Diseases

COLON of the

R E C T U M

Vol . 25 J u l y - A u g u s t 1982 No. 5

Effects of Low-dose Irradiation on Low Anterior

Preoperative Anastomosis

in Dogs*

ROBERT M. SCHAUER, M.D.,? MELVIN P. BUBRICK, M.D.,? DANIEL A. FEENEY, D.V.M.,* GARY R. JOHNSTON, D.V.M.,* ERIC S. ROLESMEYER, M.D.,? ROBERT L. STROM, M.D.,w

CLAUDE R. HITCHCOCK, M.D., PH.D.?

Schauer RM, Bubrick MP, Feeney DA, Johnston GR, Rolfsmeyer ES, Strom RL, Hitchcock CR. Effects o f low-dose preoperative irradiation on low anterior anastomosis in dogs. Dis Colon Rec. turn 1982;25:401-405.

Twenty mongrel dogs underwent preoperative radiation therapy to the colon and rectum using the Nominal Standard Dose Equa- tion to simulate treatment with 2000 rads. Each dog then under- went anterior resection of the rectosigmoid, and reconstruction was randomized into two groups consist ing o f either handsewn or EEA-stapled anastomoses. Anastomoses were examined digitally and radiographically on the day of surgery and on the seventh postoperative day. There were three radiographic leaks among the ten dogs having the handsewn anastomoses and one radio- graphic leak among the ten dogs having the EEA-stapled anas- tomoses . There was one c l i n i c a l l y s i g n i f i c a n t leak w h i c h occurred in a dog having an EEA.stapled anastomosis and was associated with peritonitis and death. The overall leak rate was 30 per cent among dogs having handsewn anastomoses and 20 per cent among dogs with stapled anastomoses. The data suggest that an anterior resection in low colorectal anastomosis can be done safely after low-dose radiation using either handsewn or stapling techniques. [Key words: Anastomosis; Carcinoma; Colon; Radia- tion therapy; Rectum; Stapling device]

PREOPERATIVE IRRADIATION has been used widely in the treatment of colorectal carcinoma. Since the first case description by Symonds in 1914,1 numerous

* Received for pubfication July 24, 1981. Address reprint requests to Dr. Bubrick: Hennepin County

Medical Center, 701 Park Avenue South, Minneapolis, Minnesota 55415.

From the Departments of ?Surgery and w Hennepin County Medical Center,

the Minneapolis Medical Research Foundation, and the :~Department of SmaU Animal

Clinical Sciences, University of Minnesota,

College of Veterinary Medicine, Minneapolis, Minnesota

treatment protocols have been established, z-7 Doses have ranged from a "low dose" of 2000 rads to the "high dose" techniques of 4500 to 5500 rads, with no general agreement as to which regimen is optimal. Five-year survival statistics have suggested im- proved survival with either dosage range. 3"5"~

The current operative treatment of rectal and rec- tosigmoid carcinoma involves a trend toward rectal sphincter-sparing resection, ~ with increasing use of low anterior anastomotic techniques. In the past, surgery following irradiation therapy to the rectum or rec tos igmoid has been l imited to ab- dominoperineal resection, presumably to avoid the risks of performing a low anastomosis in an irradiated field. 1~ Recent reports, however, suggest that small- bowel as well as colonic anastomoses can be per- formed safely following irradiation therapy utilizing both handsewn and stapling techniques.3'5" r, 11

0012-3706/82/0700/0401/$01.05 �9 American Society of Colon and Rectal Surgeons

401

Dis. Col. & Rect. 402 S C H A U E R , E T AL. July-August 1982

With this in mind, an animal study was designed (1) to assess the safety of pe r fo rming an ext raper i toneal colorectal anastomosis following low-dose irradiat ion and (2) to compare the efficacy of handsewn and sta- pling techniques u n d e r these circumstances.

Methods and Materials

Twenty mongre l dogs, weighing 40 to 54 pounds, u nde rwe n t preoperat ive irradiat ion to the rectum, using a 250 kVp, 15 mA orthovoltage x-ray unit. Sur- face source distance (SSD) was 50 cm. T h e half-value

layer o f the x-ray beam was 2.6 mm of coppe r (mm Cu). T h e radiat ion dose was measu red utilizing a " th imble" rad ia t ion c h a m b e r (Vic toreen R m e t e r Model 500 and thimble chamber Model 550-6A, Vic- toreen, Inc., Cleveland, Ohio) placed in the rec tum with the sensitive volume in the cen te r of the x-ray beam. Plastic tubing was placed in the rec tum to pro- tect the R me te r f rom damage due to peristalsis or feces. While radiat ion was adminis tered, the accumu- lated dose was moni to red for each portal to allow ac- curate dosimetry.

