Effects of Abdominal Surgery on the Development of Radiation Enteropathy

73:1093-1097, 1977 Copyright © 1977 by the American Gastroenterological Association

Vol. 73, No. 5

GASTROENTEROLOGY

Printed in U.S A.

EFFECTS OF ABDOMINAL SURGERY ON THE DEVELOPMENT OF RADIATION ENTEROPATHY THOMAS LolumcE, D. 0., DoNALD BAXTER, M.D., AND JoHN BALINT, M.B. , F.R.C.P. Departments of Medicine and Radiology, Albany Medical Center Hospital and Albany Medical College, Albany, New Y ork

A controlled, retrospective study was undertaken to ascertain whether or not abdominal surgery at any time before radiation therapy might result in an -increased incidence ofradition enteropathy. One hundred and five patients in three groups were studied; 24 had radiation enteropathy which developed an average of 12.5 months after radiation therapy, and 28 were matched controls without this complication. The records of a further group of 53 unmatched controls, who survived more than 2 years after irradiation were also examined. In the group with radiation enteropathy, 18 (75%) had a history of prior surgery, whereas only 3 (10.7%) of the matched controls had been operated before irradiation (x 2 = 3.52, P < .005). The incidence of prior surgery in the unmatched control group was 13.2%. These data indicate that abdominal surgery before radiotherapy does increase the risk of subsequent development of radiation enteropathy. Damage to the gastrointestinal tract after radiation therapy for intraabdominal malignancy is well documented and may occur in any portion of the gastrointestinal tract. Many factors have been postulated as predisposing to this injury. It has been said that radiotherapy in patients with previous abdominal surgery results in a higher incidence of radiation enteropathy. 1- 3 A retrospective controlled study was undertaken to clarify whether or not this is an accurate statement. Methods Hospital records of all 320 patients receiving radiotherapy for stage II or greater abdominal or pelvic malignancy over an 8-year period from 1968 to 1975 inclusive at the Albany Medical Center Hospital were examined. From these, all patients in whom radiation injury to the bowel occurred were selected. In addition, any patients who had radiation injury to the bowel before 1968 but were continuing treatment during the period 1968 to 1975 were included. Patients were then graded according to the method of Koffmeier and Grayt as follows: Grade I: minimal objective mucosal changes occurring shortly after treatment and having no correlation with dose or subsequent fibrosis. Grade II: necrosis, ulceration, protracted pain, bleeding and moderate stenosis. Grade III: massive hemorrhage or stenosis requiring surgical intervention; fistula formation. Patients with grade I disease were selectively excluded so as to limit the study to more serious radiation injuries. Data concerning age, sex, underlying disease process, date, dosage and type of radiotherapy, date enteritis developed, type and dates of abdominal surgery performed before radiotherapy, Received February 11, 1977. Accepted June 2, 1977. Address requests for reprints to: John A. Balint, M.B., F.R.C.P. , Division of Gastroenterology, Albany Medical College, Albany, New York 12208. This study was supported in part by the Hellman Family Foundation.

complications of the enteritis, and eventual outcome were then gleaned from the records. Matched controls with no evidence of radiation enteropathy were selected by the Albany Medical Center Tumor Registry for age, sex, underlying disease process, date, dosage, and mode and type of radiation therapy. Where more than one suitable control was found as a match for any particular patient with radiation enteropathy, both were included. For 2 patients (E. F ., A. J .), no suitable control was available. After the selection process, hospital records of controls were reviewed to see whether these patients had had abdominal surgery before their radiation therapy. Two hundred and fifty-five of the original320 patient charts were reexamined to select from them a group of unmatched controls with prolonged survival of greater than 2 years after radiotherapy in an attempt to ascertain whether such prolonged survival would afford an opportunity for an increased incidence for the development of radiation enteropathy. All 53 patients who survived 2 years or more were included in this group. All patients included in the study received their radiation dosages in a standard fashion with standard simulator and portal films before radiotherapy for monitoring purposes. Results were then analyzed statistically using the Student t-test for independent variables and the x2 test. Data are presented as mean ± standard error.

