- Email: [email protected]
Effects of fractionated doses of ionizing radiation on small intestinal motor activity
GASTROENTEROLOGY
1988;95:1249-57
Effects of Fractionated Doses of Ionizing Radiation on Small Intestinal Motor Activity MARY
F. OTTERSON,
Departments of Surgery, Center, Medical College Milwaukee, Wisconsin
SUSHIL
K. SARNA,
Physiology, and Radiation of Wisconsin: and Zablocki
The small intestinal motor effects of fractionated doses of ionizing radiation were studied in 6 conscious dogs. Eight strain-gauge transducers were implanted on the small intestine and a single gauge on the ascending colon, of each dog. After control recordings, an abdominal dose of 250 cGy was administered three times a week on alternate days for 3 successive weeks (total dose, 2250 cGy). Recordings were then made for 4 wk of follow-up. Giant migrating contractions occurred 11 times in 520 h of control recordings in the fasted and fed state, with a mean distance of origin of 55 ? 16 cm from the ileocolonic junction. Abdominal field irradiation significantly increased the incidence and distance of origin of these giant contractions to 438 in 745 recording hours and 158 2 7 cm from the ileocolonic junction, respectively. The incidence of giant migrating contractions peaked after the second dose of radiation. The amplitude ratio of radiation-induced giant migrating contractions to phase III contractions, and their duration and velocity of migration, were similar to the control state. The dogs developed diarrhea and vomiting as early as the first fraction of radiation. Irradiation also increased the incidence of retrograde giant contractions from 8 in 520 h of control recording to 42 in 745 h of recording during the radiation schedule. The radiation-induced retrograde giant contractions peaked in incidence on the day of the first fraction of radiation and were more likely to be associated with a vomiting episode than those occurring in the control period. Migrating motor complex cycling persisted during radiation and its cycle length was not different from the control or postradiation values. Our findings suggest that some of the side effects of radiation such as diarrhea, abdominal cramping, and vomiting may be related to the dramatically increased incidence of giant
Oncology. Veterans
migrating tractions.
and JOHN
E. MOULDER
and Digestive Administration
contractions
System Research Medical Center,
and
retrograde
giant
con-
R
adiation therapy is a major modality in the treatment of cancer. Current estimates are that at least 50% of patients with cancer will receive radiation therapy at some time in the course of their treatment, many with the goal of cure rather than palliation. However, as the application and efficacy of radiation therapy has increased, so has the concern for normal tissue toxicity. For the treatment of abdominal and pelvic tumors. intestinal tolerance is often a major limiting factor. The gastrointestinal side effects of abdominal radiation are both acute and late. The acute side effects include nausea, vomiting, anorexia, diarrhea, abdominal cramping, and discomfort. These side effects occur in a high percentage of patients receiving radiation therapy, but they usually subside within a few weeks after the end of the treatment. The late effects may develop from a few weeks to several years after the end of radiation therapy and include ulcerations, fistulas, strictures, obstructions, and perforations. The late effects are unpredictable in their onset and often difficult to manage. They are attributed largely to fibrosis, circulatory disturbances, or lymphatic obstruction. The aforementioned acute gastrointestinal side effects are likely due to changes in mucosal structure, epithelial transport, and motor activity, but the precise contribution of each of these factors in the manifestation of specific effects is not known. The morphologic and absorptive changes due to radia-
Abbreviations used in this paper: GMC, giant migrating contraction; MMC. migrating motor complex; RGC, retrograde giant contraction. 0 1988 by the American Gastroenterological Association 0018~5085/88/$3.50
1250
OTTERSON ET AL.
tion have previously been described. Abdominal irradiation induced decreased mitosis in the intestinal crypts, necrosis of crypt epithelial cells, and decreased height of crypts and villi. Fractionated doses of radiation decreased villus height during the second to fourth week of therapy but mucosa recovered completely 2 wk after cessation of therapy (1). Alterations in the absorptive capacity of the small intestine have also been reported with radiation. The absorption of protein, fat, carbohydrates, and vitamin Blz is decreased (2-5). The absorptive abnormalities have largely been attributed to mucosal damage. However, motility of the small intestine may also play a role. A motor pattern that hastens the passage of a meal or causes inadequate mixing of food with secretions and their exposure to the mucosal surface may also contribute to malabsorption. The interactions between absorption, secretion, and motor function are not well understood. Summers et al. (6) previously reported the effects of single sublethal doses of radiation on small intestinal myoelectric activity. They recorded myoelectric activity for 3-4 days after radiation and then killed the dogs for histologic work. The acute effects of fractionated doses of radiation on small intestinal motor activity have not been studied thus far. Our objective in this study was to determine the effects of fractionated therapeutic doses of radiation-similar to those used for ovarian cancer-on small intestinal motor activity. In particular, we sought to determine if fractionated radiation induces abnormal or unusual motor patterns that may be related to gastrointestinal side effects. The dogs were followed up for a period of 4 wk after the end of the radiation schedule to determine if the changes in motor activity normalized during this period.
Materials and Methods Experiments were performed on 6 healthy conscious dogs of either sex, each weighing 16-21 kg. Eight strain-gauge transducers were surgically implanted on the small intestine of each of the dogs under general pentobarbital anesthesia (30 mg/kg) to record circular muscle contractions. An additional strain gauge transducer was implanted on the proximal colon, 6 cm distal to the ileocolonic junction. Access to the abdominal cavity was obtained by a midventral laparotomy. The first recording site was 10-15 cm distal to the pylorus and the remaining seven were 55-70 cm apart to cover the entire length of the small intestine. The last recording site was in the small intestine 10 cm proximal to the ileocolonic junction. The strain-gauge transducer lead wires were brought out through a stainless steel cannula in the abdominal wall as previously described (7). The dogs were allowed 10 days to recover from surgery. Each dog served as its own control. Control recordings were made for 2-3 wk following recovery from surgery.
GASTROENTEROLOGY Vol. 95, No. 5
After an overnight fast, recordings were made for 4-6 h. On four occasions each dog was fed a 650-kcal solid meal (Prescription Diet, Hill’s Pet Products, Topeka, Kan.) or 480 kcal of a liquid meal (Sustacal, Mead Johnson Evansville, Ind.) at 20% of the second migrating motor complex (MMC) cycle in the duodenum (8). The postprandial recordings lasted for 4 h. During the control period, each dog underwent three mock irradiations that involved transporting the animal to the area of radiation, administering of lo-15 mgikg of sodium thiamylal (a short-acting thiobarbiturate) to achieve 5-10 min of light anesthesia (blink and swallow reflexes present), and then transporting the dog back to the laboratory where recordings were made for 6-8 h. The dogs then underwent irradiation over a period of 3 wk. Lightly anesthetized dogs were irradiated with parallel opposed lateral fields using a 250-kVp orthovoltage x-ray machine. The x-ray beam was filtered to achieve a half-value layer of 0.9 mm Cu. The focus-to-surface distance was 72 cm, and the field size at the surface of the animal was 21 X 32 cm. The dose rate of the midline of the dog was between 77 and 87 cGy/min, being lowest for the thickest animals. Animals received nine equal doses of radiation in three fractions per week on alternate days for 3 wk. Radiation dosimetry was based on the measured contours of individual dogs, and ionization chamber measurements were done in a water phantom. The midline absorbed dose was 250 cGy, and the variation in the dose across the body was less than 27%. Recordings were made on the day of radiation after the dose in a fasted state. Recordings on the day after radiation administration and for 4 wk after radiation administration were made in both the fasted and fed state. The recordings were made on a 12-channel Grass recorder (model 7, Grass Instrument Co., Quincy, Mass.). The lower and upper cutoff frequencies were set at direct current and 15 Hz, respectively. The signals were simultaneously recorded on a Hewlett-Packard magnetic FM tape recorder (model 3968A) for later replaying of data at a different speed and for filtering them electronically. The signals were filtered using Krohn-Hite filters (model 3750, Krohn-Hite Corp., Avon, Mass.) as described previously (9). The giant migrating contractions (GMCs) (lO,ll), retrograde giant contractions (RGCs) (12,13), and MMCs (14,15) were identified visually. The cycle length of the MMC was measured from the end of one phase III activity to the end of the next. Phase III activity was a group of contractions at their maximal frequency, lasting 4-8 min, that migrated caudad. The duration of postprandial MMC disruption was the interval between the time of ingestion of the meal and the first recurrence of phase III activity at any location in the small intestine. The amplitude ratio was defined as the ratio of the amplitude of the GMC or RGC and the mean maximum amplitude of contractions during the preceding phase III activity at the same recording site. The data were analyzed by Dunnett’s test following analysis of variance. All values are expressed as the mean + standard error of the mean. The mean values were determined for each experiment, from which a mean value was determined for each dog. These mean values were then used to determine
November
19RR
SG9-C6
_,J i____-__
___----___
-~~~~~
_IxI_---
SMALL INTESTINE@0 Figure
the
g
..~.._~ _~ -- -~
COLON$70
g
’
3
1. Spontaneous GMCs in the small intestine and colon. Two spontaneous GMCs originated junction. One of these contractions propagated into the proximal colon, which is represented on the left-hand side indicate the distance (in centimeters) of strain gauge (SG) transducers
overall
mean
of 50.05
p value
value
and
the
standard
was considered
error
statistically
of means.
