Small intestinal dysmotility following abdominal

irradiation in the rat small intestine

R. FRASER,1 C. FRISBY,1 L. A. BLACKSHAW,2 M. SCHIRMER,3 G. HOWARTH4

& E. YEOH3

1Departments of Medicine, 2Gastrointestinal Medicine and 3Radiation Oncology, Royal

Adelaide Hospital, Adelaide, South Australia4Cooperative Research Centre for Tissue Growth and Repair, Child Health Research Institute, Adelaide, South Australia

Abstract Abdominal symptoms such as diarrhoea,

abdominal cramps and vomiting are common during

and after abdominal radiotherapy for gynaecological

and pelvic malignancy. It has recently been recog-

nized that small intestinal dysmotility may contrib-

ute to these symptoms but the underlying

mechanisms are unclear in part because of the tech-

nical dif®culties inherent in performing studies in ir-

radiated small intestine. The aim of the current study

was to evaluate small intestinal motor activity using

perfused micromanometric techniques in 6±8-cm seg-

ments of ileum during arterial perfusion with isotonic

oxygenated ¯uorocarbon solution. Intestinal segments

from six rats were studied 4 days after treatment with

10 Gy abdominal irradiation. Ileal segments from

nine nonirradiated animals acted as controls. For each

experiment the total number of pressure waves, high-

amplitude (>20 mmHg, long-duration >6 sec) pressure

waves, and long (>20 associated) bursts of pressure

waves were determined.

Irradiation had no effect on the overall number of

pressure waves, but increased high-amplitude long-

duration (HALD) pressure waves (248 vs 7, P < 0.01).

In control animals HALD waves were localized to a

single recording site but after radiotherapy 74% of

HALD waves were temporally associated with similar

pressure waves in other manometric channels. Forty-

seven per cent of associated HALD waves migrated

aborally. Retrograde migration of HALD waves was

seen in ®ve segments following irradiation. Irradiation

abolished bursts of >20 pressure waves.

High-amplitude contractions which migrate ab-

orally are likely to contribute to diarrhoea after

abdominal irradiation. The in vitro small animal

model of radiation enteritis will permit interventions

to provide further insights into the mediation of in-

testinal dysmotility.

Keywords motility, small intestine, radiation, en-

teritis.

INTRODUCTION

Diarrhoea, vomiting and abdominal cramps are well

recognized sequelae during and following radiation

treatment of abdominal or gynaecological malignan-

cy.1,2 These symptoms compound the physical and

psychological effects of the malignant disease and may

delay completion of treatment with subsequent com-

promise of cure.3±5 Although traditionally considered to

re¯ect mucosal injury, it has more recently been appre-

ciated that small intestinal transit is also acutely dis-

turbed following radiotherapy in both humans6±8 and

animals9 and contributes to morbidity after therapy.6±8

Therapy for acute radiation enteritis is largely symptom

driven as there is limited understanding of the patho-

physiological mechanisms which underlie the illness.

Direct in vivo evaluation of changes on motor ac-

tivity in humans is dif®cult because of ethical con-

siderations. The motor patterns following both

fractionated and single dose irradiation have, however,

been studied in a variety of animal models.10±15

The development of a technique in which an isolated

small intestinal loop is arterially perfused with an oxy-

genated ¯uorocarbon solution has allowed more de-

tailed investigation of local control of in vitro ileal

motility in small laboratory animals.16,17 One of the

advantages of this preparation is that it optimizes the

viability of mucosal and enteric neural tissue

throughout the study period. This approach has been

Address for correspondenceDr R. Fraser, Department of Medicine, Royal Adelaide Hos-pital, Adelaide, SA 5000. Tel: +61 8 8222 5502; fax: +61 8 82225934.Received: 10 December 1997Accepted for publication: 2 June 1998

Neurogastroenterol. Mot. (1998) 10, 413±419

Ó 1998 Blackwell Science Ltd 413

modi®ed to incorporate the inclusion of a micro-

manometric catheter to provide point sensor data from

along the segment.18 The aim of the current study was

to examine the acute effects of abdominal radiotherapy

on motor patterns using this ileal loop.

