Gut 1995; 37: 499-504

Hyperventilation, central autonomic control, andcolonic tone in humans

M J Ford, M J Camilleri, R B Hanson, J A Wiste, M J Joyner

AbstractSymptoms attributable to hyperventila-tion are common among patients with theirritable bowel syndrome (IBS); indeed,some have suggested that hyperventila-tion may exacerbate the alimentary symp-toms of IBS. Hyperventilation changeshaemodynamic function through centraland peripheral mechanisms; its effects oncolonic motor function, however, areunknown. The aim of this study, there-fore, was to assess the effects ofhyperven-tilation on colonic tone and motility andon cardiovascular autonomic activity, andto discover if hypocapnia was critical toelicit the response. Phasic and tonicmotility of the transverse and sigmoidcolon, end tidal Pco2, pulse rate, and beatto beat pulse variability were assessedbefore, during, and after a five minuteperiod of hypocapnic hyperventilation in15 healthy volunteers; in seven othersubjects, effects of both eucapnic andhypocapnic hyperventilation were evalu-ated. Hypocapnic but not eucapnic hyper-ventilation produced an increase incolonic tone and phasic contractility in thetransverse and sigmoid regions and anincrease in pulse rate and pulse intervalvariability. The findings are consistentwith inhibition ofsympathetic innervationto the colon or direct effects ofhypocapniaon colonic smooth muscle, or both. Thesephysiological gut responses suggest thatsome of the changes in colonic functionare caused by altered brain or autonomiccontrol mechanisms.(Gut 1995; 37: 499 504)

Keywords: colon motility, irritable bowel syndrome,hyperventilation.

Physical, mental, and emotional stressors havebeen well shown to induce changes inoesophageal, gastric, small intestinal, andcolonic motility in both healthy subjects andpatients with IBS."1 Acute stress, however,bears little resemblance to the chronic distressinduced by life events and difficulties and manypatients do not associate changes in symptomlevels with changes in stress patterns. Otherphenomena, such as hyperventilation, maymediate changes in states of arousal and hencecentral autonomic outflow. It is possible thatsuch changes could not only influence colonicmotor function but, by changing tone andmotility, change the set point at which tensionreceptors are stimulated, and so determinesensory function and visceral sensitivity.Animal studies have shown the existence of a

functional subset of sympathetic neurons thatregulate gut motility; some of these neuronsare inhibited by respiratory stimulation.12 13Furthermore, studies in humans have shownthat hyperventilation increases oesophageal andgastric antral contractility14 15; colonic motility,however, has not previously been shown tochange in response to hyperventilation. 16 Giventhe association of IBS with the hyperventilationsyndrome8-10 and changed autonomic functionin some IBS patients,17-19 we wished to testthe hypothesis that hyperventilation inducessignificant changes in colonic motor function,which are mediated by the autonomic nervoussystem. Our aims were firstly, to evaluateproximal and distal colonic tone and phasicmotility together with autonomic cardiovascu-lar responses to hypocapnic hyperventilation inconscious, healthy human volunteers and sec-ondly, to discover if hypocapnia was critical ineliciting this response.

Methods

GastroenterologyResearch UnitM J FordM CamilleriR B HansonJ A Wiste

and AnesthesiologyResearch UnitM J Joyner

Mayo Clinic and MayoFoundation Rochester,USA

Correspondence to:Dr M Camilleri,Gastroenterology ResearchUnit, Mayo Clinic,Rochester, MN 55905, USA.

Accepted for publication16 February 1995

Hyperventilation produces a rise in heart rate,cardiac output, and forearm blood flow and afall in peripheral vascular resistance attribut-able respectively to vasodilatation and veno-constriction.1 Hyperventilation is, however, apotent modulator of central autonomic outflowattributable to 'central command' effects - thatis, reflex brainstem responses - and to hypocap-nic effects on the hypothalamus, hippocampus,limbic system, and locus ceruleus resultingin decreased central parasympathetic and sym-pathetic outflow.2-7 In patients with the irri-table bowel syndrome (IBS), non-alimentarysymptoms are frequently encountered, includ-ing those symptoms conventionally associatedwith the hyperventilation syndrome such aspalpitations, dizziness, and paresthesiae.7-10

HEALTHY VOLUNTEERSTwenty two healthy volunteers, aged 19-62years (mean (SEM) 31 (2* 1); 12 M and 1 0 F),were recruited by public advertisement. Nonehad had previous gastrointestinal surgery; IBS,anxiety, and depressive disorders wereexcluded using validated, screening question-naires,20 21 a clinical interview, and physicalexamination. The studies were approved byMayo Clinic's Institutional Review Board.

