Gut 1993; 34:476-482 Gastrointestinal function in chronic radiation enteritis effects of loperamide-N-oxide E K Yeoh, M Horowitz, A Russo, T Muecke, T Robb, B E Chatterton Abstract The effects of loperamide-N-oxide, a new peripheral opiate agonist precursor, on gastro- intestinal function were evaluated in 18 patients with diarrhoea caused by chronic radiation enteritis. Each patient was given, in double-blind randomised order, loperamide- N-oxide (3 mg orally twice daily) and placebo for 14 days, separated by a washout period of 14 days. Gastrointestinal symptoms; absorp- tion of bile acid, vitamin B12, lactose, and fat; gastric emptying; small intestinal and whole gut transit; and intestinal permeability were measured during placebo and loperamide-N- oxide phases. Data were compared with those obtained in 18 normal subjects. In the patients, in addition to an increased frequency of bowel actions (p<0001), there was reduced bile acid absorption, (p<0001) a higher prevalence of lactose malabsorption (p<005) associated with a reduced dietary intake of dairy products (p<002), and faster small intestinal (p<0001) and whole gut transit (p<O0O5) when compared with the normal subjects. There was no signifi- cant difference in gastric emptying between the two groups. Treatment with loperamide- N-oxide was associated with a reduced frequency of* bowel actions (p<0001), slower small intestinal (p<0001), and total gut transit (p<001), more rapid gastric emptying (p<001), improved absorption of bile acid (p<001), and increased permeability to 51Cr EDTA (p<0.01). These observations indicate Departments of Radiation Oncology, Medicine, and Nuclear Medicine, Royal Adelaide Hospital, Adelaide, South Australia E K Yeoh M Horowitz A Russo T Muecke B E Chatterton Adelaide Children's Hospital, Adelaide, South Australia T Robb Correspondence to: Associate Professor M Horowitz, Department of Medicine, Royal Adelaide Hospital, North Terrace, Adelaide, South Australia, 5000 Australia. Accepted for publication 1 September 1992 that: (1) diarrhoea caused by chronic radia- tion enteritis is associated with more rapid intestinal transit and a high prevalence of bile acid and lactose malabsorption, and (2) loperamide-N-oxide slows small intestinal transit, increases bile acid absorption, and is effective in the treatment of diarrhoea associ- ated with chronic radiation enteritis. (Gut 1993; 34: 476-482) Chronic radiation enteritis is now recognised as a frequent and clinically important sequel of abdominal and pelvic irradiation treatment for malignant disease.'1S Diarrhoea, with or with- out abdominal cramps, is the most common symptom.2 Although intestinal stricture and associated bacterial overgrowth are well recog- nised in chronic radiation enteritis, in most cases the pathophysiology of the diarrhoea is uncertain. While changes in intestinal absorp- tion and motility unrelated to bacterial over- growth have been implicated in the aetiology of diarrhoea,'2 4 6 7 there has been no comprehensive evaluation of gastrointestinal function in chronic radiation enteritis. Perhaps partly as a result of this, present approaches to treatment have often been empirical. 'I This study aimed to evaluate various aspects of gastrointestinal function in patients with diar- rhoea caused by chronic radiation enteritis and the effects of treatment with the peripheral opiate agonist precursor loperamide-N-oxide. TABLE I Charactristics of the patient group Patient Age Body weight Body mass Radiation dose Radiationfield no (y) (kg) index Diagnosis Gy/no offractionsldays (Size - cmxcm) 1 68 69-2 25-6 Carcinoma of cervix 4495 Gy/25 F/32 16-5x 18-5 (25 Gy to pt 'A' intracavitary) 2 80 71-9 24-8 Carcinomaofprostate 59 76Gy/30F/51 15 5x17-5 3 80 72-5 24-7 Carcinoma of prostate 62-15Gy/31 F/44 15-5x19-0 4 63 58 8 27-7 Carcinoma of cervix No details available (treated in Chile) 5 76 62 5 27-6 Endometrialcarcinoma 46Gy/23F/31 17-5x15-5 6* 79 66-0 23 5 Endometrial carcinoma 50Gy/25 F/41 17-5 x 18-0 (20 Gy/vaginal mould) 7 85 58 0 23-6 Carcinomaofcervix 46Gy/23F/33 16-0x16-5 (20 Gy to Pt 'A' intracavitary) 8 90 54 0 21-3 Carcinomaofcervix 50Gy/25F/33 15-0x16-0 (20 Gy to Pt 'A' intracavitary) 9 70 540 22-5 Carcinoma of cervix 34 Gy/17 F/27 14-Ox 15-0 (45 Gy to Pt 'A' intracavitary) 10 42 62-0 22-2 Carcinoma of cervix 50Gy/25 F/36 17-0x210 11 63 67-0 28-3 Endometrial carcinoma 50 Gy/25 F/32 15 x 28 5 (abdomen) 15 x 17-5 (pelvis only) 12 73 67-0 25 9 Carcinoma of cervix 50 Gy/25 F/39 14x 19-5 13* 73 65 0 21-2 Carcinoma of prostate 64 Gyl32 Ff47 13-0x 19-5 (pelvis) 8- 5 x 9-5 (prostate) 14 72 59-0 22-5 Carcinomaofcervix 50Gy/30F/45 l5-0x14-0 15 78 70-0 29-2 Endometrial carcinoma 46-8 Gyl26 F/39 15-5 x 17-0 (20 Gy vaginal mould) 16 73 54-0 21-4 Carcinoma of ovary 50 Gy/26 F/39 15-Ox 18-0 17 81 74 0 25-3 Carcinoma of prostate 64 Gy/32 F/45 l5-Ox 14-5 18 63 57-0 18-8 Carcinomaofovary 49-9Gy/25 F/35 15-Ox20-0 19 48 114-0 51-4 Carcinomafallopiantube 55Gy/30F/40 15-0x20-O 20 57 80-0 28-4 Endometrial carcinoma 50 Gy/25 F/35 Not available *Patients withdrew from study. 476 on March 21, 2022 by guest. Protected by copyright. http://gut.bmj.com/ Gut: first published as 10.1136/gut.34.4.476 on 1 April 1993. Downloaded from
