- Email: [email protected]
Effect of X-irradiation on the stomach of the rat
Inr. J. Radiarron Oncology Rio/. Phys., Vol. 17, pp. 779-784 Printed in the U.S.A. All rights reserved.
0360-3016/89 $3.00 + .oO Copyright 0 1989 Pewmon Press plc
??Original Contribution
EFFECT NORBERT
OF X-IRRADIATION BREITER,
M.D. VET.,’ KLAUS-RUDIGER AND TAREK
‘Strahlenbiologisches
ON THE STOMACH
SASSY,
TROTT,
OF THE RAT PROF.
M.D.2
M.Sc.’
Institut der UniversitPt Miinchen and Institut ftir Strahlenbiologie der GSF, Neuherberg, FRG; and ‘Department of Radiation Biology, St. Bartholomew’s Hospital, London, England
A model for localized 300 kV X-irradiation of the rat stomach was developed. After irradiation with single doses, three distinct gastric disorders were observed which occurred at different latency times. Acute death 2-3 weeks after irradiation was caused by an erosive and ulcerative gastritis and occurred in all animals given 28.5 Gy without diet, in 17% of the animals given 28.5 Gy plus diet, and in 13% of the animals given 23 Gy. Subacute to chronic fatal disorders 4 weeks to 7 months after irradiation were seen as stomach dilatation and gastroparesis, associated with the replacement of the normal gastric mucosa by a hyperkeratinized multilayered squamous epithelium. These disorders occurred in 40-100% of the animals after doses between 16 Gy and 28.5 Gy (Sdiet). An ED 50 value of 19.2 Gy (16.5-21.2 Gy, 95% confidence interval) was calculated for this gastroparesis. Late gastric obstruction exceeding 7 months after irradiation was seen in the rats because of profound changes in the gastric wall in 1318% of the animals after doses between 23 Gy and 14 Gy. In animals surviving these three periods, an atrophic mucosa and intestinal metaplasia developed. From functional and morphohistological studies, it can be concluded that there are differences in the pathogenesis of the fatal radiation damage for each of these periods after irradiation. Rat stomach, X-Irradiation, Radiation damage, Acute and chronic effects.
INTRODUCTION
ment of early and chronic radiation irradiated stomach of rats.
injuries
in the locally
The stomach is a radiosensitive part of the gastrointestinal tract (7). It does not tolerate radiation doses that are necessary to control cancer. However, some part of the stomach is often in the primary treatment field in radiotherapy (e.g. of upper abdominal lymph nodes) and may react with radiation damage such as ulceration, perforation, chronic atrophic gastritis, and depression of secretory and motor function (1, 9, 11). Very little is known about the pathogenesis and pathobiology of the various radiation effects on the stomach or on the various factors which influence the incidence, progression, and severity of these changes. This may partly be due to the fact that, until now, no animal model for subacute or chronic radiation injury to the stomach has been developed. The small number of published experiments on radiation effects on the stomach have only analyzed early injury to the gastric mucosa (2,4). This paper reports observations made during attempts to develop an animal model for the qualitative and quantitative assess-
The experiments were conducted using female Wistar/ Neuherberg rats aged 3-4 months. The rats were given local irradiation of the whole stomach with 300 kV X rays, filtered with 0.6 mm Cu and 1 mm Al at a SSD of 27 cm and a dose rate of 3.8 Gy/min. Twenty-four hours before irradiation, food was removed from the cages; however, the rats had free access to water. The stomach is surrounded by several radiosensitive organs and its shape and position is variable, so each radiation treatment had to be planned individually. Approximately 10 min prior to irradiation, the rats were anesthetized with 120 mg/kg Ketamine hydrochloride*/ Kg i.p.; 0.7 ml of a barium sulphate containing contrast medium? was given intraesophageally. The animals were fixed vertically with their heads up and their abdomen against a lead sheet containing a window 8 X 5 cm, then
Reprint requests to: Dr. Norbert Breiter, Institut ftir Strahlenbiologie der GSF-Mtinchen, Ingolst2dter Landstr. 1, 8042 Neuherberg, FRG. Acknowledgments-The authors would like to thank Pfrimmer and Co., Erlangen, FRG for subsidizing the experimental studies.
We also wish to thank Frau S. Miillenstidt from the Institute for Nuclear Biology for the help with the scintigraphic studies. Accepted for publication 13 April 1989. * Ketavet@ Parke-Davis, Freiburg, FRG. t Micropaque”, Nicholas, Sulzbach/Ts., FRG.
