Does Local Irradiation Affect Gastric Emptying in Humans?

Does Local Irradiation Affect Gastric Emptying in Humans? FREDERICK L. MAKRAUER,* ELIZABETH OATES,t JEFFREY BECKER,t ROSS ABRAMS,t TIMOTHY O'CONNOR,' RICHARD McCALLUM,:t: JAY SHUMAKER'

ABSTRACT: The authors measured gastric emptying in 13 patients undergoing radiation therapy to the chest, abdomen, or pelvis for nongastrointestinal cancer to investigate whether gastric emptying (GE) was altered by this therapy. Symptoms and weight were monitored at regular intervals. Patients served as their own controls and were compared to a group of healthy subjects. When studied prior to radiation therapy (baseline), cancer patients had a gastric emptying rate that was similar to a healthy control group (tl/2 mean ± SEM 92.0 ± 15.3 vs. 80.4 ± 8.2 min). Irradiation did not change the emptying rate, either after the first dose (early) of 180 cGy (tl/2 99.5 ± 17.9) or after 2 weeks of therapy (late) with 3000 cGy (tl/2 75.5 ± 7.3). There was no correlation of radiation field or tumor type with gastric emptying rate. Two of the 13 patients experienced nausea and vomiting during their course of radiation, but their gastric emptying was unchanged from baseline. KEY INDEXING TERMS: Gastric emptying; Nongastrointestinal cancer; Radiation therapy; Radiation field; Symptoms. [Am J Med Sci 1999;317(1):33-37.]

G

astric emptying during radiation therapy is poorly understood, but a logical choice for study. The anorexia, nausea, and vomiting which From the *Newton· Wellesley Hospital, Newton, and the tDe· partments of Radiation Oncology and Radiology, Division of Nuclear Medicine, Tufts University School of Medicine, Boston, Mas· sachusetts, and the +Department of Medicine, University of Virginia, Charlottesville, Virginia. Dr. Abrams is currently with the Division of Radiation Oncology, Johns Hopkins Hospital, Baltimore, Maryland. Dr. McCallum is currently with the University of Kansas Medical Center, Kansas City, Kansas. Received October 9, 1997; accepted in revised form May 14, 1998. Correspondence: Frederick L. Makrauer, MD, Suite 367, 2000 Washington Street, Newton, MA 02162. Supported by the Sydney F. DeYoung Foundation and the J. H. Makrauer Trust. THE AMERICAN JOURNAL OF THE MEDICAL SCIENCES

often occur during treatment are not only very unpleasant for already-ill cancer patients but also may delay or attenuate essential treatment.1-7 These symptoms have been partially attributed to delayed gastric emptying based on studies in insulin-dependent diabetes mellitus. 9 Research in animals has shown both delayed and normal emptying rates depending on the species investigated. s,lo-13 There is evidence for both a direct (peripheral) effect through impairment of gastric and intestinal motility, and an indirect (central) effect through the stimulation of the midbrain chemoreceptor trigger zone (CTZ). However, the doses of radiation used in these studies are much larger than those employed routinely in humans, so extrapolation to clinical practice is difficult. The recently demonstrated efficacy of ondansetron, a CNS serotonin antagonist, in prophylaxis for these symptoms in humans supports the hypothesis of a centrally acting humoral stimulus. 14.l5 Human study is hampered by patients' preexisting malignancy and apprehension about treatment. Current pharmacologic therapy to reduce radiation symptoms includes phenothiazines and butyrophenones,6,16 but they are not reliably effective. Ldopamine antagonists (particularly metoclopramide), which possess both central antiemetic and peripheral prokinetic properties, have been used but their efficacy is often limited by unpleasant CNS side effects. 17 The recent introduction of ondansetron may eliminate these side effects. Thus, further clarification of the role gastric emptying plays could lead to more effective and better-tolerated treatment of radiation symptoms. Methods Study Subjects. Patients were referred from the

Clinical Radiation Oncology Program of New England Medical Center Hospital and Newton-Wellesley Hospital, a 350-bed community hospital affiliated with Tufts University School of Medicine in Boston, Massachusetts. All patients were interviewed by one investigator (FLM). All subjects were treated in the Department of Radiation Oncology and all radionuclide and gastric emptying studies