COLONIC MOTOR FUNCTIONA multilumen, combined manometric andbarostat assembly was placed in the preparedcolon with the aid of colonoscopy and fluo-roscopy.22 23 Tonic and phasic contractile

499

on Decem

ber 12, 2020 by guest. Protected by copyright.

http://gut.bmj.com

/G

ut: first published as 10.1136/gut.37.4.499 on 1 October 1995. D

ownloaded from

Ford, Camilleri, Hanson, Wiste, Joyner

activity of the colon were measured by baro-stat balloons and conventional manometryrespectively. The combined assembly com-prised six manometric ports, one 10 cm oradto a 10 cm polythene balloon positioned in thetransverse colon, one 10 cm aborad to asimilar balloon in the sigmoid colon and fourports, 10 cm apart, between the two balloons,which were 50 cm apart. Each balloon waslinked to a barostat and an electromechanicalair injection device. The barostat balloonswere inflated to a minimum distending pres-sure, defined as the pressure (and volume) atwhich respiratory excursions were regularlyrecorded as changes in the barostat volume.The 'operating pressure' was set 2 mm Hgabove the minimum distending pressure(median pressure 10 mm Hg, range 8-14 mmHg). Intraballoon volumes and manometricpressure changes in response to contractileactivity were monitored continuously through-out the study. The technique and its use inhealthy and clinical states has previously beenshown to be both safe and effective when com-bined with manometry.22 23

HYPERVENTILATION AND HAEMODYNAMICFUNCTIONThe methodology of voluntary hyperventila-tion was similar to that extensively used in pre-vious pathophysiological studies.2>26 In theabsence of pulmonary disease, the end tidalPco2 (PetCo2) provides an accurate breath bybreath estimate of the arterial Pco2.26 ThePetCO2 was measured continuously throughoutthe study using a nasal airflow sensor and stan-dard capnography (CO2SMO, NovametrixMedical Systems, Wallingford, CT). Subjectswere asked to hyperventilate at 30 breaths perminute, using a mouthpiece and metronome,so that PetCo2 was maintained at less than 26mm Hg for five minutes. Colonic tone andphasic contractility were assessed over four fiveminute periods before, during, and for twoperiods after the test period of hyperventila-tion. The arterial pulse rate was monitoredcontinuously using conventional digital pulseplethysmography (CO2SMO, NovametrixMedical Systems, Wallingford, CT). Afterdemonstrating the effects of hyperventilationin the first 15 subjects, an addendum to theprotocol was written to consider whetherhypocapnia was critical to the elicitation of theresponse to hyperventilation. Thus, in sevensubjects, eucapnic hyperventilation was alsoperformed for a five minute period, 15 minutesbefore hypocapnic hyperventilation. Subjectswere instructed to hyperventilate at 30 breathsper minute during which they remainedunaware that Petco2 values were being heldconstant by the addition of 1-2 1/min of carbondioxide to the inlet of the mouthpiece. Theorder of eucapnic, followed by hypocapnic,hyperventilation was not randomised to avoidany possible carry over effects of hypocapnia.Moreover, we measured all parameters for fiveminutes before each hyperventilation periodto ensure that the baseline recordings weresimilar before the two perturbations.

EXPERIMENTAL DESIGNAll subjects were admitted to the GeneralClinical Research Center at St Mary's Hospitalfor bowel preparation comprising 1-5-2 1 oralcolonic lavage solution (polyethylene glycol andelectrolyte solution) on the evening before thestudy. All gave written informed consent andwomen of childbearing potential had a negativeplasma 1-human chorionic gonadotrophinpregnancy test. After an overnight fast,colonoscopy was performed under conscioussedation with intravenous midazolam (0.07mg/kg); sedation was reversed immediatelyafter the procedure with intravenous flumazenil(0.2-0.5 mg) to ensure return to full conscious-ness. No endoscopic abnormalities were seenand the combined manometry and balloonassembly was introduced into the colon over aguidewire and positioned under fluoroscopiccontrol. After 60 minutes' observation, theexperiment was started with continuousmonitoring of colonic tone and motility, Petco2and digital pulse plethysmography as describedabove.