A
significant.
Results Effects of Irradiation Contractions
on Giant
Migrating
Fasted state. In the normal healthy state GMCs occur very infrequently and unpredictably in In this the small intestine in the fasted state (10,ll). series we recorded a total of 10 GMCs during 330 h of control recording in all 6 dogs. These giant contractions originated at an average of 42 2 9 cm from the ileocolonic junction and migrated caudad at a speed
SGl-15
”
’
min
at 68 cm from the ileocolonic by strain gauge 9. The numbers from the pylorus.
of 1.0 ‘- 0.3 cm/s (Figure 1). The amplitude ratio of the GMCs to phase III contractions was 2.6 k 0.4. The mean duration of these contractions was 14 k 0.8 s at all sites. Before irradiation all giant contractions migrated from their point of origin to the ileocolonic junction without interruption. Five of 10 intestinal GMCs during the control recording period migrated from the ileum to the proximal colon (Figure 1). Abdominal irradiation dramatically increased the incidence and the distance of origin from the ileocolonic junction of GMCs (Figures 2 and 3). A
.y.:.:.:.:.-. ...
+
‘:::::’ . . .. .. .. .. .. ::: . . . .. .. .. .. .. . .
SG2-80
:::
x
SG3-130
......
T
SG4-180
0
hh
SG&390 Uwddd
....
CONTROL
:
1) 8 i:
:iii~.~j~~~~~~~.~.~.~.ji~,.~~~~.*.:.~:::
DURING
RADIATION
I : :
. . . . . . . . . .
POSTRADIATION
SCHEDULE
SG6-280 Bh
Figure
....
.........................
SG5-230
SG7-330_
*
....
::::i::::I:i..
520 hours
uI(
1 h
L IL
160g n
%i?
Figure -
2. Three GMCs originating in the duodenum. They occurred on the day of the seventh dose of radiation. The first two giant contractions migrated to the ileocolonic junction but the third stopped at 130 cm from the pvlorus. See Figure 1 caption for details.
745 hours
560 hours
3. This scattergram depicts each GMC as a with the x-axis referring to the experimental and the y-axis as the distance from the junction at which the contraction originated. gram illustrates both the increased incidence proximal origin of the GMCs that occurred radiation schedule. The data for the fasted states are grouped together. *p < 0.05.
dot plotted condition ileocolonic The diaand more during the and the fed
1252
OTTERSON
ET AL.
GASTROENTEROLOGY
-Day of Radiation ---Day after Radiatm ‘*‘-_ 1-4 Weeks After Radiatm
0
250
u L CONTROL 520 hours
Figure
750
1250
1750
2250
CGY DURING
RADIATION SCHEDULE 745 hours
I I I POSTRADIATION 560 hours
4. This graph illustrates the incidence of GMCs per hour by the dose of radiation. The incidence of GMCs on the day of radiation was greater than on the day after radiation at all doses. The incidence of GMCs peaked after the second dose of radiation. The incidence of GMCs decreased dramatically during the postradiation follow-up. The data for the fasted and the fed states are
grouped together.
total of 423 GMCs were recorded during the 3 wk of scheduled radiation administration (585 h). The mean distance of origin of the GMCs from the ileocolonic junction during the radiation schedule (158 f 7 cm) was significantly greater than that during the control period (p < 0.05). Before radiation administration, no GMCs originated in the proximal half of the small intestine. During the radiation schedule 155 GMCs originated in the proximal half of the small intestine (p < O.O5], and 71 of them began in the duodenum. A majority of the GMCs, go%, occurred on the day of administration of a radiation fraction (Figure 4). The incidence of these contractions peaked after the second dose (Figure 4). Radiation-induced GMCs had an amplitude ratio of 2.5 k 0.20, duration of 15 + 0.25, and velocity of 0.8 2 0.06 cm/s, similar to those seen in the preradiation state (Table 1). The GMCs seen during the radiation schedule sometimes did not migrate to the ileocoionic junction. Thirty percent of the 423 radiation-
Vol. 95, No. 5
induced giant contractions in the small intestine migrated to the proximal colon. During the 4 wk of observations after radiation administration, the incidence of GMCs (17 in 270 h) was greater than the control, but the difference was not statistically significant (Figure 4). The duration of 14 * 0.3 s, amplitude ratio of 2.4 +- 0.2, and distance of origin from the ileocecal junction of 44 ? 13 were not significantly different from those during the control period (Table 1). Fed state. Normally it takes lo-15 min after the ingestion of a meal for the ongoing MMCs to be disrupted and the fed pattern to be established. Before radiation, only one GMC was observed in the distal small intestine during 190 h of postprandial recording. This GMC occurred 15 min after feeding, at which time the MMC cycling was not yet disrupted. No giant migrating contractions were ever observed after MMC cycling was disrupted. During the radiation schedule, five GMCs occurred in the 160 h after the ongoing MMCs in the distal small intestine had been disrupted by the ingestion of a meal (Figure 5). These contractions occurred after both a liquid and a solid meal. The mean distance of origin of the postprandial GMCs from the ileocolonic junction was 32 k 22 cm. The mean time lag between the ingestion of a meal to the occurrence of GMCs during the radiation schedule was 155 + 30 min. In addition, seven GMCs were observed after the ingestion of the meal but before the disruption of the MMC cycling. These giant contractions originated at a
SGl-10
~wL.-n_A_._--.3XU
.dd
SG5-244 w-. SG6-304 !JWW.,+......‘..
Table
I. Characteristics
of Giant
Migrating Duration
Velocity
0.2
15 k 0.2
0.8 ” 0.1
2.4 ‘-’ 0.2
14 + 0.3
0.8 + 0.1
Amulitude ratio (GMC to phase III) Control During
Contractions
SG7_364
_ h
bog
JsJ,IJ_‘iru~~l~-~yi~)uI**lx~~.u___~J~~~ J &” ,._,I. /,,, /!
il
2.6 2 0.4 radiation
schedule Postradiation GMC, giant
migrating
2.5 t
contraction.
Figure
5. The
occurrence of GMCs after MMC cycling is disby the ingestion of a meal. The GMCs occurred 174 min postprandially. For details, see Figure 1 caption.
rupted
November
MOTOR
1988
Table
2. Characteristics Contractions
EFFECTS
OF RADIATION
of Retrograde
1253
Giant Episodes
Velocity (cm/s) Control During
radiation
schedule Postradiation
Figure
6. A RGC and a GMC after the third radiation dose. The two giant contractions originated at -255 cm from the pylorus and migrated in the opposite directions. The duration of the GMC is longer than that of the RGC but the RGC migrates faster than the GMC. The distances on the left side are from the pylorus. For details, see Figure 1 caption.
distance of 39 -t 13 cm from the ileocecal junction and occurred 21 + 10 min after the ingestion of the meal.
Effects of Irradiation Con&actions
on Retrograde
Giant
Retrograde giant contractions also occur very infrequently in the healthy state. Before radiation we noted a total of eight RGCs in 520 h of recording from all 6 dogs, all occurring in the fasted state. They originated an average of 219 -+ 11 cm from the pylorus and migrated orad at a speed of 12.7 t 0.1
DURING RADIATION SCHEDULE
Duration (s) 6.3 + 1.8
1.6 + 0.1
12.5
5.7 + 0.6
1.7 + 0.2
62
18.1
6.1 ? 0.4
1.7 f
83
u
*ooj::
LL 5
i
. . . .
..
:::
3004
. . . . .
POSTRAD,AT,ON
Day of Aadtation ---- Day After Radmon “““““’ l-4 Weeks After Radlatmn * . . . . . . . .
. . .
0 250
* p
0.2
cm/s. The mean duration of these contractions was 6.3 t 1.8 s at all sites. The amplitude ratio to phase III contractions was 1.6 2 0.1. Only one (12.5%) of these contractions was followed by an episode of vomiting. Irradition significantly increased the incidence of RGCs (Figure 6). Forty-two RGCs were recorded in 745 h during the 3 wk of scheduled radiation administration (p < 0.05) (Figure 6). Their point of origin, velocity, duration, and amplitude were similar to the RGCs that occurred before radiation (p > 0.05; Figure 7, Table 2). One of these RGCs occurred 107 min after the solid meal and was accompanied by vomiting of the meal. The radiation-induced RGCs were more frequently associated with vomiting (62%) than were the spontaneous RGCs. The number of RGCs was greatest on the day of the first fraction of radiation and decreased thereafter (Figure 8). No RGC was recorded on the day after radiation administration until after the sixth dose. after which the incidence peaked. During the 4-wk recovery period, 12 RGCs occurred in 560 h of recording. Three of these were during the fed state after the liquid meal (153, 200, and 260 min after ingestion of the meal), and 10 of the 12 RGCs (83%) were associated with vomiting.