METHODS

Animals

Studies were performed on 15 Sprague±Dawley rats

with mean weight 244 � 7 g. Six animals underwent

irradiation whilst nine nonirradiated animals were

used as controls. Animals were housed individually

with free access to food and water. The water con-

tained Vytrate, a concentrated electrolyte solution

(Beecham Animal Health, Dandenong, Australia). An-

imals were fed a high-carbohydrate powder containing

180 g casein plus 2.5 g of methionine per kg. A condi-

tion of the study was that any animal that lost greater

than 15% of its body weight would be killed, but in

practice no animal reached this endpoint. The study

was approved by the Animal Ethics Committees of the

University of Adelaide, the Institute of Medical and

Veterinary Science and the Animal Care and Ethics

Committee of the Women's and Children's Hospital,

Adelaide.

Protocol

Prior to abdominal irradiation, animals were anaes-

thetized with halothane (2±3%) in a purpose-built

perspex box. Irradiation at 10 Gy was administered to

the abdomen by a Phillips Deep X-Ray Unit (RT 250,

Phillips, Eindhoven, Holland) mid-separation to supine

animals. A half value layer of 2 mm copper was used

with a focus to skin distance of 50 cm. Animals were

allowed to recover fully from anaesthesia before being

returned to the cage.

Four days post-irradiation rats were anaesthetized

with intraperitoneal pentobarbitone (60 mg kg)1). (Day

four was chosen as this has previously been shown as

the time when mucosal damage is maximal.18,19) A

laparotomy was performed and the caecum located. A

6±8-cm segment of the terminal ileum was selected

adjacent to the caecum and a branch of the superior

mesenteric artery cannulated and immediately per-

fused with ¯uorocarbon emulsion, FC-43 (Green Cross

Corp., Osaka, Japan), at a rate of 1 mL min)1. The

segment of ileum was then explanted to an organ bath

containing Krebs±bicarbonate solution maintained at

37 °C with continued intra-arterial perfusion. A mul-

tilumen micromanometric assembly was inserted from

the oral end into the lumen of the intestine for man-

ometric recordings. Pressure waves were recorded via

six side holes 1 cm apart. Each channel was perfused at

0.04 mL channel)1 with degassed saline. Pressure

waves were recorded using purpose-built software

(MAD software, Royal Adelaide Hospital, Charles

Malbert) onto a Macintosh computer for later analysis.

Ileal segments were allowed 15 min to equilibrate be-

fore intestinal motility was recorded for 25 min.

Data analysis

Histology In each animal histological samples were

collected from intestinal sites adjacent to the ileal loop

for assessment of radiation damage. These were ®xed

in paraf®n and sections stained with haematoxylin and

eosin for light microscopic evaluation. A semiquanti-

tative analysis of histological damage was performed

using previously published criteria.19 A damage score

out of a possible maximum of 30 was obtained for each

animal.

Motility Pressure waves >5 mmHg in amplitude were

scored manually and motility patterns identi®ed. The

total number of pressure waves together with the mean

amplitude of pressure waves was determined for each

channel for the duration of the study. Pressure waves

with an amplitude greater than 20 mmHg and 6 s in

duration were scored as high-amplitude long-duration

(HALD) pressure waves. The distance and direction of

migration of HALD pressure waves were determined.

The organization of low-amplitude pressure waves was

also analysed. The number of bursts of associated

pressure waves >20 in sequence with a frequency of

between 10 and 15 min)1 recorded by each individual

sidehole was determined.

Statistical analysis

All results are expressed as mean � SEM. Differences

between irradiated and control animals were analysed

using an unpaired t-test. A P value of <0.05 was con-

sidered signi®cant in all analyses.

RESULTS

For the 4 days after treatment irradiated rats ate less

food and drank less water compared with the preirra-

diation period. This was associated with a mean weight

loss of 33 � 5.4 g. Consistent with a diagnosis of radi-

ation injury, ®ve of the six rats also developed diar-

rhoea characterized by loss of pellet formation.