METHODS
779
AND
MATERIALS
780
I. J. Radiation Oncology 0 Biology 0 Physics
rotated 30” to the left so that the spinal cord was not in the field during stomach irradiation. A first X-radiograph was done with 35 kV, 13 mA and 5 set exposure time on film.$ Four moveable shields were positioned, with guidance of the calibration marks around the margins of the window, to confine the treatment beam to the stomach contour. A second X ray picture was made to ensure that the stomach was properly included in this reduced field that had average dimensions of 16 X 36 mm. In addition to the stomach, only the distal few mm of the esophagus, the left adrenal gland, and the cranial pole of the left kidney were unavoidably included in the irradiation field. After finishing irradiation, a third radiograph was taken to check the constancy of the treatment field. Radiation doses during the diagnostic procedures were less than 50 mGy and were not included in the quoted exposure dose of the animals. At every irradiation, the doses were monitored by an ionization chamber placed at a fixed distance before the irradiation jig. These doses had been corrected for geometric factors and tissue absorption to estimate the dose to the stomach. In a separate experiment, the absorbed dose in the stomach was measured by a second ionization chamber surgically implanted in the abdomen of a dead animal. Groups of 7-12 rats were given single radiation doses of 14, 16, 18, 19.5, 21, 23,25, and 28.5 Gy to the stomach. Most of the 25 and the 28.5 Gy animals were fed a completely absorbable liquid diet3 in their drinking water for 3 weeks after irradiation to protect the mucosa during the acute period. Animals were observed until they developed severe clinical signs of radiation induced stomach disease. During the observation period, all rats were daily inspected, weighed, and their food consumption estimated once per week. To quantitate gastric function, the rate of gastric emptying was measured by various methods. Serial X-radiographs were taken after intraesophageal application of 0.5 ml of a contrast medium containing barium sulphate, to follow its clearance by morphometry and densitometry. This method was discontinued because anesthesia inhibited gastric motility and BaS04 adhered perferentially to the gastric mucosa and therefore was inhomogenously distributed in the gastric lumen. It is also known that BaS04 may irritate the gastric mucosa and thus influence gastric emptying rate (8). For these reasons, we used a scintigraphic method for gastric transit estimation of the unanesthetized rat. The animals had free access to water. but were held off food for 24 hr. A water soluble, nonabsorbable, pH-resistant radionuclide marker was given and followed with repeated gamma camera imaging. Five-hundred &I technetium 99 m sulfur colloid* (99m Tc-S) was diluted in 1 ml phys$ Polaroid-Polapan 485 Land Film, Type 52,Offenbach/M.. FRG. 5 Peptisorb@, Pfrimmer and Co, Erlangen, FRG.
October 1989. Volume 17, Number 4
iological saline and injected into the stomach by a gastric tube. Immediately afterwards, the animal was fixed at the neck and held vertically in front of the detector of the gamma camera? and the first image of 15-60 set duration was taken. Further images were taken every 20 min for a period of 3 hr and every 2-4 hr for the next 8 hr. To reproduce the position of the stomach in relation to the detector field, the skin of the abdomen over the stomach was marked and, at every subsequent measurement, was placed in the center of the detector field. The region of interest was set with constant size over the stomach area determined in the first image. The ratio of the activity over the region of interest to total body activity was calculated and set in relation to the first image taken as 100%. The percentage of remaining radioactivity in the stomach was determined at various times and the half time T$, which is the time in which half the activity left the stomach, was estimated. Animals were euthanatized with an overdose of ether when they developed clinical signs of acute or chronic stomach injury and subjected to an extensive postmortem examination. At autopsy, the stomach was opened along the greater curvature, cleaned with saline, and fixed in a mixture (1:3) of ethanol (99%) and acetic acid (99%) for 24 hr. It was then stored in ethanol (70%) for at least 24 hr. Longitudinal strips were cut along the total length of the stomach including cardia, rumen, fundus, antrum, and pylorus. They were embedded in paraffin wax. Sections of 3 pm thickness were stained with hematoxylin and eosin for standard histology and with PAS to study surface and neck cells of the gastric mucosa. Survival curves were constructed using the product limit method which corrects for concurrent risks ( 10) and the ED 50 was calculated using a probit program. Intercurrent deaths were mostly due to esophageal perforation which occurred independent of radiation dose in seven out of 80 (9%) irradiated animals.
RESULTS Irradiation caused three distinct gastric disorders that occurred at different latency times. Higher doses resulted in shorter latency times (Fig. 1).
Acute period Stomach irradiation caused lethality 2-3 weeks after irradiation in all four animals given 28.5 Gy without diet, in two out of 12 animals given 28.5 Gy plus diet, and in one out of eight animals given 23 Gy (Fig. 1). At that time the rats showed sudden and irreversible hypophagia, cachexia, dehydration, and regurgitation. Patches of mucus and ingesta adhered to the chin and to the feet. The number of fecal pellets was decreased and they became * Tesuloid Technetium t Normalfeld Gamma FRG.
Kit@, von Heyden, Miinchen. FRG. Kamera MK2, Philips, Miinchen.
X-irradiation of the rat stomach 0 N.
% Ll”lrnO,S
BREITER
781
et a/.
w
surYIva1
.
100
““lrradloted
cootrol
rut5
20 weeks after20 Gy Istomorhdllototlon I
80 60 40
.
20
..
.
0 I
5
IO
15
20 weeks
25
40
after
\rradmt,on
Fig. 1. Survival curves of rats locally X-irradiated
50
60
70
to the stomach.
. .
. .
smaller and were dark stained. Serial radiographs showed distension of the esophagus and delayed entry of the contrast medium into the stomach. However, once the stomach was filled with contrast medium, its emptying rate was increased. At autopsy, the stomach did not differ in size and shape from an unirradiated rat stomach, but the esophagus was distended from the pharynx to the gastric cardia and filled with ingesta. In the opened stomach, the whole surface was covered with a fibrinous layer. Histologically submucosal edema was seen in all regions of the stomach. In the rumen, areas of hyperkeratinized, multilayered squamous epithelium were interspersed with areas of erosion. The fundic mucosa showed inflammation of the propria, depletion of the surface and neck cells, and degeneration of the parietal and chief cells. The antral mucosa was totally eroded and ulcerated near the pylorus.