33

Irradiation and Gastric Emptying

were performed in the Division of Nuclear Medicine at the New England Medical Center Hospitals. The study population consisted of patients over 21 years of age who were to have radiation therapy to the chest, abdomen, or pelvis for histologically proven primary or metastatic carcinoma or lymphoma. None of the patients received chemotherapy within 2 weeks of the study. Patients were excluded from the study for the following reasons: mechanical disorders of the upper gastrointestinal tract, including prior gastric surgery, primary gastric or duodenal malignancy, and active or chronic peptic disease; symptomatic gastro-esophageal reflux; metabolic disorders known to affect gastric emptying, including diabetes mellitus and untreated hypothyroidism; medications and drugs such as narcotics, Ldopa, antiemetics including tetrahydrocannabinol, and tobacco. Females who were pregnant were not eligible for the study. All patients and control subjects gave informed consent. The protocol was approved by the Human Investigation Committees of Newton-Wellesley Hospital and New England Medical Center Hospitals. External Controls. A noncancer control group was studied to isolate any possible effect of cancer itself on gastric emptying. A control group of healthy subjects consisting of four male and two female volunteers ranging in age from 22 to 45 years was recruited from the metropolitan Boston area in order to develop normal data for the radionuclide gastric emptying study. They were of normal weight and build. One subject was studied three times and the remaining subjects were each studied once. In the subject studied more than once, the average tV2 was considered the subject's gastric emptying rate. Study Design and Protocol. The first study, baseline, was done immediately after the radiation therapy simulation to determine if the psychological stress of having the diagnosis of cancer and anticipating radiotherapy affected gastric emptying. Sustacal (Mead Johnson, Evansville, IN) was chosen as the test meal in an attempt to facilitate digestion after radiation exposure. Sustacal has widespread acceptability to cancer patients as a routinely used nutritional supplement. We also chose it because it contains a high nutrient load and is more a modified solid than a liquid. We tested our hypothesis that this was truly a modified solid meal by studying a control group of five normal subjects. Their individual emptying curves as well as an average curve of all data points demonstrate that emptying occurs after a lag phase and is linear, both of which are features typical of the emptying of a solid meal. The caloric constituents of Sustacal, therefore, mimic those of a modified solid meal with resultant stimulation of small bowel receptors and production of gastric contractility. The second study, early, was performed after the 34

first radiation fraction (180 cGy) to assess any acute effect. The third and final study, late, occurred after delivery of 3000 cGy, approximately 2 weeks after the early study. This point was chosen to study the possible effects of cumulative radiation. All subjects who completed the three study sessions were included for analysis. Patient information included date of examinations, age, sex, weight, tumor type and stage, radiation field, symptoms, and medications. Radiation. Radiation was given to multiple areas. The fraction sizes ranged from 150 cGy to 300 cGy given at the rate of approximately 100 cGy per minute. The dose rate to the treated area was constant for all patients. Fractional scatter dose to the stomach was calculated according to the size of the treated field, the distance from the stomach, and the fractional dose of radiation. Gastric Emptying. Sixty minutes after completion of simulation or radiation therapy, patients consumed a liquid caloric meal over a 15-minute period (80 mL every 5 min) in the sitting position. The meal consisted of 240 mL room temperature vanilla-flavored Sustacal, labeled with 1 mCi of 99mTc sulfur colloid (volume 0.1-0.2 mL). It was consumed at 9:45 AM after a lO-hour overnight fast. Imaging was performed in the upright position using a standard-field-of-view gamma camera. Planar and posterior static images of the stomach were obtained in a paired fashion at 15-minute intervals. Imaging was terminated when less than 40% of the radioactive meal remained in the stomach. Quantitative analysis of gastric emptying as a geometric mean was performed on a standard nuclear medicine computer system. Each pair of images yielded one data point. The t1J2 was defined as the time at which 50% of initial activity (counts) had emptied from the stomach into the small intestine. Symptom Methodology. A symptom score was kept for each subject and completed at the conclusion of each study. The degree of nausea, vomiting, and diarrhea was graded from 0 (none) to 3 (severe) and was recorded separately. Clinical and symptom information was not available to the investigators interpreting the gastric emptying studies (EO, JB). Analysis of Data. The mean emptying time (tv.) was compared between external controls and baseline, and tested for statistical significance using the independent sample t test. Differences in mean gastric emptying rate between baseline, early, and late for all patients and by radiation field were compared and tested using a one factor repeated measures analysis of variance. Each subject's symptom score was measured at baseline, early, and late. Differences from 0 in the median Symptom Score were tested for radiation field (chest, abdomen, pelvis) at each study (baseJanuary 1999 Volume 317 Number 1

Makrauer et al

Table 1. Patient Population Patient No.