414 Ó 1998 Blackwell Science Ltd

R. Fraser et al. Neurogastroenterology and Motility

Histology

Radiation damage to the mucosa was apparent histo-

logically in all irradiated animals with moderate crypt

hypoplasia, fusion and stunting of villi and in®ltration

of in¯ammatory cells. There was a signi®cant reduc-

tion in the villus to crypt ratio of irradiated compared

to nonirradiated control animals (1.08 � 0.83 vs

2.5 � 0.118, P < 0.001). The histological damage score

was signi®cantly higher in the irradiated rats when

compared to the control animals (14.0 � 0.271 vs

0.278 � 0.08, P < 0.001).

Motility

In segments from control animals the motor pattern

was composed predominantly of low-amplitude pres-

sure waves which were frequently organized in bursts

of >20 (Fig. 1). A total of seven high-amplitude long-

duration (HALD) pressure waves were seen in the nine

ileal segments from control rats. These were always

localized to a single pressure sensor. Following irradi-

ation there was no change in the total number of

pressure waves (Fig. 2) but the organization of motor

activity was markedly different. The most obvious

change was the increase in the number of HALD

pressure waves (Figs 3 and 4). After irradiation a total

of 248 HALD pressure waves were recorded. Of these,

66 were localized to a single sidehole and 182 were

recorded across multiple sideholes. The mean spatial

association length of HALD pressure waves was

2.63 � 0.58 cm. Association between HALD waves was

seen in all irradiated ileal loops. HALD waves fre-

quently migrated in an aboral direction (n � 33) or

occurred simultaneously in a number of pressure

channels (n � 17) (Table 1). In addition retrograde

migration of HALD pressure waves (n � 15) was ob-

served in ®ve of the six ileal segments from irradiated

animals (Fig. 5). The mean amplitude of HALD pres-

sure waves in the segments from irradiated rats was

33.7 � 2.6 mmHg vs 40.0 � 14.3 mmHg in segments

from control animals (P ns). There was no difference in

the duration of HALD pressure waves between irradi-

ated and control animals (10.8 � 0.95 s vs

10.1 � 10.81 s, respectively, P ns). There was no dif-

ference in the amplitudes (33.2 � 0.9 mmHg vs

33.4 � 1.3 mmHg, P � ns), duration (9.0 � 1.1 sec vs

9.3 � 2.2 sec, P � ns), distance of migration (2.4 �

0.3 cm vs 2.2 � 0.3 cm, P � ns) or velocity of migra-

Figure 1 Example of manometric tracingfrom isolated arterially perfused ileal seg-ment of a control rat. In addition to frequentsporadic pressure wave activity, there arealso prolonged bursts (>20) of pressure wavesin the second and third channels. Sideholesare located 1 cm apart with most proximalchannel (labelled oral) at the top.

Figure 2 Group data for total number of pressure waves perchannel during entire study for isolated arterially perfusedileal segments from control and irradiated rats. There was nosigni®cant difference between the two groups. Data aremeans � SEM.

Ó 1998 Blackwell Science Ltd 415

Volume 10, Number 5, October 1998 Dysmotility in acute radiation enteritis

tion (4.9 � 0.6 cm sec)1 vs 4.5 � 1.0 cm sec)1, P � ns)

between antegrade and retrograde HALD waves. No

long bursts of low-amplitude pressure waves were re-

corded after irradiation but the reduction was not sta-

tistically signi®cant when compared to control

animals (Fig. 6).

DISCUSSION

The current study shows that moderate radiation en-

teritis is associated with an alteration in the patterning

but not rate of small intestinal pressure waves. The

most obvious change was an increased number of high-

amplitude long-duration pressure waves compared to

control animals. In addition the organization of high-

amplitude pressure waves was also markedly changed

following irradiation, with both antegrade and retro-

grade migration as well as simultaneous onset of

pressure waves along the ileal segment, in contrast to

the ileal loops from control animals where migration

did not occur. There was also a trend for a decrease in

the number of bursts of contractions.

