contrasting effects of subtotal enteric bypass...

7
[CANCER RESEARCH 40, 538-543, March 1980] 0008-5472/80/0040-0000$02.00 Contrasting Effects of Subtotal Enteric Bypass, Enterectomy, and Colectomy on Azoxymethaneinduced Intestinal Carcinogenesis1 Robin C. N. Williamson,2 Frederick L. R. Bauer,3 Onno T. Terpstra,3 Jeffrey S. Ross,4 and Ronald A. Malt5 Surgical Services, Shriners Burns Institute, Massachusetts General Hospital, Massachusetts 02115 ABSTRACT Compensatory hyperplasia after extensive loss of functioning small or bargeintestine might predispose to the development of neoplasia in the residual adapted bowel. To test this hypothe sis, male Fischer rats were randomized to receive 85 to 90% jejunoibeal resection on bypass, subtotal colectomy, or no op eration (controls). One week later, the first of six weekly s.c. injections of azoxymethane (i S mg/kg/week) was given. At the 36th week postoperatively, mean body weight after enteric bypass was 62% of control values, and after enterectomy or cobectomy it was 78 to 79% of control. Adaptation after all three operations was characterized by 22 to 84% increments in vilbous height and crypt depth in the residual functioning ileum ( p = 0.05 to 0.001 ); the depth of cobonic crypts was unchanged. Fewer rats developed intestinal tumors after en tenic bypass (36%) than after any of the other treatments (80 to 91%) (p = 0.01 to 0.001 ). Compared with controls, bypass reduced the number of cobonic tumors by 77% ( p < 0.001). Although resection did not affect cobonic tumor yield, it tripled the incidence of tumors in the duodenum and jejunum (p = 0.025). Colectomy promoted rectal carcinogenesis (p < 0.05). Anastomotic tumors were commoner after intestinal resection. The lower frequency of tumors after jejunoibeal bypass con trasts with enhanced carcinogenesis after entenectomy on co bectomy. Profound reduction in body weight may prevent the promotional effect of adaptive hyperplasia. INTRODUCTION Enhanced susceptibility of proliferating cells to carcinogenic agents probably explains the increased prevalence of expeni mental cobonic tumors after either proximal or distal enterec tomy (25, 35) or pancreaticobiliary diversion to mid-small bowel (37), since all 3 operations produce at least transient cobonic hyperplasia (23, 24, 32, 36). Following jejunal bypass as opposed to jejunab resection, hyperplasia of the small bowel remaining in circuit is slightly slower in origin but soon develops to the same degree (i 0, i i , 13, 20, 22, 36); compensatory hyperplasia in the colon, however, is much less easy to detect (6, 36). Subtotal resection of the colon itself elicits adaptive growth of the ileum (1 , 39, 40). ‘Supportedby the USPHS through the National Cancer Institute and the National Large Bowel Cancer Project (Grant CA 17324), and by the Stanley Thomas Johnson Foundation. 2 Present address: Department of Surgery, Bristol Royal Infirmary, Bristol B52 8HW, England. 3 Present address: Department of Surgery, University of Rotterdam, Rotter dam, The Netherlands. 4 Present address: Department of Pathology, Berkshire Medical Center, PiUs field, Mass. 01201. 5 To whom requests for reprints should be addressed, at Massachusetts General Hospital, Boston, Mass. 021 14. Received January 8, 1979; accepted November 12, 1979. Boston 02 114, and Department of Surgery, Harvard Medical School, Boston, Since compensatory hyperplasia of the residual small bowel (and, by implication, the colon) is proportional to the extent of partial enterectomy performed (16, 32, 33), subtotal removal of functioning small bowel might cause maximal enhancement of coborectabcarcinogenesis. Patients losing a large proportion of functioning intestine, for example in the surgical treatment of morbid obesity or inflammatory bowel disease, might there fore be at risk of developing bowel cancer in later years. This possibility was tested in rats exposed to the selective intestinal carcinogen, azoxymethane. In fact, jejunoileal and cobonic resections increased the frequency of tumors in adjacent sec tions of bowel, but jejunoileab bypass inhibited carcinogenesis. MATERIALS AND METHODS Male Fischer rats (Charles River Breeding Laboratories, WiI mington, Mass.) were received into the animal quarters 7 to 10 days before operation. Quarters were lighted in 12-hr cycles, and animals were housed in cages with open wine-mesh bot toms. Purina rat chow and water were given ad !ibitum, except that food was withdrawn overnight following operation. Rats weighing 126 ±9.3 g (S.D.) (N = 190) were randomly assigned to 1 of 4 groups. Controls had no operation (Chart 1A). In the other 3 groups, lapanotomy was carried out through a midline incision under light ether anesthesia. The small bowel was delivered and measured between the ligament of Tneitz and the ileocecal valve by gently stretching S-cm segments against a ruler. The combined length of jejunum and ibeum varied from 85 to 90 cm. In subtotal (85 to 90%) entenic bypass (Chart 1B), the small bowel was divided S cm distal to the ligament of Treitz and the distal cut end was closed and invaginated; the proximal transected end of the jejunum was anastomosed end-to-side to the terminal ileum, S cm proximal to the ileocecal valve. The operation provided a long (75- to 80-cm) self-emptying blind loop, similar to an end-to-side je junoileabbypass procedure for obesity in humans (26). Subtotal (85 to 90%) entenectomy (Chart 1C) was carried out by nesec tion of a similar length (75 to 80 cm) of mid-small bowel, leaving S-cm stumps of jejunum and ileum, which were joined by end to-end anastomosis. Subtotal colectomy (Chart 10) required division of the small bowel immediately proximal to the ileocecal valve and of the large bowel at the level of the pelvic brim (rectosigmoid junction); intestinal ôontinuitywas then restored by end-to-end ileorectal anastomosis. All anastomoses were performed with 6-0 silk sutures. Rats in each group received either an aqueous solution of azoxymethane (i S mg/kg/week) or sterile water by 6 weekly s.c. injections, beginning 1 week after operation or control observation. Dilute solutions of azoxymethane (Ash Stevens, Inc., Detroit, Mich.) were stoned at —20° until needed and then thawed and further diluted to a concentration of i to 2 mg/mI 538 CANCERRESEARCHVOL. 40 on June 22, 2018. © 1980 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Upload: phungtu