Gender

Age

Weight (lbs)

Tumor/Stage

Radiation Field

1 2 3 4 5 6 7 8 9 10 11 12 13

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

73 25 60 62 68 65 57 75 79 69 77 63 45

145 138 192 194 126 165 137 200 180 176 133 127 182

ProstateIB2 TesticularllIA seminoma RectumIB2 BreastIN BreastlII Lung, large cellllV BreastII MelanomalII Lung, large cellllV Prostate/C ProstateIB2 BreastlIV BreastIN

Pelvis Whole abdomen Pelvis L2-S2 L supraclavicular/chest Mediastinum and chest Right chest Right axilla/right chest T11-L4 Pelvis Pelvis L4-S3 Right hip

line, early, late) using the Wilcoxon signed-rank test, where appropriate. Results

Thirteen of sixteen patients enrolled completed the study, 8 male and 5 female, with an age range of twenty-five to seventy-nine years (Table 1). Three patients were withdrawn for reasons unrelated to the side effects of radiation: patient #11 wished to avoid the extra radiation exposure from emptying studies, patient #12 died from her primary disease, and patient #13 wished to avoid the extra time the study required. Six potential subjects were interviewed but not entered in the study: three declined to participate and the other three were rejected. All three patients who declined felt they were too ill to undergo a research study; one patient, age 88, with carcinoma of the lung, declined before being interviewed, and the other two patients, with carcinoma of the prostate and breast, respectively, declined after being interviewed. The three subjects who were rejected did not meet inclusion criteria for the following reasons: age 76 with adenocarcinoma of the lung and Type I diabetes mellitus, carcinoma of the lung Stage D with a radiation field inappropriate for study, and age 70 carcinoma of the colon who had commenced radiation therapy without the investigator's (FLM) knowledge. The baseline study showed no difference between patients and healthy external controls (t% mean ± SEM 92.0 ± 15.3 vs. 80.4 ± 8.2 min), indicating that the presence of cancer alone did not change gastric emptying (Figure 1). No significant change from the baseline group was noted at either the early (t% mean ± SEM 99.5 ± 17.9 min) or late study (t% mean ± SEM 75.5 ± 7.3 min) (Figure 1). There was a trend toward a delay at the early study (Figure 1) in the pelvis group (Figure 2), but this did not attain statistical significance. No measurement for any patient group during any study period reached statisTHE AMERICAN JOURNAL OF THE MEDICAL SCIENCES

tical significance when compared with the healthy external control group. The majority of patients tolerated radiation therapy well regardless of field, and none required reduction or cessation of therapy because of symptoms (Table 2). Patients in the abdomen group experienced the most severe symptoms, but their symptom scores did not vary significantly from the other groups. Two patients (#2 and #6) became clinically ill with symptoms that could be attributed to their radiation therapy. The half emptying time and emptying curves of these two patients were similar to the other patients' results when compared at each 1~

r--------------------------------------,

125

25

BASElINE

EARLY

LATE

Cumulative Radiation Dose (cGy)

Figure 1. Changes in gastric emptying rate with cumulative radiation (n = 13). Bars represent mean gastric emptying rate (GER) tl/2 minutes. Lines show the standard error of the mean GER tl/2 times. There was no statistically significant difference between the mean times at baseline and those at the early or late studies.

35

Irradiation and Gastric Emptying

18) . - - - - - - - - - - - - - - : : - - - - - - - - - , •

. . . . . . . . . . . . . . . . . .+. . . . . . . . . . . . . . . . . . . . . . . . . .

·······························1

150

,.., 1-.......•.............................................