‘:’
5
(%)
12.2 t 0.1 22.1 2 6.6 ? 3.6
of
vomiting
560 hours
745 hours
. . . . . . . . . . . . . . . ..
Amplitude ratio
depicts each RGC as a dot plotted with the x-axis as the experimental condition and the y-axis as the distance of origin from the pylorus. There was a statistically significant increase in incidence during the radiation schedule but the point of origin was not significantly different. The data for the fasted and the fed states are aroused together.
I CONT:OL 520 hours
7. This scattergram
Figure
750
1250
1750
2250
CGY
DURING RADIATION SCHEDULE 745 hours
II
I
POSTRADIATION 560 hours
8. The incidence of RGCs per hour by dose of radiation. The incidence of RGCs peaked with the first dose of radiation. There is a second peak after the sixth dose of radiation. This may reflect two different mechanisms of initiation of RGCs.
1254
OTTERSON
ET AL.
GASTROENTEROLOGY Vol. 95, No.
5
requirement varied from 2 days to 5 wk. Some of the dogs showed restlessness and discomfort during the occurrence of a GMC in the proximal small intestine.
Discussion
Figure
9. A. Small bowel phase III activity before radiation. B. Same animal and same strain gauge transducers depicted in A. Phase III activity is made up of contractions that have become fused.
Efiects of Irradiation Complexes
on Migrating
Motor
The MMCs persisted during the radiation schedule and during the 4 wk of follow-up observations. The MMC cycle length during the radiation schedule (88 * 5 min) showed a trend toward more frequent cycling than during the control period (101 k 6 min) or after irradiation (109 + 11 min), but the differences were not statistically significant (p > 0.05). The phase III activity originated in the duodenum and migrated caudad as usual, but there was a noticeable change in the characteristics of its contractions during and after irradiation. The phasic contractions often fused to form long-duration and large-amplitude contractions (Figure 9). Before radiation exposure, the average duration of MMC disruption after a liquid meal was 154 5 13 min. This significantly increased to 204 + 16 min during the radiation schedule and remained increased during postradiation follow-up 261 k 19 min (p < 0.05). The period of MMC disruption after a solid meal was generally longer than the scheduled postprandial recording period of 4 h and therefore could not be compared. Clinical
Symptoms
Liquid stools occurred as early as the day of the first fraction of radiation. Before radiation the volatile liquid proportion of the dog feces was 62% * 5% and it increased to 73% + 2% during the second week of radiation (p > 0.05). On average the dogs developed diarrhea 5.5 + 2.1 days after the start of the radiation schedule. Blood appeared in the feces as early as the day after the first fraction. Anorexia was variable and all dogs experienced some weight loss. No dog lost >lO% of his original body weight. At some time during the course of the experiment all 6 dogs had to be force-fed. This
Our findings show that therapeutic doses of fractionated irradiation of the abdomen have dramatic effects on small intestinal motor activity. These effects include the following. [a) The incidence of GMCs and their distance of origin from the ileocolonic junction are significantly increased. (b) The increased incidence and more proximal origin of GMCs occur as early as a few hours after the first fraction of radiation is administered and return to near normal levels within days of cessaiion of radiation administration. (c) The incidence of RGCs is significantly increased and peaks with the first dose of radiation. (d) Migrating motor complex cycling persists during radiation and there is a trend toward a decrease in its cycle length. (e) The duration of postprandial disruption of the MMC after a liquid meal is significantly increased. Abdominal radiation exposure causes a variety of symptoms that have been temporally divided into a prodromal phase, acute radiation sickness, and the late effects of radiation. The prodromal phase of radiation sickness occurs hours to days after radiation exposure and is manifested by nausea, vomiting, diarrhea, and abdominal cramping. It is distinct from acute radiation sickness in that the absorptive, secretory, and anatomic changes associated with radiation damage cannot be identified. Acute radiation sickness has the same symptoms, occurs later in the course of radiation administration, and has a well-documented series of anatomic and physiologic abnormalities. Acute radiation sickness generally subsides within several weeks after radiation exposure has ceased. Late radiation sickness (ulceration, fistula formation, and perforation) is believed to be the result of microvascular changes in the bowel. Only 20% of patients receiving abdominal radiation exhibit the late sequelae of radiation in a clinically apparent fashion. Our experimental design allowed us to investigate the prodromal and acute phases of radiation sickness but did not allow us to address the problems of late effects of radiation. Our most dramatic finding was an increase in the incidence and proximal origin of the GMCs. The GMC has a very large amplitude and long duration when compared with the normal phasic activity of the small intestine (10). The giant contractions usually migrate uninterrupted from their point of origin to the ileocolonic junction and frequently into the proximal colon. The amplitude of the giant contraction is 2-3 times greater than the phasic contractions
MOTOREFFECTS OF RAIIIATION 1255
Novemberl98H
of the migrating motor complex and its duration is 4-6 times longer. Physically, the GMC represents a ZO-SO-cm-long segment of bowel that is spasmodically contracted and moves caudad at a velocity of -1 cm/s. Kruis et al. (11) reported that GMCs are several-fold more propulsive than phase III contractions. These powerful contractions are an efficient way of rapidly dumping small intestinal contents into the colon. Giant migrating contractions have also been reported to produce abdominal pain and discomfort. Sarna (10) reported that when these contractions occurred in the middle to upper small intestine, the dogs often whined and were restless. Kellow and Phillips (16) reported that GMCs were associated with pain in patients with irritable bowel syndrome. Although pain perception is quite individual, the irradiated animals who whined and were restless were frequently experiencing GMCs. Often these contractions appeared to arouse the animals from sleep. The GMC may well be the small intestinal motor correlate of cramping after exposure to abdominal radiation. In our study, the dogs developed diarrhea and exhibited GMCs as early as the first dose of radiation. The diarrhea continued for as long as 4 wk after radiation, at which time the animals were killed. Mucosal damage and the resulting changes in absorption and secretion have previously been suggested as the factors contributing to radiationinduced diarrhea. However, mucosal damage does not explain the early onset of diarrhea. Trier and Browning (1) and Tarpila (17) reported little damage to the small intestinal mucosa and change in absorptive capacity during the first week of an up to 1000 cGy total dose of fractionated radiation. Our findings suggest that the generation of abnormal motor patterns, such as an excessive number of proximally originating GMCs, may be one of the major factors in early radiation-induced diarrhea. An excessive number of proximally originating GMCs may contribute to diarrhea by rapidly dumping bile salts, pancreatic and intestinal enzymes, and intestinal secretions into the colon. A GMC takes <5 min to migrate from the duodenum to the ileocolonic junction. The rapid transit time due to GMCs would allow very little time for the reabsorption of bile salts from the ileum. Ileal mucosal damage may further aggravate this situation. Bile salts and pancreatic enzymes are secreted cyclically with every MMC cycle in the fasted state. The secretions peak during phase II activity, when most of the GMCs occurred. Cholerheic enteropathy has previously been reported to be a major factor in the pathogenesis of radiation-induced diarrhea (18-22). Although GMCs have not been thoroughly charac-
for very long, their occurrence after feeding is a very rare event and they have never been reported
terized
to occur after MMC cycling has been disrupted and fed motor patterns established throughout the small intestine. After abdominal irradiation, we documented postprandial GMCs that occurred after MMC cycling had been disrupted. These postprandial GMCs may dump undigested food into the colon, and colonic bacterial decomposition of the nutrients may compound the problem of diarrhea. Early in the course of radiation, diarrhea may be entirely due to GMCs; however, late diarrhea, particularly that occurring during the postradiation period, is probably due to small intestinal mucosal damage, enteritis, and the effects of radiation on the colon. The incidence of GMCs decreased rapidly within 24 h of the cessation of radiation exposure and was near normal during the 4-wk postradiation follow-up. The underlying mechanisms of a significantly higher incidence of GMCs and their more proximal origination are not known. Several possibilities exist. The giant contractions may be stimulated by direct injury to the enteric neurons, by the release of endogenous peptides, neurotransmitters. or free radicals from the gut wall, or by irritation of the mucosa. Abdominal irradiation also significantly increased the incidence of RGCs. The RGC, like the GMC, is a powerful spasm of the small intestine. It begins in the middle of the small intestine, obliterates the lumen, and rapidly migrates orad without interruption, dumping intestinal contents into the stomach (12.13).In the dogs in preparation for vomiting receiving radiation, each vomiting episode was preceded by an RGC, but not each RGC resulted in emesis. Frequently, when an RGC occurred, the animals appeared to be nauseated and would lick and swallow frequently. This activity occurred whether or not a vomiting episode followed. Retrograde giant contractions occurring during the radiation schedule and after radiation were more likely to be associated with an emetic episode than those occurring in the control state. Also, RGCs occurred without vomiting postprandially during the radiation schedule and afterwards, which was never observed during the control period. The occurrence of RGCs without vomiting may be a contributory factor in the delay of gastric emptying reported by Dubois et al. (23) and Dorval et al. (24),in the sensations of fullness and bloating, and in anorexia. The RGCs without vomiting would dump the intestinal contents into the stomach. The effects of frequent dumping of intestinal and pancreaticobiliary secretions into the stomach in the fasted state are not well understood in the acute situation. The incidence of RGCs peaked the day of the first
1256
OTTERSON
ET AL.
