Screening by Colonoscopy for Colonic Epithelial Dysplasia in Inflammatory Bowel Disease
Total Page:16
File Type:pdf, Size:1020Kb
Gut: first published as 10.1136/gut.28.11.1489 on 1 November 1987. Downloaded from Gut, 1987, 28, 1489- 1494 Screening by colonoscopy for colonic epithelial dysplasia in inflammatory bowel disease A P MANNING, 0 R BULGIM, M F DIXON, AND A T R AXON From the Gastroenterology Unit, General Infirmary at Leeds and Department ofPathology, University of Leeds, Leeds SUMMARY We report the results of a prospective study of screening for colorectal epithelial dysplasia by regular colonoscopy in patients with longstanding, extensive colitis (DET group: 112 patients, 366 colonoscopies) together with the findings in all other patients with colitis who have undergone colonoscopy in our unit (non-DET group: 77 patients, 196 colonoscopies). Thirty six DET patients had dysplasia on at least one examination: two patients with high grade dysplasia (HGD) were colonoscoped on suspicion of carcinoma, one asymptomatic patient had HGD at first colonoscopy and one patient had HGD on his sixth colonoscopy, all having carcinomas resected at surgery; the remainder had low grade dysplasia (LGD). Of the DET patients, 100 constituted an ongoing surveillance group (354 colonoscopies) in which LGD was common, being seen on at least one occasion in 33% of patients (16-4% of examinations), but HGD was noted only once with a Dukes A cancer found at surgery. Six non-DET patients had dysplasia diagnosed, this being LGD in all. Even in a carefully selected group of colitics the incidence of HGD is low, but its detection may enable the removal of a colorectal carcinoma at an early and curable stage. http://gut.bmj.com/ Patients with longstanding, extensive ulcerative carcinoma. Preliminary results were published in colitis (UC) have an increased risk of colorectal 1980.9 We have now analysed the results of this study carcinoma." The recognition of this association led to the end of 1984, together with the findings at in the 1960's to the practice of 'prophylactic' procto- colonoscopy of all other patients with colonic on September 29, 2021 by guest. Protected copyright. colectomy for patients with extensive UC of 10 or inflammatory bowel disease to determine the more years duration.4" Because of the observation prevalence and clinical associations of mucosal that invasive carcinoma may be accompanied, or dysplasia in our patient population. even preceded, by mucosal changes termed precancer or dysplasia,67 however, routine 'prophylactic' Methods proctocolectomy has been largely abandoned in favour of regular colonic or rectal examinations with PATI ENTS multiple biopsy." Although this 'surveillance' At the Gastroenterology Unit, General Infirmary at approach has been widely embraced, the epidemiol- Leeds, since 1978 we have colonoscoped regularly, ogy and significance of mucosal dysplasia in UC is usually annually, all patients with colitis of eight poorly understood. As such an understanding years duration or longer in whom colonic involve- is crucial to the development of an effective manage- ment is known to be extensive or total to screen for ment policy for patients with UC, a prospective study dysplasia or carcinoma. In addition other patients was begun at this hospital in 1978 in which patients with colitis have undergone colonoscopy for a variety with longstanding, extensive colitis undergo annual of clinical indications, biopsies usually being taken to colonoscopy for the detection of dysplasia and clarify diagnosis or to determine extent of disease. Information on all these patients up to the end of Address for correspoiideflcc: D)r A P Mi.nning. Gistroenterology Unit. The General Infirm.ary. Iceds LS I 3[X. 1984 was obtained from departmental records and Received for public;t[ion 9 April 1987. supplemented by case note review. 1489 Gut: first published as 10.1136/gut.28.11.1489 on 1 November 1987. Downloaded from 149() M19AIanning, Bldgimi, D)ixon, (aid A.xsoi Colonoscopy was usually carried out under light Table 2 IJ)IvspliasI /)aliIlat: )lS('l( (' of(If/)Iawl( i'l sedation after a standard laxative and enema bowel colo1)oscopi.sp)er' orined 1(78-1984 preparation. Biopsy specimens were taken with Yoa of standard forceps from flat mucosa ait approximately 8 (colono(o/)p to 10 cm intervals along the colon and rectum and Pati(ent Group IO,aplo.si.,i.17 /979197() 19811MS I1) 19834 also from any identifiable mass lesions. The biopsies I SG U(C ( /()I(' were mounted on filter paper before being fixed in 2 S( LJC (1±+ + () + ++ by formalin and processed routine techniques. They 3 SG LC' (1 + (I ( were sectioned at three levels, stained with haemra- 4 S; LJ(' toxylin and eosin and examined by a single patholo- s SG Ilndict. + 6 SG tC' + gist with an interest in gastrointestinal pathology. For 7 SG LJC + + + the purpose of this study, only samples with definite 8 Si tiC + dysplasia were recorded and divided into low (LGD) 9 S,G In(dct. + and high grade dysplasia (HGD) using criteria which 1() SG lJC (1 + 11 SG UC + + (1 correspond to those recently put forward by an 1 2 SG LJC' (1 (1 + + (1 international study group." 