Ventral dorsal right-to-left and left- to-right oblique portals were used to adminis ter 915 to 920 rads to a 4-cm length of the rec tum in one dose. T h e x-ray beam was cen te red along a line pe rpend icu la r to the long axis of the body, 7 cm anter ior to the anus. Each portal exposed an area 4 x 6 cm on the surface o f each dog's skin on the side o f the x-ray tube, and each portal was used to adminis ter one-ha l f o f the above- ment ioned dep th dose. No surface overlap was used. T h e angle (off vertical) fo r each por ta l was deter- mined specifically for each animal d e p e n d i n g on body weight and shape.

Calculations were p lanned to simulate a 2000 rad, 12-day, ten-fract ion low-dose protocol for p reopera- tive radiat ion therapy using the nominal s tandard dose equat ion (NSD) as follows: 12"1a NSD =dose (to- tal) x (days on t rea tment) --11 x (number o f frac- tions) -'z4. According to the NSD equation, a total dose o f 875.6 rads was to be delivered by two portals (437.8 rads per portal). A 5 per cent increase in dep th dose was adminis tered to compensate for absorpt ion of the plastic tubing on the dorsal side of the R meter.

After a three-week resting per iod, the dogs were randomly assigned to one of two operat ive groups. All dogs unde rwen t mechanical bowel cleansing, and each dog was given 6 g carbenicillin (Geopen | in divided doses over a f o u r - h o u r per ioperat ive period.

Surgery was done u n d e r aseptic technique t h roug h a vertical midline incision, and anesthesia was by in- t ravenous administrat ion o f sodium thiopental . At o p e r a t i o n , the s igmoid was d issec ted f r ee o f its

peri toneal at tachments. T h e rectum was mobilized anter ior ly and poster ior ly , and both lateral rectal stalks were divided. A 5-to-10cm segment of rec- tosigmid was then resected, and end- to -end anas- tomosis was done between the proximal sigmoid and ext raper i tonea l rec tum using one of the two anas- tomotic techniques.

G ro u p I dogs u n d e r w e n t a handsewn inver t ing two-layer anastomosis. T h e inner layer consisted of a cont inuous full-thickness 3-0 chromic catgut suture utilizing the Connell stitch anter ior ly to secure cir- cumferent ia l mucosal inversion. T h e outer layer con- sisted of i n t e r rup ted Lember t - type sutures of 4-0 silk. Group II dogs u n d e r w e n t a stapled anastomosis using the EEA s tapl ing device ( A u t o s u t u r e C o m p a n y , Division of United States Surgical Corp., Stanford, Connecticut) . Proximal and distal bowel loops were secured to the stapling device with 2-0 Prolene con- t inuous full-thickness suture, and a complete inner ring of resected bowel wall was obta ined f rom both colonic and rectal segments af ter the stapling device had been fired. Each anastomosis was evaluated by digi tal e x a m i n a t i o n on the o p e r a t i n g table. Im- mediate postoperat ive ba r ium-enema examinat ions were p e r f o r m e d using a volume sufficient to outl ine the rec tum, proximal colon, and anastomotic site, as described by Goligher et al. 14 A second ba r ium-enema examina t ion was p e r f o r m e d on the seventh post- operat ive day and then once weekly if a leak was demonst ra ted .

T h e level of anastomosis was eva lua ted by digital examinat ion at the time of the ba r ium-enema exami- nation on the seventh postoperat ive day. This level was measured and recorded. T h e abdominal wound was also inspected at this time. All dogs were killed o n e to two m o n t h s p o s t o p e r a t i v e l y a n d we re e x a m i n e d f o r a n a s t o m o t i c d e h i s c e n c e , p e r i a n a - stomotic abscess, and intraper i toneal abscess.

Results

Effects of Irradiation: T h e irradiat ion was toler- ated well by all dogs. Some evidence o f skin damage and alopecia was noted in all 20 animals. Grossly, no radiation changes were seen at the level o f the rec- tosigmoid, proximal colon, or small bowel. Histologi- cally, all resected specimens of small bowel were found to have some changes consistent with irradia- tion effect. Some were more severe than others. T h e spec t rum of findings noted included (1) diminished mucosal height , (2) abnormal mucosal archi tecture , (3) plasma cell mucosal infiltration, (4) ectatic mucosal and submucosal vessels, (5) increase in mitoses in the mucosa, (6) thickening o f the submucosa, (7) rem-

Volume 25 Nt, mber 5 L O W - D O S E I R R A D I A T I O N 403

TABLE l. Site of ,4nastomosis

4 cm 4.5 cm 5 cm 5.5 cm 6 cm 6,5 cm 7 cm

Group I (sutured) 2 I Group II (stapled) 1 1 2

TOTAL ] 3 3

3 3 1 4 1 1

7 3 2 1

nants o f s loughed mucosa, and (8) submucosal in- f l ammato ry infiltration.