Transcript
Gut 1993; 34:476-482
Gastrointestinal function in chronic radiationenteritis effects of loperamide-N-oxide
E K Yeoh, M Horowitz, A Russo, T Muecke, T Robb, B E Chatterton
AbstractThe effects of loperamide-N-oxide, a new
peripheral opiate agonist precursor, on gastro-intestinal function were evaluated in 18patients with diarrhoea caused by chronicradiation enteritis. Each patient was given, indouble-blind randomised order, loperamide-N-oxide (3 mg orally twice daily) and placebofor 14 days, separated by a washout period of14 days. Gastrointestinal symptoms; absorp-tion of bile acid, vitamin B12, lactose, and fat;gastric emptying; small intestinal and wholegut transit; and intestinal permeability were
measured during placebo and loperamide-N-oxide phases. Data were compared with thoseobtained in 18 normal subjects. In the patients,in addition to an increased frequency of bowelactions (p<0001), there was reduced bile acidabsorption, (p<0001) a higher prevalence oflactose malabsorption (p<005) associatedwith a reduced dietary intake of dairy products(p<002), and faster small intestinal (p<0001)and whole gut transit (p<O0O5) when comparedwith the normal subjects. There was no signifi-cant difference in gastric emptying betweenthe two groups. Treatment with loperamide-N-oxide was associated with a reducedfrequency of* bowel actions (p<0001),slower small intestinal (p<0001), and total guttransit (p<001), more rapid gastric emptying(p<001), improved absorption of bile acid(p<001), and increased permeability to 51CrEDTA (p<0.01). These observations indicate
Departments ofRadiationOncology, Medicine, andNuclear Medicine, RoyalAdelaide Hospital,Adelaide, South AustraliaE K YeohM HorowitzA RussoT MueckeB E Chatterton
Adelaide Children'sHospital, Adelaide,South AustraliaT RobbCorrespondence to:Associate Professor MHorowitz, Department ofMedicine, Royal AdelaideHospital, North Terrace,Adelaide, South Australia,5000 Australia.Accepted for publication1 September 1992
that: (1) diarrhoea caused by chronic radia-tion enteritis is associated with more rapidintestinal transit and a high prevalence of bileacid and lactose malabsorption, and (2)loperamide-N-oxide slows small intestinaltransit, increases bile acid absorption, and iseffective in the treatment of diarrhoea associ-ated with chronic radiation enteritis.(Gut 1993; 34: 476-482)
Chronic radiation enteritis is now recognised as a
frequent and clinically important sequel ofabdominal and pelvic irradiation treatment formalignant disease.'1S Diarrhoea, with or with-out abdominal cramps, is the most common
symptom.2 Although intestinal stricture andassociated bacterial overgrowth are well recog-nised in chronic radiation enteritis, in mostcases the pathophysiology of the diarrhoea isuncertain. While changes in intestinal absorp-tion and motility unrelated to bacterial over-
growth have been implicated in the aetiology ofdiarrhoea,'2 4 6 7 there has been no comprehensiveevaluation of gastrointestinal function in chronicradiation enteritis. Perhaps partly as a result ofthis, present approaches to treatment have oftenbeen empirical. 'I
This study aimed to evaluate various aspects ofgastrointestinal function in patients with diar-rhoea caused by chronic radiation enteritis andthe effects of treatment with the peripheralopiate agonist precursor loperamide-N-oxide.
TABLE I Charactristics ofthe patient group
Patient Age Body weight Body mass Radiation dose Radiationfieldno (y) (kg) index Diagnosis Gy/no offractionsldays (Size - cmxcm)
1 68 69-2 25-6 Carcinoma of cervix 4495 Gy/25 F/32 16-5x 18-5(25 Gy to pt 'A' intracavitary)
2 80 71-9 24-8 Carcinomaofprostate 59 76Gy/30F/51 15 5x17-53 80 72-5 24-7 Carcinoma of prostate 62-15Gy/31 F/44 15-5x19-04 63 58 8 27-7 Carcinoma of cervix No details available (treated in Chile)5 76 62 5 27-6 Endometrialcarcinoma 46Gy/23F/31 17-5x15-56* 79 66-0 23 5 Endometrial carcinoma 50Gy/25 F/41 17-5 x 18-0
Gastrointestinalfunction in chronic radiation enteritis- effects ofloperamide-N-oxide
Methods
SUBJECTSTwenty patients (16 women, four men, medianage 73 years (range 42-90), median body weight65 kg (54-114 kg) and median body mass index(BMI) 25d1 (18-8-31-4)) with persistent diar-rhoea 3-22 years after therapeutic pelvic irradia-tion for carcinoma of the genitourinary tract (seeTable I for individual patient characteristics)were studied. Persistent diarrhoea was defined asa chronic doubling ofthe frequency of defecationbefore irradiation and a minimum frequency ofat least 14 bowel actions a week. Patients whohad had gastrointestinal surgery, apart fromuncomplicated appendicectomy, were excluded.Other criteria for exclusion were inflammatorybowel disease, pernicious anaemia, evidence ofhepatic or renal dysfunction (any biochemicalparameter for hepatic function exceeding twicethe normal range or plasma creatinine >0-12mmol/l, or both), and a constant requirement formedication (such as anticholinergics, antidepressants, or non-steroidal anti-inflammatorydrugs) which might influence gastrointestinalmotility, or otherwise interfere with the evalua-tion of the test drug. Significant small intestinalstrictures were excluded by small bowel radio-logical investigations.