DATA ANALYSIS

Colonic motorparametersPhasic manometric pressure activity andchanges in both pressure and volume ofthe two barostat balloons were sampled asanalogue signals at 8 Hz and converted todigital signals before entry into a computer. Asin previous studies, movement and respiratoryartefacts were filtered out using a modifiedVAX LAB filtering program (Digital Equip-ment Corporation, Boston, MA) to removewaveforms with a frequency of greater than 10per minute, pressure changes of less than 10mm Hg amplitude or less than four secondsduration.22 23 Phasic pressure activity measuredmanometrically was expressed as the activityindex - that is, the area under the contractioncurve per five minute period. The mean mano-metric phasic activity in the transverse and sig-moid colon was calculated from the meanvalues recorded from two sensors positionedproximal and distal to the barostat balloons.The barostat balloon volumes recorded in thetransverse and sigmoid colon were similarlycomputer analysed to separate baseline balloonvolume from phasic volume events. Phasicvolume events were defined as changes involume greater than 10% of baseline volumeand occurring at a frequency of one to four perminute. Baseline volumes were calculated bycomputerised exclusion of the phasic volumeevents from the barostat recordings and aver-aged over 60 second time frames during eachfive minute period to represent colonic tone.

Haemodynamic and respiratory parametersFor each 30 second time frame of the 20minute study, mean pulse rates, respiratoryrates, and end tidal Petco2 values werecalculated; similarly, mean beat to beat pulseintervals were calculated using a modificationof the VAX LAB peak finding program. To

500

on Decem

ber 12, 2020 by guest. Protected by copyright.

http://gut.bmj.com

/G

ut: first published as 10.1136/gut.37.4.499 on 1 October 1995. D

ownloaded from

Effect of hyperventilation on colonic tone

TABLE I Effect of hypocapnic hyperventilation on regional colonic tone, end tidal Pco2pulse rate, and pulse interval variability (n=22, mean (SEM))

Afterfirst After secondBefore HVT time period time period

ToneBaseline barostat volume inTC (ml) 146-3 (10) 131-4 (10) 134 (9) 143 (10)

Mean % change in TC tone* 10.3 (2. 1)t 718 (2.3)t 2 (3)Baseline barostat volume inSC (ml) 118 (9)4 109 (8) 107 (6) 115 (8)

Mean % change in SC tone* 6.5 (1.4)jt 7s4 (1.8)t 116 (1.3)MotilityTC phasic activity index(AUC mm Hg/5 min) 202 (54) 153 (36) 361 (69)t 205 (51)

SC phasic activity index(AUC mm Hg/5 min) 782 (208) 1011 (244) 1360 (341)t 948 (228)

OthersEnd tidal Pco2 (mm Hg) 41.8 (0.7) 23-5 (0.6)t 31-6 (0.8)t 38-9 (0.8)Pulse rate (per min) 65-2 (1-8) 81-2 (1l8)t 65-7 (1-8) 61-5 (1-9)Coefficient of variation of

pulse interval (/o) 16-9 (1-6) 36-8 (4.5)t 29-3 (5.2)t 14-2 (1-4)

HVT=hyperventilation; *relative to pre-hyperventilation; TC=transverse colon; SC=sigmoidcolon; AUC=area under curve.tp<001 v pre-hyperventilation by paired analysis; tp<005 v transverse colon.

obtain a dynamic measure of cardiovascularautonomic activity, the coefficient of variationof the pulse interval per minute (standard devi-ation/mean pulse peak interval) was calculated.Thus, an increase in the coefficient ofvariationin pulse interval implies increased parasym-pathetic activity or decreased sympatheticactivity, or both.27

Statistical analysisEffects of hypocapnic hyperventilation were

assessed on the entire group of 22 healthysubjects. The five minute period before hyper-ventilation and the next three five minuteperiods were compared by paired t tests fornormally distributed data, that is colonic tone(baseline balloon volume), end tidal Pco2value, and pulse rate. For data that were notnormally distributed - that is, colonic phasicactivity and the coefficient of variation of thepulse interval - we used the non-parametricpaired Wilcoxon rank sum tests. In view of theperformance of pairwise comparisons of dataduring hyperventilation and post-hyperventila-tion periods with the pre-hyperventilationperiod, we applied Bonferroni's correction formultiple tests (p<0.05+3), and thus p<0017rather than p<005 was considered significant.