Results Data concerning matched control and experimental groups are summarized in tables 1 and 2, respectively. The matched control group consisted of 27 females and 1 male. The experimental group was comprised of 23 females and 1 male. Ages in the matched control group varied from 30 to 75 years with a mean of60.5 years. The experimental group varied in age from 41 to 80 years with a mean of 61 years. There was no significant statistical difference in this respect between these two groups (P > 0.5). External radiation dosages were similar in the two

1093

1094 TABLE

Patient

Vol . 73, No.5

LOIUDICE ET AL.

Age

Primary tumor

Sex

1. Patient profiles (contro ls) Total radiation (dose and type)

Previous abdominal surgery

Survival after radiation

None None None None None None None None None None None None None None None None None None None None Small bowel resection Hysterectomy None Hysterectomy None None None None

lyr" lyr 6 yrs, 6 mob 3yr" 13 yr 5 yt' 3 yrb 1 yr, 6 mo b 3 yrb 2 yrb 8yr 2yr 5yr 6 mob 1 yrb 6 mob 2 yrb 1 yrb 1 yrb 2 mo b 6 mob 6 mob 1 yr 1 yrb 3 yrb 3 mob 3 yr 5 yr

yr

K. G. D. P. B. F . M. M. M. M. M.A. C. F . C. H . M. P. M.H. L. P . R. S. E. B. E. H . R. C. P . G. B. G. B. P . B. B. L. K. V.E. C. V. R. F . R. H . E. F . M.L. L. S. J. R.

61 60 57 61 58 59 56 61 60 64 60 51 50 66 61 61 68 74 59 75 64 66 46 62 72 67 66 30

F F F F F F F F F F F F F F F F F F F F F F F F F F F M

Squamous cell carcinoma cervix Squamous cell carcinoma cervix Squamous cell carcinoma cervix Squamous cell carcinoma cervix Squamous cell carcinoma cervix Squamous cell carcinoma cervix Squamous cell carcinoma cervix Squamous cell carcinoma cervix Squamous cell carcinoma cervix Squamous cell carcinoma cervix Squamous cell carcinoma cervix Squamous cell carcinoma cervix Squamous cell carcinoma cervix Squamous cell carcinoma cervix Adenocarcinoma uterus Adenocarcinoma uterus Adenocarcinoma uterus Adenocarcinoma cervix Adenocarcinoma cervix Adenocarcinoma cervix Adenocarcinoma ovary Adenocarcinoma ovary Adenocarcinoma ovary Adenocarcinoma ovary Anaplastic carcinoma, cervix Anaplastic carcinoma, uterus Uterine sarcoma Seminoma testes

4830 rads" 6000 radsa 6000 radsa + 5150 rads < 5205 rads" + 3500 mg-hy
a 6°Co b

c d

e

teletherapy. Expired. Radium implant. Cesium implant. 4 Mev, supervoltage irradiation.

groups (P > 0.1) and ranged from 2700 to 6000 rads (mean 4619 ± 174) in the matched control group, and from 2500 to 8750 rads (5128 ± 266) in the experimental group. Thirteen patients in the experimental group and 14 in the control group, in addition to external radiation, received either a radium (7 experimentals, 9 controls) or a cesium (6 experimentals, 5 controls) implant. Doses of radium in experimental patients ranged from 3500 to 5150 rads (mean 4901) and that in controls from 3600 to 5250 rads (mean 4818). Doses of cesium in experimental patients ranged from 3500 to 5556 mg-hr (mean 4011) and that in controls from 3500 to 5750 mg-hr (mean 4050). Patients receiving both external and internal radiation sources, with an internal radiation dose > 5000 rads or mg-hr, received external radiation only to the paramedian areas. No significant difference for total tissue dose or radiation dose owing to implants was noted (P > 0.1). The predominant underlying disease in both groups was squamous cell carcinoma of the uterine cervix (50% of controls; 41.7% of experimentals), followed in frequency by adenocarcinoma of the uterus (10.7% of controls; 20.8% of experimentals), adenocarcinoma of the uterine cervix (10.7% of controls; 12.5% of experimentals), adenocarcinoma of the ovary (14.3 % of controls;