dose of radiation. A second peak occurred after the sixth dose of radiation. The second peak corresponds with the time frame during which one would expect to see the anatomic and absorptive changes beginning to occur. The first peak occurred much earlier than when these changes have been documented. This bimodal distribution may reflect two separate mechanisms of RGC induction due to radiation. The mode of action by which abdominal irradiation causes vomiting is postulated to be through peripheral irritation similar to vomiting induced by copper sulfate (25). Borison (26) reported that abdominal exposure was necessary for vomiting and that supradiaphragmatic vagotomy and cord section above T-8 prevented vomiting in cats. The afferent signals from the abdomen to the vomiting center require an intact chemoreceptor trigger zone to cause radiation-induced vomiting. It is not known, however, whether the stimulation of RGCs requires a central connection; they may be initiated entirely by the enteric and spinal cord neural mechanisms. Summers et al. (6) reported that a single sublethal dose of 938 cGy disrupted the normal cycling of the MMC. Our findings show that this was not so with the fractionated doses of radiation, even though the total fractionated dose was more than twice the sublethal dose used by Summers et al. With fractionated irradiation there was no significant change in the frequency of MMC cycling, although they cycled somewhat faster. The only change we observed was that the phasic contractions during phase III activity often fused to form larger-amplitude and longerduration single contractions. Summers et al. (6) did not observe the occurrence of GMCs in their myoelectric data but reported bizarre patterns representing migrating clustered contractions. The myoelectric correlates of GMCs were defined only recently by Sarna (10). We did not observe any consistent occurrence of migrating clustered contractions with fractionated irradiation. The duration of postprandial disruption of MMC cycling was significantly prolonged by radiation and the prolongation was sustained during the 4-wk recovery period. The prolongation may be due to radiation-induced delayed gastric emptying (23,24) or to delayed intestinal transit, or both. The MMC cycling does not resume until digestion is complete and the upper gastrointestinal tract is nearly empty. Most studies to date report that mucosal damage and absorption changes due to fractionated irradiation reverse within 4-6 wk after the end of radiation therapy. Our findings show that the incidence of GMCs and RGCs in the small intestine is also closer to normal values 4 wk after the end of fractionated irradiation. However, some of the clinical symp-
GASTROENTEROLOGY
Vol. 95, No. 5
toms, such as diarrhea, continued. Prolonged damage to the colon or the persistence of GMCs in the colon may account for these effects. Karaus and Sarna (27) reported that GMCs in the colon precede defecation and may be associated with mass movements. An excessive number of these contractions may induce diarrhea. In conclusion, therapeutic doses of ionizing radiation to the abdomen, which are commonly used for various gynecologic, gastrointestinal, and lymphoid malignancies, produce an abnormally large number of proximally originating GMCs and normally originating RGCs. The GMCs may contribute to the symptoms of diarrhea, abdominal cramping, and discomfort, whereas the RGCs may contribute to delayed gastric emptying, a feeling of fullness, and abdominal discomfort. Selective pharmacologic inhibition of the giant contractions may alleviate or minimize some of these symptoms. The prevention of these acute side effects of radiation may also be significant for health care and defense personnel in accidental or belligerent nuclear exposure. References 1. Trier JS. Browning TH. Morphologic response of the mucosa of human small intestine to x-ray exposure. J Clin Invest 1966; 45:194-204. 2. Vatistas S. Hornsey S. Radiation-induced protein loss into the gastrointestinal tract. Br J Radio1 1966;34:547-50. 3. Kinsella TJ, Bloomer WD. Tolerance of the intestine to radiation therapy. Surg Gynecol Obstet 1980;151:273-84. 4. Weiss RG, Stryker JA. “C-lactose breath tests during pelvic radiotherapy: the effect of the amount of bowel irradiated. Radiology 1982;142:507-10. 5. Kinn A. Lantz B. Vitamin B,, deficiency after irradiation for bladder carcinoma. J Urol 1984;131:888-90. 6. Summers RW, Flatt AJ. Prihoda M, Mitros FA. Small intestinal motility in dogs after irradiation injury. Dig Dis Sci 1987;32:1402-10. 7. Sarna S. Northcott P, Belbeck L. Mechanism of cycling of migrating myoelectric complexes: effect of morphine. Am J Physiol (Gastrointest Liver Physiol 5) 1982;242:G588-95. 8. Sarna S, Condon RE. Morphine-initiated migrating myoelectric complexes in the fed state in dogs. Gastroenterology 1984; 86:662-g. 9. Sarna SK. Myoelectric correlates of colonic motor complexes. Am J Physiol (Gastrointest Liver Physiol 13) 1986;250:G21320. 10. Sarna SK. Giant migrating contractions and their myoelectric correlates in the small intestine. Am J Physiol (Gastrointest Liver Physiol 16) 1987;253:G697-705. 11. Kruis W, Azpiroz F. Phillips SF. Contractile patterns and transit of fluid in canine terminal ileum. Am J Physiol (Gastrointest Liver Physiol 12) 1985;249:G264-70. motor corre12. Lang IM, Sarna SK, Condon RE. Gastrointestinal lates of vomiting in the dog: quantification and characterization as an independent phenomenon. Gastroenterology 1986; 90:40-7. activity of the 13. Lang IM. Sarna SK, Motor and myoelectric digestive tract associated with vomiting, regurgitation and nausea. In: Wood J. ed. Handbook of physiology. Bethesda: Am Physiol Sot [in press).
November
14. 15. 16.
17.
18.
19.
20.
21. 22.
23.
MOTOR
1988
Sarna SK. Cyclic motor activity; migrating motor complex: 1985. Gastroenterology 1985;89:894-913. Szurszewski JH. A migrating electric complex of the canine small intestine. Am J Physiol 1969;217:1757-63. Kellow JE, Phillips SF. Altered small bowel motility in irritable bowel syndrome is correlated with symptoms. Gastroenterology 1987;92:1885-93. Tarpila S. Morphological and functional response of the human small intestinal mucosa to ionizing radiation. Stand J Gastroenterol 1971;6(Suppl 12):1-52. Hofmann AF. The syndrome of ileal disease and the broken enterohepatic circulation: cholerheic enteropathy. Gastroenterology 1967:52:752-7. Berk RN, Seng DG. Cholerheic enteropathy as a cause of diarrhea and death in radiation enteritis and its prevention with cholestyramine. Radiology 1972;104:153-6. Jackson KL, Entenman C. The role of bile secretion in the gastrointestinal radiation syndrome. Radiation Res 1959;lO: 67-79. Sullivan MF. Dependence of radiation diarrhea on the presence of bile in the intestine. Nature 1962;195:1217-8. Chary S, Thomson DH. A clinical trial evaluating cholestyramine to prevent diarrhea in patients maintained on low-fat diets during pelvic radiation therapy. J Radiat Oncol Biol Phys 1984;10:1885-90. Dubois A, Jacobus JP. Grissom MP. Conklin JJ. Altered gastric emptying and prevention of radiation-induced vomiting in dogs. Gastroenterology 1987;86:444-8.
24. Dorval
ED, Mueller
EFFECTS
OF RADIATION
GP, Eng RR. Durakovec
A, Conklin
1257
JJ,
Dubois A. Effect of ionizing radiation on gastric secretion and gastric motility in monkeys. Gastroenterology 1985;89:37480. 25. Young RW. Mechanisms and treatment of radiation-induced nausea and vomiting. In: Davis CJ, Lake-Bakaar GV, GrahameSmith DC, eds. Nausea and vomiting: mechanisms and treatment. New York: Springer-Verlag. 1986:101-2. 26. Borison HK. Site of emetic actions of x-radiation in the cat. J Comp Neurol 1957;197:439-553. 27. Karaus M. Sarna SK. Giant migrating contractions during defecation in the dog colon. Gastroenterology 1987;92:92533.
Received January 8. 1988. Accepted June 6, 1988. Address requests for reprints to: Dr. Sushi1 K. Sarna, Zablocki Veterans Administration Medical Center, Surgical Research 151. 5000 West National Avenue, Milwaukee, Wisconsin 53295. This work was supported in part by grants from the National Institutes of Health (DK 32346). the Veterans Administration Medical Research Service (7722-OlP). and the Cancer Center of the Medical College of Wisconsin. The authors thank Dr. John Kalbfleisch for help in the statistical analysis of data and Mary Farrar for expert secretarial assistance in the preparation of the manuscript. This work was done with the technical assistance of Dawn Kubacki and Robert Ryan.