CheoI

D~~

·····11-·

130 120

~

110

~ 100

f;l

90

0:: CD

80

c:

i

70

W

o

8)

ii

(3

50

.., 30 20 10

BASEUNE

EARLY

LATE

Cumulative Radiation Dose (cGy)

Figure 2. Changes in gastric emptying rate with cumulative radiation by radiation field (n = 13). Bars represent mean gastric emptying rate (GER) t1J2 minutes. Lines represent one standard error of the mean for mean GER t1J2 times.

study time and in each phase of gastric emptying. Patient 2 lost 15 pounds (11% of body weight) secondary to nausea and vomiting which occurred during his early and late studies. His field was paraaortic with a reverse Y inverted mantle. Patient 6 had anorexia continuously, but nausea and vomiting only during the late study. His weight dropped 13 pounds (7% of body weight). His field was right lung and mediastinum. Discussion

The gastrointestinal symptoms that follow irradiation are well known. Subacute effects have been reduced with newer radiation therapy techniques.1 8 ,19 Nevertheless, early symptoms remain significant and may lead to reduction or cancellation of treatment. The cause of these early symptoms is not well understood. We studied gastric emptying in a group of cancer patients undergoing radiation therapy to determine if the results of prior research investigation in animals could apply to humans. We assumed that a delay in gastric emptying would be found, as has been seen in diabetic patients with similar symptoms. Indeed, if the mechanism of gastrointestinal symptoms could be better understood, more rational and effective pharmacotherapy could be developed. Delayed gastric emptying rate has been observed in patients with small cancer of the lung who have not been irradiated. 20 - 22 This is felt to be a paraneoplastic syndrome and has been associated with a more diffuse "pseudo-obstruction" picture and serum an36

tineuronal antibodies. We did not observe this phenomenon in our patients and antibody levels were not measured. The delay in gastric emptying seen at the time of the early study in the pelvis group was no more than expected with day-to-day variability in gastric emptying rate, but it is possible that with a larger patient sample a statistically significant difference might have been reached. Our study population represented a heterogeneous sample of cancer types reflecting the typical experience of a tertiary radiation oncology center. We grouped them by radiation field because exposure of abdominal viscera represents a higher level of risk for symptoms based on prior animal studies. 5 Some symptom variation was expected based on the differences in port size, disease stage, and anatomic location within the abdomen. We attempted to eliminate all possible factors that could affect symptoms except the patient's cancer and his or her radiation. The small number of patients developing significant side effects reflects the level of sophistication of modem radiation therapy techniques. Two of the three patients who became most seriously ill were remarkable for the size of the radiation field (whole abdomen), as well as prior pelvic surgery, which is felt to increase the risk of small-bowel injury. Neither had other recognized risk factors for excessive radiation side effects. It is possible that with a larger number of patients, significant radiation side effects would have been recognized, particularly in the abdomen or pelvis groups. Also, we cannot exclude the possibility that higher radiation dosing might have affected gastric emptying. However, it was not our intention to favor (or weight) the study toward any particular patient group at a potentially higher risk for symptoms, but rather to look at the routine experience of our unit.

Table 2. Mean Symptom Score by Radiation Field Radiation Field Chest (n = 4) Baseline Early Late Abdomen (n = 3) Baseline Early Late Pelvis (n = 6) Baseline Early Late

Nausea

Vomiting

Diarrhea

0.00 0.00 0.50

0.00 0.00 0.25

0.00 0.00 0.00

0.00 0.67 1.00

0.00 0.00 0.00

0.00 0.00 0.00

0.00 0.00 0.17

0.00 0.00 0.00

0.00 0.00 1.00

Symptom score scale: 0 = none to 3 = severe. January 1999 Volume 317 Number 1

Mokrouer et 01

Conclusion

Gastric emptying was not delayed immediately after irradiation of the chest, abdomen, or pelvis in a small sample of patients experiencing only minor nausea, and no vomiting or diarrhea, from therapy. A similar observation was made 1 hour after completion of a full course of therapy, when symptoms remained minimaL It is not possible to extrapolate from these observations any conclusion about the cause of major symptoms such patients may experience during therapeutic irradiation. Acknowledgments