4.2% of experimentals), ·anaplastic carcinoma of the uterus (3 .6% of controls; 4·.2 % ofexperimentals), uterine sarcoma (3.6% of controls; 4.2% of experimentals), testicular seminoma (3. 6% of controls; 4.2% of experimentals), and basal cell carcinoma of the uterine cervix (0% controls; 4.2% of experimentals). Three patients in the matched control group (10.7%) had had abdominal surgery varying from 1 to 5 years before radiation therapy. Eighteen patients in the experimental group (75%). had had abdominal surgery from 2 months to 43 years 'before radiotherapy. The association of previous abdominal surgery with the development of radiation enteropathy was found to be highly significant (x 2 = 35.2, P < 0.005). Of the 24 patients noted to have radiation enteropathy, 16 were grade III disease and 8 were grade II disease. Six of the 8 patients (75%) with grade II disease and 12 of 16 patients (75% with grade III disease had had previous abdominal surgery before radiation. There was no difference between these two groups statistically. Radiation injury in the experimental group affected virtually every level of the small bowel and colon. Eighteen patients (75%) had colonic injury, 20 (83%) had injury to the small bowel, and 14 (58%) had injury to both the large and small bowel. In 17 patients (71 %)

TABLE

Patient

Age

1095

SURGERY AND RADIATION ENTERITIS

November 1977

Sex

Primary tumor

2. Patient profiles (radiation enteropathy) Total radiation (dose and type)

Previous abdominal surgery

4400 rads" + 5082 radsb 5000 rads"

Appendectomy, cophorectomy Appendectomy

6000 rads"

Appendectomy, tectomy None

Portion of bowel involved by enteropathy

Survival after radiation

yr

H.M.

68 F

H. D.

60 F

D.P.

60 F

E. R.

55 F

A.M.

59 F

E. W.

63 F

D. L.

62 F

R. D.

51 F

S. S.

67 F

E. F.

70 F

A. J.

46 F

P. S.

42 F

Squamous cell carcinoma cervix Squamous cell carcinoma cervix Squamour cell carcinoma cervix Squamous cell carcinoma cervix Squamous cell carcinoma cervix Squamous cell carcinoma cervix Squamous cell carcinoma cervix Squamous cell carcinoma cervix Squamous cell carcinoma cervix Squamous cell carcinoma cervix, adenocarcinoma uterus Basal cell carcinoma cervix Adenocarcinoma uterus

J. R.

59 F

Adenocarcinoma uterus

R. E.

61 F

Adenocarcinoma uterus

B. L.

77 F

Adenocarcinoma uterus

J. M.

79 F

Adenocarcinoma cervix

M. C.

41 F

Adenocarcinoma cervix

E. F.

80 F

Adenocarcinoma cervix

H. N.

64 F

Adenocarcinoma ovary

I. s. G. G.

56 F 72 F

M.P.

70 F

F. L.

57 F

Adenocarcinoma ovary Anaplastic carcinoma cervix Anaplastic carcinoma uterus Uterine sarcoma

54 M

Seminoma testes

J.

w.