1988;95:1249-57
Effects of Fractionated Doses of Ionizing Radiation on Small Intestinal Motor Activity MARY
F. OTTERSON,
Departments of Surgery, Center, Medical College Milwaukee, Wisconsin
SUSHIL
K. SARNA,
Physiology, and Radiation of Wisconsin: and Zablocki
The small intestinal motor effects of fractionated doses of ionizing radiation were studied in 6 conscious dogs. Eight strain-gauge transducers were implanted on the small intestine and a single gauge on the ascending colon, of each dog. After control recordings, an abdominal dose of 250 cGy was administered three times a week on alternate days for 3 successive weeks (total dose, 2250 cGy). Recordings were then made for 4 wk of follow-up. Giant migrating contractions occurred 11 times in 520 h of control recordings in the fasted and fed state, with a mean distance of origin of 55 ? 16 cm from the ileocolonic junction. Abdominal field irradiation significantly increased the incidence and distance of origin of these giant contractions to 438 in 745 recording hours and 158 2 7 cm from the ileocolonic junction, respectively. The incidence of giant migrating contractions peaked after the second dose of radiation. The amplitude ratio of radiation-induced giant migrating contractions to phase III contractions, and their duration and velocity of migration, were similar to the control state. The dogs developed diarrhea and vomiting as early as the first fraction of radiation. Irradiation also increased the incidence of retrograde giant contractions from 8 in 520 h of control recording to 42 in 745 h of recording during the radiation schedule. The radiation-induced retrograde giant contractions peaked in incidence on the day of the first fraction of radiation and were more likely to be associated with a vomiting episode than those occurring in the control period. Migrating motor complex cycling persisted during radiation and its cycle length was not different from the control or postradiation values. Our findings suggest that some of the side effects of radiation such as diarrhea, abdominal cramping, and vomiting may be related to the dramatically increased incidence of giant
Oncology. Veterans
migrating tractions.
and JOHN
E. MOULDER
and Digestive Administration
contractions
System Research Medical Center,
and
retrograde
giant
con-
R
adiation therapy is a major modality in the treatment of cancer. Current estimates are that at least 50% of patients with cancer will receive radiation therapy at some time in the course of their treatment, many with the goal of cure rather than palliation. However, as the application and efficacy of radiation therapy has increased, so has the concern for normal tissue toxicity. For the treatment of abdominal and pelvic tumors. intestinal tolerance is often a major limiting factor. The gastrointestinal side effects of abdominal radiation are both acute and late. The acute side effects include nausea, vomiting, anorexia, diarrhea, abdominal cramping, and discomfort. These side effects occur in a high percentage of patients receiving radiation therapy, but they usually subside within a few weeks after the end of the treatment. The late effects may develop from a few weeks to several years after the end of radiation therapy and include ulcerations, fistulas, strictures, obstructions, and perforations. The late effects are unpredictable in their onset and often difficult to manage. They are attributed largely to fibrosis, circulatory disturbances, or lymphatic obstruction. The aforementioned acute gastrointestinal side effects are likely due to changes in mucosal structure, epithelial transport, and motor activity, but the precise contribution of each of these factors in the manifestation of specific effects is not known. The morphologic and absorptive changes due to radia-
Abbreviations used in this paper: GMC, giant migrating contraction; MMC. migrating motor complex; RGC, retrograde giant contraction. 0 1988 by the American Gastroenterological Association 0018~5085/88/$3.50
1250
OTTERSON ET AL.
tion have previously been described. Abdominal irradiation induced decreased mitosis in the intestinal crypts, necrosis of crypt epithelial cells, and decreased height of crypts and villi. Fractionated doses of radiation decreased villus height during the second to fourth week of therapy but mucosa recovered completely 2 wk after cessation of therapy (1). Alterations in the absorptive capacity of the small intestine have also been reported with radiation. The absorption of protein, fat, carbohydrates, and vitamin Blz is decreased (2-5). The absorptive abnormalities have largely been attributed to mucosal damage. However, motility of the small intestine may also play a role. A motor pattern that hastens the passage of a meal or causes inadequate mixing of food with secretions and their exposure to the mucosal surface may also contribute to malabsorption. The interactions between absorption, secretion, and motor function are not well understood. Summers et al. (6) previously reported the effects of single sublethal doses of radiation on small intestinal myoelectric activity. They recorded myoelectric activity for 3-4 days after radiation and then killed the dogs for histologic work. The acute effects of fractionated doses of radiation on small intestinal motor activity have not been studied thus far. Our objective in this study was to determine the effects of fractionated therapeutic doses of radiation-similar to those used for ovarian cancer-on small intestinal motor activity. In particular, we sought to determine if fractionated radiation induces abnormal or unusual motor patterns that may be related to gastrointestinal side effects. The dogs were followed up for a period of 4 wk after the end of the radiation schedule to determine if the changes in motor activity normalized during this period.
Materials and Methods Experiments were performed on 6 healthy conscious dogs of either sex, each weighing 16-21 kg. Eight strain-gauge transducers were surgically implanted on the small intestine of each of the dogs under general pentobarbital anesthesia (30 mg/kg) to record circular muscle contractions. An additional strain gauge transducer was implanted on the proximal colon, 6 cm distal to the ileocolonic junction. Access to the abdominal cavity was obtained by a midventral laparotomy. The first recording site was 10-15 cm distal to the pylorus and the remaining seven were 55-70 cm apart to cover the entire length of the small intestine. The last recording site was in the small intestine 10 cm proximal to the ileocolonic junction. The strain-gauge transducer lead wires were brought out through a stainless steel cannula in the abdominal wall as previously described (7). The dogs were allowed 10 days to recover from surgery. Each dog served as its own control. Control recordings were made for 2-3 wk following recovery from surgery.
GASTROENTEROLOGY Vol. 95, No. 5
After an overnight fast, recordings were made for 4-6 h. On four occasions each dog was fed a 650-kcal solid meal (Prescription Diet, Hill’s Pet Products, Topeka, Kan.) or 480 kcal of a liquid meal (Sustacal, Mead Johnson Evansville, Ind.) at 20% of the second migrating motor complex (MMC) cycle in the duodenum (8). The postprandial recordings lasted for 4 h. During the control period, each dog underwent three mock irradiations that involved transporting the animal to the area of radiation, administering of lo-15 mgikg of sodium thiamylal (a short-acting thiobarbiturate) to achieve 5-10 min of light anesthesia (blink and swallow reflexes present), and then transporting the dog back to the laboratory where recordings were made for 6-8 h. The dogs then underwent irradiation over a period of 3 wk. Lightly anesthetized dogs were irradiated with parallel opposed lateral fields using a 250-kVp orthovoltage x-ray machine. The x-ray beam was filtered to achieve a half-value layer of 0.9 mm Cu. The focus-to-surface distance was 72 cm, and the field size at the surface of the animal was 21 X 32 cm. The dose rate of the midline of the dog was between 77 and 87 cGy/min, being lowest for the thickest animals. Animals received nine equal doses of radiation in three fractions per week on alternate days for 3 wk. Radiation dosimetry was based on the measured contours of individual dogs, and ionization chamber measurements were done in a water phantom. The midline absorbed dose was 250 cGy, and the variation in the dose across the body was less than 27%. Recordings were made on the day of radiation after the dose in a fasted state. Recordings on the day after radiation administration and for 4 wk after radiation administration were made in both the fasted and fed state. The recordings were made on a 12-channel Grass recorder (model 7, Grass Instrument Co., Quincy, Mass.). The lower and upper cutoff frequencies were set at direct current and 15 Hz, respectively. The signals were simultaneously recorded on a Hewlett-Packard magnetic FM tape recorder (model 3968A) for later replaying of data at a different speed and for filtering them electronically. The signals were filtered using Krohn-Hite filters (model 3750, Krohn-Hite Corp., Avon, Mass.) as described previously (9). The giant migrating contractions (GMCs) (lO,ll), retrograde giant contractions (RGCs) (12,13), and MMCs (14,15) were identified visually. The cycle length of the MMC was measured from the end of one phase III activity to the end of the next. Phase III activity was a group of contractions at their maximal frequency, lasting 4-8 min, that migrated caudad. The duration of postprandial MMC disruption was the interval between the time of ingestion of the meal and the first recurrence of phase III activity at any location in the small intestine. The amplitude ratio was defined as the ratio of the amplitude of the GMC or RGC and the mean maximum amplitude of contractions during the preceding phase III activity at the same recording site. The data were analyzed by Dunnett’s test following analysis of variance. All values are expressed as the mean + standard error of the mean. The mean values were determined for each experiment, from which a mean value was determined for each dog. These mean values were then used to determine
November
19RR
SG9-C6
_,J i____-__
___----___
-~~~~~
_IxI_---
SMALL INTESTINE@0 Figure
the
g
..~.._~ _~ -- -~
COLON$70
g
’
3
1. Spontaneous GMCs in the small intestine and colon. Two spontaneous GMCs originated junction. One of these contractions propagated into the proximal colon, which is represented on the left-hand side indicate the distance (in centimeters) of strain gauge (SG) transducers
overall
mean
of 50.05
p value
value
and
the
standard
was considered
error
statistically
of means.