13 SG LJ' + (1 ( 14 SG UtC 0+ + + + (1 1 5C ('iSohn's + (1 + (1 (1 (1 Results 1 6 SG lJC + (1 ( 1 7 SG tc' (1 (1 (1 (1I+ PATIENT GROUPS 18 SG UC + (1 ( Of 189 patients with colitis who had undergone 1 9 SG UC' (} 1) + ( 21( SG UC' (} + + (1 (1 ( colonoscopy, 1 12 had had their diseatse for eight years 2 SCJ LJC + (1 ( or longer in duration, which was extensive or total by SG UC' () + +I+ ( (1 at least one of (a) barium enema (b) colonoscopic 23 SG UC () + + (1 (1 24 S, tiC (} + + + appearances (c) colonic histology (DET group); 77 25 SG UC 1) + (1 had colitis of less than eight years' duration and/or 26 S(G C'iohil's + (1 disease that was not extensive or total by any 27 SG LJC + + (1 28 S(G tLJc criterion (non-DET group). The characteristics of + 29 Sc; tLC the two groups are shown in Table 1. The 112 patients http://gut.bmj.com/ 31) SG lJC () (1 (I (1 in group the DET had undergone 366 colonoscopies, 31 SGc UC' (1 + 1) + (1 nine having had a further 13 examinltions before the 32 SG UC duration of their disease had reached eight years. The 33 SG UC 34 DET UC group non-DET had undergone 106 colonoscopies T3h DET UC' on including the 13 examinations subsequent DET 36 DET UC ++ group patients. 37 ilon-DLET ('rohnl's ++ 38 non-DET UC + 39 moti on September 29, 2021 by guest. Protected copyright. I)YSPLASIA -DET UC 4(0 non-DET UC' on one Of the 189 patients, 42 had dysplasia at least 41 inon-DET UC + (1 occasion (Table 2), 36 being in the DET group and six 42 non-DE' tJC' in the non-DET group (X'= 14-5; p<0)0 l). Of the 36 with dysplasia in the DET group, two D)ET=colitis>cight ycars in Duraition aind Extcsiis lc - Totil; patients 6 aind underwent Sc; =)ET paticr'tsalso in thc suirsc illatncc group; non-DEL= (patients 36) colonoscopy colitis<cight ycars and/or noni cxtcnsi,sc or tot(l tIJC=lccrIatiVsc coolitis; Indct inIctcrilinaitc' idliopathiccolitis()= no dvsplasia +=lowgradcdysplasia; ++=highgradcdysplasia; I=two Table 1 (Characte-istic.s of patie1ts wnil/i (cOlitiS ulnidergoinlg colonoscopics pcrlormcd within a ycar indicatcd thus colontscopv1, 2= paticnt 6 had survcillarncc colonoscopics aii'tcr suhtotal colctonly for caircinoma inititilly suspcctcd ol 1Ba cncmrn. 1)LET1group Non-)D 'l (ni //12) group) (n = 77) Malc:fcIal1c 57:55 36:41 for confirmation of carcinoma suspected on barium Agcatonscti(yr) (nican±SD) 2'93±11.2 381+16-7 enema. Patient 6 had a polypoid lesion in the sigmoid Diagnosis: colon both on barium enema and at colonoscopy with Ulccrativc colitis 98 50) [IGD on biopsy; patient 36, with a caecal lesion Crohn'sdiscasc 5 12 radiologically, had an incomplete colonoscopy but 9 1 'Indeterminatc' idiopathic colitis with lIGD on biopsies taken from the transverse DET=colitis>cight vcaii-s in l)uration and Fxtcnisisc or Total; colon. A carcinoma was subsequently resected non-DET= colitis<cight scars and/or niot cxtcnsisic or totail. surgically in each case. In a further 11 patients Gut: first published as 10.1136/gut.28.11.1489 on 1 November 1987. Downloaded from Screening by colonoscopyfior colonic epithelial dysplasia ill inflattnsitorv bowel disease 1491 dysplasia was detected at first colonoscopy; in 10 this In order to assess the prevalence of dysplasia with was LGD and in one IIGD (patient 34). In this increasing duration of disease, the number of asymptomatic patient with colitis of 15 years duration dysplastic examinations per decade of disease c7olectomy revealed a Dukes' A carcinoma. duration in the surveillance group was determined and is shown in Table 3. SURVEII I ANCE GROUP Of the 112 patients in the DET group, 13 underwent CO- ECTOMY FINDINGS colectomy after a single colonoscopy, one of whom In the DET group 17 patients had resective surgery (patient 6) had subtotal colectomy and subsequent for colitis after colonoscopy. Of these, dysplasia had colonoscopies, leaving 100 DET patients who can be been diagnosed in five: in two (patients 6 and 36) the regarded as an ongoing surveillance group (SG). colonoscopies had been done with a prior suspicion In the surveillance group, a total of 354 examina- of carcinoma that was subsequently confirmed; in tions had been carried out and the number of one (patient 34) HGD was noted at first colonoscopy colonoscopies per patient is shown in the Figure. Thirty three patients in the surveillance group had dysplasia on at least one occasion, in 59 out of 152 TIablc 3 Incidence ofdvsplasia acc(ordinig to decade of colonoscopies.