Eighteen normal volunteers (11 women, sevenmen, median age 45 years (19-75), median bodyweight 75 kg (45-109kg), and median BMI 26-0(17-8-38-7) who had no history of gastro-intestinal disease and were not taking anymedication were also studied. The median age inthe normal volunteers was less than that in thepatient group (p<0-01) but there was no signifi-cant difference in BMI between the two groups.All of the patients and all but one of the controlswere white.
PROTOCOLEach of the 20 patients was given, in double-blind randomised order, loperamide oxidetablets (Janssen Pharmaceutica Beerse,Belgium) in a dose of 3 mg twice daily by mouth(0800 h and 2000 h) and placebo (identical tabletswithout loperamide oxide) for 14 days, separatedby a wash-out period of 14 days. Between days 3and 14 after starting either loperamide oxide orplacebo, each patient underwent measurementsof the following aspects of gastrointestinalfunction: (i) gastrointestinal symptoms (day 14);(ii) absorption of bile acid and vitamin B12 (days3 to 10), lactose (day 6), dietary fat and stoolweight (day 10); (iii) gastrointestinal transit:gastric emptying, small intestinal and whole guttransit (day 7); and (iv) intestinal permeability(days 13, 14). Because of methodologicalpracticalities, absorption of bile acid and vitaminB12 was determined before the measurements ofgastrointestinal transit and intestinal perme-ability was measured last.Each patient had a complete blood examina-
tion, multiple biochemical analysis of plasma,and ECG on recruitment into the study and atthe end of each phase of the study. Any adverseeffects from the trial medication were also docu-mented at the end of each phase of the study.
The control subjects did not receive eitherplacebo or loperamide oxide tablets but under-went an identical series of measurements on oneoccasion.The study protocol was approved by the
Ethics Committee of the Royal AdelaideHospital on March 30, 1989 and each subjectgave written informed consent.
MEASUREMENTS
Gastrointestinal symptomsThe following symptoms were assessed byquestionnaire on entry into the study and at theend of each phase of the study8: (i) nausea, (ii)vomiting, (iii) abdominal pain, (iv) frequency ofbowel actions (number of bowel actions perweek) and stool consistency (hard formed, softformed, loose or watery). Except for (iv), eachsymptom was scored according to the followingclassification: 0 - symptom absent; 1 - mild,symptom could be ignored if the subject did notthink about it; 2 - moderate, symptom could notbe ignored but did not influence daily activities;3 - severe, symptom influenced daily activities.8The total score (maximum 9) was then calculatedfor each phase of the study. The frequency andconsistency of bowel actions for the 2 weekperiods on loperamide oxide and placebo andduring the wash-out period were recorded in adiary. The number of bowel actions each weekwas calculated. The number of bowel actionsin each three day faecal collection was alsorecorded.A questionnaire was also used to record
dietary intake of milk products and fat and thepresence or absence of intolerance to milk ormilk products (symptoms of nausea, abdominalbloating, abdominal cramps, or diarrhoea) aftereach subject had been enrolled into the study.Intake of dairy foods was expressed as calciumintake/week and fat intake was classified as1=low, 2=medium, or 3=high.9
AbsorptionBile acid and vitamin B12 absorption weredetermined by measuring the whole bodyretention at 7 days of selenium-75 labelled homo-taurocholic acid (SeHCAT) and cobalt-58 label-led vitamin B12 (58Co vitamin B12).46 On day 1,after a light breakfast of tea and toast at 0700 h,one capsule of 58Co vitamin B12 (AmershamInternational) containing 30 kBQ 58Co wasswallowed with 30 ml water at 0900 h. Thephoto-peak counts corresponding to the 808 keVphoton energy of 58Co and the 265 keV photonenergy of 75Se (to correct for down scatter fromthe higher energy, 58Co window into the lowerenergy 75Se window) were recorded from awhole body counter. One capsule of SeHCAT(Amersham International) containing 37 kBQ75Se was then swallowed with 30 ml water. Fourhours later the number of counts in both the 58Coand 75Se photo peak windows were determ-ined.67 This figure, after correction for roombackground counts (and down-scatter countsfrom the 58Co into the 75Se photo peak windows)was considered to represent 100% whole body
477
on March 21, 2022 by guest. P
rotected by copyright.http://gut.bm
j.com/
Gut: first published as 10.1136/gut.34.4.476 on 1 A
retention. There was no significant differencebetween the 58Co counts at baseline and at4 hours. Seven days later at 0900 h the counts inboth 58Co and 75Se photo peak windows wereagain measured. This number, after correctionfor room background counts, radionuclidedecay, and Compton scatter was used to derivethe percentage whole body retention of bile acidand vitamin B 12.67 Although the secondmeasurement of vitamin B12 and bile acidabsorption was made at least 4 weeks later, acorrection for patient background counts wasmade by first counting the residual counts inboth the 58Co and 75Se photo peak windowsbefore the administration of the second doses of58Co vitamin B12 and SeHCAT. Otherwise anidentical protocol was observed.
Dietary fat absorption was assessed bymeasuring the fat content in a 3 day collection offaeces, using a standard method modified forextraction of fatty acids into chloroform.'0Lactose absorption was assessed by measure-
ment of breath hydrogen concentrations after anoral lactose load. Each subject fasted overnightand at 0900 h was given 50 g lactose dissolved in200 ml water, after an antiseptic mouth wash anda baseline end expiratory breath sample had beentaken. Subsequent end expiratory breathsamples were obtained at least every 30 minutesfor up to 4 hours. All breath samples wereanalysed for hydrogen concentration using asensitive chromatographic method, capable ofdetecting concentrations as low as 1 ppm. Asustained peak in breath hydrogen greater than10 ppm was taken to indicate malabsorption oflactose." The day after the performance of thistest, patients were asked (by one of the investi-gators who was unaware of the hydrogen analy-sis) whether symptoms of nausea, abdominalbloating, cramps, or diarrhoea occurred within8 hours of the lactose load."