Antegrade migrating high-amplitude contractions

have been previously described in arterially perfused

isolated rat ileum where they were reported to be the

major mechanism for propulsion of intraluminal con-

tents from the ileal segments.16,17 In these studies, the

frequency of HALDs increased dramatically in re-

sponse to intraluminal distension. Their appearance in

the current study where intraluminal distension was

minimal (restricted to the volume of manometric per-

fusate, » 0.5 mL h)1) was unexpected. The mecha-

nisms responsible for stimulation are unclear. In the

intact animal giant migrating contractions (GMC) have

been reported in response to noxious stimuli.20,21

GMCs have also been reported following irradia-

tion10,11 and may be attributable to a heightened sen-

sitivity of intestinal re¯exes to intraluminal bile.12

Whether or not previously in vivo reported giant

migrating contractions correspond to the in vitro high-

amplitude pressure waves observed in our study is

unclear. In a previous study, it was proposed that

similar contractions in isolated rat ileum were in vitro

correlates of the GMC.17 In dogs and humans, GMCs

typically have an amplitude and duration 2±3 times

that of normal contractions and migrate for long dis-

tances along the intestine.22 Although increased fol-

lowing radiotherapy, GMCs are still relatively

uncommon. The increased amplitude and duration of

Figure 3 Example of manometric tracingfrom arterially perfused segment of rat ileum4 days after abdominal irradiation. Frequentsporadic pressure waves are seen but there areno prolonged bursts (>20) of pressure waves.High-amplitude long-duration (HALD) pres-sure waves can be seen which migrate in anantegrade fashion. The tracing is arranged asin Fig. 1.

Figure 4 Group data for the number of high-amplitude long-duration (HALD) pressure waves per channel during the entirestudy for isolated arterially perfused ileal segments fromcontrol and irradiated rats. There was a signi®cant increase inthe number of HALD pressure waves after irradiation(P < 0.003). Data are means � SEM.

416 Ó 1998 Blackwell Science Ltd

R. Fraser et al. Neurogastroenterology and Motility

the HALD pressure waves in our study is similar to

that previously reported in GMCs. In addition the

HALD pressure waves migrated at a rapid rate. It is

thus tempting to suggest that HALD pressure waves

correspond to GMCs. However, the short extent over

which migration could be assessed (due to the limited

length of the ileal segment studied) makes it dif®cult

to be certain that HALD pressure waves represent the

local manifestation of GMCs. Thus whilst it is possible

that HALD contractions may be related to these motor

activities, they may also be one component of the local

motor response which is normally combined with

nonpropulsive mixing motor activities. It is possible

that the minor degree of luminal distension by perfu-

sate may result in the stimulation of HALD type

pressure waves. Direct mechanical stimulation of the

mucosa by the micromanometric assembly itself may

also be a factor in the production of the HALD con-

traction, although the effect in control rats was mini-

mal.

In ®ve of the six preparations from irradiated animals

retrograde migration of the high-amplitude contrac-

tions was observed. This activity did not occur in any

of the control preparations. Retrograde migration of

high-amplitude contractions has not been previously

described in an isolated small intestinal preparation.

Contractions which migrate in a retrograde fashion

have been described in association with aboral pro-

pulsion of contents as occurs with vomiting in the

intact animal.10,23,24 Our data suggest, however, that

retrograde migration of motor activity can be coordi-

nated at least over short distances by neural networks

in the small intestine.

Because of the complex hierarchical arrangement of

the control mechanisms underlying small intestinal

motor function it is dif®cult to evaluate the disruption

of individual pathways in the intact animal. Previous

studies on both intestinal mechanics and the motor

effects of hormones have utilized arterially perfused

ileal segments16,17,25±28 but such techniques have not

been previously applied to the study of dysmotility

associated with radiation enteritis. The use of an in

vitro preparation offers a number of advantages. As-

sessment of the disruption of individual control path-

ways can be performed more easily in an in vitro

preparation when effects of central neural or humoral

counter-regulatory mechanisms are removed. In addi-

tion an in vitro preparation can be used to study

compounds which are too toxic or expensive for use in

Figure 5 Example of manometric tracingfrom isolated arterially perfused segment ofrat ileum 4 days after abdominal irradiationshowing retrograde migration of high-amplitude long-duration (HALD) pressurewaves. The tracing is arranged as in Fig. 1.

Figure 6 Group data for the number of prolonged bursts (>20)of pressure waves per channel during the entire study for ar-terially perfused ileal segments from control and irradiatedrats. There was a trend for a decrease in the number of burstsafter irradiation (P � 0.09). Data are means � SEM.