A
significant.
Results Effects of Irradiation Contractions
on Giant
Migrating
Fasted state. In the normal healthy state GMCs occur very infrequently and unpredictably in In this the small intestine in the fasted state (10,ll). series we recorded a total of 10 GMCs during 330 h of control recording in all 6 dogs. These giant contractions originated at an average of 42 2 9 cm from the ileocolonic junction and migrated caudad at a speed
SGl-15
”
’
min
at 68 cm from the ileocolonic by strain gauge 9. The numbers from the pylorus.
of 1.0 ‘- 0.3 cm/s (Figure 1). The amplitude ratio of the GMCs to phase III contractions was 2.6 k 0.4. The mean duration of these contractions was 14 k 0.8 s at all sites. Before irradiation all giant contractions migrated from their point of origin to the ileocolonic junction without interruption. Five of 10 intestinal GMCs during the control recording period migrated from the ileum to the proximal colon (Figure 1). Abdominal irradiation dramatically increased the incidence and the distance of origin from the ileocolonic junction of GMCs (Figures 2 and 3). A
.y.:.:.:.:.-. ...
+
‘:::::’ . . .. .. .. .. .. ::: . . . .. .. .. .. .. . .
SG2-80
:::
x
SG3-130
......
T
SG4-180
0
hh
SG&390 Uwddd
....
CONTROL
:
1) 8 i:
:iii~.~j~~~~~~~.~.~.~.ji~,.~~~~.*.:.~:::
DURING
RADIATION
I : :
. . . . . . . . . .
POSTRADIATION
SCHEDULE
SG6-280 Bh
Figure
....
.........................
SG5-230
SG7-330_
*
....
::::i::::I:i..
520 hours
uI(
1 h
L IL
160g n
%i?
Figure -
2. Three GMCs originating in the duodenum. They occurred on the day of the seventh dose of radiation. The first two giant contractions migrated to the ileocolonic junction but the third stopped at 130 cm from the pvlorus. See Figure 1 caption for details.
745 hours
560 hours
3. This scattergram depicts each GMC as a with the x-axis referring to the experimental and the y-axis as the distance from the junction at which the contraction originated. gram illustrates both the increased incidence proximal origin of the GMCs that occurred radiation schedule. The data for the fasted states are grouped together. *p < 0.05.
dot plotted condition ileocolonic The diaand more during the and the fed
1252
OTTERSON
ET AL.
GASTROENTEROLOGY
-Day of Radiation ---Day after Radiatm ‘*‘-_ 1-4 Weeks After Radiatm
0
250
u L CONTROL 520 hours
Figure
750
1250
1750
2250
CGY DURING
RADIATION SCHEDULE 745 hours
I I I POSTRADIATION 560 hours
4. This graph illustrates the incidence of GMCs per hour by the dose of radiation. The incidence of GMCs on the day of radiation was greater than on the day after radiation at all doses. The incidence of GMCs peaked after the second dose of radiation. The incidence of GMCs decreased dramatically during the postradiation follow-up. The data for the fasted and the fed states are
grouped together.
total of 423 GMCs were recorded during the 3 wk of scheduled radiation administration (585 h). The mean distance of origin of the GMCs from the ileocolonic junction during the radiation schedule (158 f 7 cm) was significantly greater than that during the control period (p < 0.05). Before radiation administration, no GMCs originated in the proximal half of the small intestine. During the radiation schedule 155 GMCs originated in the proximal half of the small intestine (p < O.O5], and 71 of them began in the duodenum. A majority of the GMCs, go%, occurred on the day of administration of a radiation fraction (Figure 4). The incidence of these contractions peaked after the second dose (Figure 4). Radiation-induced GMCs had an amplitude ratio of 2.5 k 0.20, duration of 15 + 0.25, and velocity of 0.8 2 0.06 cm/s, similar to those seen in the preradiation state (Table 1). The GMCs seen during the radiation schedule sometimes did not migrate to the ileocoionic junction. Thirty percent of the 423 radiation-
Vol. 95, No. 5
induced giant contractions in the small intestine migrated to the proximal colon. During the 4 wk of observations after radiation administration, the incidence of GMCs (17 in 270 h) was greater than the control, but the difference was not statistically significant (Figure 4). The duration of 14 * 0.3 s, amplitude ratio of 2.4 +- 0.2, and distance of origin from the ileocecal junction of 44 ? 13 were not significantly different from those during the control period (Table 1). Fed state. Normally it takes lo-15 min after the ingestion of a meal for the ongoing MMCs to be disrupted and the fed pattern to be established. Before radiation, only one GMC was observed in the distal small intestine during 190 h of postprandial recording. This GMC occurred 15 min after feeding, at which time the MMC cycling was not yet disrupted. No giant migrating contractions were ever observed after MMC cycling was disrupted. During the radiation schedule, five GMCs occurred in the 160 h after the ongoing MMCs in the distal small intestine had been disrupted by the ingestion of a meal (Figure 5). These contractions occurred after both a liquid and a solid meal. The mean distance of origin of the postprandial GMCs from the ileocolonic junction was 32 k 22 cm. The mean time lag between the ingestion of a meal to the occurrence of GMCs during the radiation schedule was 155 + 30 min. In addition, seven GMCs were observed after the ingestion of the meal but before the disruption of the MMC cycling. These giant contractions originated at a
SGl-10
~wL.-n_A_._--.3XU
.dd
SG5-244 w-. SG6-304 !JWW.,+......‘..
Table
I. Characteristics
of Giant
Migrating Duration
Velocity
0.2
15 k 0.2
0.8 ” 0.1
2.4 ‘-’ 0.2
14 + 0.3
0.8 + 0.1
Amulitude ratio (GMC to phase III) Control During
Contractions
SG7_364
_ h
bog
JsJ,IJ_‘iru~~l~-~yi~)uI**lx~~.u___~J~~~ J &” ,._,I. /,,, /!
il
2.6 2 0.4 radiation
schedule Postradiation GMC, giant
migrating
2.5 t
contraction.
Figure
5. The
occurrence of GMCs after MMC cycling is disby the ingestion of a meal. The GMCs occurred 174 min postprandially. For details, see Figure 1 caption.
rupted
November
MOTOR
1988
Table
2. Characteristics Contractions
EFFECTS
OF RADIATION
of Retrograde
1253
Giant Episodes
Velocity (cm/s) Control During
radiation
schedule Postradiation
Figure
6. A RGC and a GMC after the third radiation dose. The two giant contractions originated at -255 cm from the pylorus and migrated in the opposite directions. The duration of the GMC is longer than that of the RGC but the RGC migrates faster than the GMC. The distances on the left side are from the pylorus. For details, see Figure 1 caption.
distance of 39 -t 13 cm from the ileocecal junction and occurred 21 + 10 min after the ingestion of the meal.
Effects of Irradiation Con&actions
on Retrograde
Giant
Retrograde giant contractions also occur very infrequently in the healthy state. Before radiation we noted a total of eight RGCs in 520 h of recording from all 6 dogs, all occurring in the fasted state. They originated an average of 219 -+ 11 cm from the pylorus and migrated orad at a speed of 12.7 t 0.1
DURING RADIATION SCHEDULE
Duration (s) 6.3 + 1.8
1.6 + 0.1
12.5
5.7 + 0.6
1.7 + 0.2
62
18.1
6.1 ? 0.4
1.7 f
83
u
*ooj::
LL 5
i
. . . .
..
:::
3004
. . . . .
POSTRAD,AT,ON
Day of Aadtation ---- Day After Radmon “““““’ l-4 Weeks After Radlatmn * . . . . . . . .
. . .
0 250
* p
0.2
cm/s. The mean duration of these contractions was 6.3 t 1.8 s at all sites. The amplitude ratio to phase III contractions was 1.6 2 0.1. Only one (12.5%) of these contractions was followed by an episode of vomiting. Irradition significantly increased the incidence of RGCs (Figure 6). Forty-two RGCs were recorded in 745 h during the 3 wk of scheduled radiation administration (p < 0.05) (Figure 6). Their point of origin, velocity, duration, and amplitude were similar to the RGCs that occurred before radiation (p > 0.05; Figure 7, Table 2). One of these RGCs occurred 107 min after the solid meal and was accompanied by vomiting of the meal. The radiation-induced RGCs were more frequently associated with vomiting (62%) than were the spontaneous RGCs. The number of RGCs was greatest on the day of the first fraction of radiation and decreased thereafter (Figure 8). No RGC was recorded on the day after radiation administration until after the sixth dose. after which the incidence peaked. During the 4-wk recovery period, 12 RGCs occurred in 560 h of recording. Three of these were during the fed state after the liquid meal (153, 200, and 260 min after ingestion of the meal), and 10 of the 12 RGCs (83%) were associated with vomiting.