I~vel of Anastomosis: All anas tomoses were per - f o r m e d between 4 and 7 cm f r o m the den ta te line. T h e r e were no significant d i f ferences in the level o f a n a s t o m o s i s b e t w e e n the h a n d s e w n a n d s t a p l e d g roups (Table 1).

Anastomotic Leaks: T h r e e leaks were ident i f ied radiographical ly a m o n g the ten G r o u p I dogs (30 pe r cent) and one leak a m o n g the ten G r o u p I I dogs (10 pe r cent) (Table 2). These di f ferences are not statisti- cally significant. None o f the leaks were demons t r ab l e in the i m m e d i a t e pos topera t ive study, and all o f the leaks were pos ter ior in position. In all four cases, the leaks were clinically insignificant and closed spon- taneously in seven to 28 days.

T h e r e was one clinically significant leak which ap- pea r ed in one o f the dogs in the s tapled group. This leak was not radiographical ly ev iden t on e i ther o f the two ba r ium studies but became a p p a r e n t when the dog deve loped peri tonit is and subsequent ly died on the 16th pos topera t ive day. At autopsy, the anasto- motic site was intact, and the leak was found in the colon p rox imal to the anastomosis .

Complications: T h e r e were four wound infect ions (20 per cent), and one dog had a fascial dehiscence which requ i red reopera t ion (Table 3). T h e r e were no addi t ional deaths for an overall mortal i ty o f 5 per cent. Autopsy evaluat ion o f the anas tomoses revealed one p e r i a n a s t o m o t i c h e m a t o m a , o n e a n a s t o m o t i c s tr icture, and the previous ly descr ibed p reanas to - motic leak with peri tonit is (Table 4). T h e r e were no statistically s ignif icant d i f f e rences in mor ta l i ty o r morbid i ty between the two groups.

Discussion

T h e NSD equat ion 12"13"15 was utilized to simulate the effects o f a two-week p reope ra t i ve i r radia t ion the rapy schedule (2000 rads) at a rate of 200 rads per fract ion and five fractions pe r week. T h e NSD equa- tion was initially r epo r t ed as the nomina l "single" dose equa t ion ~ and was des igned to ex t rapo la te the the rapeu t ic effect o f various dose / t ime fract ionat ion schedules back to an equivalent single dose. T h e NSD c o n c e p t is a lso u s e f u l in c o m p a r i n g v a r i o u s rad io therapeu t ic schedules and in fo rmula t ing treat- m e n t s c h e d u l e s b a s e d on k n o w n n o r m a l t i ssue tolerance)2,13

T h e NSD concep t does have shor tcomings. These include a lack o f compensa t ion for the i r radia ted vol- ume, an a t endency to underes t imate t r e a t m e n t t ime effect for low l inear ene rgy t r ans f e r (LET) irradia- tion, TM and a fai lure to consider changes in vascular condit ions which may occur dur ing the course o f a fractional the rapy p r o g r a m ) 3 An addit ional criticism is that the NSD equat ion was based on data der ived f rom a m i n i m u m o f four fractions, and this lends u n c e r t a i n t y to the e x t r a p o l a t i o n back to a single equivalent dose. '~''5 It was o u r feeling, however, tha t despi te its drawbacks, using the NSD was far super io r to a t t empt ing a f rac t ionated therapy reg imen on ex- pe r imenta l animals. Logistic and technical considera- tions made the lat ter al ternative epecially undesirable .

