The intraluminal pressure patterns diverticulosis ofthe colon · Guit, 1964, 5, 201 Theintraluminal pressure patterns in diverticulosis ofthe colon Part I Resting patterns ofpressure

Guit, 1964, 5, 201

The intraluminal pressure patterns in diverticulosisof the colon

Part I Resting patterns of pressure

NEIL STAMFORD PAINTER AND S. C. TRUELOVE

From the Nuffield Department of Clinical Medicine, The Radcliffe Infirmary, Oxford

EDITORIAL SYNOPSIS It might be anticipated that patients with diverticulosis might have increasedintraluminal colonic pressures but this could not be demonstrated under basal conditions (Part 1).With morphine there is an increased intraluminar pressure, particularly in the affected part of

the colon (Part 1I). This differential response is of special interest in relation to the development andprogression of diverticulosis.The use of morphia is clearly contraindicated in acute diverticulitis. Similarly prostigmine

(neostigmine methylsulphate) increases the frequency and amplitude of pressure waves in the colon(Part III).These increased pressures can be abolished by the intravenous injection of probanthine (propan-theline bromide) and it is likely that this may have a part to play in the treatment of acute diverti-culitis either alone or to counteract the unwanted effects of morphine (Part IV). In therapeutic dosespethidine (Demerol) does not cause the sigmoid colon to generate high pressures and is recommendedas the analgesic of choice for acute diverticulitis.

Colonic diverticula consist of herniations of themucous membrane of the colon through the musclecoat that normally contains it. As herniation impliesthe existence of a propelling or pulsion force, withor without an initial defect in the integument, it hasbeen suggested that colonic diverticula are causedby abnormal pressures in the lumen of the colon.However, at the time the present study began nosystematic observations had been made to test thishypothesis and the pressures in the colon indiverticulosis were unknown (Thompson, 1959).The main object of this study was to measure the

pressure waves produced in the lumen of the humansigmoid colon in diverticulosis under basal con-ditions, and to compare the results with thoseobtained from control subjects studied underidentical conditions.

METHOD OF STUDY

The intracolonic pressures were measured and recordedby the method of Chaudhary and Truelove (1961). Threewater-filled open-ended polythene tubes were insertedinto the sigmoid colon through a sigmoidoscope, whichwas then withdrawn, leaving the tubes in situ. The tubeswere bound together so that their tips were 7 5 cm. apart.Each tip was enclosed in a metal cuff to permit radio-

logical visualization of the point of recording. Theposition of the tubes was ascertained radiologically byfilling the tube located farthest up the colon with Hypaque.When a patient with diverticulosis was being studied, thisradiograph was compared with the barium enema inorder to determine which of the recording tips weresituated in segments of the bowel that bore diverticula.The polythene tubes were coupled to Shillingford-

Muller transducers whose signals were fed into aCambridge three-channel pressure recorder with directpen-writers. This apparatus was calibrated against astandard sphygmomanometer, so that a pressure changeof 2 mm. Hg caused a 1mm. deflection of the recordingpen. Thus each large square on the recording papercorresponded to a pressure of 10 mm. Hg.The Cambridge three-channel pressure recorder has a

stable base line which allows accurate pressure recordingto continue for several hours and yields a permanenttracing that can be studied at leisure. The recordingpaper moved at a standard speed so that the duration ofany pressure wave could be measured easily.

After the introduction of the polythene tubes, theposition of the recording tips was ascertained radiologic-ally. The patient was then made comfortable in bed andallowed to settle down for a time, after which the pressuresin the sigmoid colon were recorded for one hour. Duringthis time, the patient abstained from food and tobacco,and was only engaged in conversation when this wasessential to the performance of the experiment.

