The Effect of Pelvic Irradiation on Ileal Function

The Effect of Pelvic Irradiation on Ileal Function Therapeutic Radiology 1 John A. Stryker, M.D., Gershon W. Hepner, M.D.,2 and Rodrigue Mortel, M.D. Thirty-three patients with gynecological neoplasms undergoing radiotherapy to the pelvis had cholyl] 1_ 14C] glycine breath tests to assess ileal function. Breath tests were performed on each patientin the first andfifth weeksof treatment and 19 of the patients hada thirdtest threemonths post-treatment. Inthe first test, 29.9 ± 16.8% (mean ± SD) of the administered dosewasexcreted in breath 14C in 24 hours. Thisroseto 47.3 ± 15.9% (t = 6.08; p < .001) in the fifth week andfell to 36.3 ± 16% (t = 2.29; p < .05) at three months post-treatment. Eight patients hadbreath tests performed one year post-treatment andthe test percentages were 32.7 ± 7.8% (t = 1.19; P > .10). The increase in 14C02 excretion in the fifth week of treatment occurred at a time whenmost patients were having diarrhea. Thedatasuggest that bile acid malabsorption due to ileal dysfunction may be a factor in radiation-induced diarrheawhich occurs in nearly all patientsduringpelvic irradiation. Bile. Cholylglycine breath test. Intestines, effects of irradiation TERMS: 7 [4].470 • Radiations, injurious effects, gastrointestinal INDEX Radiology 124:213-216, July 1977 TABLE I: ADIATION THERAPY to the pelvis and lower abdomen causes diarrhea because the small intestine is included in the treatment fields (1). The mechanism of diarrhea production is unknown, but patients with diseases involving the ileum develop cholerheic enteropathy (2) due to decreased bile acid reabsorption, with passage of bile acids into the colon which induce water and electrolyte secretion by the colonic mucosa. Animal experiments (3) have suggested that ileal dysfunction with cholerheic enteropathy may be an important factor in the diarrhea occurring in irradiated patients. Newman et al. (4) used cholylglycine breath tests to evaluate ileal function in patients 1-24 years following completion of radiotherapy for pelvic malignancies. Sixteen of 17 patients, 12 of whom were symptomatic, were found to have abnormal cholylglycine breath tests. It was concluded that alteration in stool habit and interruption of enterohepatic circulation caused by radiation injury to the small bowel is an almost inevitable late complication of all forms of pelvic radiotherapy. We have utilized the cholylglycine breath test in a prospective manner and have evaluated patients with gynecological malignancies undergoing radiotherapy in an attempt to evaluate the role of ileal dysfunction as a factor in the etiology of radiation-induced diarrhea. In addition, we have attempted to confirm Newman's conclusions by obtaining breath tests in some of our patients one year following completion of the radiotherapy. R PATIENTS WITH GYNECOLOGIC MALIGNANCIES UNDERGOING RADIOTHERAPY No. of Patients Primary Site Stage 2 8 1 1 2 8 3 2 5 1 Cervix Cervix Cervix Cervical stump Cervix Endometrium Endometrium Ovary Ovary Ovary I A (Post-cone) II B III B II B Recurrent I II I II III ~- ---- TABLE II: PATIENTS STUDIED AS CONTROLS Patient Age/Sex Primary Site JB RM JO OS SR 49M 56M 70M 75M 19M Esophagus Esophagus Stomach Bladder Skin of face 1st Week Cholylglycine 5th Week Breath Cholylglycine Test Breath Test ----------------------------- 21.6 8.1 4.3 29.9 21.7 30.0 8.3 7.7 6.9 33.2 MATERIALS AND METHODS (58 ± 10) (mean ± SD). Sixteen of the patients had previous pelvic surgery but none were known to have. inflammatory bowel disease. In addition, 5 patients undergoing radiotherapy with techniques which did not require irradiation of the ileum were selected as controls (TABLE II). Patient Population Radiotherapy Techniques Thirty-three patients undergoing radiation therapy for malignant gynecological neoplasms were selected for study (TABLE I). The patients ranged in age from 44 to 78 The patients with gynecologic malignancies received external beam therapy by four different techniques (TABLE 1 From the Milton S. Hershey Medical Center (J.A.S.; R.M.), The Pennsylvania State University, Hershey, Pa. Presented at the Sixty-second Scientific Assembly and Annual Meetingof the Radiological Society of North America, Chicago, III., Nov. 14-19. 1976. Supported in part by GrantAM 17303 of the NationalInstitutesof Health. 2 Present address: Department of Medicine, UCLA School of Medicine, HarborGeneral Hospital, Torrance, Calif. 