T h e t iming of surgery a f te r radiat ion the rapy is o f considerable impor tance . Radiat ion changes in the bowel may be e i ther acute or chronic. Crowley et al. ' z

found that the acute changes seen one to three days af ter i r radia t ion consisted o f loss o f epithelial ceils

TABLE 2. Anastomotic Leaks

Group I (sutured)

Number of Leaks/ Number of Dogs (Per Cent)

Group II (stapled)

Number of Leaks/ Number of Dogs (Per Cent)

Radiographic leaks Clinical leaks

3/10 (30) 0/10 (0)

l l lO (lO) 1/lO (lO)

TOTAL 3 / 1 0 (30) 2 / 1 0 (20)

Dis, Col, & Rect. 404 SCHAUER, ET AL. July-August 1982

TABLE 3. Complications and Deaths

Number/Number of Dogs Per Cent

Wound infections 4/20 20 Wound dehiscence 1/20 5 Colon perforation/peritonitis 1/20 5 Total complications 6/20 30 Deaths 1/20 5

lining the villi, reduction in the height of the villi, disappearance of mitotic figures, and edema of the submucosa and muscularis. They noted a marked re- covery by the ninth day after irradiation and found that the bowel appeared quite normal by the 22nd day except for occasional crypt abscesses and inflam- matory cells in the submucosa. Chronic changes have been described by Black et aL ~8 and include mucosal ulcerat ion, deep displacement o f epithelial cells within the wall of the rectum (colitis cystica pro- funda), atypical epithelial regeneration, fibrosis, and vascular sclerosis. The severity of these lesions de- pends upon the radiation dose, the number of frac- tions, and the time interval from the irradiation. The use of fewer fractions for a given dose produces more chronic damage. Also, bowel injury may look more severe at 12 months than at four months a f te r irradiation.

The ideal time for surgery should be after the acute mucosal lesions have healed but before the chronic changes have begun to develop. We chose a three- week interval with this in mind. Histologic findings in our surgical specimens at three weeks were consistent with those seen by Crowley et al. lr at 22 days. Mucosal regeneration was complete and irradiation changes were minimal.

Fear of anastomotic dehiscence has always been a major factor influencing the choice of operation for patients with colorectal cancer. The mortality rate for patients with anastomotic dehiscence can be as high as 33 per cent compared with 2.6 per cent among pa-

TABkE 4. Autopsy Findings

Number/Number of Dogs Per Cent

No leak/no abscess/ patent anastomosis 17/20 85

Perianastomotic hematoma 1/20 5 Anastomotic stricture 1/20 5 Colonic perforation/

peritonitis 1/20 5

tients with intact anastomoses. This is especially im- portant when considering low colorectal anastomosis because these anastomoses are more prone to break- down than are in t raper i tonea l colocolic anasto- moses. ~~ The main factors implicated in the higher failure rate of low colorectal anastomoses are poor blood supply and exclusion of the anastomosis from the peritoneal cavity. 14 In addition to these factors, the administration of radiation therapy is known to cause a further increase in the risk of dehiscence? ~

Anastomotic dehiscence rates for anterior colonic resection vary widely in the literature from I per cent 7 to 51 per centt4; this wide range may reflect the diligence with which these breakdowns are sought. Gol igher et al . 14 p e r f o r m e d low-volume bar ium enema examinations in patients I4 days after anas- tomosis and demonstrated the difference between clinically significant leaks and radiographic leaks that heal without clinical consequence. Radiographic ani- mal studies from our laboratory by Bubrick et al. ~

showed that leaks after colorectal anastomoses were demonstrable between the third and seventh day. Based on these data, we chose to perform barium- enema examinations on the seventh postoperative day. Although radiographic leaks may not be signifi- cant clinically, they do serve as an objective parameter with which to make comparative statements about dif- ferent anastomotic techniques and different clinical circumstances. In the current study, the radiographic leak rates of 30 per cent in the handsewn group and 10 per cent in the stapled group are actually lower than the leak rates seen in the nonirradiated control dogs from the previous study by Bubrick et al. ~9

One clinically significant leak appeared in this study in addition to the radiographically def ined leaks. This leak occurred in a stapled anastomosis and involved an anterior perforat ion of the colon just proximal to the anastmosis. The radiographically de- fined leaks were all posterior in position. The anterior leakage is more serious, and the anterior colon may be more vulnerable to perforation in this setting. In- flation studies performed by Crowley et al. ~ demon- strated that rupture of irradiated anastomoses occurs along the antimesenteric border of the colon, whereas rupture of nonirradiated colon occurs at the mesen- teric border. He also demonstrated a rupture away from the suture line in almost all cases, presumably following the principle of Laplace's law (T --- P x R), where wall tension (T) is directly related to the radius of the lumen (R).

The most striking finding of this study, however, is the fact that the technical results were unchanged when extraperi toneal colorectal anastomoses were made after low-dose irradiation. The leak rates and mortality rates both compare favorably with data or

Volume 25 Number 5 LOW-DOSE IRRADIATION 4 0 5

n o n i r r a d i a t e d dogs as well as p u b l i s h e d c l in i ca l

s tudies . T h e l o n g - t e r m safety o f these a n a s t o m o s e s a f t e r i r r a d i a t i o n t h e r a p y is p r e s u m e d to be accepta-

ble, bu t the possibi l i ty o f late s t r i c t u r i n g or s tenosis

may n e e d to be r u l e d o u t by f u r t h e r s tudies .