Gastrointestinal transitAt 0830 h, after an overnight fast (from 2200 h on
TABLE II Results in patients during the placebo phase ofthe study and in control subjects
Parameter Control Patients (placebo) p Value
No of subjects 18 18Calcium intake (mg/wk) 3155 (65-9486) 1475 (0-8190) <0 02Gastrointestinal symptoms:
Data are median values and ranges (analysis by Mann-Whitney U test).
the evening before the test) each subject ingested200 ml water containing 10g lactulose (Duphalac-Duphar, Holland), 20 g dextrose, 40 mBQ99mTc-sulphur colloid and 50 radio-opaqueplastic markers over a period of 2 minutes."24The subject then laid supine under a wide fieldview gamma camera. A cobalt marker was placedat the right anterior superior iliac crest to aidalignment of images. Data were acquired for aminimum of 5 hours. For the first 30 minutesafter meal completion data were sampled everyminute, followed by 3 minute frames for theremainder of the study. End expiratory breathsamples were taken at 10-15 minute intervalsafter ingestion of the meal for breath hydrogenanalysis. During the study the subject wasallowed to sit up, or walk around briefly if theywished every 60 minutes.
Regions of interest were drawn around thestomach and colon. 12 4 After data were correctedfor radionuclide decay, scatter, and gammaray attenuation, using previously describedmethods," time-activity curves characteris-ing gastric emptying and colonic filling werederived. By a process of computer subtraction atime activity curve representing small intestinaltransit was also obtained. 'ITime zero was considered the time of comple-
tion of the test meal. A sustained rise in thebreath hydrogen concentration >5 ppm wasconsidered to represent the arrival of the meal atthe caecum. 16 For gastric emptying, the lag phasebefore any of the meal emptied from the stomachand the 50% emptying time were determined.For small intestinal transit the following para-meters were determined: (i) start of colonicfilling (arrival of head of meal assessed scinti-graphically and by breath hydrogen measure-ments)'6; (ii) small intestinal transit (start ofcolonic filling minus the lag phase for gastricemptying); (iii) small intestinal residence (areaunder derived small intestinal transit timeactivity curve); (iv) time for 50% colonic filling. 14
Whole gut transit was determined radiographic-ally by the first arrival of the radio-opaquemarker(s) and the time when 50% or more of theradio-opaque markers were in the stool. 2
Intestinal permeabilityThis was assessed by two tests.(1) Urinary excretion of51CrEDTA. Each subjectwas initially asked to collect their urine for 24hours. At 0930 h after an overnight fast (from2400 h the subject ingested 2-10 ml of 51CrEDTA (Australian Atomic Energy Commission,code no CRIMI) containing 750 kBQ of 5'Cr,followed by 300 ml water. Fasting was main-tained for a further 2 hours, after which a normaldiet was permitted. Patients were instructed totake no alcohol and to collect all urine for thenext 24 hours. The entire 24 hour collection wascounted in a large volume counter. The 24 hoururine collection before 51Cr EDTA was alsocounted, to correct for any background radio-activity derived from 75Se from the bile acidabsorption test. The. result was expressed as apercentage of administered dose. " l
(2) Urinary excretion ofa double sugar-lactuloselrhamnose. After an overnight fast (from 2400 h,
478
on March 21, 2022 by guest. P
rotected by copyright.http://gut.bm
j.com/
Gut: first published as 10.1136/gut.34.4.476 on 1 A
Gastrointestinalfunction in chronic radiation enteritis - effects ofloperamide-N-oxide
each subject- was instructed to urinate at 0900 hand was then given 22-6 g glucose, 5 g lactulose,and 1 g L-rhamnose in 100 ml of distilled water.Thirty minutes after ingestion ofthe test solutionthe patient was encouraged to drink and wasinstructed to collect urine over the next 5 hours.The concentration of sugars in urine wasmeasured by high performance thin layerchromotography. After correction of the rham-nose values for urea concentration (the latter wasdetermined in a 1 ml aliquot of urine), the resultswere expressed as a ratio of the 5 hour urinaryexcretions of lactulose and rhamnose.'9
STATISTICAL ANALYSISData were evaluated using the Wilcoxon ranksum test (paired measurements), the Mann-Whitney U test (unpaired measurements), the x2test, and linear regression analysis. A p value of<0 05 was considered significant. Data areexpressed as median values and ranges.
ResultsEighteen of 20 patients completed the study.The two patients who withdrew (numbers 6 and13) felt they could no longer cope with theprogramme of evaluations of gastrointestinalfunction required in the protocol. No significantadverse effects were reported and no significantchanges in haematological, biochemical, andelectrocardiographc parameters were docu-mented. One scintigraphic assessment of gastricemptying and intestinal transit (in patient 5), oneurine collection for the double sugar lactulose/rhamnose permeability test (patient 10) and onefaecal fat estimation (in patient 11) were lost. Inconsequence only 17 paired measurements wereused in some cases for the comparison betweenplacebo and loperamide oxide phases of thestudy.
