ov er 60 vs. those younger , of the adenomas of the right colon, especially of AHGD and MP . A denomatous polyps hav e variable dimensions, fr om less than 1 mm to ov er 5 cm of diameter . Small polyps (< 5 mm that w e indicated as ver y ver y small polyps, also kno wn as “diminutiv e polyps ”) ar e rar ely pedunculated. In our casistics w e repor t 843 sessile forms on 1175 polypectomies which includes 508 “di -minutiv e polyps ”. The adenomas ar e classified accor ding to the W orld H ealth O rganization (12) in tubular adenomas, villous adenomas and tubulo-villous adenomas. T o be classified as villous, the adenoma should hav e a villous component of at least 75%. They account for 5 to 15% of adenomas. T ubulo-villous adenomas, having 26 to 75% villous component, account 5 to 15% of adenomas (13). In the curr ent study , the per centage of tubular adenomas was 70,65%, of villous 15,65%, and of tubulo-villous 13,70%. H istologically , adenomatous polyps ar e made up of epithelial packed tubules, divided b y their own lamina, which tendentially dev elop and ramify in a horiz ontal way regar ding to the lev el of the muscularis mucosae (14). V illous polyps ar e, instead, made up fr om a cen -tral nucleus of connettiv e tissue fr om which gr ow up many villi reco ver ed b y epithelium that dev elop in ver -tical way regar ding to the intestinal lumen. The reco ve -ring epithelium pr esents some characteristics similar in all adenomas, independently fr om tubular or villous str uctur e. The aspects of dysplasia or atypia ar e schema -tically repr esented fr om an incr ease of the mitosis and of the pluristratification of the cells that in the case of adenomatous polyps pr oject in the lumen of the tubu -les, while the villous polyps inv ade the connettiv e nu -cleus of the polyp . The grade of dysplasia can be lo w , moderate or high, accor ding to the nuclear alterations 414 G. B enfatto e Coll. T ABLE 7 -DISTRIB UTION OF COL ORECT AL MP ENDOSCOP ICALL Y REMO VED. Age Right colon Left colon R ectum T otal < 60 years (%) 6 (35,3%) 13 (48,15%) 11 (44%) 30 (53,5%) > 60 years (%) 11 (64,7%) 14 (51,85%) 14 (56%) 39 (56,5%) T otal 17 27 25 69 T ABLE 8 -DISTRIB UTION OF COL ORECT AL AL GD, AHGD AND MP SUR GICALL Y REMO VED IN P A TIENT S UNDER 60 YEARS. Right colon Left colon R ectum T otal AL GD (%) 6 (35,3%) 7 (33,35%) 6 (37,5%) 19 (35,2%) AHGD (%) 7 (41,2%) 9 (42,85%) 6 (37,5%) 22 (40,75%) MP 4 (23,5%) 5 (23,8%) 4 (25%) 13 (24,05%) T otal 17 21 16 54 T ABLE 9 -DISTRIB UTION OF COL ORECT AL AL GD, AHGD AND MP SUR GICALL Y REMO VED IN P A TIENT S O VER 60 YEARS. Right Left R ectum T otal AL GD (%) 7 (29,15%) 8 (29,65%) 7 (31,8%) 22 (30,15%) AHGD (%) 11 (45,85%) 12 (44,45%) 9 (40,9%) 32 (43,85%) MP 6 (25%) 7 (25,9%) 6 (27,3%) 19 (26%) T otal 24 27 22 73