Conclusion

T w e n t y dogs u n d e r w e n t p r e o p e r a t i v e r a d i a t i o n

t h e r a p y to the r e c t u m us ing the NSD e q u a t i o n to s imula te t r e a t m e n t with 2000 rads. All dogs u n d e r - w e n t a n t e r i o r r e s e c t i o n wi th low c o l o r e c t a l ana s - tomosis u s ing e i the r h a n d s e w n or s t ap l ing t echn iques .

T h e overal l leak rate was 30 pe r c e n t a m o n g the dogs h a v i n g h a n d s e w n a n a s t o m o s e s a n d 20 p e r c e n t a m o n g the dogs wi th s tap led anas tomoses . T h e r e was o n e dea th in the series. T h e data sugges t that a n t e r i o r

resec t ion a n d a n d low colorectal anas tomos i s can be d o n e safely a f t e r low-dose r a d i a t i o n a n d tha t e i the r h a n d s e w n o r s t ap l ing t e c h n i q u e s may be used.

References

1. Symonds CJ. Cancer of rectum: excision after application of radium. Proc R Soc Med 1913;7:152.

2. GreenJP. Rectal cancer adjuvant radiation therapy. In: What's new in cancer care? 1979;3:1.

3. Higgins GA Jr, Conn JH, Jordan Ph Jr, Humphrey EW.,Ros- w~tz B, Keehn RJ. Preoperative radiotherapy for colorectal cancer. Ann Surg 1975;181:624-31.

4. Romsdahl MM, Withers HR. Radiotherapy combined with curative surgery: its use as therapy for carcinoma of the sigmoid colon and rectum. Arch Surg 1978; 113:446-53.

5. Roswit B, Higgins GA Jr, Keehn RJ. Preoperative irradiation for carcinoma of the rectum and rectosigmoid colon: report of a national Veterans Administration randomized study. Cancer 1975;35:1597-1602.

6. Stearns MWJr, Deddish MR, Quan SH. Preoperative roentgen therapy for cancer of the rectum. Surg Gynecol Obstet 1959; 109:225-9.

7. Stevens KR Jr, Fletcher WS, Allen CV. Anterior resection and primary anastomosis following high dose preoperative ir- radiation for adenocarcinoma of the recto-siglnoid. Cancer 1978;41:2065-71.

8. Learning RH, Stearns MW Jr, Deddish MR. Preoperative ir- radiation in rectal carcinoma. Radiology 1961;77:257-63.

9. Slanetz CA Jr, Herter FP, Grinnell RS. Anterior resection ver- sus abdominoperineal resection for cancer of the rectum and rectosigmoid: an analysis of 524 cases. Am J Surg 1972; 123:110-7.

10. Schrock TR, Deveney CW, Dunphy J. Factors contributing to leakage of colonic anastomoses. Ann Surg 1973; 177:513-8.

11. Photopnlos GJ, Delgado G, Fowler WC Jr, Walton LA. Intesti- nal anastomoses after radiation therapy by surgical stapling instruments. Obstet Gynecol 1979;54:515-8.

12. Ellis F. Fractionation in radiotherapy. Mod Trends Radio- therapy 1967; 1:34-51.

13. Ellis F. The relationship of biological effect to dose-time- fractionation factor in radiotherapy. Curr Top Radiat Res 1968;357-97.

14. Goligher JC, Graham NG, de Dombal FT. Anastomotic dehis- cence after anterior resection of rectum and sigmoid. Br J Surg 1970;57:109-18.

15. Orton CG, Ellis F. A simplification in the use of the NSD con- cept in practical radiotherapy. BrJ Radiol 1973;46:529-37.

16. Fletcher GH, ed. Textbook of radiotherapy. Philadelphia: Lea & Febiger, 1973;121-51.

17. Crowley LG, Anders CJ, Nelsen T, Bagshaw M. Eftect of radi- ation on canine intestinal anastomoses. Arch Surg 1968;96:423-8.

18. Black WC, Gomez LS, Yuhas JM, Kligerman MM. Quantita- tion of the late effects of x-radiation on the large intestine. Cancer 1980;45:444-51.

19. B~lbrick MP, Lundeen JW, Hitchcock CR. A comparative ra- diographic study of low anterior colon anastomoses in dogs. Surgery 1981;89:454-9.