Subacute to chronic period After surviving the acute phase the following number of irradiated animals (corrected for intercurrent deaths) had to be killed between 1 and 7 months after irradiation (Fig. 1): ten out of 12 animals after 28.5 Gy plus diet, all seven animals after 25 Gy plus diet, six out of eight animals after 23 Gy, seven out of ten animals after 21 Gy, all nine animals after 19.5 Gy, six out of ten animals after 18 Gy, and four out of ten animals after 16 Gy. Clinically, these animals showed a progressive and irreversible weight loss up to more than 25% of the original
Fig. 2. Radiograph marked dilatation
of a rat taken 20 weeks after 20 Gy. showing of the stomach.
. I
. .
Fig. 3. Rate of clearance of 99-m Tc-S-colloid labelled physiological saline from the stomach of unirradiated control rats and irradiated rats. Ninety percent confidence limits are given for control rats.
body weight, at which point the animals were euthanatized. Palpation and abdominal x-ray examinations after a barium meal revealed an enormous dilatation of the stomach (Fig. 2). Gastric transit times were measured in four animals that had received a single dose of 20 Gy X rays locally to the stomach 4 months previously and had lost about 10% of their body weight. A group of five unirradiated animals of similar age were measured as controls. Figure 3 shows the typical gastric emptying curve in unirradiated rats. The T$ was calculated to be 40 + 10 min. In irradiated rats, values showed enormous scatter and little tendency to decrease. For the first 3 hr after the radionuclide application, the amount of marker retained in the irradiated stomach exceeded the 100% level. This was caused by the slow entrance of the marker into the irradiated stomach, which was not completely filled during the first images. At autopsy, the mucosal surface of the distended stomach was covered with a whitish layer. Histologically this correlated with the replacement of the gastric epithelium
Fig. 4. Eight weeks after 20 Gy: Erosion (E) of the antral mucosa and beginning replacement by a hyperkeratinized multilayered squamous epithelium (S). Mm = muscularis mucosae; Sm = submucosa: Mp = muscularis propria.
782
I. J. Radiation Oncology 0 Biology 0 Physics
October
1989. Volume
17. Number
4
Fig. 5. Twenty weeks after 20 Gy: complete replacement of the gastric epithelium by a squamous epithelium with abnormal keratin (k) production.
Fig. 7. Forty weeks after 20 Gy: healed ulceration with hyperkeratinized squamous epithelium and complete destruction of the mucosa propria, muscularis mucosae, and submucosa.
by a multilayered squamous epithelium that was often hyperkeratinized and covered with bacteria. In the second month, this new surface was only seen near the border of the rumen and extending to the fundus region (Fig. 4). In the following weeks, it progressed into the antral region and reached the pylorus in animals that were euthanatized between the fourth and fifth month (Fig. 5). It showed an enormous production of keratin. In animals killed later,
new layer was occasionally ulcerated (Fig. 6) and sometimes extended to the musculature. indicating a healed ulcer (Fig. 7). In the large vessels of the mucosa, submucosa, and subserosa hyalinoid and fibrinoid. degenerative changes of the vessel walls and sometimes thrombosis were observed.
Fig. 6. Thirty weeks after 20 Gy: ulceration
of the new surface.
this
Chronic period One out of eight animals after 23 Gy, three out of ten animals in each of the groups that received 2 1 Gy, 18 Gy, and 16 Gy, and eight out of ten animals after 14 Gy were killed after a latency exceeding 7 months (Fig. 1). These animals, over a period of a few weeks, developed progressive signs of cachexia and regurgitation similar to those in the subacute/chronic period. At necropsy, the stomach had a normal size or was slightly smaller. Adhesions between stomach and liver were often seen. The stomach walls were stiff and thickened and the stomach lumen was narrowed. Histologically (Fig. 8) cystic infiltrations of the gastric wall by regenerating crypts were seen often penetrating the muscularis mucosae and submucosa. These regenerated glands consisted exclusively of PAS-positive cells. The submucosa was massively hyalinized. Animals surviving these three periods were euthanatized long after irradiation because of general debilitation
X-irradiation of the rat stomach 0 N. BREIWRet al.
Fig. 8. One year after wall.
16 Gy: cystic infiltration
of the gastric
or development of spontaneous mammary or thymic tumors. These rats had chronic gastric mucosal atrophy in which the fundic glands were replaced by short crypts containing only PAS-positive surface and neck cells (Fig. 9).
Fig. 9. Two years after 16 Gy: chronic intestinal metaplasia. P = PAS positive and neck cells.
Functional and morphohistological studies indicate that for each period after irradiation there are differences in the pathogenesis of radiation injury. These differences are related to the great variability in turnover times of the various cell populations in the mucosa, but may also be influenced by alterations in the autonomous nerve function and humoral factors such as increases in ph and by bacterial contamination. The radiographically detectable stenosis at the cardia during the acute period may be caused by acute gastritis which has been described in earlier studies by Engelstad (4). Only animals fed liquid diet can survive this acute erosive and ulcerative gastritis. Experimental (4) and clinical (1, 6, 9) observations suggest that the antral region near the pylorus is the most sensitive part of the stomach in the early phase. Most ulcers were seen in this region, whereas in other parts of the stomach only erosions occurred.
atrophic mucosa and (dark stained) surface
The intermediate, subacute to chronic gastric symptoms may have been caused by the gastroparesis of the dilated stomach which is accompanied by profound histological
incidence
DISCUSSION
783
of stomach
dilatation
%
.