Ó 1998 Blackwell Science Ltd 417

Volume 10, Number 5, October 1998 Dysmotility in acute radiation enteritis

live animals. With the recent development of micro-

manometric techniques, it is also feasible to measure

intraluminal pressures from multiple points along

segments from different regions of rat small intestine

to provide greater insights into region-speci®c dys-

functions.

There are limited data about the effect of small in-

testinal pressure waves on localized patterns of transit.

Transit has been shown to be most rapid when con-

tractions migrate distally and slowed when bursts of

contractions are present29. The effect of irradiation on

luminal ¯ow was not determined in the current study.

Single sublethal doses of abdominal irradiation have

been shown in previous studies to accelerate small

intestinal transit.9 In addition the occurrence of high-

amplitude contractions which migrate in an antegrade

direction results in luminal expulsion of ¯uid.16 It

therefore seems likely that the motor activity seen in

the segments from irradiated animals (HALD pressure

waves and reduced bursts) would result in more rapid

aboral movement of intraluminal contents.

In the current study animals received a single high

dose of irradiation to produce intestinal damage in

contrast to the more usual fractionated radiotherapy

used in clinical practice. Repeated anaesthesia in ro-

dents via the usual intraperitoneal injection of barbi-

turates is, however, associated with an unacceptably

high mortality rate which increases with the number

of injections. In addition the injections themselves

may result in intestinal damage. Studies which have

examined the effects of fractionated radiotherapy on

motility have previously been performed in dogs,

which are expensive to purchase and maintain. It has

been reported that rats can safely undergo repeated

inhaled anaesthesia with halothane which permits

administration of fractionated doses of irradiation30

which would allow us to also evaluate the in vitro ef-

fects of fractionated irradiation using rats.

ACKNOWLEDGMENTS

This work was supported by grants from the Anti

Cancer Foundation of the Universities of South Aus-

tralia and the Government Employees Medical Re-

search Fund. We gratefully acknowledge the technical

advice provided by Dr P. Bercik and Dr L. Bouley on

explantation of the ileal segments.

REFERENCES

1 Kinsella TJ, Bloomer WD. Tolerance of the intestine toradiation therapy. Surg Gynecol Obstet 1980; 151: 273±84.

2 Yeoh E, Horowitz M. Radiation Enteritis. Surg GynecolObstet 1987; 165: 373±9.

3 Fyles A, Keane T, Barton M, Simm J. The effect of treat-ment duration in the local control of cervix cancer. Ra-diotherapy Oncol 1992; 25: 273±9.

4 Gill P, Denham J, Jamieson G, Devitt P, Yeoh E, Ol-weny C. Patterns of treatment failure and prognosticfactors associated with treatment of oesophageal carci-noma with chemotherapy and radiotherapy either as soletreatment or followed by surgery. J Clin Oncol 1992; 10:1037±43.

5 Amdur R, Parsons J, Fitzgerald L, Million R. The effect ofoverall treatment time on local control in patients withadenocarcinoma of the prostate treated with radiationtherapy. Int J Radiation Oncol Biol Phys 1990; 19: 1377±82.

6 Yeoh E, Horowitz M, Russo A, et al. A retrospective studyof the effects of pelvic irradiation for carcinoma of thecervix on gastrointestinal function. Int J Radiation OncolBiol Phys 1993; 26: 229±37.

7 Yeoh E, Horowitz M, Russo A, et al. Effect of pelvic irra-diation on gastrointestinal function: a prospective longi-tudinal study. Am J Med 1993; 95: 397±406.

8 Yeoh E, Horowitz M, snm>Russo A, Muecke T, Robb T,Chatterton B. Gastrointestinal function in chronic radia-tion enteritis-effects of loperamide-N-oxide. Gut 1993; 34:476±82.

9 Summers R, Kent T, Osborne J. Effects of drugs, ileal ob-struction and irradiation on rat gastrointestinal propulsion.Gastroenterology 1970; 59: 731±9.

10 Otterson M, Sarna S, Moulder J. Effects of fractionateddoses of ionizing radiation on small intestinal motor ac-tivity. Gastroenterology 1988; 95: 1249±57.