201

on July 4, 2021 by guest. Protected by copyright.

http://gut.bmj.com

/G

ut: first published as 10.1136/gut.5.3.201 on 1 June 1964. Dow

nloaded from

http://gut.bmj.com/

Neil Stamford Painter and S. C. Truelove

The pressure tracings thus obtained were later analysedto determine the number of pressure waves of variousdimensions that had been recorded by each tube. Thetube farthest up the colon was called 'lead 1', the middletube 'lead 2', and that tube which was nearest the anuswas labelled 'lead 3'. Data obtained from leads whosetips were situated below the recto-sigmoid junction wereexcluded from this analysis. Leads whose tips weresituated in segments of the colon that bore diverticulawere said to be 'related to diverticula', while leads whosetips recorded from apparently normal segments of colonin subjects with diverticulosis were designated leads 'notrelated to dixerticula'. The reason for this subdivisionwill emerge later.The numbers of subjects studied, observations made,

and leads from which pressure tracings were obtainedare given in Table I. These totals differ as some patientswere observed twice. On average, two leads reached thesigmoid in normal subjects, while the structure of thediseased bowel often prevented more than one leadpassing beyond the recto-sigmoid junction.

TABLE INUMBER OF SUBJECTS STUDIED, OBSERVATIONS

MADE, AND LEADS FROM WHICH PRESSURE TRACINGS WEREOBTAINED

Number Number of Number of Leads Analysedof ObservationsSubjects

Normal subjectsPatients withdiverticulosis

3228

33 6629 20 (related to diverticula)

31 (not related todiverticula)

SUBJECTS STUDIED

NORMAL SUBJECTS These were persons who satisfiedthe following criteria. 1 They had no history ofdisease or dysfunction of the gut. 2 Their colons hadbeen shown to be normal radiologically; a few of theyounger patients did not have barium enemas, butclinically and sigmoidoscopically their colons werenormal.

PATIENTS WITH DIVERTICULOSIS These patients hadbeen examined by barium enema in the previoustwo years and found to have diverticulosis. Some hadremained symptom-free while others had sufferedattacks of diverticulitis.

It was realized that the control subjects were notideal because their average age was different fromthat of the diverticulosis patients, but this criticismmay be answered by stating that no correlation wasfound between age and the number or dimensions ofthe pressure waves.

RESULTS

RESTING PRESSURE PATTERNS IN THE NORMAL SIGMOIDCOLON The pressure tracings showed that eachlead recorded a base line pressure, which was found

to be within a few millimetres of mercury of atmo-spheric pressure. This basal pressure results from theinterplay of various factors, including the state ofthe bowel wall, the condition of the abdominalmusculature, the height of the recording tip inrelation to the manometer, and the weight of theviscera overlying the loop of bowel being studied(Rowlands, 1962). This basal pressure remainedalmost constant throughout any one observation,apart from the small rhythmic variations due torespiration.The usual resting pattern of pressure waves

consisted of the irregular occurrence of waves,sometimes singly and sometimes in series (Fig. 1).A series of waves might consist of two or threewaves or a succession of several waves. Pressurewaves at one level of the gut were frequently seen tooccur independently of the pressure at another level.The gut might be producing pressures on all threeleads at once, or, alternatively, one or more of theleads might remain at the basal level for long periodsof time. These waves of pressure usually occurredindependently on each lead without any stricttemporal relationship to waves on other leads.When waves were recorded on all three leadssimultaneously, their form might be similar ordifferent and it was extremely rare to find waves ofexactly similar dimensions occurring simultaneouslyon all three leads. No evidence of the analwardprogression of a pocket of high pressure from theregion of lead 1 down the sigmoid so as to affect theother leads in their turn was obtained except whenflatus was passed (Fig. 2).

There was considerable variation in the pressurewaves recorded in different subjects, and in the samesubjects if different periods of the same hour wereconsidered. Some subjects were studied on twoseparate days and the number of pressure wavesrecorded in one hour was different on the two days.Sometimes the intraluminal pressure remained

approximately at atmosphere for long periods oftime, only changing as a few waves of pressure wereproduced during an hour. On the other hand, veryoccasionally one or more pens were seen to moverhythmically almost continually for several minuteswith little interruption.The majority of waves recorded from normal

subjects at rest were simple waves representingpositive changes of pressure. These simple waveswere of two main types. The first type showed agentle rise in pressure to a height that was less than10 mm. Hg and an equally gentle decline to theoriginal level; the duration of these waves wasbetween 10 and 30 seconds. The second typediffered in that the rise of pressure was more rapidand reached a greater amplitude (10 to 30 mm. Hg)