COMPARISON BETWEEN CONTROL SUBJECTS ANDPATIENTS DURING PLACEBO TREATMENTTable II summarises the results for the patients
TABLE III Results in patients during the placebo and loperamide phases
Data are median values and ranges (analysis by Wilcoxon test).
and the control subjects during the placebophase of the study. There was no evidence ofsmall intestinal bacterial overgrowth in thepatients - that is, no early rise in breath hydrogenconcentrations after oral lactose or lactulose.Both the number of bowel actions per week(assessed by diary) and stool frequency per 3days (assessed by 3 day faecal collection) weregreater in the patients than in the control group(p<0-001). There was no significant differencein stool weight. Except for the frequency ofbowel actions, there was no difference in gastro-intestinal symptoms between the patients andcontrols. Bile acid absorption was considerablyreduced in the patients (p<0-001), but there wasno significant difference between the two groupsin vitamin B12 absorption. Faecal fat excretionwas less in the patients (p<0-01). Dietary intakeof diary products was lower (p<0 02) and fatintake also tended to be less (p=0 09) in thepatients. Ten patients had lactose malabsorp-tion, compared with two of the control subjects(p<005). Of the patients with lactose mal-absorption, only two had a history of gastro-intestinal intolerance to milk or milk products.Despite this, dietary intake ofdairy products wasless in patients with lactose malabsorption thanin the remainder ofthe group (213 mg (0-4940) v2354 mg (320-8190) p<005). Gastrointestinalsymptoms after the lactose load were reported byfour patients. There was no significant differencein gastric emptying between the two groups, butsmall intestinal transit (p<0001) and whole guttransit (p<005) were faster in the patients thanin the control subjects. There was no significantdifference in intestinal permeability between thetwo groups. In the control group there was asignificant correlation (r>0-48, p<0 05)between bile acid absorption and whole guttransit (first radio-opaque marker and 50%markers). In the patients, stool frequency per 3days was inversely related to bile acid absorption(r=-0 49, p<005). Otherwise there were nosignificant relationships between the frequencyof bowel actions or stool weight and otherparameters of gastrointestinal function in eitherthe patients or the control subjects.
EFFECTS OF LOPERAMIDE OXIDETable III summarises the results during theplacebo and loperamide oxide phases of thestudy. Loperamide oxide decreased stool fre-quency (p<0-001) and weight (p<001). Aftertreatment with loperamide oxide the number ofbowel actions was still greater in the patientsthan in the control subjects (p<0 05) but thestool weight was less (p<0-01) (Table IV). Bileacid absorption was increased appreciably byloperamide oxide (p<001) (Figure) and, aftertreatment, was not significantly different fromvalues in the control group (Table IV). Some-what surprisingly, treatment with loperamideoxide was associated with a small but significantdecrease (p<005) in vitamin B12 absorption.Loperamide oxide did not affect faecal fat excre-tion, but during treatment with this drug therewas an inverse relationship between bile acidabsorption and faecal fat excretion (r=-0-63,p<0-01). There was no difference between the
479
on March 21, 2022 by guest. P
rotected by copyright.http://gut.bm
j.com/
Gut: first published as 10.1136/gut.34.4.476 on 1 A
Data are median values and ranges (analysis by Mann-Whitney U test).
two groups in the lag phase, but the time for 50%gastric emptying was less during treatment withloperamide oxide (p<0 01) and gastric emptyingwas faster in patients after treatment withloperamide oxide (p<0 05) than in control sub-jects (Table IV). In contrast, loperamide oxidedelayed both small intestinal (p<0 01) andwhole gut (p<001 first marker, p<0 05 50%markers) transit. During treatment with lopera-mide oxide, small intestinal transit and whole guttransit were not significantly different from thecontrol group (Table IV). There were significantrelationships between stool weight and bothwhole gut transit (r=-0-58, p<0 01) and stoolfrequency (r=0-45, p<005) during treatmentwith loperamide oxide. Intestinal permeability,as assessed by the 5'Cr EDTA test, was greater(p<O-Ol) after treatment loperamide oxide butthere was no difference in the absorption oflactulose/rhamnose (Table III). 51Cr EDTAabsorption was related to small bowel residenceduring both the placebo and loperamide oxidephases (r>0-45, p<0 05).
Absorptionbile acid
Absorption ofbile acid incontrol subjects and patientswith radiation enteritisduring treatment withplacebo and loperamideoxide. Data are shown asmedian values andinterquartile ranges.
40
3030
0CDo 20
CD4._
0
10
0Control
I- pc<0.01
II0 001|
Placebo Loperamideoxide
DiscussionDuring abdominal or pelvic irradiation, or both,acute radiation enteritis, manifest as diarrhoeawith or without abdominal cramps, is almostinevitable.' 6 The prevalence of chronic radiationenteritis, defined as the persistence of symptomsbeyond 3 months of the completion of radiationtreatment, is, however, uncertain. ' Several retro-spective surgical series suggest that the incidenceof 'significant' damage is only 5-15%.2)21 Thesestudies have considerable limitations, however.In particular, they do not include patients whodied or were lost to follow up in the prolongedinterval between treatment and the onset ofsymptoms. In contrast, a sustained increase inbowel frequency was noted in 12 of 17 unselectedwomen who had received pelvic radiotherapy forgynaecological malignancy 1-26 years pre-viously.2 It therefore seems probable that theincidence of chronic radiation enteritis has beengenerally underestimated (and may be more than50%), that increased bowel frequency is the mostfrequent symptom of chronic radiation enteritis,and that most patients with this condition do notseek medical help until the occurrence ofa severecomplication such as stricture, perforation, oranaemia caused by blood loss.' There is someevidence that the incidence of radiation boweldisease has increased in recent years as a result ofchanges in long established techniques.3'5For example, although radium has now beenreplaced by safer isotopes such as cesium andcobalt, this has also entailed new or modifiedapplicators, altered dose rates, and changes inthe number of treatments.The rectum, sigmoid colon, and terminal