90509. shan 213 214 JOHN A. STRYKER AND OTHERS TABLE III: RADIOTHERAPY TECHNIQUES IN PATIENTS STUDIED No. of Patients Technique of External Therapy 17 8 7 Parallel opposed 16 x 16 cm (AP-PA) 4250-5100 rads.' 5-6 wks/850 reds/week 5 fractions/week Four field "box" 16 x 16 cm & 10 x 16 cm 4420-5100 rads/ 5-6 wks/850 reds/wk. 5 fraction/week July 1977 TABLE IV: CHOLYLGLYC1NE BREATH TEST RESULTS AND STOOL FREQUENCY IN PATIENTS STUDIED Additional Therapy Intracavitary (8) Intracavitary (4) "boost" 2400-2800 rads (3) "boost" 1400 rads + T.V. 3000 rads (I) Parallel opposed (AP-PA) 2000 rads/2 112 weeks + moving strip (2800 rads) 5 fractions/week Total abdomen (3000 rads/4 wks) + parallel opposed (2000 rads/2 wks) 5 fractions/week III) each of which would be expected to irradiate at least a portion of the terminal ileum. Twenty of the patients with gynecologic malignancies were treated with cobalt-60 gamma rays from a Picker C9 unit and 13 were treated with 10 MeV x rays from a Clinac 18. All fields were treated daily in all patients except in the "box" technique in which two of the four fields were treated daily. Cholylglycine Breath Test The cholylglycine breath test was performed by Hepner's (5) modification of the test described by Sherr et 81.(6) and Fromm and Hofmann (7). It involves oral administration of 1.5 JlCi cholyl [ 1-14C]glycine (New England Nuclear, Boston, Mass.), and collection of breath samples in ethanolic hyamine 2, 4, 6, 8, 12, 18 and 24 hours later for determination of interval breath 14C02 specific activity. A zero-hour sample is also taken and used as a blank whenever a subject has had a previous 14C02 breath test. Total breath 14C0 2 excretion in 24 hours was determined and expressed as a percentage of administered cholyl [ 114C]glycine excreted in breath 14C02 in 24 hours. The normal value for 24-hour breath 14C0 2 excretion is 26 ± 11% (mean ± SD), and a value greater than 48 % is considered abnormal. The initial cholylglycine breath test was performed on the first day of radiotherapy in 17 patients, on the second dayin 12 patients, on the third day in 1 patient, on the fourth day in 1 patient, and on the sixth day in 2 patients. The second cholylglycine breath test was performed in the fifth week of radiotherapy in 31 patients and in the sixth week in 2 patients. Nineteen patients had a third test performed approximately three months after completion of radiotherapy. In addition, 2 patients had their third tests performed approximately nine months after completion of radiotherapy and one had her third test one year postradiotherapy. Three patients had a fourth test approximately one year post-radiotherapy and another patient had a fourth test 18 months post-radiotherapy. The times of performance of the cholylglycine breath tests will be referred to as the first-week tests, the fifth-week tests, the three-month tests, and the one-year tests. 1st Week 5th Week 3Month Percent of administered cholyl[l-14C] glycine excreted in breath 14C02 in 24 hours. (mean ± S.D.) 29.9 ± 16.8 47.3 ± 14.9 36.3 16 Stools per day (mean ± S.D.) 1.5 ± .7 4.2 4 1.5 1 TABLE V: ± ± ± 1Year 32.7 ± 7.8 1. 7 ± 1 ABNORMAL CHOLYLGLYCINE BREATH TESTS (>48%) IN PATIENTS STUDIED 1st Week 5th Week 3-Month l-Year 3/33 (9%) 15/33 (45%) 4/19 (21%) 0/8 The five control patients had their initial cholylglycine breath tests in the first week of radiotherapy and their second test in the fifth week. Stool Frequency The patients undergoing pelvic irradiation were asked at least once per week during radiotherapy to describe the number and consistency of their stools for the preceding 24-hour period. Statistical Analysis Means and standard deviations were determined by standard methods. For tests of comparison, Student's paired t-test was used; for further data analysis the coefficient of correlation was employed (8). RESULTS The results of the cholylglycine breath tests and the stool frequency at the time of the tests for the patients with gynecologic malignancies are shown in TABLE IV. In the first-week test, the percentage of administered 14C excreted in 24 hours was 29.9 ± 16.8% (mean ± SD). In the fifth-week test the percentage rose to 47.3 ± 15.9 %. The difference between the first week and fifth week values was highly significant (t = 6.08; P < .001). In the 19 patients studied three months after radiotherapy the percentage had dropped to 36.3 ± 16 %. The three-month tests differed significantly from the first-week tests (t = 2.29; P < .05). The 8 patients studied one year post-radiotherapy had test percentages of 32.7 ± 7.8 %. The difference between the first-week tests and the one-year tests was not significant (t = 1.19; P > .10). The stool frequency for the 26 patients questioned in the first week of treatment was 1.5 ± .7 stools/day. In the fifth week the stool frequency increased to 4.2 ± 4 stools/day in the 33 patients questioned. The difference between