202


http://gut.bmj.com

/G


nloaded from

http://gut.bmj.com/

The intraluminal pressure patterns in diverticulosis of the colon

FIG. 1. Resting pressure patterns in the normal sigmoid colon. Tracings obtained from normal sigmoid colons; onlylead 3 in the top tracing was not located above the rectosigmoidjunction. Each tracing represents six minutes of recordingtime; this scale has been adhered to throughout this study unless otherwise stated. The variation in the number anddimensions of the pressure waves occurring in the sigmoid can be seen. The highest wave in the top tracing represents30 mm. Hg pressure, while in the lowest tracing the intraluminal pressure hardly altered except for the small changescaused by respiration.

and its fall steeper, so that its duration was 10 to40 seconds. Between these two forms, every gradationhas been seen. Exceptionally a similar form of wavethat represented negative pressure was recorded.

These small waves corresponded in form to thetype I wave described in the literature. These simplewaves were the main type of pressure wave en-countered in this study, although they were not so inreports of earlier observations made with the aid ofballoons.

It was unusual to record types of waves other thanthese simple waves; only occasionally were complexwaves, or waves of higher amplitude, recorded innormal subjects at rest. The configuration of the

complex pressure waves was similar to the type IIwaves that have been described by those workerswho used balloons to study colonic motility. It isimportant to emphasize that the use of the termstype I and type II is merely a convenient method ofdescribing the form of such waves and is not intendedto imply that balloon tracings correspond to theexact pressure measurements that are recorded byopen-ended tubes.The considerable variation in the number of

pressure waves produced by normal subjects in agiven hour was noted by Chaudhary and Truelove(1961), and confirmed in this study, and consequentlyany quantitative data relating to colonic pressures

203


http://gut.bmj.com

/G


nloaded from

http://gut.bmj.com/


must be derived from a large number of observations.In this study, intraluminal pressure tracings wereobtained from a considerable number of controlsubjects to provide a yardstick against which thelevels of pressure produced by patients with diverti-culosis could be compared.

RESTING PRESSURE PATTERNS IN THE SIGMOID COLONIN DIVERTICULOSIS As diverticula have been thoughtto owe their origin to an abnormality of the intra-colonic pressure, the resting pressure patternsobtained from subjects with diverticulosis wereexamined to discover whether they differed from thenormal resting pattern, but no obvious differencewas found.The variability of the wave forms, their irregular

or rhythmic occurrence and their temporal andspatial independence of the pressure existing in aneighbouring part of the bowel that were describedin the section dealing with the normal resting patternof pressures apply equally to the resting pressurepattern of diverticulosis coli. No obvious increase inthe height of the pressure waves or in the steepnessof their rise and fall from the basal pressure wasnoted, and it was apparent that, in diverticulosis aswell as in health, the intrasigmoid pressure sometimesremained at about atmospheric pressure for longperiods of time (Figs. 2 and 3). Detailed analysisof the pressure tracings revealed that these impres-sions were correct as no significant difference in theshape or number of the pressure waves of variousheights could be detected.

This failure to detect any difference between thenormal resting pressure pattern and that of diverti-culosis led to a further analysis of the pressuretracings. When it was remembered that the segmentsof the sigmoid that actually bear diverticula aresometimes structurally different from the neighbour-ing colon, from which they may be sharply demarc-ated (Fig. 4), it was considered that this anatomicaldifference might be reflected in an equally abruptchange in function as regards the generation ofintraluminal pressures. Consequently it was decidedto consider those leads whose tips were situated insegments that bore diverticula separately from thoselocated in apparently normal bowel in subjects withdiverticulosis. Therefore the resting patterns indiverticulosis were divided into two groups, 'restingpatterns not related to diverticula' and 'restingpatterns related to diverticula'. It will be appreciatedthat leads in the same patient might be distributedbetween the two groups, while all the leads inanother patient might fall into only one of thesegroups. Thus some subjects acted as their owncontrols in the subdivision of the resting patterns ofdiverticulosis.