ileum seem to be particularly at risk from pelvicradiotherapy.4 While the histological features ofchronic radiation enteritis, characterised byobliterative endarteritis of the small vessels inthe intestinal wall, submucosal oedema, andlymphatic dilatation are well documented,22there is little information about the patho-physiology of diarrhoea caused by chronic radia-tion enteritis.' 4 It has been suggested that diar-rhoea reflects malabsorption because of eitherinfiltration of the mucosa by inflammatory cellsand luminal narrowing of the submucosalarterioles,23 or intestinal bacterial overgrowth.4The possibility that abnormalities in intestinalmotor function may be important has also beenrecognised.624 In the first comprehensive studyof intestinal function in patients with diarrhoeacaused by chronic radiation enteritis, we haveshown a high prevalence of abnormalities inintestinal transit and absorption. Both smallintestinal and whole gut (mainly colonic) transit,were faster in the patients than the controlsubjects. Bile acid absorption was reduced inmany of the patients (although vitamin B12absorption was usually normal), and 10 patientshad malabsorption of lactose. All of these factorsmay theoretically contribute to diarrhoea. Theseobservations were not the result of intestinalstricture or bacterial overgrowth. The relativelysmall difference in median age between thepatients and the control subjects is unlikely tobe important. The observation that despiteincreased stool frequency, stool weight was notgreater in the patients is likely to reflect dietary
480
I<
on March 21, 2022 by guest. P
rotected by copyright.http://gut.bm
j.com/
Gut: first published as 10.1136/gut.34.4.476 on 1 A
Gastrointestinalfunction in chronic radiation enteritis - effects ofloperamide-N-oxide
changes, which inevitably occur in patients witha long history of diarrhoea. Intake of dairyproducts was less in the patients and probablycontributed to the normal faecal fat excretion. Itis also possible that in some patients concurrentradiation proctitis caused rectal irritability andcontributed to a stool weight that was lower thanwould be predicted from stool frequency.The SeHCAT test has made intestinal bile acid
malabsorption much more easy to detect.2526 Ahigh prevalence of bile acid malabsorption hasbeen reported in patients with otherwiseunexplained diarrhoea2627 and it has been sug-gested that this is of pathogenetic importanceand predictive of improvement with cholestyra-mine. Our observation of a high prevalence ofbile acid malabsorption in patients with chronicradiation enteritis is consistent with the results ofa previous study.4 While vitamin B12 malabsorp-tion has been shown in some patients,4 it isrecognised that bile acid absorption is a moresensitive test of terminal ileal function thanvitamin B12 absorption.28 The reduced bile acidabsorption is likely to reflect damage to theterminal ileal mucosa. It should, however, berecognised that more rapid intestinal transit, byreducing the time available for absorption, couldcontribute to decreased bile acid absorption anddiarrhoea.62729 In particular, the terminal ileumseems to have an important role in the regulationof small intestinal transit.30The lactose malabsorption shown in 10
patients probably reflects mucosal damage, assmall intestinal bacterial overgrowth was essen-tially excluded and all of the patients were white.A high prevalence of lactose malabsorption hasbeen reported previously in patients withchronic radiation enteritis who required pare-nateral nutrition.3' The observation that lactosemalabsorption is a common long term effect ofabdominal or pelvic irradiation, or both, isimportant and consistent with the findings of aprospective study conducted by our group whichsuggests that mucosal damage in acute radiationenteritis does not resolve in many patients.7 It isnot surprising that a history of milk intolerancewas seldom obtained in patients with lactosemalabsorption, although dietary intake of dairyproducts was reduced."We have shown for the first time that small
intestinal and whole gut transit is faster inpatients with chronic radiation enteritis. Bothsingle large doses and fractionated doses ofradiation have been shown to alter small intesti-nal motility in animals.2432 Fractionated doses ofabdominal radiation increase the frequency ofgiant migrating contractions and retrogradegiant contractions.24 We and others have reportedthat both small intestinal and total gut transit isfaster in acute radiation enteritis.733 Althoughthere is unequivocal evidence of damage to theintestinal smooth muscle in chronic radiationenteritis,34 we are unaware of any studies ofintestinal motor function in chronic radiationenteritis in either animals or humans.
Medical therapy of diarrhoea caused bychronic radiation enteritis has been largelyempirical and there have been no adequatecontrolled studies. ' Antidiarrhoeal agents, broadspectrum antibiotics, cholestyramine, sulpha-
salazine, and oral corticosteroids have all beenused33536 but anecdotal experience suggeststhat satisfactory control of diarrhoea is rarelyachieved. Most series have focused on surgicalapproaches to complications such as strictureand fistula formation. The results of our studyindicate that loperamide-N-oxide is effective inthe treatment of diarrhoea associated withchronic radiation entritis.