6000 rads" + 5150 radsb 4160 rads" + 5150 radsb 7000 rads"

cholecys-

None Appendectomy

3420 rads" + 5150 radsb 5200 radsd + 3500 mg-hr• 6000 radsd

Appendectomy

5000 rads"

Exploratory laparotomy

6000 radsd + 3500 mg-hr• 5000 rads"

None

8750 radsd + 3000 mg-hr• 5000 rads d + 4006 mg-hr• 5000 rads" + 4000 mg-hr• 6100 rads" 3060 rads" + 4768 radsb 4500 rads" + 4900 radsb 4200 rads• 2500 rads<~ 4750 rads" + 5556 mg-hr• 6120 rads• 5400 radsd + 3500 radsb 4500 rads•

None

Gastric resection

Hysterectomy, appendectomy Hysterectomy appendectomy Appendectomy None

Sigmoid colon

8 yr

Jejunum, ileum, sigmoid colon Jejunum, ileum

34 yr

Sigmoid colon

6 yr

Jejunum, sigmoid colon Jejunum, sigmoid colon Jejunum, transverse colon Jejunum, colon

7 yrc

Ileum, sigmoid and transverse colon Sigmoid colon and jejunum, ileum

4 yrc

Jejunum, ileium, sigmoid colon Ileum, sigmoid and ascending colon Ileum, sigmoid colon Ileum , duodenum, sigmoid colon Jejunum

2 yr, 6 mo

1 yr<

2 yrc 1 yrc

2 yr

2 yrc

7 yr

1 yr<

3 yr '" 1 yr, 6 mo< 1 yr<

None

Sigmoid colon, jejunum Sigmoid colon

1 yr'"

Nephrectomy

Sigmoid colon

3 yr

Hysterectomy, appendectomy Appendectomy Appendectomy, subtotal hysterectomy Hysterectomy

Ileum, jejunum

4 yr

Ileum, jejunum Ileum, JeJunum, sigmoid colon Sigmoid colon, ileum Ileum, jejunum

2 yr 8 yr

Ileum, jejunum

28 yr

Vagotomy, pyloroplasty

Hysterectomy, cholecystectomy Radical pelvic node dissection

7 yr

1 yr, 6 mo

a 6°Co

teletherapy. Radium implant. ' Expired. d 4 MeV, super voltage irradiation ' Cesium implant. b

of this group, surgical resection or bypass of some portion of the bowel was necessary because of radiation injury. Survival in the matched control group after radiotherapy varied from 2 months to 13 years (mean 2.8 ± 0.55 years). That of the experimental group ranged from

1 to 34 years. With the exception of 2 extraordinary patients who survived 34 and 28 years, respectively, after radiation therapy, the mean survival was 3.4 ± 0.54 years. With this exception, no significant difference in survival between the two groups was noted (P > 0.4). The number of patients who expired in the matched

1096

LOIUDICE ET AL .

control group during the course of the study exceeded that of the experimental group. Their average length of survival, however, exceeded the average length oftime necessary for the development of radiation enteropathy, i.e., 12.5 months. Eleven patients in the matched control group survived 1 year or less. Because of this apparent discrepancy in survival between the experimental group and matched control group, the unmatched control group of 53 patients who survived longer than 2 years was studied. Mean survival in this group was 3.6 years. Of the 53 long survivors, 1 had carcinoma of the ovary and had had abdominal surgery before radiotherapy, 3 had carcinoma of the uterus and none had prior abdominal surgery, and 49 had carcinoma of the cervix, 6 with prior abdominal surgery. The mean external radiation dose for this entire group was 4784 ± 220 rads. Patients in this group, with either endometrial or cervical carcinoma, were all treated, in addition, with either radium (28 patients) or cesium (25 patients) implants. Average radium dosage was 4653 rads and average cesium dosage was 4000 mg-hr. Only 7 of 53 patients (13.2%), then, with no significant difference in radiation dose, mode of radiation administration, or type of disease from either the matched control or experimental groups, had abdominal surgery before radiotherapy. No further cases of radiation enteropathy were found among these long survivors and no significant difference was noted between this group and the matched control group with respect to the incidence of abdominal surgery before radiation (P > 0.1). Discussion Radiation therapy is the treatment of choice for invasive cancer of many abdominal viscera in most institutions. With supervoltage irradiation, the over-all5-year survival, particularly for gynecological cancer, has significantly increased. 1•5 However, the incidence of major complications attributable directly to radiation therapy has also risen.""'7 The small intestine, in particular, is susceptible to radiation injury. Severe radiation damage to this portion of the bowel may be responsible for more acute and long term suffering and risk of mortality than damage to any other abominopelvic organ.&-lo It has the least tolerance of radiation, especially in the duodenal segment, of any of the gastrointestinal viscera. 11 Were it not for the dynamic and perpetual free movement of the loops, small bowel injury would occur in nearly all patients treated with a full course of radiotherapy for abdominopelvic neoplasia. When loops are fixed in position by abdominal adhesions, the same segments of bowel may be exposed to very high doses of irradiation and the chances for development of radiation enteropathy may, therefore, rise markedly. The colon and rectum are not as susceptible to radiation injury. Indeed, the rectum is probably the most resistant of the gastrointestinal viscera to the development of severe radiation enteropathy. Severe complications, however, do occur in this region, usually consisting of shortening, ulceration, stenosis, and fistulization. The incidence of