Once again, no obvious difference was apparent inthe resting patterns, and quantitative tests that willbe described later confirmed that this impressionwas correct. It was found that the resting patterns ofpressure waves produced by the normal sigmoid, bythe sigmoid that was beset with diverticula, and bythose segments of the sigmoid that did not beardiverticula in subjects afflicted with diverticulosis,were essentially similar. In addition, no difference inthe levels of basal intraluminal pressure wasdemonstrated.

QUANTITATIVE ASPECTS OF THE PRESSURE WAVES INTHE SIGMOID COLON IN NORMAL SUBJECTS AND INTHOSE WITH DIVERTICULOSIS Chaudhary and True-love (1961) were unable to classify pressure wavesinto the types I, II, III, and IV that previous workershad described on the basis of balloon studies.Confronted with this situation, they decided toconfine their analysis of pressure patterns to makingsimple measurements of the waves, regardless oftheir configuration, thus avoiding some of thedifficulties that are inherent in those methods ofdescription that rely entirely on the subjectivejudgement of the observer. In this way theyestablished some values by which the number andcharacter of the pressure waves produced by any twoseries of sigmoid colons in a given time may becompared. Their methods were used in this study.The number of waves occurring in the standard

time of one hour was counted. The height andduration of these waves were measured withoutregard to their form; the height was taken as thevertical distance between the basal pressure and thehighest point of the wave, whether its form wassimple or complex. Thus we determined for eachgroup of subjects the total number of waves ofvarious specific dimensions occurring on each leadlocated in the sigmoid colon. Thus a value could beobtained for the mean number of waves of anyspecific dimensions per lead per 60 minutes ofrecording time. These results are shown in TablesII and III.When these values were examined, it was apparent

that there was little difference in the three groupswhen the amplitude of the waves was considered(Table II). Fewer waves were recorded from leadsthat were in relation to diverticula. This differencewas due to the smaller number of waves of thelowest amplitude that occurred in the affectedsegments. In the normal colon, fewer waves ofbetween 10 and 19 mm. Hg were recorded comparedwith the other two groups, but waves of between20 and 29 mm. Hg were commoner in this group.When waves greater than 30 mm. Hg were con-sidered, there was little to choose between the three

204


http://gut.bmj.com

/G


nloaded from

http://gut.bmj.com/

The intraluminal pressure patterns in diverticulosis of the colon

FIG. 2. The effect ofpassingflatus. All leads situated in sigmoid colon ofpatient with diverticulosis, leads 1 and 2 beingrelated to diverticula. The tracing remained at the resting basal level ofpressure except when flatus was passed. This'progression' ofpressure towards the anus was not seen at any other time.

FIG. 3. Resting pressure pattern in diverticulosis. All three leads were in the sigmoid colon but had not reached thelevel of diverticula. The sudden drop in the basal level on the left of the tracing was due to the deliberate re-setting ofthe recording pens. Waves of over 60 mm. Hg are shown on lead 1, but these high pressures were confined to the vicinityof this lead, as lead 2 recorded lower pressures, although it was only 7-5 cm. distant. Comparison of this tracing withFigure 2 shows the variation in pressure pattern that occurs in the sigmoid colon in diverticulosis.

FIG. 4. Diverticulosis coli. This longitudinal section of a human colon shows two diverticula. The bowel below them(left) is apparently normal, and the change in structure between this and the diseased bowel is abrupt.