Loperamide-N-oxide is a peripheral opiateagonist precursor which is converted to lopera-mide in the intestinal lumen37 and liver.38Loperamide-N-oxide is itself inactive and hasnegligible affinity for opiate receptors.39 Plasmaconcentrations of loperamide after oral adminis-tration of loperamide-oxide are twofold lowerthan the equivalent oral doses of loperamide,"resulting in a twofold increase in the already highsafety margin ofloperamide.40 41 At the same timeanti-diarrhoeal efficacy is maintained, since adose of even 1 mg has been shown to be effectivein a randomised double blind study of acutediarrhoea in adults.42 As a result of the gradualactivation of loperamide oxide, associated withlessened absorption of loperamide into thesystemic circulation, it has been suggested thatthe incidence of adverse effects such as constipa-tion may be less.4'
Previous studies43 I which suggest that lopera-mide may be effective in the treatment of diar-rhoea associated with acute radiation enteritisprovide no insights into the mechanisms ofaction, as only symptoms were evaluated. Otherstudies indicate that loperamide may prolongintestinal transit, normalise intestinal fluidand electrolyte movement, and increase analsphincter tone.45" Loperamide treatment ofdiarrhoea associated with chronic radiationenteritis was associated with significant changesin stool frequency and volume, gastrointestinaltransit, intestinal absorption, and permeability.Stool frequency and volume decreased and bileacid absorption increased after treatment. Weare unable to explain why vitamin B12 absorp-tion decreased during treatment with loperamideoxide, but can only comment that the magnitudeof the change was relatively small. Loperamideoxide increased the rate of gastric emptying butslowed both small intestinal and whole guttransit. Intestinal permeability to 51Cr EDTAbut not lactulose/rhamnose was increased byloperamide oxide. The beneficial effects of thedrug on stool frequency, bile acid absorption,and intestinal permeability are likely to reflectthe slowing of small intestinal and colonictransit.45 The fact that permeability, as evaluatedby the double sugar test, was not increased isconsistent with this hypothesis, as results of thistest are independent of intestinal transit.49 Aneffect of loperamide oxide on intestinal secretioncannot be excluded as this was not examined.44The faster gastric emptying after treatment is notunexpected.50 This effect is likely to reflect theabsence of central opiate properties of lopera-mide compared with other opiates such as mor-phine which slow gastric emptying. Althoughmost of our patients had minimal upper gastro-intestinal symptoms and normal gastric empty-ing, the effect of loperamide on gastric emptyingmay be beneficial in some patients with chronic
481
on March 21, 2022 by guest. P
rotected by copyright.http://gut.bm
j.com/
Gut: first published as 10.1136/gut.34.4.476 on 1 A
radiation enteritis, as chronic gastroparesis is arecognised sequel to abdominal irradiation.'We thank Miss T Piscioneri for tvping the manuscript. This studywas supported by the National Health and Medical ResearchCouncil of Australia and Janssen Pharmaceutica.
2 Newman A, Katsaris J, Blendis LM, et al. Small-intestinalinjury in women who have had pelvic radiotherapv. Lancet1973; ii: 1471-3.
3 Schofield PF, Carr NO, Holden D. Pathogenesis and treat-ment of radiation bowel disease. ] Roy Soc Med 1986; 79:30-2.
4 Ludgate SM, Merrick MV. The pathogenesis of post-irradiation chronic diarrhoea: measurement of SeHCAT andB,2 absorption for differential diagnosis determines treat-ment. Clin Radiol 1985; 36: 275-8.
5 Galland RB, Spencer J. Natural history and surgical manage-ment of radiation enteritis. Br]7 Surg 1987; 74: 742-7.
6 Yeoh EK, Lui D, Lee NY. The mechanism of diarrhearesulting from pelvic and abdominal radiotherapy; a pros-pective study using selenium-75 labelled conjugated bileacid and cobalt-58 labelled cyanocobalanin. Br J Radiol1984; 57: 1131-36.
7 Yeoh E, Horowitz M, Maddox A, et al. Effects of abdominalirradiation on gastrointestinal function. Gastroenterology1989; 96: A560.
8 Horowitz M, Harding PE, Chatterton BE, et al. Acute andchronic effects of domperidone on gastric emptying indiabetic autonomic neuropathy. Dig Dis Sci 1985; 30: 1-9.
9 Angus RM, Eisman JA. Osteoporosis: the role of calciumintake and supplementation. Med]7 Aust 1988; 148: 630-3.
10 Varley H. Determination of the fat content of faeces. PracticalClinical Biochemistry, 4th ed. London: Heinemann, 1967:419-21.
11 Horowitz M, Wishart J, Mundy L, et al. Lactose and calciumabsorption in postmenopausal osteoporosis. Arch Intern Med1987; 147: 534-6.
12 Read NW, Al-Janabi MN, Holgate AM, Barber DC,Edwards CA. Simultaneous measurement of gastric empty-ing, small bowel residence and colonic filling of a solid mealby the use of the gamma camera. Gut 1986; 27: 300-8.
13 Read NW, Miles CA, Fischer D, et al. Transit of a mealthrough the stomach, small intestine and colon in normalsubjects and its role in the pathogenesis of diarrhea.Gastroenterology 1980; 79: 1276-82.
14 Malagelada JR, Robertson JS, Brown ML, et al. Intestinaltransit of solid and liquid components of a meal in health.Gastroenterology 1984; 87: 1255-63.
15 Collins PJ, Horowitz M, Cook DJ, et al. Gastric emptying innormal subjects - a reproducible technique using a singlescintillation camera and computer system. Gut 1983; 24:1117-25.
16 Caride VJ, Prokop EK, Troncale FJ, et al. Scintigraphicdetermination of small intestinal transit time: comparisonwith the hydrogen breath technique. Gastroenterologv 1984;86:714-20.
17 Hetzel PAS, LaBrooy JT, Bellon M, et al. The 5"Cr-EDTAabsorption test in coeliac patients treated with a gluten freediet. IRCS Med Sci 1986; 14: 1183-4.
19 Pounder RE, Gomes MF, Lokschin F, et al. the L-rhamnose/lactulose permeability test: non-invasive assessment of smallintestinal mucosal damage. Gastroenterology 1983; 84: 1276.
20 Cosse De JJ, Rhodes RS, Wentz WB. The natural history andmanagement of radiation induced injury of the gastro-intestinal tract. Surg Gynecol Obstet 1969; 170: 369-82.
21 Schofield FF, Holden D, Carr HD. Bowel disease afterradiotherapy. ] Roy Soc Med 1983; 76: 463-6.
22 Wellwood JM, Jackson BT. The intestinal complications ofradiotherapy. Br7 Surg 1973; 60: 814-8.
23 Carr ND, Pullen RB, Hasleton PS, et al. Microvascular studiesin human radiation bowel disease. Gut 1984; 25: 448-54.
24 Otterson MF, Sarna SK, Moulder JE. Effects of fractionateddoses of ionizing radiation on small intestinal motor activitv.Gastroenterologv 1988; 95: 1249-57.