Post on 13-May-2018

223 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Contrasting Effects of Subtotal Enteric Bypass ...cancerres.aacrjournals.org/content/canres/40/3/538.full.pdf · Contrasting Effects of Subtotal Enteric Bypass, Enterectomy, and Colectomy

[CANCER RESEARCH 40, 538-543, March 1980]0008-5472/80/0040-0000$02.00

Contrasting Effects of Subtotal Enteric Bypass, Enterectomy, andColectomy on Azoxymethaneinduced Intestinal Carcinogenesis1

Robin C. N. Williamson,2 Frederick L. R. Bauer,3 Onno T. Terpstra,3 Jeffrey S. Ross,4 and Ronald A. Malt5

Surgical Services, Shriners Burns Institute, Massachusetts General Hospital,Massachusetts 02115

ABSTRACT

Compensatory hyperplasia after extensive loss of functioningsmall or bargeintestine might predispose to the development ofneoplasia in the residual adapted bowel. To test this hypothesis, male Fischer rats were randomized to receive 85 to 90%jejunoibeal resection on bypass, subtotal colectomy, or no operation (controls). One week later, the first of six weekly s.c.injections of azoxymethane (i S mg/kg/week) was given. Atthe 36th week postoperatively, mean body weight after entericbypass was 62% of control values, and after enterectomy orcobectomy it was 78 to 79% of control. Adaptation after allthree operations was characterized by 22 to 84% incrementsin vilbous height and crypt depth in the residual functioningileum ( p = 0.05 to 0.001 ); the depth of cobonic crypts wasunchanged. Fewer rats developed intestinal tumors after entenic bypass (36%) than after any of the other treatments (80to 91%) (p = 0.01 to 0.001 ). Compared with controls, bypassreduced the number of cobonic tumors by 77% ( p < 0.001).Although resection did not affect cobonic tumor yield, it tripledthe incidence of tumors in the duodenum and jejunum (p =0.025). Colectomy promoted rectal carcinogenesis (p < 0.05).Anastomotic tumors were commoner after intestinal resection.The lower frequency of tumors after jejunoibeal bypass contrasts with enhanced carcinogenesis after entenectomy on cobectomy. Profound reduction in body weight may prevent thepromotional effect of adaptive hyperplasia.

INTRODUCTION

Enhanced susceptibility of proliferating cells to carcinogenicagents probably explains the increased prevalence of expenimental cobonic tumors after either proximal or distal enterectomy (25, 35) or pancreaticobiliary diversion to mid-small bowel(37), since all 3 operations produce at least transient cobonichyperplasia (23, 24, 32, 36). Following jejunal bypass asopposed to jejunab resection, hyperplasia of the small bowelremaining in circuit is slightly slower in origin but soon developsto the same degree (i 0, i i , 13, 20, 22, 36); compensatoryhyperplasia in the colon, however, is much less easy to detect(6, 36). Subtotal resection of the colon itself elicits adaptivegrowth of the ileum (1 , 39, 40).

‘Supportedby the USPHS through the National Cancer Institute and theNational Large Bowel Cancer Project (Grant CA 17324), and by the StanleyThomas Johnson Foundation.