IOO-
SO-
/
01
0
5
IO
I5
20
25
30
35
dose [Gy 1 Fig. 10. Dose effect curve for the actuarial risk of stomach dilatation during the subacute period. The figure shows the probit curve, the observed values (0) and the 95% confidence interval for the calculated ED 50 value.
784
I. J. Radiation Oncology 0 Biology 0 Physics
changes. These changes were related to hyperregeneration of the squamous epithelium of the forestomach, which also migrated into the mucosal part of the stomach. Sometimes this altered epithelium was ulcerated, which may be due to submucosal fibrosis and vascular injury. The squamous metaplasia of the mucosa is comparable to that observed after irradiation of the rectum of rats ( 12) when the anus was included in the irradiation field. Epidermal cells were found to migrate from the anal margin into the rectum. Abnormal keratinization in the squamous epithelium, which resulted in a thicker than normal squamous epithelium, was also clinically observed 3 months2 years after esophageal radiotherapy (5). Similar histological changes in the stomach associated with gross distension of the stomach and loss in body weight were reported after irradiation of the mouse thorax with high doses (3). The relationship of the functional changes and the slow emptying rate to the histomorphological changes remains to be proven. However, similar functional disorders such as decreased motility, gastroparesis, and dilatation have been noted clinically without any morphological changes that would have been observed by gastroscopy and fluoroscopy (11). In addition to the histomorphological changes, other explanations for the functional disorders should be discussed, such as changes in the neurohumoral transmitters and receptors and other physiological alterations. The late deaths are related to profound changes of the
October 1989, Volume 17, Number 4
gastric wall, causing obstruction or rigidity of the stomach. These changes were similar to experimental observations made in the rat colon (14) in which animals were given a laxative to prevent radiation ulceration and fatal rectal obstruction. Animals showed periods of reversible subileus and symptom-free intervals, which became progressively shorter. The animals finally developed an irreversible and fatal rectal obstruction that was associated with the invasion of glands into the musculature and into the nerve plexus of the intestinal wall as ulcers healed. Animals irradiated to the stomach showed periods of ulceration during the acute and subacute/chronic phase, long before animals were dying from gastric obstruction in the chronic period. The ED 50 was estimated for animals developing gastroparesis (Fig. 10). The incidence of stomach dilatation rose steeply with increasing dose. We calculated an ED 50 value of 19.2 Gy (16.5-21.2 Gy, 95% confidence interval) for animals killed in this period. The complexity of the structural and the functional organization of the stomach is reflected in the sequence of very different clinical reactions to high radiation doses which are similar to those reported in the radiopathological literature on individual observations in patients (5, 13). Although the pathogenesis of the various syndromes are not clear, the presented model appears useful in studying a variety of problems which are of interest to clinical radiotherapy.
REFERENCES 1. Brick, J. B. Radiation effects on the human stomach. Gastroenterol. 13:363-370; 1946. 2. Chen, K. Y.; Withers, H. R. Survival characteristics of stem cells of gastric mucosa in C3H mice subjected to localized gamma irradiation. Int. J. Radiat. Biol. 21521-534; 1972. 3. Down, J. D.; Easton, D. F.; Steel, G. G. Repair in the mouse lung during low dose-rate irradiation. Radiother. Oncol. 6: 29-42; 1986. 4. Engelstad, R. B. The effect of roentgen rays on the stomach in rabbits. Am. J. Roentgenol. 40:243-263; 1938. 5. Fajardo, L. F. Alimentary tract. In: Fajardo, L. F., ed. Pathology of radiation injury. Paris and New York: Masson Publ.; 1982:47-76. 6. Feiring, W.; Jampol, M. L. Perforation of a gastric ulcer following intensive radiation therapy. N. Engl. J. Med. 242: 751-753; 1950. 7. Friedmann, M. Calculated risks of radiation injury of normal tissue in the treatment of cancer of the testis. Proc. 2nd Nat. Cancer Conf. 1:390-400; 1952. 8. Griffith, G. H.; Owen, G. M.; Campbell, H.; Shields, R.
9. 10.
11.
12.
13.
14.
Gastric emptying in health and in gastroduodenal disease. Gastroenterol. 54: 1-7; 1968. Hamilton, F. E. Gastric ulcer following radiation. Arch. Surg. 55:394-399; 1947. Kaplan. E. L.: Meier, P. Nonparametric estimation from incomplete observations. Am. Stat. Assoc. J. 53:457-481; 1958. Layer, P.; Demol, P.; Hotz, J.; Goebell, H. Gastroparesis after radiation. Successful treatment with carbachol. Dig. Dis. Sci. 31:1377-1380; 1986. Ross, G. A.: Hopewell, J. W. The influence of field positioning on the effects of localized pelvic irradiation. Radiat. Res. Proc. 8th ICRR Edinburgh, Vol. I:253; 1987. Rubin, P.; Casarett, G. W. Clinical radiation pathology, Vol. 1. Philadelphia, London, Toronto: W. B. Saunders Co.; 1968. Trott, K.-R.: Breiter, N.; Spiethoff, A. Experimental studies on the pathogenesis of the chronic radiation ulcer of the large bowel in rats. Int. J. Radiat. Oncol. Biol. Phys. 12: 1637-1643; 1986.