(incr eased dimensions, pleomorphism, loss of polarity , stratification and incr easing of the mitosis) (15, 16). M or e recently , this classification scheme has been replaced b y a system that recogniz es two grades of dy -splasia (high and lo w). The two-grade system is pr efe -rable since it is associated with less inter-obser ver va -riation. H igh grade dysplasia should be consider ed as a car cinoma in situ and it is thought to repr esent an intermediate step in the ev olution fr om adenomatous polyp to cancer ,and is a significant risk factor for sub -sequent color ectal malignancy . Lesions that infiltrated the submucosa w er e consider ed to be cancer .The ade -nomatous polyps ar e b y definition dysplastic, but only a small minority of adenomas pr ogr ess to cancer . The location of adenoma was found to be similar to that of cancer in the colon-r ectum, especially for AHGD (17). F rom the T ables of our study it is sho w ed ho w the siz e of adenomas and the age of the patients ar e corr e-lated with a remar kable incr ease of lesions with high grade of dysplasia. P ar ticularly villous histology and incr easing polyp siz e ar e independent risk factors for high-grade dysplasia within an adenoma (18). The ri -sk for high-grade dysplasia incr eases fr om 1 per cent in “diminutiv e polyps ” to 6% in small adenomas and 21% in larger adenomas (18). This study sho ws an in -cr eased incidence of AHGD as the siz e of lesion in -cr eases. It can be evidenced fr om the data repor ted ho w those per centages ar e higher in the subjects ov er 60 years. Ther efor e, older age is another risk factor for hi -gh-grade dysplasia within adenoma, independent of si -ze and histology . E ndoscopic polypectomy/mucosectomy for large color ectal polyps is a difficult method of tr eatment, although it is safe in experienced hands and pr ev ent fr om undergoing unnecessar y surger y. The potential risk of malignancy and technical difficulty in achie -ving complete remo val of large color ectal polyps repr e-sent the most serious pr oblem for the endoscopist. In our study ther e ar e included 127 patients in which the endoscopically complete remo ving of the lesion wasn ’t done because of the siz e and/or the sites of polyps, and ther efor e surger y was necessar y. Also a suite of biopsies negativ e for car cinoma do not ex clude the possibility of a canceriz ed polyp . Some polyps of incr eased consistence, with a large implanting base, ulcerated or that cannot be lifted with injections in the submucosa, must be consider ed potentially malignant. P olyp ’s site, besides its exten -sion, is another factor to be notable. A polyp placed betw een nine and tw elv e o’clock of the field of vision sho ws a bigger difficulty of tr eatment regar ding to a corr espondent position at fiv e o’clock. The identification of the polyps could be easier mo ving the abdomen or changing patient ’s position. Some polyps that ar e placed in both sides of a hau -stral bend (clamshell polyps) can giv e some pr oblems during the endoscopical remo ving pr ocedur e. P lated “carpet polyps ” that dev elop ov er 4 cm cannot be en -doscopically tr eated and need surgical pr ocedur e. B e-sides, fr om the data of the literatur e, it is clear that the polyps 31 to 40 mm in siz e hav e a malignant potential of 40% and in those larger than 40 mm the rate is mo -re than 60%. These data suggest that endoscopic the -rapy is not oncological adeguate in such cases and that primar y surgical therapy should be adv ocated (19). Some difficulties can be also found in the endosco -pical remo ving of polyps in patients with div er ticulitis, “dolichocolon ” or surgical adhesions. As concerns to the rectal polyps, the so-called “pr o-blematic ” rectal polyps ar e repr esented b y those placed near the anus and in the rectal-sigmoid junction or H ouston ’s valv es. Among those with a large implan -ting base (large or giant polyps) ar e consider ed pr oble -matic those expanding for ov er 1/3 of the rectal cir -cumfer ence and/or those with an extension into two continuous bends (20). The choice of surgical options in pr oblematic rectal polyps for esees a transanal ap -pr oach (generally possible till 12-13 cm fr om the anus) or a transabdominal one (in the higher lesions).

Conclusions

This study ,confirming the findings of sev eral other A uthors (5, 7, 10, 19), sho ws that the majority of polyps can be remo ved thr ough a colonoscope (90,25% in our series). E ndoscopic snar e polypectomy per formed on an outpatient basis b y an exper t endo -scopist is the best pr ocedur e in the tr eatment of the majority of the color ectal adenomas. Surger y must be per formed when the histologic examination sho ws the pr esence of an inv asiv e car cinoma; it is mandator y in urgency when an endoscopic polypectomy determines a colic per foration or a bleeding other wise not solv able (0,30% in our casistic). Those “accidents ” ar e mor e fr equent in the elderly because the anatomical and morphological regr essiv e modifications of the colon (“ pr esb yopic-colon ”) and because of the fr equency of div er ticular disease. The location, the siz e, the grade of dysplasia of the polyps and the anatomical and morphologic condi -tions of the colon should repr esent a limit to the feasi -bility and efficacy of the endoscopic tr eatment. Clinical and endoscopical follo w-up must be per formed in all the patients at differ ent inter vals ac -cor ding to the histologic findings and the results of the successiv e endoscopies. 415 P athological ev aluation in color ectal polyps endoscopic tr eatment