2 Present address: Department of Surgery, Bristol Royal Infirmary, Bristol B52

8HW, England.3 Present address: Department of Surgery, University of Rotterdam, Rotter

dam, The Netherlands.4 Present address: Department of Pathology, Berkshire Medical Center, PiUs

field, Mass. 01201.5 To whom requests for reprints should be addressed, at Massachusetts

General Hospital, Boston, Mass. 021 14.Received January 8, 1979; accepted November 12, 1979.

Boston 02 114, and Department of Surgery, Harvard Medical School, Boston,

Since compensatory hyperplasia of the residual small bowel(and, by implication, the colon) is proportional to the extent ofpartial enterectomy performed (16, 32, 33), subtotal removalof functioning small bowel might cause maximal enhancementof coborectabcarcinogenesis. Patients losing a large proportionof functioning intestine, for example in the surgical treatmentof morbid obesity or inflammatory bowel disease, might therefore be at risk of developing bowel cancer in later years. Thispossibility was tested in rats exposed to the selective intestinalcarcinogen, azoxymethane. In fact, jejunoileal and cobonicresections increased the frequency of tumors in adjacent sections of bowel, but jejunoileabbypass inhibited carcinogenesis.

MATERIALS AND METHODS

Male Fischer rats (Charles River Breeding Laboratories, WiImington, Mass.) were received into the animal quarters 7 to 10days before operation. Quarters were lighted in 12-hr cycles,and animals were housed in cages with open wine-mesh bottoms. Purina rat chow and water were given ad !ibitum, exceptthat food was withdrawn overnight following operation.

Rats weighing 126 ±9.3 g (S.D.) (N = 190) were randomlyassigned to 1 of 4 groups. Controls had no operation (Chart1A). In the other 3 groups, lapanotomy was carried out througha midline incision under light ether anesthesia. The small bowelwas delivered and measured between the ligament of Tneitzand the ileocecal valve by gently stretching S-cm segmentsagainst a ruler. The combined length of jejunum and ibeumvaried from 85 to 90 cm. In subtotal (85 to 90%) entenic bypass(Chart 1B), the small bowel was divided S cm distal to theligament of Treitz and the distal cut end was closed andinvaginated; the proximal transected end of the jejunum wasanastomosed end-to-side to the terminal ileum, S cm proximalto the ileocecal valve. The operation provided a long (75- to80-cm) self-emptying blind loop, similar to an end-to-side jejunoileabbypass procedure for obesity in humans (26). Subtotal(85 to 90%) entenectomy (Chart 1C) was carried out by nesection of a similar length (75 to 80 cm) of mid-small bowel, leavingS-cm stumps of jejunum and ileum, which were joined by endto-end anastomosis. Subtotal colectomy (Chart 10) requireddivision of the small bowel immediately proximal to the ileocecalvalve and of the large bowel at the level of the pelvic brim(rectosigmoid junction); intestinal ôontinuitywas then restoredby end-to-end ileorectal anastomosis. All anastomoses wereperformed with 6-0 silk sutures.

Rats in each group received either an aqueous solution ofazoxymethane (i S mg/kg/week) or sterile water by 6 weeklys.c. injections, beginning 1 week after operation or controlobservation. Dilute solutions of azoxymethane (Ash Stevens,Inc., Detroit, Mich.) were stonedat —20°until needed and thenthawed and further diluted to a concentration of i to 2 mg/mI

538 CANCERRESEARCHVOL. 40

on June 22, 2018. © 1980 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Page 2: Contrasting Effects of Subtotal Enteric Bypass ...cancerres.aacrjournals.org/content/canres/40/3/538.full.pdf · Contrasting Effects of Subtotal Enteric Bypass, Enterectomy, and Colectomy

Mortalityrates in animalswith carcinogen orvehicleSubse

quentRats surEarly post deaths (bevivingNo.

of ratsoperativefore 27(after30Treatmentatoutsetdeathswk)wk)AzoxymethaneControl26422Enteric

bypass325522Entericresection328420Colectomy306618VehicleControl14212Enteric

bypass193610Entericresection194312Colectomy1

84410

C ENTERECTOMY

Intestina! Bypass, Resection, and Cancer

D COLECTOMYB ENTERICSYPASSA CONTROl.

Chart 1. Operations performed. 5haded segments, site of specimens taken for histological measurement of villous height and crypt depth. LIG.. bigament.

for injection. All animals received azoxymethane from the samebatch.

Rats were weighed weekly, observed for evidence of intestinal tumor formation, and killed 36 weeks after operation. Atautopsy, tumors were charted, measured, and sent for histological examination, together with possible metastases. Oneparaffin block was prepared from each tumor, and severalsections were cut at various levels to assess the maximal depthof invasion. In addition, 1-cm segments of macroscopicablynormal intestine were stored in fonmalin for subsequent histological measurement of vilbousheight and crypt depth. Thesesegments were obtained from the small bowel 3 (terminal ibeum)and 35 (midileum) cm proximal to the ileocecal valve, and fromthe large bowel 3 (right colon) and 15 (left colon) cm distal tothe cecum (Chart 1). Segments were coded to eliminate observer bias, and 10 small bowel vibliand cnypts (sectioned fromtop to bottom without interruption) were measured penslide byocular micrometry. In the colon, crypt depth was measured.