0360-3016/89 $3.00 + .oO Copyright 0 1989 Pewmon Press plc
??Original Contribution
EFFECT NORBERT
OF X-IRRADIATION BREITER,
M.D. VET.,’ KLAUS-RUDIGER AND TAREK
‘Strahlenbiologisches
ON THE STOMACH
SASSY,
TROTT,
OF THE RAT PROF.
M.D.2
M.Sc.’
Institut der UniversitPt Miinchen and Institut ftir Strahlenbiologie der GSF, Neuherberg, FRG; and ‘Department of Radiation Biology, St. Bartholomew’s Hospital, London, England
A model for localized 300 kV X-irradiation of the rat stomach was developed. After irradiation with single doses, three distinct gastric disorders were observed which occurred at different latency times. Acute death 2-3 weeks after irradiation was caused by an erosive and ulcerative gastritis and occurred in all animals given 28.5 Gy without diet, in 17% of the animals given 28.5 Gy plus diet, and in 13% of the animals given 23 Gy. Subacute to chronic fatal disorders 4 weeks to 7 months after irradiation were seen as stomach dilatation and gastroparesis, associated with the replacement of the normal gastric mucosa by a hyperkeratinized multilayered squamous epithelium. These disorders occurred in 40-100% of the animals after doses between 16 Gy and 28.5 Gy (Sdiet). An ED 50 value of 19.2 Gy (16.5-21.2 Gy, 95% confidence interval) was calculated for this gastroparesis. Late gastric obstruction exceeding 7 months after irradiation was seen in the rats because of profound changes in the gastric wall in 1318% of the animals after doses between 23 Gy and 14 Gy. In animals surviving these three periods, an atrophic mucosa and intestinal metaplasia developed. From functional and morphohistological studies, it can be concluded that there are differences in the pathogenesis of the fatal radiation damage for each of these periods after irradiation. Rat stomach, X-Irradiation, Radiation damage, Acute and chronic effects.
INTRODUCTION
ment of early and chronic radiation irradiated stomach of rats.
injuries
in the locally
The stomach is a radiosensitive part of the gastrointestinal tract (7). It does not tolerate radiation doses that are necessary to control cancer. However, some part of the stomach is often in the primary treatment field in radiotherapy (e.g. of upper abdominal lymph nodes) and may react with radiation damage such as ulceration, perforation, chronic atrophic gastritis, and depression of secretory and motor function (1, 9, 11). Very little is known about the pathogenesis and pathobiology of the various radiation effects on the stomach or on the various factors which influence the incidence, progression, and severity of these changes. This may partly be due to the fact that, until now, no animal model for subacute or chronic radiation injury to the stomach has been developed. The small number of published experiments on radiation effects on the stomach have only analyzed early injury to the gastric mucosa (2,4). This paper reports observations made during attempts to develop an animal model for the qualitative and quantitative assess-
The experiments were conducted using female Wistar/ Neuherberg rats aged 3-4 months. The rats were given local irradiation of the whole stomach with 300 kV X rays, filtered with 0.6 mm Cu and 1 mm Al at a SSD of 27 cm and a dose rate of 3.8 Gy/min. Twenty-four hours before irradiation, food was removed from the cages; however, the rats had free access to water. The stomach is surrounded by several radiosensitive organs and its shape and position is variable, so each radiation treatment had to be planned individually. Approximately 10 min prior to irradiation, the rats were anesthetized with 120 mg/kg Ketamine hydrochloride*/ Kg i.p.; 0.7 ml of a barium sulphate containing contrast medium? was given intraesophageally. The animals were fixed vertically with their heads up and their abdomen against a lead sheet containing a window 8 X 5 cm, then
Reprint requests to: Dr. Norbert Breiter, Institut ftir Strahlenbiologie der GSF-Mtinchen, Ingolst2dter Landstr. 1, 8042 Neuherberg, FRG. Acknowledgments-The authors would like to thank Pfrimmer and Co., Erlangen, FRG for subsidizing the experimental studies.
We also wish to thank Frau S. Miillenstidt from the Institute for Nuclear Biology for the help with the scintigraphic studies. Accepted for publication 13 April 1989. * Ketavet@ Parke-Davis, Freiburg, FRG. t Micropaque”, Nicholas, Sulzbach/Ts., FRG.