11 Otterson M, Sarna S, Lee M. Fractionated doses of ionisingradiation alter postprandial small intestinal motor activity.Dig Dis Sci 1992; 37: 709±15.

12 Otterson M, Leming S, Moulder J, Dawes L. Intraluminalbile enhances radiation induced giant contractions. Gas-troenterology 1996; 110: A728.

13 Summers R, Flatt A, Prihoda M, Mitros F. Effect of irra-diation on morphology and motility of canine small in-testine. Dig Dis Sci 1987; 32: 1402±10.

14 Summers R, Glenn C, Flatt A, Elahmady A. Radiation andindomethacin effects on morphology, prostaglandins andmotility in dog jejunum. Am J Physiol (Gastrointest LiverPhysiol 24) 1991; 261: G145±G151.

15 Summers R, Glenn C, Flatt A, Elahmady A. Does irradia-tion produce irreversible changes in canine jejunal myo-electric activity? Dig Dis Sci 1992; 37: 716±72.

16 Bercik P, Armstrong D, Fraser R, Dutoit P, Blum A, KuceraP. Luminal distension, motility patterns and luminaltransport in isolated arterially perfused rat ileum. Gastro-enterology 1993; 104: A476.

17 Bercik P, Armstrong D, Fraser R, et al. Origin of motilitypatterns in isolated arterially perfused rat intestine. Gas-troenterology 1994; 106: 649±57.

18 Fraser R, Frisby C, Schirmer M, et al. Effect of fractionatedabdominal irradiation on small intestinal motility. Gas-troenterology 1996; 110: A666.

19 Fraser R, Frisby C, Yeoh E, Schirmer M, Howarth G. Ef-fects of IGF-I on mucosal response to radiation-inducedinjury. Gastroenterology 1996; 110: A1072.

418 Ó 1998 Blackwell Science Ltd

R. Fraser et al. Neurogastroenterology and Motility

20 Otterson M, Sarna S. Neural control of small intestinalgiant migrating contractions. Am J Physiol 1994; 266:G576±G584.

21 Kruis W, Azpiroz F, Phillips S. Contractile patterns andtransit of ¯uid in canine terminal ileum. Am J Physiol(Gastrointest Liver Physiol 12) 1985; 249: G264±70.

22 Sarna S. Giant migrating contractions and their myoelec-tric correlates in the small intestine. Am J Physiol 1987;253: G697±G705.

23 Lang I, Marvig J, Sarna S. Comparison of gastrointestinalresponses to CCK-8 and associated with vomiting. Am JPhysiol (Gastrointest Liver Physiol 17) 1988; 254: G254±63.

24 Lang I, Sarna S. Motor and myoelectric activity associatedwith vomiting, regurgitation, and nausea. In: Handbook ofPhysiology ± the Gastrointestinal System 1, Chapter 32:1179±98. 1989.

25 Weems W, Seygal G. Fluid propulsion by cat intestinalsegments under conditions requiring hydrostatic work.

Am J Physiol (Gastrointest Liver Physiol 3) 1981; 240:G147±56.

26 Weems W. Intestinal wall motion, propulsion, and ¯uidmovement: trends towards a uni®ed theory. Am J Physiol1982; 243: G177±G188.

27 Weems W, Seidel E, Johnson L. Induction of a speci®cpattern of jejunal propulsive behaviour by cholecystokinin.Am J Physiol 1985; 248: G470±G478.

28 Weems W, Weisbrodt N. Ileal and colonic propulsive be-haviour: contribution of enteric neural circuits. Am JPhysiol 1986; 250: G653±G59.

29 Schemann M, Ehrlein H-J. Postprandial patterns of caninejejunal motility and transit of luminal content. Gastroen-terology 1986; 90: 991±1000.

30 Ang K, van der Kogel A, van der Schurfen E. Inhalationalanesthesia in experimental radiotherapy: a reliable andtime-saving system for multifractionation studies in aclinical department. Int J Radiation Oncol Biol Phys 1982;8: 145±8.

Ó 1998 Blackwell Science Ltd 419

Volume 10, Number 5, October 1998 Dysmotility in acute radiation enteritis