Tumors were classified into 3 histological types, papillary(exophytic, vilboglandular, verrucous), tubular (adenomatous,glandular), or colboid(mucinous). In addition, 3 levels of malignant invasion were determined, absent (carcinoma in situ),superficial (submucosab),or deep (transmunab).

Statistical tests included Student's t test for analysis ofmorphometnic data and the 1-tailed x2 test and the MannWhitney U test for comparisons of tumor yields and histologicalpatterns.

RESULTS

Mortalfty. The penioperative mortality rates varied between16 and 25% (Table I ). In each of the 4 groups receivingazoxymethane, a further 4 to 6 rats died before the 27th week,either of respiratory infection or (in operated animals) of wasting and progressive loss of weight. No tumors were found inany of these animals, which were excluded from further analysis. The yields of rats surviving a minimum of 30 weeks areshown in Table i.

Table 1

Weight Loss. In control animals, administrationof azoxymethane did not prevent a weight gain of about 20 g/weekduring the first few weeks. Entenic bypass produced a mean12% bossof total body weight within 48 hr of operation (Chart2). Thereafter, weight remained approximately static for 7 to10 days before gradually returning to preoperative levels after2 to 3 weeks. Body weight after bypass remained substantiallylower than in any of the other 3 groups for the remainder of theexperiment, though differences from animals with intestinalresection did not reach significance during the course of azoxymethane injections. When sacrificed at 36 weeks, rats withbypass weighed 62% of the amount of control rats and 78 to80% that of rats with enterectomy or colectomy.

Enterectomy and colectomy were associated with a slightlygreater postoperative weight loss (20%) than was enteric bypass, but surviving rats gained weight more rapidly and appeared healthier thereafter. By the end of the experiment, ratswith subtotal resection of the small on large intestine weighed78 to 79% of the amount of unopenated controls. Throughoutthe experiment, body weights of animals receiving water were

MARCH1980 539

on June 22, 2018. © 1980 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Page 3: Contrasting Effects of Subtotal Enteric Bypass ...cancerres.aacrjournals.org/content/canres/40/3/538.full.pdf · Contrasting Effects of Subtotal Enteric Bypass, Enterectomy, and Colectomy

R. C. N. Wi!!iamson et a!.

400

300

200

100

00

WEEKSAFTEROPERATIONChart 2. Body weight after operation In rats receiving azoxymethane.

BODYWEIGHT

Ig)

5 10 15 20 25 30 35

similar to those receiving azoxymethane, regardless of theexperimental procedure.

All 3 operations markedly increased the intake of food andwater during the early weeks. Colectomy caused a particularlyprofuse diarrhea initially, but after a few weeks movementsreturned to a semisolid consistency.

Intestinal Adaptation. The jejunaband ileal segments remaining in circuit after midbowebresection or bypass showedmacroscopic hypertrophy, elongation, and dilation. By contrast, the caliber of the bongblind loop after bypass showedprogressive tapering away from the site of anastomosis. Nogross changes in cobonicappearance were detected.

Entenic bypass caused slightly greater morphometnicchanges in the ibeumthan did either entenectomy or colectomy.Vilbiwere 39 to 84% taller after bypass than those in any of theother 3 groups, and crypts were 22% deepen than those ineither controls or rats with cobectomy. Although entenectomyand cobectomy increased vilbous heights in the terminal ileumby 24 to 32% of control values, neither operation altered cryptdepth at this site (Chart 3). Midileab vibbousheight was notaffected by small bowel bypass on bargebowel resection, butboth procedures caused a slight (16%) increase in crypt depth(Chart 3).

Neither resection nor bypass of the small bowel significantlyaltered the depth of cobonic crypts at 36 weeks as comparedwith control values.

Prevalence of Small Bowel Tumors. Approximately 1 inevery 3 control animals had an entenic tumor at autopsy (Table2). These tumors were scattered through the small bowel, 3each in the duodenum and terminal ibeumand 2 in the upperjejunum (Chart 4A). Neither entenic bypass nor colectomysignificantly altered the total number of rats with small boweltumors; however, entenic bypass did reduce the number oftumors in the duodenojejunal segment ( p < 0.05). After cobectomy, only 2 entenic tumors were found, both in the duodenum(Chart 4D). After bypass, 3 tumors were encountered at theanastomosis and a fourth arose nearby in the distal ileum(Chart 4B).