Vol . 73 , No.5

such complications is higher in those patients who receive as part of their therapy either a radium or cesium implant. These changes, as in injury to the small bowel, may be progressive and irreversible. The latent period between completion of radiotherapy and the development of symptoms of chronic intestinal damage varies from a few months to many years. In our study, this latent period averaged 12.5 months (fig. 1). Prolonged survival after radiotherapy does not seem to allow for more cases to develop. Major factors thought to play an etiological role include the dose of radiation, diabetes mellitus and hypertension, vascular occlusive disease,7• 12 • 13 and as previously discussed, intraabdominal adhesions. Some investigators have attempted to correlate the incidence of radiation injury with dose with variable success. Attempts have been made to define tolerable doses or dose rate for rectal mucosa.4 • 14 Others have tried to show a correlation between the calculated dose and the incidence of bowel injury. 15 • 16 Still others have failed to show any increased incidence of this complication in those patients shown to have received higher doses. 17- 19 From'data collected during this study, it appears that radiation enteropathy can occur throughout the range ofn1diation dose from as low a dose as 2500 rads to as · high a dose as 8700 rads. Diabetes mellitus and hypertension (systolic pressure greater than 140 mm Hg and diastolic pressure greater than 90 mm Hg) did not appear to be significant factors in the development of radiation enteropathy in our study. Two patients in the control group and 3 patients with radiation enteropathy had diabetes. Hypertension was present in 4 controls and 3 patients with radiation enteropathy . .Atherosclerosis waspresent in several of the older individuals in both the control and experimental groups. We fou!ld this a difficult parameter to evaluate. No premature vascular occlusive disease was noted. Our study was designed to evaluate the one remaining major etiological factor for radiation enteropathy, i.e., previous abdominal surgery. Conjecture has surrounded the issue for many years. Studies have appeared in the literature which propose that it may be a factor in the development of this condition. 1- 3 There are other studies which deny its importance. 13 • 20 However, none of these studies was sufficiently controlled or designed specifically to answer the question of the importance of previous abdominal surgery. We believe that 10, - - - . - - - - - -- -----,

FIG. 1. Number of patients developing radiation enteropathy during specific time intervals after radiotherapy.