25 Ferraris R, Jazrawi R, Bridges C, et al. Use of a labelled bileacid (75SeHCAT) as a test of ileal function. Gastroenterologv1986;90: 1129-36.
26 Merrick MV, Eastwood MA, Ford MJ. Is bile acid malabsorp-tion underdiagnosed? An evaluation of accuracy of diagnosisby measurement of SeHCAT retention. BM] 1985; 290:665-8.
27 Sciarretta G, Fagioli G, Furno A, et al. 75SeHCAT test in the
detection of bile acid malabsorption in functional diarrhoeaand its correlation with small bowel transit. Gust 1987; 28:970-5.
28 Fromm MD, Thomas PJ, Hofmann AF. Sensitivity andspecificity in tests of distal ileal function: prospectiecomparison of bile acid and vitamin B12 absorption in ilealresection patients. Gastroenterologv 1973; 64: 1077-90.
29 Schiller LR, Hogan RB, Moranski SG, et al. Studies of theprevalence and significance of radiolabelled bile acid absorp-tion in a group of patients with idiopathic chronic diarrhea.Gastroenterologv 1987; 92: 151-60.
30 Read NW, McFarlane A, Kinsman RI, et al. Effects ofinfusion of nutrient solutions into the ileum on gastro-intestinal transit and plasma levels of neurotensin andenteroglucagon. Gastroenterologv 1984; 86: 714-20.
32 Summers RW, Kent TH, Osbourne JW. Effects of drugs, ilealobstruction and irradiation on rat gastrointestinal propul-sion. Gastroenterologv 1970; 59: 731-9.
33 La Manna MM, Parker JA, Wolodzko BS, et al. Radionuclideoesophageal and intestinal transit scintigraphy in patientsundergoing radiation therapy. Radiat Med 1985; 3 (no 1):13-6.
34 Rubin P, Casarett GW. Alimentary tract small and largeintestines and rectum. In: Rubin P, Casarett GW, eds.Clinical radiation pathology, vol 1. Philadelphia: Saunders,1968: 224-6.
35 Heusinkveld RS, Manning MR, Aristizabal SA. Controlof radiation-induced diarrhea with cholestyramine.Int3 Radiat Oncol Biol Phys 1978; 4: 687-90.
36 Goldstein F, Khoury J, Thornton JJ. Treatment of chronicradiation enteritis and colitis with salicylazosulfapyridineand systemic cortico-steroids. Am ] Gastroenterol 1979; 65:201-8.
37 Lavriisen K, Meurdermans W, Hendrick J, et al. Thereductive in-vitro metabolism of loperamide-N-oxide by ratliver homogenates, red blood cells and gut contents. Arch IntPharmodvn Ther 1984; 270: 174-5.
38 Niemegeers CJE, Awouters F, Lenaerts FM, et al. Anti-diarrheal specificity and safety of the N-oxide of loperamide(R58425) in rats. Drug Development Research 1986; 8: 279-86.
39 Gommeren W, Leysen J. In vitro receptor binding and amineuptake inhibition profile ofR58425 and loperamide (R18533).Janssen Research Foundation PRR 13, October 1988.
40 Van De Velde V, LeGrand M, Woestenborgus R, et al. Oralbioavailabilitv of loperamide-N-oxide after single intake of a 4mg capsule in healthv subjects. Janssen Clinical ResearchReport 3, April 1987.
41 Van De Velde V, LeGrand M, Van Rooy P, et al. Dose-proportionalitv of the pharmnacokinetics of loperamide oxide andloperamide after oral dosing of2 to 16 mg ofloperamide-oxide tohealthv volunteers. Janssen Clinical Research Report 6,February 1988.
42 Spaepen W, Lemmens PJ, Verlinden M. Ffficacy (floperamnideoxide in acute diarrhea in adults. A douible-blind, randomized,dose-effect comparison with loperamide and placebo. JanssenResearch Foundation (data on file).
43 Chapaux J, Chapaux P, Royer E. Loperamide in patients withradiotherapy-induced diarrhea. Arznein-Forsch Drug Res1974;28: 1:864-6.
44 Geginat G, Behdjati F, Green E, et al. Study of loperamide inpatients with diarrhea following telecobalt irradiation formalignant abdominal tumors. Strahlenther 1978; 154: 11-5.
45 Schiller LR, SantaAna CA, Morawski SG, Fordtran JS.Mechanism of the antidiarrheal effect of loperamide. Gastro-enterology 1984; 86: 1475-80.
46 Basilisco G, Bozzani A, Camboni G, et al. Effect of loperamideand naloxone on mouth-to-caecum transit time evaluated bylactulose hydrogen breath test. Gut 1985; 26: 700-3.
47 Read M, Read NW, Barber DC, et al. Effects of loperamide onanal sphincter function in patients complaining of chronicdiarrhea with fecal incontinence and urgency. Dig Dis Sci1982; 27: 807-14.
48 Kachel G, Ruppin H, Hagel J, et al. Human intestinal motoractivity and transport: effects of a synthetic opiate. Gastro-enterology 1986; 90: 85-93.
49 Brunetto AL, Pearson ADJ, Gibson R, et al. the effect ofpharmacological modification of gastric emptying andmouth-to-caecum transit time on the absorption of sugarprobe marker molecules of intestinal permeability in normalmam. Eur]7 Clin Invest 1990; 20: 279-84.
50 Cann PA, Read NW, Holdsworth CD. Loperamide improvesdyspeptic symptoms and accelerates gastric emptying inpatients with irritable bowel syndrome. Gut 1982; 25: A437.
51 Layer P, Demol P, Hotz J, et al. Gastroparesis after radiation.DigDisSci 1986; 31: 377-80.
on March 21, 2022 by guest. P
rotected by copyright.http://gut.bm
j.com/
Gut: first published as 10.1136/gut.34.4.476 on 1 A