METHODS
779
AND
MATERIALS
780
I. J. Radiation Oncology 0 Biology 0 Physics
rotated 30” to the left so that the spinal cord was not in the field during stomach irradiation. A first X-radiograph was done with 35 kV, 13 mA and 5 set exposure time on film.$ Four moveable shields were positioned, with guidance of the calibration marks around the margins of the window, to confine the treatment beam to the stomach contour. A second X ray picture was made to ensure that the stomach was properly included in this reduced field that had average dimensions of 16 X 36 mm. In addition to the stomach, only the distal few mm of the esophagus, the left adrenal gland, and the cranial pole of the left kidney were unavoidably included in the irradiation field. After finishing irradiation, a third radiograph was taken to check the constancy of the treatment field. Radiation doses during the diagnostic procedures were less than 50 mGy and were not included in the quoted exposure dose of the animals. At every irradiation, the doses were monitored by an ionization chamber placed at a fixed distance before the irradiation jig. These doses had been corrected for geometric factors and tissue absorption to estimate the dose to the stomach. In a separate experiment, the absorbed dose in the stomach was measured by a second ionization chamber surgically implanted in the abdomen of a dead animal. Groups of 7-12 rats were given single radiation doses of 14, 16, 18, 19.5, 21, 23,25, and 28.5 Gy to the stomach. Most of the 25 and the 28.5 Gy animals were fed a completely absorbable liquid diet3 in their drinking water for 3 weeks after irradiation to protect the mucosa during the acute period. Animals were observed until they developed severe clinical signs of radiation induced stomach disease. During the observation period, all rats were daily inspected, weighed, and their food consumption estimated once per week. To quantitate gastric function, the rate of gastric emptying was measured by various methods. Serial X-radiographs were taken after intraesophageal application of 0.5 ml of a contrast medium containing barium sulphate, to follow its clearance by morphometry and densitometry. This method was discontinued because anesthesia inhibited gastric motility and BaS04 adhered perferentially to the gastric mucosa and therefore was inhomogenously distributed in the gastric lumen. It is also known that BaS04 may irritate the gastric mucosa and thus influence gastric emptying rate (8). For these reasons, we used a scintigraphic method for gastric transit estimation of the unanesthetized rat. The animals had free access to water. but were held off food for 24 hr. A water soluble, nonabsorbable, pH-resistant radionuclide marker was given and followed with repeated gamma camera imaging. Five-hundred &I technetium 99 m sulfur colloid* (99m Tc-S) was diluted in 1 ml phys$ Polaroid-Polapan 485 Land Film, Type 52,Offenbach/M.. FRG. 5 Peptisorb@, Pfrimmer and Co, Erlangen, FRG.
October 1989. Volume 17, Number 4
iological saline and injected into the stomach by a gastric tube. Immediately afterwards, the animal was fixed at the neck and held vertically in front of the detector of the gamma camera? and the first image of 15-60 set duration was taken. Further images were taken every 20 min for a period of 3 hr and every 2-4 hr for the next 8 hr. To reproduce the position of the stomach in relation to the detector field, the skin of the abdomen over the stomach was marked and, at every subsequent measurement, was placed in the center of the detector field. The region of interest was set with constant size over the stomach area determined in the first image. The ratio of the activity over the region of interest to total body activity was calculated and set in relation to the first image taken as 100%. The percentage of remaining radioactivity in the stomach was determined at various times and the half time T$, which is the time in which half the activity left the stomach, was estimated. Animals were euthanatized with an overdose of ether when they developed clinical signs of acute or chronic stomach injury and subjected to an extensive postmortem examination. At autopsy, the stomach was opened along the greater curvature, cleaned with saline, and fixed in a mixture (1:3) of ethanol (99%) and acetic acid (99%) for 24 hr. It was then stored in ethanol (70%) for at least 24 hr. Longitudinal strips were cut along the total length of the stomach including cardia, rumen, fundus, antrum, and pylorus. They were embedded in paraffin wax. Sections of 3 pm thickness were stained with hematoxylin and eosin for standard histology and with PAS to study surface and neck cells of the gastric mucosa. Survival curves were constructed using the product limit method which corrects for concurrent risks ( 10) and the ED 50 was calculated using a probit program. Intercurrent deaths were mostly due to esophageal perforation which occurred independent of radiation dose in seven out of 80 (9%) irradiated animals.
RESULTS Irradiation caused three distinct gastric disorders that occurred at different latency times. Higher doses resulted in shorter latency times (Fig. 1).
Acute period Stomach irradiation caused lethality 2-3 weeks after irradiation in all four animals given 28.5 Gy without diet, in two out of 12 animals given 28.5 Gy plus diet, and in one out of eight animals given 23 Gy (Fig. 1). At that time the rats showed sudden and irreversible hypophagia, cachexia, dehydration, and regurgitation. Patches of mucus and ingesta adhered to the chin and to the feet. The number of fecal pellets was decreased and they became * Tesuloid Technetium t Normalfeld Gamma FRG.
Kit@, von Heyden, Miinchen. FRG. Kamera MK2, Philips, Miinchen.
X-irradiation of the rat stomach 0 N.
% Ll”lrnO,S
BREITER
781
et a/.
w
surYIva1
.
100
““lrradloted
cootrol
rut5
20 weeks after20 Gy Istomorhdllototlon I
80 60 40
.
20
..
.
0 I
5
IO
15
20 weeks
25
40
after
\rradmt,on
Fig. 1. Survival curves of rats locally X-irradiated
50
60
70
to the stomach.
. .
. .
smaller and were dark stained. Serial radiographs showed distension of the esophagus and delayed entry of the contrast medium into the stomach. However, once the stomach was filled with contrast medium, its emptying rate was increased. At autopsy, the stomach did not differ in size and shape from an unirradiated rat stomach, but the esophagus was distended from the pharynx to the gastric cardia and filled with ingesta. In the opened stomach, the whole surface was covered with a fibrinous layer. Histologically submucosal edema was seen in all regions of the stomach. In the rumen, areas of hyperkeratinized, multilayered squamous epithelium were interspersed with areas of erosion. The fundic mucosa showed inflammation of the propria, depletion of the surface and neck cells, and degeneration of the parietal and chief cells. The antral mucosa was totally eroded and ulcerated near the pylorus.