Following subtotal entenectomy, the number of rats with smallbowel tumors (70%) was greatenthan in any other group (Table2). Although much of this increase was due to a high yield ofanastomotic tumors (Chart 4C), enterectomy also producedmany more duodenojejunab tumors pen rat (0.85) than eitherentenic bypass (0), colectomy (0.1 1), oncontrols (0.23) (Chart4). In the remaining small bowel, the increased tumor yieldafter entenic resection (Chart 4C) made a particularly strikingcontrast to the decreased tumor yield after entenic bypass(Chart 4B).

Prevalence of Large Bowel Tumors. Eighty-six% of controlanimals had one barge bowel tumor or more, as opposed to27% of animals with entenic bypass (Table 2; p < 0.001 ). Allthese tumors were distributed throughout the colon and upperrectum, but spared the cecum (Chart 4). Twenty-two controlrats produced a total of 30 cobonectabtumors (Chart 4A); thesame number of rats with small bowel bypass produced only 7tumors (Chart 4B; p < 0.001 ). Following entenic bypass, theoverall number of rats with tumors situated either in the largebowel alone on throughout the intestinal tract was less thanone-half that occurring in any other group (Table 2).

Compared with controls, enterectomy altered neither thenumber nor the distribution of coborectal tumors (Chart 4C).After subtotal colectomy, many tumors arose at the site ofileonectabanastomosis, but in addition the number of rectaltumors was increased 3-fold (Chart 4D). Only after colectomywere tumors encountered in the middle and lower rectum.

Tumor Histology.The 3 differenthistologicaltypesofcancerdid not correlate with the different geographical regions of theintestinal tract, except that colboid cancer did not affect therectum. Papillary tumors predominated in controls and in ratswith entenic bypass, but after enterectomy and colectomytubular carcinomas were slightly more common, occurringespecially at anastomoses and in the rectum. In control animals, only 13% of tumors were noninvasive; operation increased the proportion of carcinoma in situ to 33 to 45%.There was no correlation between the size of tumor and depthof invasion. All colboidcancers were deeply invasive, but many

540 CANCERRESEARCHVOL. 40

on June 22, 2018. © 1980 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Page 4: Contrasting Effects of Subtotal Enteric Bypass ...cancerres.aacrjournals.org/content/canres/40/3/538.full.pdf · Contrasting Effects of Subtotal Enteric Bypass, Enterectomy, and Colectomy

%withtumorsEntericEnterec

Tumors and meControlBypasstomyColectomytastases(N= 22)(N = 22)(N = 20)(N =18)Intestinal

tumors9136a8083Smallboweltumors

Large bowel tumorsAnastomotic tumors32

8614

27c14d70b

70508367Metaetasea2392517

Intestina! Bypass, Resection, and Cancer

MID ILEUM

Chart 3. Mean vilbousheight and crypt depth in the ileum after operation. Bars, SE. Significance (Student's t test): a, p < 0.001 versus control, enterectomy,colectomy; b. p < 0.05 versus control, colectomy; c, p < 0.005 versus control; d, p < 0.05 versus control.

TERMINAL ILEUM

400

300

200

100

VILLOUSHEIGHT(pm)

CRYPTDEPTH

(pm)

0

100

200

300CONTROL INTLKIC ENTERECTOMYCOLECTOMY CONTROL@ COLECTOMY

BYPASS BYPASS

Table 2Percentage of rats with intestinal tumors and metastases

the number of rectal tumors. The findings after intestinal resection are consistent with our recent reports that coborectalcarcinogenesis can be enhanced by partial enterectomy onpancreaticobibiary diversion (25, 35, 37). Indeed, the presentobservation that small bowel resection promotes cancer inbowel proximal to the site of resection supports the previoushypothesis (37) that postoperative increments in the number ofdistal tumors depend on adaptive mucosal hyperplasia ratherthan on greater excretion of bile acids. Partial enterectomyproduces some compensatory growth of the duodenum andjejunum as well as of the ileum (8, 16, 32—34).

The contrasting effects of entenic excision and exclusion onsmall bowel carcinogenesis can scarcely be attributed to anytransient differences in adaptive response (36). Bypass causedslightly greater morphometnic increments in the ileum at 36weeks than did equivalent resection. Despite reflecting grossalterations in mucosabcellularity (36), however, vibbousheightand crypt depth are not very sensitive indices of adaptation (4).The absence of any demonstrable atrophy in the lower part ofthe blind loop might therefore be due eIther to the inadequacyof these simple histological measurements or to the fact thatileum is less dependent than jejunum on the presence ofbuminalnutrients for preservation of mucosabintegrity (34, 36).Differential adaptation might perhaps contribute to the respective tumor yields in the colon, since structural adaptation of thebargebowel follows partial entenectomy but probably does notfollow entenic bypass (23, 31 , 36). In the present study thedepth of cobonic crypts was unchanged 36 weeks after eitherprocedure.