205


http://gut.bmj.com

/G


nloaded from

http://gut.bmj.com/


TABLE IINUMBER OF WAVES BY AMPLITUDE OF WAVE IN SIGMOID

Amplitude ofWaves (mm. Hg)

1-910-1920-2930-3940-4950-5960+Total

COLON1

Resting ValuesNormal Diverticulosis

Leads not Related to Leads Related toDiverticula Diverticula

34-72-3090303030-02

38-6

35-23-80-60-20-20-10-1

40-2

2524-40-6050303005

31-3'Mean values per lead per 60 minutes' recording time.

groups of leads. Over 80% of the waves in eachgroup were of an amplitude less than 10 mm. Hgand few waves exceeded 20 mm. Hg. No markeddifference was seen in the three groups, and nopreponderance of waves of high pressure wasproduced by those segments that were affected bydiverticulosis.When the duration of the waves was considered

(Table III), it was found that over 80 %0 of the wavesproduced by the three types of colonic segments

TABLE IIINUMBER OF WAVES BY DURATION OF WAVE

IN SIGMOID COLONLResting Values

Duration of Normal DiverticulosisWaves (sec.)

Leads not Related Leads Related toto Diverticula Diverticula

bounded by the wave and the basal pressure line.This area is proportional to the work done (i.e.,pressure x time) by the colon in generating orwithstanding this wave of pressure.They calculated this product for every wave

recorded by one lead in an hour and called the sumof these products the 'colonic motility index'. Thisindex is not a measure of the motility of the colonas Chaudhary and Truelove first thought, but ameasure of the 'total pressure' or total work pro-duced by the colon in the vicinity of one recordinglead in one hour. It has the advantage of being asimple number that can be plotted on a scale. Theleft half of Fig. 7 in Part II of this study shows thecolonic motility indices belonging to the leads fromwhich the mean values that have already been givenwere derived. They are expressed in thousands anddivided into the same three groups, according to thetype of colonic segment in which the recording leadwas situated. The average value of the indices ineach column is shown by the horizontal line in thatcolumn.

It is at once apparent that despite the variationof the values of the indices in each column, theaverage value in each group is almost the same. Thisfinding supports our previous impression that theresting patterns of pressure in health and diverti-culosis are essentially similar and suggests that underbasal conditions the 'total pressure' produced by thecolon in diverticulosis is no greater than thatgenerated by the normal colon.

DISCUSSION

1-9 10 7 13-1 8-310-19 17-7 16 9 11-920-29 7-3 7-3 6-630-39 1-73 1-48 2-440-49 05 094 1-250-59 0 4 0 39 0-860+ 0-3 0-13 0-15Total 38-6 40-2 31-3

'Mean values per lead per 60 minutes' recording time.

lasted less than 30 seconds. Those leads that wererelated to diverticula recorded more waves of aduration greater than 30 seconds than those thatwere situated in normal segments; such waves wereabout half as common again in the affected segments.

THE COLONIC MOTILITY INDEX Chaudhary andTruelove (1961) proposed that this index be used as ameasure that would enable the motor activity of twogroups of sigmoid colons to be compared. As mostpressure waves are roughly triangular in shape, theyargued that the product of the height (in mm. Hg)and the duration (in seconds) of a given wave wasapproximately proportional to twice the area

The resting patterns of intraluminal pressure in thehuman sigmoid colon show considerable variationboth in health and diverticulosis, not only indifferent subjects but in the same subjects when theyare studied at different times. Consequently it isnecessary to study a considerable number of subjectsin order to compare the activity of groups of colonsunder basal conditions.The resting pressure patterns produced by normal

colons that have been obtained in this study areessentially similar to those reported by Chaudharyand Truelove (1961). No previous systematic studyof the intraluminal pressure patterns in diverticulosiscoli has been reported so that the present results forthis disease cannot be discussed in relation to thework of others.No obvious difference has been found in the

resting pressure patterns produced by the sigmoidcolon in health and in diverticulosis. This was trueboth for segments of the sigmoid that actually borediverticula as well as for apparently normal segmentsof the colon in subjects with diverticulosis. Simple

206


http://gut.bmj.com

/G


nloaded from

http://gut.bmj.com/

The intraluminal pressure patterns in diverticulosis of the colon 207

quantitative analysis of the pressure tracings failedto reveal any preponderance of waves of highintracolonic pressure in diverticulosis. The segmentsthat actually bore diverticula did produce morewaves of longer duration than other segments, butsuch waves represented only a small proportion ofthe pressure waves recorded and would not appear tobe important in the genesis of diverticula.