Subacute to chronic period After surviving the acute phase the following number of irradiated animals (corrected for intercurrent deaths) had to be killed between 1 and 7 months after irradiation (Fig. 1): ten out of 12 animals after 28.5 Gy plus diet, all seven animals after 25 Gy plus diet, six out of eight animals after 23 Gy, seven out of ten animals after 21 Gy, all nine animals after 19.5 Gy, six out of ten animals after 18 Gy, and four out of ten animals after 16 Gy. Clinically, these animals showed a progressive and irreversible weight loss up to more than 25% of the original
Fig. 2. Radiograph marked dilatation
of a rat taken 20 weeks after 20 Gy. showing of the stomach.
. I
. .
Fig. 3. Rate of clearance of 99-m Tc-S-colloid labelled physiological saline from the stomach of unirradiated control rats and irradiated rats. Ninety percent confidence limits are given for control rats.
body weight, at which point the animals were euthanatized. Palpation and abdominal x-ray examinations after a barium meal revealed an enormous dilatation of the stomach (Fig. 2). Gastric transit times were measured in four animals that had received a single dose of 20 Gy X rays locally to the stomach 4 months previously and had lost about 10% of their body weight. A group of five unirradiated animals of similar age were measured as controls. Figure 3 shows the typical gastric emptying curve in unirradiated rats. The T$ was calculated to be 40 + 10 min. In irradiated rats, values showed enormous scatter and little tendency to decrease. For the first 3 hr after the radionuclide application, the amount of marker retained in the irradiated stomach exceeded the 100% level. This was caused by the slow entrance of the marker into the irradiated stomach, which was not completely filled during the first images. At autopsy, the mucosal surface of the distended stomach was covered with a whitish layer. Histologically this correlated with the replacement of the gastric epithelium
Fig. 4. Eight weeks after 20 Gy: Erosion (E) of the antral mucosa and beginning replacement by a hyperkeratinized multilayered squamous epithelium (S). Mm = muscularis mucosae; Sm = submucosa: Mp = muscularis propria.
782
I. J. Radiation Oncology 0 Biology 0 Physics
October
1989. Volume
17. Number
4
Fig. 5. Twenty weeks after 20 Gy: complete replacement of the gastric epithelium by a squamous epithelium with abnormal keratin (k) production.
Fig. 7. Forty weeks after 20 Gy: healed ulceration with hyperkeratinized squamous epithelium and complete destruction of the mucosa propria, muscularis mucosae, and submucosa.
by a multilayered squamous epithelium that was often hyperkeratinized and covered with bacteria. In the second month, this new surface was only seen near the border of the rumen and extending to the fundus region (Fig. 4). In the following weeks, it progressed into the antral region and reached the pylorus in animals that were euthanatized between the fourth and fifth month (Fig. 5). It showed an enormous production of keratin. In animals killed later,
new layer was occasionally ulcerated (Fig. 6) and sometimes extended to the musculature. indicating a healed ulcer (Fig. 7). In the large vessels of the mucosa, submucosa, and subserosa hyalinoid and fibrinoid. degenerative changes of the vessel walls and sometimes thrombosis were observed.
Fig. 6. Thirty weeks after 20 Gy: ulceration
of the new surface.
this
Chronic period One out of eight animals after 23 Gy, three out of ten animals in each of the groups that received 2 1 Gy, 18 Gy, and 16 Gy, and eight out of ten animals after 14 Gy were killed after a latency exceeding 7 months (Fig. 1). These animals, over a period of a few weeks, developed progressive signs of cachexia and regurgitation similar to those in the subacute/chronic period. At necropsy, the stomach had a normal size or was slightly smaller. Adhesions between stomach and liver were often seen. The stomach walls were stiff and thickened and the stomach lumen was narrowed. Histologically (Fig. 8) cystic infiltrations of the gastric wall by regenerating crypts were seen often penetrating the muscularis mucosae and submucosa. These regenerated glands consisted exclusively of PAS-positive cells. The submucosa was massively hyalinized. Animals surviving these three periods were euthanatized long after irradiation because of general debilitation
X-irradiation of the rat stomach 0 N. BREIWRet al.
Fig. 8. One year after wall.
16 Gy: cystic infiltration
of the gastric
or development of spontaneous mammary or thymic tumors. These rats had chronic gastric mucosal atrophy in which the fundic glands were replaced by short crypts containing only PAS-positive surface and neck cells (Fig. 9).
Fig. 9. Two years after 16 Gy: chronic intestinal metaplasia. P = PAS positive and neck cells.
Functional and morphohistological studies indicate that for each period after irradiation there are differences in the pathogenesis of radiation injury. These differences are related to the great variability in turnover times of the various cell populations in the mucosa, but may also be influenced by alterations in the autonomous nerve function and humoral factors such as increases in ph and by bacterial contamination. The radiographically detectable stenosis at the cardia during the acute period may be caused by acute gastritis which has been described in earlier studies by Engelstad (4). Only animals fed liquid diet can survive this acute erosive and ulcerative gastritis. Experimental (4) and clinical (1, 6, 9) observations suggest that the antral region near the pylorus is the most sensitive part of the stomach in the early phase. Most ulcers were seen in this region, whereas in other parts of the stomach only erosions occurred.
atrophic mucosa and (dark stained) surface
The intermediate, subacute to chronic gastric symptoms may have been caused by the gastroparesis of the dilated stomach which is accompanied by profound histological
incidence
DISCUSSION
783
of stomach
dilatation
%
.