‘:’
5
(%)
12.2 t 0.1 22.1 2 6.6 ? 3.6
of
vomiting
560 hours
745 hours
. . . . . . . . . . . . . . . ..
Amplitude ratio
depicts each RGC as a dot plotted with the x-axis as the experimental condition and the y-axis as the distance of origin from the pylorus. There was a statistically significant increase in incidence during the radiation schedule but the point of origin was not significantly different. The data for the fasted and the fed states are aroused together.
I CONT:OL 520 hours
7. This scattergram
Figure
750
1250
1750
2250
CGY
DURING RADIATION SCHEDULE 745 hours
II
I
POSTRADIATION 560 hours
8. The incidence of RGCs per hour by dose of radiation. The incidence of RGCs peaked with the first dose of radiation. There is a second peak after the sixth dose of radiation. This may reflect two different mechanisms of initiation of RGCs.
1254
OTTERSON
ET AL.
GASTROENTEROLOGY Vol. 95, No.
5
requirement varied from 2 days to 5 wk. Some of the dogs showed restlessness and discomfort during the occurrence of a GMC in the proximal small intestine.
Discussion
Figure
9. A. Small bowel phase III activity before radiation. B. Same animal and same strain gauge transducers depicted in A. Phase III activity is made up of contractions that have become fused.
Efiects of Irradiation Complexes
on Migrating
Motor
The MMCs persisted during the radiation schedule and during the 4 wk of follow-up observations. The MMC cycle length during the radiation schedule (88 * 5 min) showed a trend toward more frequent cycling than during the control period (101 k 6 min) or after irradiation (109 + 11 min), but the differences were not statistically significant (p > 0.05). The phase III activity originated in the duodenum and migrated caudad as usual, but there was a noticeable change in the characteristics of its contractions during and after irradiation. The phasic contractions often fused to form long-duration and large-amplitude contractions (Figure 9). Before radiation exposure, the average duration of MMC disruption after a liquid meal was 154 5 13 min. This significantly increased to 204 + 16 min during the radiation schedule and remained increased during postradiation follow-up 261 k 19 min (p < 0.05). The period of MMC disruption after a solid meal was generally longer than the scheduled postprandial recording period of 4 h and therefore could not be compared. Clinical
Symptoms
Liquid stools occurred as early as the day of the first fraction of radiation. Before radiation the volatile liquid proportion of the dog feces was 62% * 5% and it increased to 73% + 2% during the second week of radiation (p > 0.05). On average the dogs developed diarrhea 5.5 + 2.1 days after the start of the radiation schedule. Blood appeared in the feces as early as the day after the first fraction. Anorexia was variable and all dogs experienced some weight loss. No dog lost >lO% of his original body weight. At some time during the course of the experiment all 6 dogs had to be force-fed. This
Our findings show that therapeutic doses of fractionated irradiation of the abdomen have dramatic effects on small intestinal motor activity. These effects include the following. [a) The incidence of GMCs and their distance of origin from the ileocolonic junction are significantly increased. (b) The increased incidence and more proximal origin of GMCs occur as early as a few hours after the first fraction of radiation is administered and return to near normal levels within days of cessaiion of radiation administration. (c) The incidence of RGCs is significantly increased and peaks with the first dose of radiation. (d) Migrating motor complex cycling persists during radiation and there is a trend toward a decrease in its cycle length. (e) The duration of postprandial disruption of the MMC after a liquid meal is significantly increased. Abdominal radiation exposure causes a variety of symptoms that have been temporally divided into a prodromal phase, acute radiation sickness, and the late effects of radiation. The prodromal phase of radiation sickness occurs hours to days after radiation exposure and is manifested by nausea, vomiting, diarrhea, and abdominal cramping. It is distinct from acute radiation sickness in that the absorptive, secretory, and anatomic changes associated with radiation damage cannot be identified. Acute radiation sickness has the same symptoms, occurs later in the course of radiation administration, and has a well-documented series of anatomic and physiologic abnormalities. Acute radiation sickness generally subsides within several weeks after radiation exposure has ceased. Late radiation sickness (ulceration, fistula formation, and perforation) is believed to be the result of microvascular changes in the bowel. Only 20% of patients receiving abdominal radiation exhibit the late sequelae of radiation in a clinically apparent fashion. Our experimental design allowed us to investigate the prodromal and acute phases of radiation sickness but did not allow us to address the problems of late effects of radiation. Our most dramatic finding was an increase in the incidence and proximal origin of the GMCs. The GMC has a very large amplitude and long duration when compared with the normal phasic activity of the small intestine (10). The giant contractions usually migrate uninterrupted from their point of origin to the ileocolonic junction and frequently into the proximal colon. The amplitude of the giant contraction is 2-3 times greater than the phasic contractions
MOTOREFFECTS OF RAIIIATION 1255
Novemberl98H
of the migrating motor complex and its duration is 4-6 times longer. Physically, the GMC represents a ZO-SO-cm-long segment of bowel that is spasmodically contracted and moves caudad at a velocity of -1 cm/s. Kruis et al. (11) reported that GMCs are several-fold more propulsive than phase III contractions. These powerful contractions are an efficient way of rapidly dumping small intestinal contents into the colon. Giant migrating contractions have also been reported to produce abdominal pain and discomfort. Sarna (10) reported that when these contractions occurred in the middle to upper small intestine, the dogs often whined and were restless. Kellow and Phillips (16) reported that GMCs were associated with pain in patients with irritable bowel syndrome. Although pain perception is quite individual, the irradiated animals who whined and were restless were frequently experiencing GMCs. Often these contractions appeared to arouse the animals from sleep. The GMC may well be the small intestinal motor correlate of cramping after exposure to abdominal radiation. In our study, the dogs developed diarrhea and exhibited GMCs as early as the first dose of radiation. The diarrhea continued for as long as 4 wk after radiation, at which time the animals were killed. Mucosal damage and the resulting changes in absorption and secretion have previously been suggested as the factors contributing to radiationinduced diarrhea. However, mucosal damage does not explain the early onset of diarrhea. Trier and Browning (1) and Tarpila (17) reported little damage to the small intestinal mucosa and change in absorptive capacity during the first week of an up to 1000 cGy total dose of fractionated radiation. Our findings suggest that the generation of abnormal motor patterns, such as an excessive number of proximally originating GMCs, may be one of the major factors in early radiation-induced diarrhea. An excessive number of proximally originating GMCs may contribute to diarrhea by rapidly dumping bile salts, pancreatic and intestinal enzymes, and intestinal secretions into the colon. A GMC takes <5 min to migrate from the duodenum to the ileocolonic junction. The rapid transit time due to GMCs would allow very little time for the reabsorption of bile salts from the ileum. Ileal mucosal damage may further aggravate this situation. Bile salts and pancreatic enzymes are secreted cyclically with every MMC cycle in the fasted state. The secretions peak during phase II activity, when most of the GMCs occurred. Cholerheic enteropathy has previously been reported to be a major factor in the pathogenesis of radiation-induced diarrhea (18-22). Although GMCs have not been thoroughly charac-
for very long, their occurrence after feeding is a very rare event and they have never been reported
terized
to occur after MMC cycling has been disrupted and fed motor patterns established throughout the small intestine. After abdominal irradiation, we documented postprandial GMCs that occurred after MMC cycling had been disrupted. These postprandial GMCs may dump undigested food into the colon, and colonic bacterial decomposition of the nutrients may compound the problem of diarrhea. Early in the course of radiation, diarrhea may be entirely due to GMCs; however, late diarrhea, particularly that occurring during the postradiation period, is probably due to small intestinal mucosal damage, enteritis, and the effects of radiation on the colon. The incidence of GMCs decreased rapidly within 24 h of the cessation of radiation exposure and was near normal during the 4-wk postradiation follow-up. The underlying mechanisms of a significantly higher incidence of GMCs and their more proximal origination are not known. Several possibilities exist. The giant contractions may be stimulated by direct injury to the enteric neurons, by the release of endogenous peptides, neurotransmitters. or free radicals from the gut wall, or by irritation of the mucosa. Abdominal irradiation also significantly increased the incidence of RGCs. The RGC, like the GMC, is a powerful spasm of the small intestine. It begins in the middle of the small intestine, obliterates the lumen, and rapidly migrates orad without interruption, dumping intestinal contents into the stomach (12.13).In the dogs in preparation for vomiting receiving radiation, each vomiting episode was preceded by an RGC, but not each RGC resulted in emesis. Frequently, when an RGC occurred, the animals appeared to be nauseated and would lick and swallow frequently. This activity occurred whether or not a vomiting episode followed. Retrograde giant contractions occurring during the radiation schedule and after radiation were more likely to be associated with an emetic episode than those occurring in the control state. Also, RGCs occurred without vomiting postprandially during the radiation schedule and afterwards, which was never observed during the control period. The occurrence of RGCs without vomiting may be a contributory factor in the delay of gastric emptying reported by Dubois et al. (23) and Dorval et al. (24),in the sensations of fullness and bloating, and in anorexia. The RGCs without vomiting would dump the intestinal contents into the stomach. The effects of frequent dumping of intestinal and pancreaticobiliary secretions into the stomach in the fasted state are not well understood in the acute situation. The incidence of RGCs peaked the day of the first
1256
OTTERSON
ET AL.