Despite the approximations inherent in itsdefinition, the colonic motility index is the onlymeasure so far available that allows the motoractivity of two groups of colons to be compared withregard to the production of pressures. These indiceswere calculated and showed a considerable generalmeasure of agreement in each group of leads, whiletheir average values in each group of leads were verysimilar. This suggests that the total pressuresgenerated or withstood by the colon in health and indiverticulosis are essentially the same under restingconditions.However, a variety of factors influences the motor

behaviour of the human colon, such as eating,drinking, defaecation, and emotion (Chaudhary andTruelove, 1961; Connell, 1961). Therefore the effectsof these stimuli on the pressures in the sigmoidcolon deserve to be studied as it is possible that they

play an important part in the aetiology of thedisease by altering the intracolonic pressure patterns.

SUMMARY

The herniation of the colonic mucosa in diverti-culosis may be the result of abnormally highintracolonic pressures, of weakness of the muscularispropria, or of a combination of these two factors.A systematic study has been made of the intra-

luminal sigmoid pressures in diverticulosis coli and inthe healthy sigmoid colon employing open-endedwater-filled polythene tubes coupled to a Cambridgemulti-channel pressure-recorder.Under basal conditions no evidence was obtained

of any major difference in the intraluminal pressuresin diverticulosis and in health when the recordingswere analysed in various ways.

REFERENCES

Chaudhary, N. A., and Truelove, S. C. (1961). Human colonicmotility: A comparative study of normal subjects, patients withulcerative colitis, and patients with the irritable colon syndromeGastroenterology, 40, 1-36.

Connell, A. M. (1961). Personal communication.Rowlands, E. N. (1962). In a Symposium on Small Intestinal Motility

given at the Annual General Meeting of the British Societyof Gastroenterology, 1961. Gut, 3, 94.

Thompson, H. R. (1959). Diverticulitis of the colon. Postgrad. med. J.,35, 86-91.

Part II The effect of morphine

Morphine is still the most important analgesic.Much conflicting evidence has accumulated over thepast 70 years regarding its effect on the intestine.Vaughan Williams and Streeten (1950, 1951) havepointed out that the opposing views of previousinvestigators are largely the result of the limitationsof the techniques they employed, and that many ofthe differing conclusions that have been expressedcan be reconciled when this fact is appreciated.Nevertheless, even though its mode of action and,in particular, its effect on the human intestine, areso little understood, the drug is very widely used.As it seemed possible that morphine might alter thepattern of the intracolonic pressures, we measuredits effect on the pressures in the human colon both inhealth and in diverticulosis coli.

METHOD

The method of recording the intracolonic pressure hasbeen described in Part I of this study. A large number ofthe subjects whose restin g patterns have been describedwere given 10 mg. of morphine sulphate, either intra-venously or intramuscularly, after their resting patternshad been recorded for one hour. The intracolonic

pressures were recorded for a further hour following thisinjection and the tracing thus obtained was called the'post-morphine' pattern. Hence it was possible tocompare the resting patterns of intrasigmoid pressureswith those observed after morphine in both health anddiverticulosis.

RESULTS

EFFECT OF MORPHINE ON THE PRESSURE PATTERNS INTHE NORMAL SIGMOID COLON When given intra-venously, the effect of morphine became apparentalmost at once in almost every patient (Fig. 1).Initially the basal intraluminal pressure nearlyalways rose one or two millimetres of mercury and,superimposed on this rise, there occurred a success-ion of waves of high pressure, which continued forseveral minutes. These dramatic changes wereusually followed by a period lasting a few minutesduring which the pressure tracing was more or lessflat before another series of waves was generated.The height of the waves in the initial complexvaried from lead to lead. The height of these initialpressures was seldom exceeded by the waves seen


http://gut.bmj.com

/G


nloaded from
http://gut.bmj.com/

Documents

The intraluminal pressure patterns diverticulosis ofthe colon · Guit, 1964, 5, 201 Theintraluminal pressure patterns in diverticulosis ofthe colon Part I Resting patterns ofpressure