The authors acknowledge the kind assistance of Mrs. Susan Chaisson and Mrs. Lorraine Maxwell in transcription, and John Griffith, PhD, for biostatistics. References 1. Roswit B, Malsky SJ, Reid CB. Radiation tolerance of the gastrointestinal tract. Front Radiat Ther Oncol. 1972;6:16081. 2. Roswit B, Malsky SJ, Reid CB. Severe radiation injuries of the stomach, small intestine, colon, and rectum. Am J Roentgenol. 1972;114:460-75. 3. Kellum JM, Jaffe BM, Calhoun TR, Ballinger WF. Gastric complications after radiotherapy for Hodgkins disease and other lymphomas. Am J Surg. 1977;134:314-7. 4. Berthong M, Fajardo LF. Radiation injury in surgical pathology. Am J Surg Pathol. 1981;5(2):153-78. 5. Salazar OM, Rubin P, Keller B, Scarantino C. Systemic (half-body) radiation therapy: response and toxicity. Radiat Oncol Biophys J. 1978;4:937-50. 6. Pradhan DG, Brown CD. Gastroduodenal morbidity of para-aortic and iliac lymph node irradiation using a single anterior 16 MY X-ray field. Clin Radiol. 1988;39:438-41. 7. Padilla GV. Gastrointestinal side effects and quality of life in patients receiving radiation therapy. Nutrition. 1990;6: 367-70. 8. Cooper JR, Mattsson JL. Control of radiation-induced emesis with promethazine, cimetidine, thiethylperazine, or naloxone. Am J Vet Res. 1979;40(8):1057-61. 9. Loo FD, Palmer DW, Soergel, KH, Kalbfleisch JH, Wood CM. Gastric emptying in patients with diabetes mellitus. Gastroenterology. 1984;86:485-94.

THE AMERICAN JOURNAL OF THE MEDICAL SCIENCES

10. Dubois A, Jacobus JP, Grissom MP, Eng RR, Conklin JJ. Altered gastric emptying and prevention of radiationinduced vomiting in dogs. Gastroenterology. 1984;86:444-8. 11. Dorval ED, Mueller GP, Eng R, Durakovic A, Conklin JJ, Dubois A. Effect of ionizing radiation of gastric secretion and gastric motility in monkeys. Gastroenterology. 1985;89: 374-80. 12. Summers RW, Glenn CE, Flatt AJ, Elahinedy A. Radiation and indomethacin effects on morphology, prostaglandins and motility in dog jejunum. Am J Physiol. 1991;261:G14551. 13. Dubois A, Walker RI. Prospects for management of gastrointestinal injury associated with the acute radiation syndrome. Gastroenterology. 1988;95(2):500-7. 14. Denriksson R, Lomberg D, Israelsson G, Zackrisson B, Franzet L. The effect of ondansetron on radiation-induced emesis and diarrhoea. Acta Oncol. 1992;31(7):767-9. 15. Spitzer TR, Bryson J, Cirenza E, Foelber R, Wallerstadt M, Stout C, et al. Randomized double-blind placebocontrolled evaluation of oral ondansetron in the prevention of nausea and vomiting associated with fractionated total body irradiation. J Clin Oncol. 1994;12(11):2432-8. 16. Anonymous. Ondansetron to prevent vomiting after cancer chemotherapy. Med Lett Drugs Ther. 1991;33:63-4. 17. Reynjens A. Domperidone as an anti-emetic: summary of research reports. Postgrad Med J. 1979;55(suppll):50-4. 18. Gallagher MJ, Brereton DD, Rostock RA, Zero JM, Zekoski DA, Poyss LF et al. A prospective study of treatment techniques to minimize the volume of pelvic small bowel with reduction of acute and late effects associated with pelvic irradiation. Int J Radiat Oncol BioI Phys. 1986;12: 1565-73. 19. Gunderson LL, Russell AD, Llewellyn DJ, Doppke KP, Tepper JE. Treatment planning for colorectal cancer: radiation and surgical techniques and value of small-bowel films. Int J Radiat Oncol BioI Phys. 1985;11:1379-93. 20. Schuffler MD, Baird HW, Fleming CR, Bell EC, Bouldin TW, Malagelada JR, et al. Intestinal psudo-obstruction as the presenting manifestation of small-cell carcinoma of the lung. Ann Intern Med. 1983;98(2):129-34. 21. Chinn JS, Schuffler MD. Paraneoplastic visceral neuropathy as a cause of severe gastrointestinal motor dysfunction. Gastroenterology. 1988;95:1279-86. 22. Sodhi N, Camilleri M, Camoriano JK, Low PA, Fealey RD, Perry MC. Autonomic function and motility in intestinal pseudoobstruction caused by paraneoplastic syndrome. Dig Dis Sci. 1989;34(12):1937-42.

37