Subtotal small bowel bypass probably protects against cancer by sharply reducing body weight. Since resection andbypass leave the same length of functioning small bowel in

:@ <o.o@versusenterectomy,cobectomy;p<0.001versuscontrol(x2test).p < 0.05 versus control; p < 0.001 versus enteric bypass, colectomy (x'

test).C@ <@ versus enterectomy; p < 0.OOi versus control, colectomy (x' test).d@ < 0.05 versus enterectomy; p < 0.OOi versus cobectomy (@ test).

papillary and tubular cancers grew copiously into the lumenwithout Invading the muscubarispropria.

The number of metastases was not affected by operation(Table 2); secondary deposits were almost always located inthe regional lymph nodes, with occasional serosal and pulmonary metastases. No tumors were found outside the intestinaltract. No neoplasms occurred in any rats receiving water,although one granuboma and one inflammatory ulcer werediscovered on suture lines.

DISCUSSION

Jejunoileal bypass not only failed to promote carcinogenesisIn the residual adapted bowel but actually inhibited the devebopment of intestinal tumors. By contrast, jejunoileal resectiontrebled the yield of tumors in the duodenum and remainingjejunum, and subtotal cobectomy caused a similar increase in

MARCH 1980 541

on June 22, 2018. © 1980 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Page 5: Contrasting Effects of Subtotal Enteric Bypass ...cancerres.aacrjournals.org/content/canres/40/3/538.full.pdf · Contrasting Effects of Subtotal Enteric Bypass, Enterectomy, and Colectomy

R. C. N. Wi!!iamson et a!.

DCOLECTOMYIN. 181

A B

CONTROL(N=221 ENTERICBYPASS(N—22l ENTERECTOMY(N=20l

Chart 4. Number and distribution of intestinal tumors after operation. Each dot represents a singletumor. Significance (Mann-Whitney U test) as follows.Duodenojejunal tumors: enterectomy p = 0.025 versus control, p < 0.005 versus colectomy, p < 0.OOiversus enteric bypass; enteric bypass p < 0.05 versuscontrol. Total small bowel tumors: enterectomy p <0.001 versus control, enteric bypass, colectomy.Large bowel tumors: enteric bypass p < 0.OOi versuscontrol, p < 0.002 versus enterectomy. Rectal tumors:colectomy p < 0.05 versus control, enterectomy, p= 0.001 versus enteric bypass.

circuit, the poorer rate of weight gain after bypass presumablyrebates to bacterial contamination of the bong self-emptyingblind loop, with possible interference in the mucosal absorptionof nutrients or release of toxic substances that impair appetiteor hepatic function. Entenocolic mucosa is extremely sensitiveto p.o. food intake, and fasting rapidly reduces intestinal cellproliferation (i 5, 32). Caloric restriction inhibits the development of many different experimental cancers (3, 5, i 7, 30),and the protective effect of high-fiber diets (12) may be partlyattributable to associated reductions in nutrient intake andbody weight (18). Moreover, actuarial records suggest thatmalignant disease is generally more common in the obese (29),and this association may hold true for intestinal cancer (9, i 7,4i).

Modest reduction in body weight could also explain theabsence of the anticipated increase in coborectabtumors after85 to 90% small bowel resection. Alternatively, rapid intestinaltransit may have diminished mucosal exposure to intraluminalcocarcinogens, such as bile acid (2, 28). Lessenenterectomies,which do promote colon carcinogenesis, entail neither substantial long-term reductions in body weight nor severe diarrhea (25, 35). Since compensatory hyperplasia is maximal inthe bowel immediately adjacent to a resected segment (32,38), adaptation of the jejunum (after midenterectomy) on therectum (after colectomy) might have been sufficiently intenseto overcome any overall inhibitory effects.

The high incidence of suture line tumors after entenicor colicresection confirms our previous experimental findings (35, 37).Even after entenic bypass, the few small bowel tumors that didoccur were confined to the region of the jejunoileal anastomosis. Mechanisms might include maximal hyperplasia adjacent to the anastomosis and chronic irritation by nonabsorbableforeign bodies (27).

All 3 operations caused persistent hyperplasia of the ibeabmucosa without promoting cancer at this site. This observationaccords both with our previous experience of chemically induced cancer in rats (35, 37) and with the rarity of adenocan

cinoma of the ibeumin humans, except in Cnohn's disease (7,i 9, 21). The tumors at the site of ibealanastomoses probablyarose from the adjacent jejunum or rectum, since ibeabmucosaappears to possess local protective mechanisms against cancer. No tumors at all were encountered in the excluded intestineafter small bowel bypass, although about one-third of Crohn'scancers affect short-circuited loops of intestine (14).