416 G. B enfatto e Coll. 1. N eugut I, Jacobson JS, D eviv o I. E pidemiology of color ectal adenomatous polyps. Cancer E pidemiol B iomar ker P rev 1993; 3: 159-176. 2. Bond JH. Colon polyps and cancer . E ndoscopy 1999; 31: 60-65. 3. M uto T , N agawa H, W atanabe T , M asaki T , Sawada T . Colo -rectal car cinogenesis: historical revie w . D is Col R ectum 1997; 40 (suppl.): 580-585. 4. T ada S, Y ao T , Iida M, K oga H, H izaea K, F ujishima M. A cli -nicopathologic study of small flat color ectal car cinoma. Can -cer 1994; 74: 2430-2435. 5. W inaw er SJ, Z auber A G, H o MN, O’B rien M, G ottlieb LS, Stemberg SS, et al. P rev ention of color ectal cancer b y colono -scopic polypectomy . N E ngl J M ed 1993; 329: 1977-1981. 6. A tkin WS, C uzick J, N or tho ver JMA, Whynes DK. P rev en -tion of color ectal cancer b y once-only sigmoidoscopy . Lancet 1993; 341: 736-740. 7. Ikeda Y , M ori M, Shibahara K, Iwashita A, H araguchi Y , Saku M. The role of adenoma for color ectal cancer dev elopment: differ ence in the distribution of adenoma with lo w-grade dy -splasia, high-grade dysplasia, and cancer that inv ades the sub -mucosa. Surger y 2002; 131 (1): 105-108. 8. R ex DK, Lehman GA, H aw es RH, et al. Scr eening colono -scopy in asymptomatic av erage-risk persons with negativ e fe -cal occult blood tests. G astr oenter ology 1991; 100: 64. 9. R ex DK, Lehman GA, U lbright TM, et al. Colonic neoplasia in asymptomatic persons with negativ e fecal occult blood te -sts: influence of age, gender , and family histor y. Am J G a-str oenter ol 1995; 88: 825. 10. R ex DK. Colonoscopy: a revie w of its yeald for cancers and adenomas b y indication. Am J G astr oenter ol 1995; 90: 353. 11. P atel K, H offman NE. The anatomical distribution of color ec -tal polyps at colonoscopy . J Clin G astr oenter ol 2001; 33: 222. 12. M orson C, Sobin LH. International histological classification of tumours .15. H istological typing of intestinal tumours. G e-nev a: WHO. 1976. 13. B ensen SP , Cole BF , M ott L A, et al. Color ectal hyperplastic polyps and risk of recurr ence of adenomas and hyperplastic polyps. Lancet 1999; 354: 1873. 14. D ay W , M orson BC. P athology of adenomas. The pathogene -sis of color ectal cancer . P hiladephia: WB Saunders. 1978; pp . 43-75. 15. E kelund G, Lindstr om C. H istopathological analysis of beni -gn polyps in patients with car cinoma of the colon and rectum. G ut 1974; 15: 654. 16. K ozuka S. P remalignancy of the mucosal polyp in the large in -testine. H istological gradation of the polyp on the basis of epithelial pseudostratification and glandular branching. D is Colon R ectum 1975; 18: 483. 17. E ide TI. The age-, sex-, and site-specific occurr ence of adeno -mas and car cinomas of the large intestine within a defined po -pulation. Scan J G astr oenter ol 1986; 21: 1083-1088. 18. O’B rien MJ, W inav er SJ, Z auber A G, et al. The national polyp study: patient and polyp characteristics associated with high-grade dysplasia in color ectal adenomas. G astr oenter ology 1990; 98: 371. 19. D ell ’A bate P, Iosca A, G alimber ti A, P iccolo P, Soliani P, F og -gi E. E ndoscopic tr eatment of color ectal benign-appearing le -sions 3 cm or lager: techniques and outcome. D is Colon R ec -tum 2001; 44 (1): 112-118. 20. W ay e JD. H ow big is too big? G astr ointest E ndosc 1996; 43: 256-257.

References