the stool frequency in the first week and the fifth week was significant (t = 4.68; P < .001). The stool frequency at three months post-treatment was 1.5 ± 1 in the 13 patients Therapeutic Vol. 124 EFFECT OF PELVIC IRRADIATION ON ILEAL FUNCTION questioned and was 1.7 ± 1 in the 6 patients questioned at one year post-treatment. Although the increased choIylglycine breath tests in the fifth week of treatment occurred at the time of significantly increased stool frequency, it appeared that there was poor correlation between the degree of elevation of the cholylglycine breath tests and the severity of diarrhea. A statistical test of correlation demonstrated a significant negative correlation between the fifth week cholylglycine breath tests and the stool frequency at that time (r = - .36; p < .05) suggesting that the patients with the greatest stool frequency tended to have the lowest cholylglycine breath tests. The results of the cholylglycine breath tests for the control patients are seen in TABLE II. For the first-week tests, the results were 17.1 ± 10.6% and for the fifth-week tests 17.2 ± 13.4%. The difference between the firstweek and fifth-week tests was not significant (t = .01; P > .10). The number of patients with gynecologic malignancies having cholylglycine breath tests above 48 % is presented in TABLE V. Three patients had abnormal cholylglycine breath tests in the first week. One of the 3 patients whose first-week test was 82 % had had a cholecystectomy which is known to increase the cholylglycine breath test. This patient also had abnormal fifth-week and three-month tests. A second patient whose first-week test was 60.5 % had her first-week test performed on the sixth day of radiotherapy suggesting that ileal dysfunction had occurred in this patient at this early time. The 15 patients who had abnormal cholylglycine breath tests in the fifth week of radiotherapy were analyzed for factors which might have contributed to these results. Eight of the 15 patients had had previous pelvic surgery. However, only 8/16 (50%) who had had previous pelvic surgery had abnormal fifthweek tests suggesting that pelvic fixation of bowel loops had little effect on the breath test results. Also, there was little evidence that there was any difference in the breath test results according to the different radiotherapy techniques utilized. Eight of 17 (47%) patients treated by the AP-PA technique and 4/8 (50 % of patients treated by the "box" technique and 3/7 (43 %) of patients treated by the moving strip technique had abnormal cholylglycine breath tests in the fifth week of treatment. In addition, there was little evidence that the patients treated on the two different treatment units had significant differences in breath test results since 8/20 (40%) of patients treated with cobalt-60 gamma rays had abnormal breath tests in the fifth treatment week and 7/13 (54 %) of patients treated on the Clinac 18 had abnormal breath tests in the fifth week. DISCUSSION Our study suggests that ileal dysfunction occurs in a significant number of patients undergoing radiotherapy to the pelvis and lower abdomen since 45 % of the patients had abnormal cholylglycine breath tests in the fifth treatment week. The abnormal breath tests suggest ileal dysfunction because bile salts are normally absorbed in the ileum and the effect of radiotherapy is unlikely to be due 215 Radiology to a change in intestinal flora which can also alter the test. Also, an abscopal effect of irradiation is unlikely since the control patients undergoing radiotherapy had normal fifth .week tests. The data are consistent with the concept that ileal dysfunction is a factor in the etiology of radiationinduced diarrhea since the abnormal cholylglycine breath tests occurred at a time when most of the patients were having diarrhea. The negative correlation between the degree of elevation of the cholylglycine breath test and the stool frequency at the time of the tests is paradoxical and may have been due to the excretion of 14C in diarrheal stool in some patients rather than in breath 14C02 (9). The time course of ileal dysfunction in our patients does not correlate precisely with reported histopathologic studies performed by Trier and Browning (10) in patients receiving abdominal radiotherapy. Jejunal biopsies in these patients were normal in the first week of treatment but in the second week, shortening of villae was seen, and by completion of treatment, marked villous shortening and reduced mucosal thickness had occurred. Within two weeks after radiotherapy completion the mucosal histology had returned to normal. Because of the necessity of starting radiotherapy promptly, a number