ACKNOWLEDGMENTS

We thank Peter Davies for statistical advice and Gary James and ElizabethHurst for supplying the Illustrations.

REFERENCES

1. Buchholtz, T. W., Mabamud,D., Ross, J. S., and Malt, R. A. Onset of cellproliferation in the shortened gut: growth after subtotal colectomy. surgery.80: 601-607, 1976.

2. Burkitt, D. P., Walker, A. A. P., and Painter, N. S. Effect of dietary fibre onstools and transit-times, and its role in the causation of disease. Lancet, 2:1408—1412,1972.

3, Carroll, K. K. Experimental evidence of dietary factors and hormone-dependent cancers. Cancer Res., 35: 3374-3383. 1975.

4, Clarke, R. M. Control of intestinal epithebialreplacement: lack of evidencefor a tissue-specific blood-borne factor. Cell Tissue Kinet., 7: 241—250,1974.

5. Clayson, D. B. Nutrition and experimental carclnogenesis: a review. CancerRes., 35: 3292-3300, 1975.

6. Currier, B., Levitt. M. D., Buchwabd,H., and Bond, J. H. Cobonicconservationof carbohydrate in jejuno-ibealbypass patients. Gastroenterology, 62: i 042,1977.

7. DarlIng, R. C.. and Welch, C. E. Tumors of the small intestine. N. EngI. J.Med.. 260: 397-408, 1959.

8. Dowllng, R. H., and Gleeson, M. H. Cell turnover following small bowelresection and by-pass. Digestion, 8: 176—i90, 1973.

9, Dublin, L. I., and Marks. H. Proceedings of the Association of Ufe bnsuranceMedical Directors of America, 24: 47, 1938 [CIted by Tannenbaum, A.(29)].

10. Dudrlck, S. J., Daly, J. M.. Castro, G.. and Akhtar, M. Gastrointestinaladaptation following small bowel bypass for obesity. Ann. Surg., 185: 642—648, 1977.

11. Fenyö,G., Backman, L, and Halbberg, D. Morphological changes of thesmall intestine following jejuno-Ileal shunt In obese subjects. Acta Chir.Scand.,142:154—i59,1976.

12. Fleiszer, D., Murray, D., MacFarlane, J., and Brown, R. A. Protective effectof dietary fibre against chemically induced bowel tumors in rats. Lancet, 2:

CANCERRESEARCHVOL. 40542

on June 22, 2018. © 1980 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Page 6: Contrasting Effects of Subtotal Enteric Bypass ...cancerres.aacrjournals.org/content/canres/40/3/538.full.pdf · Contrasting Effects of Subtotal Enteric Bypass, Enterectomy, and Colectomy

genesis in rats. Cancer Res., 35: 3824—3830,1975.28. Reddy, B. S. Role of bile metabobites in colon carcinogenesis. Animal

models. Cancer, 36: 2401—2406,1975.29. Tannenbaum, A. Relationship of body weight to cancer incidence. Arch.

Pathol., 30: 509-Si 7, 1940.30. Tannenbaum, A., and Silverstone, H. Nutrition and the genesis of tumours.

In: A. W. Raven (ed), Cancer, pp. 306—334.London: Bufterworths Pub.,Inc., 1957.

31 . Tibson,M. D., Michaud, J. T., and Livstone, E. M. Early proliferative activityin the left colon of the rat after partial small-bowel resection. Surg. Forum,27:445-446, 1976.

32. Williamson, R. C. N. Intestinal adaptation. I. Structural, functional andcytokinetic aspects. N. Engb.J. Med., 298: 1393-1 402, i 978.

33, Williamson, R. C. N. Intestinal adaptation. 2. Mechanisms of control. N. Engi.J, Med., 298: 1444-1450, 1978.

34, Williamson, R. C. N., and Bauer, F. L. R. Evidence for an enterotropichormone: compensatory hyperplasia in defunctioned bowel. Br. J. Surg.,65: 736-739, 1978.

35, Williamson, R. C. N., Bauer, F. L. R., Oscarson, J. E. A., Ross, J. S., andMalt, R. A. Promotion of azoxymethane-induced cobonicneoplasia by resection of the proximal small bowel. Cancer Res., 38: 3212—3217, 1978.

36. Williamson, R. C. N., Bauer, F. L. A., Ross, J. S., and Malt, R. A. Proximalenterectomy stimulates distal hyperplasia more than bypass or pancreaticobiliary diversion. Gastroenterobogy, 74: 16—23,1978.

37, Williamson, R. C. N.. Bauer, F. L R., Ross, J. S., Watkins, J. B., and Malt,R. A. Enhanced cobonic carcinogenesis with azoxymethane in rats afterpancreaticobillary diversion to mid small bowel. Gastroenterology, 76:1386—1392,1979.