SURGERY AND RADIATION ENTERITIS

November 1977

1097

our study has clarified this issue. There was a highly 5. Villa Santa V: Complications of radiotherapy for carcinoma of the uterine cervix. Am J Obstet Gynecol114:717-726, 1972 significant difference between the matched control group and the group that developed radiation enteropa- 6. Buchler DA, Kline JC, Peckham BM, et al: Radiation reactions in cervical cancer therapy. Am J Obstet Gynecol 111:745-750, thy, showing conclusively that previous abdominal sur1971 gery is indeed a factor in the development of radiation 7. Calame RJ, Wallach RC: An analysis of the complications of enteropathy. Of the total of 28 patients in both control radiologic treatment of carcinoma of the cervix. Surg Gynecol groups and the experimental group with surgical operaObstet 125:39-44 , 1967 tions before radiotherapy, 18 (61.5%) eventually went 8. Marshak RH , Lindner AE: Radiology of the Small Intestine. on to develop radiation enteropathy. The findings of this Philadelphia, WB Saunders, 1970 study do not provide an answer as to the reason for the 9. Perkins DE, Spjut HJ: Intestinal stenosis following radiation therapy: a roentgenologic-pathologic study. Am J Roentgenol high incidence of radiation enteropathy in patients with Radium Ther Nucl Med 88:953-966, 1962 previous abdominal surgery. The most likely explanation, particularly for small bowel injury, is that pelvic 10. Requarth W, Roberts S: Intestinal injuries following irradiation of pelvic viscera for malignancy. Arch Surg 73:682-688, 1956 adhesions result in a fixation of bowel. This concept is 11. Rowsit B, Malsky SJ , Reid CB: Severe radiation injuries of the supported by the observation that 16 of the 18 patients stomach, small intestine , colon and rectum. Am J Roentgenol with radiation enteropathy and previous surgery had Radium Ther Nu~l Med 114:460-475, 1972 appendectomy or pelvic surgery (table 2). 12. DeCosse JJ, Rhodes RS , Wentz WB , et al: The natural history These observations raise the question of what can be and management of radiation induced injury of the gastrointesdone to reduce the risk of intestinal damage after irratinal tract. Ann Surg 170:369-384, 1969 diation in patients with previous abdominal surgery. 13 . Well wood JM, Jackson BT: The intestinal complications of radiotherapy . Br J Surg 60:814-818, 1973 Where alternative satisfactory modes of treatment are 14. Ingelman-Sundberg A: Rectal injuries following radium treatavailable they should be used in preference to radiation ment of cancer of the cervix uteri. Acta Radiol [Diag] [Suppl) therapy. When no such alternatives are available, con(Stockh) 64:1, 1947 sideration should be given to defining possible fixation 15. Brack CB, Everett HS, Dickison R: Irradiation therapy for carciof bowel by radiological studies. If fixation is demonnoma of the cervix- its effect on the urinary tract. Obstet Gynestrated in the region to be irradiated, especial care in col 7:196-202 , 1956 planning radiation fields to try to minimize bowel irra- 16. Cowell M: The hazards of supplementary x-ray therapy in the diation is indicated. radiation treatment of carcinoma of the cervix uteri . Br J Radiol REFERENCES 1. Strockbine MF, Hancock JE, Fletcher GH: Complications in 831

patients with squamous cell carcinoma of the intact uterine cervix treated with 3000 rads or more whole pelvic irradiation. Am J Roentgenol Radium Ther Nucl Med 108:293-304, 1970 2. Powell-Smith C: Factors influencing the incidence of radiation injury in cancer of the cervix. J Can Assoc Radiol 16:132-137, 1965 3. Localio SA, Stone A, Friedman M: Surgical aspects of radiation enteritis. Surg Gynecol Obstet 129:1163-1172, 1969 4. Kottmeier HL, Gray MJ: Rectal and bladder injuries in relation to radiation dosage in carcinoma of the cervix- a five year follow-up . Am J Obstet Gynecol 82:74-78, 1961

26:652-657' 1953 17. Fletcher GH, Brown TC , Rutledge FN: Clinical significance of rectal and bladder dose measurements in radium therapy of cancer of the uterine cervix. Am J Roentgenol Radium Ther Nucl Med 79:421-426, 1958 18. Strickland P: Damage to the rectum in the radium treatment of carcinoma of the cervix. Br J Radio! 27:630-635·, 1974 19. Strickland P , Gregory C: Rectal dose and rectal damage in the intracavitary treatment of uterine cancer. Acta Radiol [Ther] (Stockh) 56:289-298, 1961 20. Van Nagell JR, Maruyama Y, Parker JC , et al: Small bowel injury following radiation therapy for cervical cancer. Am J Obstet Gynecol118:163-167, 1974