of our patients had their initial cholylglycine breath test performed late in the first week of treatment. It is possible that one of the three abnormal first-week cholylglycine breath tests was caused by the effect of irradiation on ileal function at this early time when histologic changes would have been absent. Furthermore, 4 patients had abnormal cholylglycine breath tests three months post-treatment suggesting that ileal dysfunction persists in some patients longer than would have been predicted by histologic studies. Thus, it appears that ileal dysfunction induced by irradiation may occur prior to morphologic change and that morphologic recovery is more rapid than functional recovery. The results of our study do not support the conclusion of Newman et a/. (4) that nearly all patients who receive pelvic radiotherapy develop ileal dysfunction as a late effect of treatment since none of our patients studied one year following radiotherapy had abnormal cholylglycine breath tests and the bowel habit in the patients questioned at that time was normal. It should be emphasized, however, that late-occurring lesions in the gastrointestinal tract which are thought to be due primarily to vascular damage (11) can become evident many years following radiotherapy. Most of these injuries, however, are seen within the first year after radiation therapy (12). Also, Newman did not state the dose-time-fraction regimen or the size of the treatment fields which may have differed from ours. We could detect no difference in the breath test results in our patients according to the various treatment techniques or treatment units (cobalt-60 versus Clinac 18). However, it is unlikely that the cholylglycine breath test is sufficiently quantitative to detect small differences in acute intestinal epithelial injury (13). Since ileal dysfunction may be a factor in radiationinduced diarrhea, cholestyramine (14), a resin which sequesters bile salts, may be of value in treatment. Recent studies, however, have suggested that release of prosta- 216 JOHN A. STRYKER AND OTHERS glandins from irradiated intestine may be another factor since radiation-induced diarrhea can be reversed in some patients by treatment with the prostaglandin inhibitor acetylsalicylate (15). It is clear that multiple factors may be involved in the etiology of radiation-induced diarrhea and additional investigation is needed to elucidate them. John A. Stryker, M.D. Department of Radiotherapy M.S. Hershey Medical Center Hershey, Pa. 17033 REFERENCES 1. Rubin P, Casarett GW: Clinical Radiation Pathology. Philadelphia, Saunders, 1968, pp 193-240 2. Hofmann AF: The syndrome of ileal disease and the broken enterohepatic circulation: cholerheic enteropathy. Gastroenterology 52:752-757, Apr 1976 3. Berk RN, Seay DG: Cholerheic enteropathy as a cause of diarrhea and death in radiation enteritis and its prevention with cholestyramine. Radiology 104:153-156, Jul 1972 4. Newman A, KatsarisJ, BlendisLM, et al: Small-intestinalinjury in women who had had pelvic radiotherapy. Lancet 2:1471-1473,29 Dec 1973 July 1977 5. Hepner GW: Increased sensitivity of the cholylglycine breath test for detecting ileal dysfunction. Gastroenterology 68:8-16, Jan 1975 6. Sherr HP, Sasaki Y, Newman A, et al: Detection of bacterial deconjugation of bile salts by a convenient breath-analysis techniqueN Engl J Med 285-656-661, 16 Sep 1911 7. Fromm H, Hofmann AF: Breath test for altered bile-acid metabolism. Lancet 2:621-625, 18 Sep 1971 8. Colton T: Statistics in Medicine. Boston, Little, Brown and Co., 1974, chap 6 9. Pedersen L, Arnfred T, Thaysen EH: Rapid screening of increased bile acid deconjugation and bile acid malabsorption by means of the glycine-1-[14C] cholylglycine assay. Scand J Gastroenterol 8:665-672, 1973 10. Trier JS, Browning TH: Morphologic response of the mucosa of human small intestine to x-ray exposure. J Clin Invest 45: 194-204, Feb 1966 11. Warren S. Friedman NB: Pathology and pathologic diagnosis of radiation lesions in the gastrointestinal tract. Am J Pathol 18: 499-507, May 1972 12. DeCosse JJ, RhodesRS, Wentz WB, et al: The natural history and managementof radiation-induced injury of the gastrointestinaltract. Ann Surg 170:369-384, 1970 13. Hepner GW, Hofmann AF, Thomas PJ: Metabolism of steroid and amino acid moieties of conjugated bile acids in man. J. Cholylglycine. J Clin Invest 51:1889-1897, Jul 1972 14. Hofmann AF, Poley JR: Cholestyramine treatment of diarrhea associated with ileal resection. N Engl J Med 281:397-402,21 Aug 1969 15. Mennie AT, Dalley VM, Dinneen LC, et al: Treatment of radiation-induced gastrointestinal distress with acetylsalicylate. Lancet 2:942-943, 15 Nov 1975