38. Williamson, R. C. N., Buchholtz, T. W., and Malt, A. A. Humoral stimulationof cell proliferation in small bowel after transection and resection. Gastroenterobogy, 75: 249—254,1978.

39. Wright, H. K., Cleveland, J. C., Tilson, M. D., and Herskovic, T. Morphologyand absorptive capacity of the ileum after ibeostomyIn man. Am. J. Surg.,117: 242—245,1969.

40. Wright, J. K., Poskitt, T., Cleveland, J. C.. and Herskovic, T. The effect oftotal colectomy on morphology and absorptive capacity of the ileum in rats.J, Surg. Res., 9: 301—304,1969.

4i . Wynder, E. L., and Shigematsu, T. Environmental factors in cancer of thecolon and rectum. Cancer, 20: 1520—i561 , 1967.

552—553.1978.13. Gleason,M. H., Cublen,J., and Dowbing.R. H. Intestinalstructureand

functionaftersmallbowelby-passin therat.Clin.Sci.(Oxf.),43: 731-742,1972.

14. Greensteln, A. J., Sachar, D.. Pucibbo,A., Kreel, I.. Gelber,5., Janowitz. H.D.. and Aufses, A., Jr. Cancer in Crohn's disease after diversionary surgery.A report of seven carcinomas occurring in excluded bowel. Am. J. Surg.,135:86—90,1978.

15. Hagemann, R. F., and Stragand, J. J. Fasting and refeeding: cell kineticresponseof jejunum,ileum,andcolon.CellTissueKinet.,10:3—i4, 1977.

16. Hanson, W. R., Osborne, J. W., and Sharp, J. G. Compensation by theresidual intestine after intestinal resection in the rat. I. Influence of amountoftissueremoved.Gastroenterobogy,72:692-700,1977.

17. Heaton, K. W. Cancer of the large bowel: dietary factors. In: S. C. Trueloveand E. Lee (ads.), Topics in Gastroenterobogy,vol. 5, pp. 29—44.Oxford,England: Blackwell, 1977.

18. Heaton, K. W., and Williamson, R. C. N. Dietary fibre and experimental coloncancer. Lancet, 2: 784—785,1978.

19. Hoffman, J. P.. Taft, D. A., Whellis, R. F., and Walker, J. H. Adenocarcinomain regional enterltbsof the small Intestine. Arch. Surg., 112: 606—611, 1977.

20. Mange, H., Bloch, R., schaumbdffel, E., and Riecken, E. 0. Jejunal bypassand reconstitution. In: A. H. Dowling and E. 0. Riecken (eds.), IntestinalAdaptation, pp. 61—67.Stuttgart: Schattauer Verlag, 1974.

21. Nesbit, R. R., Ebbadawi,N. A., Morton, J. H., and Cooper, R. A. Carcinomaof the small bowel. A complication of regional enteritis. Cancer (Phiba.),37:2948—2959.1976.

22. Nygaard, K. Resection of the small intestine in rats. Ill. Morphologicalchangesin theintestinaltract.ActaChir.Scand.,133:233—248,1967.

23. Nundy, S., Malamud, 0., Obertop, H., Sczerban, J., and Malt, R. A. Onsetof cell proliferation in the shortened gut: colonic hyperplasia after ilealresection. Gastroenterobogy, 72: 263—266,1977.

24. Obertop, H., Nundy, S., Malamud, D., and Malt, R. A. Onset of cell proliferatlonin theshortenedgut:rapidhyperplasiaafterjejunalresection.Gastroenterology, 72: 267—270,1977.

25. Oscarson, J. E. A., Veen, H. F., Ross, J. S., and Malt, A. A. blealresectionpotentlates1,2-dimethylhydrazine-inducedcoboniccarcinogenesis.Ann.Surg.,189:503-508. 1979.

26. Payne, J. H., and DeWind, L. T. Surgical treatment of obesity. Am. J. Surg.,118:141—147,1969.

27. Pozharlsski, K. M. The significance of nonspecific injury for colon carcino

543

!ntestina! Bypass, Resection, and Cancer

MARCH1980

on June 22, 2018. © 1980 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Page 7: Contrasting Effects of Subtotal Enteric Bypass ...cancerres.aacrjournals.org/content/canres/40/3/538.full.pdf · Contrasting Effects of Subtotal Enteric Bypass, Enterectomy, and Colectomy

1980;40:538-543. Cancer Res   Robin C. N. Williamson, Frederick L. R. Bauer, Onno T. Terpstra, et al.   Carcinogenesisand Colectomy on Azoxymethane-induced Intestinal Contrasting Effects of Subtotal Enteric Bypass, Enterectomy,

  Updated version

  http://cancerres.aacrjournals.org/content/40/3/538

Access the most recent version of this article at:

   

   

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  [email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/40/3/538To request permission to re-use all or part of this article, use this link

on June 22, 2018. © 1980 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from