dose of radiation. A second peak occurred after the sixth dose of radiation. The second peak corresponds with the time frame during which one would expect to see the anatomic and absorptive changes beginning to occur. The first peak occurred much earlier than when these changes have been documented. This bimodal distribution may reflect two separate mechanisms of RGC induction due to radiation. The mode of action by which abdominal irradiation causes vomiting is postulated to be through peripheral irritation similar to vomiting induced by copper sulfate (25). Borison (26) reported that abdominal exposure was necessary for vomiting and that supradiaphragmatic vagotomy and cord section above T-8 prevented vomiting in cats. The afferent signals from the abdomen to the vomiting center require an intact chemoreceptor trigger zone to cause radiation-induced vomiting. It is not known, however, whether the stimulation of RGCs requires a central connection; they may be initiated entirely by the enteric and spinal cord neural mechanisms. Summers et al. (6) reported that a single sublethal dose of 938 cGy disrupted the normal cycling of the MMC. Our findings show that this was not so with the fractionated doses of radiation, even though the total fractionated dose was more than twice the sublethal dose used by Summers et al. With fractionated irradiation there was no significant change in the frequency of MMC cycling, although they cycled somewhat faster. The only change we observed was that the phasic contractions during phase III activity often fused to form larger-amplitude and longerduration single contractions. Summers et al. (6) did not observe the occurrence of GMCs in their myoelectric data but reported bizarre patterns representing migrating clustered contractions. The myoelectric correlates of GMCs were defined only recently by Sarna (10). We did not observe any consistent occurrence of migrating clustered contractions with fractionated irradiation. The duration of postprandial disruption of MMC cycling was significantly prolonged by radiation and the prolongation was sustained during the 4-wk recovery period. The prolongation may be due to radiation-induced delayed gastric emptying (23,24) or to delayed intestinal transit, or both. The MMC cycling does not resume until digestion is complete and the upper gastrointestinal tract is nearly empty. Most studies to date report that mucosal damage and absorption changes due to fractionated irradiation reverse within 4-6 wk after the end of radiation therapy. Our findings show that the incidence of GMCs and RGCs in the small intestine is also closer to normal values 4 wk after the end of fractionated irradiation. However, some of the clinical symp-
GASTROENTEROLOGY
Vol. 95, No. 5
toms, such as diarrhea, continued. Prolonged damage to the colon or the persistence of GMCs in the colon may account for these effects. Karaus and Sarna (27) reported that GMCs in the colon precede defecation and may be associated with mass movements. An excessive number of these contractions may induce diarrhea. In conclusion, therapeutic doses of ionizing radiation to the abdomen, which are commonly used for various gynecologic, gastrointestinal, and lymphoid malignancies, produce an abnormally large number of proximally originating GMCs and normally originating RGCs. The GMCs may contribute to the symptoms of diarrhea, abdominal cramping, and discomfort, whereas the RGCs may contribute to delayed gastric emptying, a feeling of fullness, and abdominal discomfort. Selective pharmacologic inhibition of the giant contractions may alleviate or minimize some of these symptoms. The prevention of these acute side effects of radiation may also be significant for health care and defense personnel in accidental or belligerent nuclear exposure. References 1. Trier JS. Browning TH. Morphologic response of the mucosa of human small intestine to x-ray exposure. J Clin Invest 1966; 45:194-204. 2. Vatistas S. Hornsey S. Radiation-induced protein loss into the gastrointestinal tract. Br J Radio1 1966;34:547-50. 3. Kinsella TJ, Bloomer WD. Tolerance of the intestine to radiation therapy. Surg Gynecol Obstet 1980;151:273-84. 4. Weiss RG, Stryker JA. “C-lactose breath tests during pelvic radiotherapy: the effect of the amount of bowel irradiated. Radiology 1982;142:507-10. 5. Kinn A. Lantz B. Vitamin B,, deficiency after irradiation for bladder carcinoma. J Urol 1984;131:888-90. 6. Summers RW, Flatt AJ. Prihoda M, Mitros FA. Small intestinal motility in dogs after irradiation injury. Dig Dis Sci 1987;32:1402-10. 7. Sarna S. Northcott P, Belbeck L. Mechanism of cycling of migrating myoelectric complexes: effect of morphine. Am J Physiol (Gastrointest Liver Physiol 5) 1982;242:G588-95. 8. Sarna S, Condon RE. Morphine-initiated migrating myoelectric complexes in the fed state in dogs. Gastroenterology 1984; 86:662-g. 9. Sarna SK. Myoelectric correlates of colonic motor complexes. Am J Physiol (Gastrointest Liver Physiol 13) 1986;250:G21320. 10. Sarna SK. Giant migrating contractions and their myoelectric correlates in the small intestine. Am J Physiol (Gastrointest Liver Physiol 16) 1987;253:G697-705. 11. Kruis W, Azpiroz F. Phillips SF. Contractile patterns and transit of fluid in canine terminal ileum. Am J Physiol (Gastrointest Liver Physiol 12) 1985;249:G264-70. motor corre12. Lang IM, Sarna SK, Condon RE. Gastrointestinal lates of vomiting in the dog: quantification and characterization as an independent phenomenon. Gastroenterology 1986; 90:40-7. activity of the 13. Lang IM. Sarna SK, Motor and myoelectric digestive tract associated with vomiting, regurgitation and nausea. In: Wood J. ed. Handbook of physiology. Bethesda: Am Physiol Sot [in press).
November
14. 15. 16.
17.
18.
19.
20.
21. 22.
23.
MOTOR
1988
Sarna SK. Cyclic motor activity; migrating motor complex: 1985. Gastroenterology 1985;89:894-913. Szurszewski JH. A migrating electric complex of the canine small intestine. Am J Physiol 1969;217:1757-63. Kellow JE, Phillips SF. Altered small bowel motility in irritable bowel syndrome is correlated with symptoms. Gastroenterology 1987;92:1885-93. Tarpila S. Morphological and functional response of the human small intestinal mucosa to ionizing radiation. Stand J Gastroenterol 1971;6(Suppl 12):1-52. Hofmann AF. The syndrome of ileal disease and the broken enterohepatic circulation: cholerheic enteropathy. Gastroenterology 1967:52:752-7. Berk RN, Seng DG. Cholerheic enteropathy as a cause of diarrhea and death in radiation enteritis and its prevention with cholestyramine. Radiology 1972;104:153-6. Jackson KL, Entenman C. The role of bile secretion in the gastrointestinal radiation syndrome. Radiation Res 1959;lO: 67-79. Sullivan MF. Dependence of radiation diarrhea on the presence of bile in the intestine. Nature 1962;195:1217-8. Chary S, Thomson DH. A clinical trial evaluating cholestyramine to prevent diarrhea in patients maintained on low-fat diets during pelvic radiation therapy. J Radiat Oncol Biol Phys 1984;10:1885-90. Dubois A, Jacobus JP. Grissom MP. Conklin JJ. Altered gastric emptying and prevention of radiation-induced vomiting in dogs. Gastroenterology 1987;86:444-8.
24. Dorval
ED, Mueller
EFFECTS
OF RADIATION
GP, Eng RR. Durakovec
A, Conklin
1257
JJ,
Dubois A. Effect of ionizing radiation on gastric secretion and gastric motility in monkeys. Gastroenterology 1985;89:37480. 25. Young RW. Mechanisms and treatment of radiation-induced nausea and vomiting. In: Davis CJ, Lake-Bakaar GV, GrahameSmith DC, eds. Nausea and vomiting: mechanisms and treatment. New York: Springer-Verlag. 1986:101-2. 26. Borison HK. Site of emetic actions of x-radiation in the cat. J Comp Neurol 1957;197:439-553. 27. Karaus M. Sarna SK. Giant migrating contractions during defecation in the dog colon. Gastroenterology 1987;92:92533.
Received January 8. 1988. Accepted June 6, 1988. Address requests for reprints to: Dr. Sushi1 K. Sarna, Zablocki Veterans Administration Medical Center, Surgical Research 151. 5000 West National Avenue, Milwaukee, Wisconsin 53295. This work was supported in part by grants from the National Institutes of Health (DK 32346). the Veterans Administration Medical Research Service (7722-OlP). and the Cancer Center of the Medical College of Wisconsin. The authors thank Dr. John Kalbfleisch for help in the statistical analysis of data and Mary Farrar for expert secretarial assistance in the preparation of the manuscript. This work was done with the technical assistance of Dawn Kubacki and Robert Ryan.