IOO-
SO-
/
01
0
5
IO
I5
20
25
30
35
dose [Gy 1 Fig. 10. Dose effect curve for the actuarial risk of stomach dilatation during the subacute period. The figure shows the probit curve, the observed values (0) and the 95% confidence interval for the calculated ED 50 value.
784
I. J. Radiation Oncology 0 Biology 0 Physics
changes. These changes were related to hyperregeneration of the squamous epithelium of the forestomach, which also migrated into the mucosal part of the stomach. Sometimes this altered epithelium was ulcerated, which may be due to submucosal fibrosis and vascular injury. The squamous metaplasia of the mucosa is comparable to that observed after irradiation of the rectum of rats ( 12) when the anus was included in the irradiation field. Epidermal cells were found to migrate from the anal margin into the rectum. Abnormal keratinization in the squamous epithelium, which resulted in a thicker than normal squamous epithelium, was also clinically observed 3 months2 years after esophageal radiotherapy (5). Similar histological changes in the stomach associated with gross distension of the stomach and loss in body weight were reported after irradiation of the mouse thorax with high doses (3). The relationship of the functional changes and the slow emptying rate to the histomorphological changes remains to be proven. However, similar functional disorders such as decreased motility, gastroparesis, and dilatation have been noted clinically without any morphological changes that would have been observed by gastroscopy and fluoroscopy (11). In addition to the histomorphological changes, other explanations for the functional disorders should be discussed, such as changes in the neurohumoral transmitters and receptors and other physiological alterations. The late deaths are related to profound changes of the
October 1989, Volume 17, Number 4
gastric wall, causing obstruction or rigidity of the stomach. These changes were similar to experimental observations made in the rat colon (14) in which animals were given a laxative to prevent radiation ulceration and fatal rectal obstruction. Animals showed periods of reversible subileus and symptom-free intervals, which became progressively shorter. The animals finally developed an irreversible and fatal rectal obstruction that was associated with the invasion of glands into the musculature and into the nerve plexus of the intestinal wall as ulcers healed. Animals irradiated to the stomach showed periods of ulceration during the acute and subacute/chronic phase, long before animals were dying from gastric obstruction in the chronic period. The ED 50 was estimated for animals developing gastroparesis (Fig. 10). The incidence of stomach dilatation rose steeply with increasing dose. We calculated an ED 50 value of 19.2 Gy (16.5-21.2 Gy, 95% confidence interval) for animals killed in this period. The complexity of the structural and the functional organization of the stomach is reflected in the sequence of very different clinical reactions to high radiation doses which are similar to those reported in the radiopathological literature on individual observations in patients (5, 13). Although the pathogenesis of the various syndromes are not clear, the presented model appears useful in studying a variety of problems which are of interest to clinical radiotherapy.
REFERENCES 1. Brick, J. B. Radiation effects on the human stomach. Gastroenterol. 13:363-370; 1946. 2. Chen, K. Y.; Withers, H. R. Survival characteristics of stem cells of gastric mucosa in C3H mice subjected to localized gamma irradiation. Int. J. Radiat. Biol. 21521-534; 1972. 3. Down, J. D.; Easton, D. F.; Steel, G. G. Repair in the mouse lung during low dose-rate irradiation. Radiother. Oncol. 6: 29-42; 1986. 4. Engelstad, R. B. The effect of roentgen rays on the stomach in rabbits. Am. J. Roentgenol. 40:243-263; 1938. 5. Fajardo, L. F. Alimentary tract. In: Fajardo, L. F., ed. Pathology of radiation injury. Paris and New York: Masson Publ.; 1982:47-76. 6. Feiring, W.; Jampol, M. L. Perforation of a gastric ulcer following intensive radiation therapy. N. Engl. J. Med. 242: 751-753; 1950. 7. Friedmann, M. Calculated risks of radiation injury of normal tissue in the treatment of cancer of the testis. Proc. 2nd Nat. Cancer Conf. 1:390-400; 1952. 8. Griffith, G. H.; Owen, G. M.; Campbell, H.; Shields, R.
9. 10.
11.
12.
13.
14.
Gastric emptying in health and in gastroduodenal disease. Gastroenterol. 54: 1-7; 1968. Hamilton, F. E. Gastric ulcer following radiation. Arch. Surg. 55:394-399; 1947. Kaplan. E. L.: Meier, P. Nonparametric estimation from incomplete observations. Am. Stat. Assoc. J. 53:457-481; 1958. Layer, P.; Demol, P.; Hotz, J.; Goebell, H. Gastroparesis after radiation. Successful treatment with carbachol. Dig. Dis. Sci. 31:1377-1380; 1986. Ross, G. A.: Hopewell, J. W. The influence of field positioning on the effects of localized pelvic irradiation. Radiat. Res. Proc. 8th ICRR Edinburgh, Vol. I:253; 1987. Rubin, P.; Casarett, G. W. Clinical radiation pathology, Vol. 1. Philadelphia, London, Toronto: W. B. Saunders Co.; 1968. Trott, K.-R.: Breiter, N.; Spiethoff, A. Experimental studies on the pathogenesis of the chronic radiation ulcer of the large bowel in rats. Int. J. Radiat. Oncol. Biol. Phys. 12: 1637-1643; 1986.