Transversemanometry

Sigmoidmanometry'ai1A.IL A -A a 1

[20

mm 119

so

ml

Transversebarostat _ f

Sigmoidbarostat

Figure 1: A typical recording of the effects of hypocapnic hyperventilation (shaded area) on

phasic and tonic contractility in the transverse and sigmoid colon. Increases in colonic toneare reflected by decreases in baseline barostat volume.

In the eucapnic hyperventilation experiment,two pairwise comparisons with the pre-hyperventilation period were performed,Bonferroni's correction was again applied,and p<0 025 (p<005+2) was consideredsignificant.

Results

COLONIC MOTOR FUNCTIONBaseline barostat balloon volumes in thetransverse and sigmoid colon were different(Table I, p<005) under the conditions of theexperiment, where the operating pressure wasdetermined by the intrabag pressure thatresulted in recording of respiratory artefact.The mean operating pressures in the twoballoons were significantly different (9.8 (0.3)mm Hg in transverse, and 1 14 (0.5) mm Hgin sigmoid colon; p<0 01).

Transverse and sigmoid colonic toneincreased during hypocapnic hyperventilation(Figs 1, 2, and 3). This effect persisted duringthe five minute period after cessation of hyper-ventilation, at a time when PetCO2 was stillsignificantly below baseline (Table I, Fig 4).The mean reductions in barostatic balloonvolumes (consistent with increased colonictone) during hypocapnic hyperventilation were103% and 6.5% in the transverse and sigmoidcolon respectively (see Table I).

Colonic phasic pressure activity was notsignificantly changed during hypocapnichyperventilation (Fig 5). Phasic pressureactivity increased, however, immediately aftercessation of hypocapnic hyperventilation inboth the transverse and sigmoid colon (Fig 5).As anticipated, baseline sigmoid colon motilitywas greater than that seen in the transversecolon.

It is worth noting that the mean baselinebarostat volume during the five minutes beforeeucapnic and hypocapnic hyperventilationwere similar (167 (23) ml and 166 (24) mlrespectively in the transverse colon, and 106(11) and 105 (10) ml respectively in the sig-moid colon). Neither tone or phasic motilitywas significantly changed by eucapnic hyper-ventilation in contrast with significantincreases in both tone and motility afterhypocapnic hyperventilation (Table II).

CARDIOVASCULAR AUTONOMIC FUNCTIONDuring hypocapnic hyperventilation, PetCO2was significantly lower than baseline and thispersisted for five minutes after cessation ofhyperventilation (Table I, Fig 3). Duringhyperventilation, the mean pulse rate increasedover baseline and fell rapidly to normal oncessation of hyperventilation. There was astatistically significant increase in pulse intervalvariability over baseline in both the five minuteperiods during and immediately after hypo-capnic hyperventilation. This effect was notseen in response to eucapnic hyperventilation(Table II). There was a significant increasein pulse rate, however, during eucapnichyperventilation (69 (3) bpm) compared with

.-

501

on Decem

ber 12, 2020 by guest. Protected by copyright.

http://gut.bmj.com

/G

ut: first published as 10.1136/gut.37.4.499 on 1 October 1995. D

ownloaded from

Ford, Camilleri, Hanson, Wiste, Joyner

~ 4 '' U

,'.i,'0''!,'t'''; lime (mm) v-

Figure 2: Effect of hypocapnic hyperventilation (shaded area) on baseline barostat volun- that is, colonic tone - in the transverse and sigmoid colon in 22 healthy subjects.Increased tone, reflected by decreased balloon volumes, are plotted as positive integers. D6show the mean (SEM) increase in tone (%). *p<001 v values before the study.

prehyperventilation (63 (3) bpm, p=0.01 bypaired analysis).

DiscussionThe physiological responses to hyperventilationare complex. At least three components arerecognised in the visceral responses to hyper-ventilation. Firstly, there is an immediateresponse attributable to the effects of 'centralcommand' - that is, reflex brainstem responsesto real or anticipated muscular effort, causingcardiovagal inhibition, venoconstriction, and a

tachycardia with an increase in cardiac out-put.' 3 28 29 A second component of the hyper-ventilation response is the decrease invasoconstriction of the skeletal muscle arterio-lar bed resulting in increased limb bloodflow. 1 This is thought to be the result ofhypocapnia either acting peripherally or cen-

trally. Peripheral effects of hyperventilationinclude decreased intracellular calcium ion

. 10 20

Time (min):Figure 3: Effect of hypocapnic hyperventilation (shaded area) on mean minute barostatvolumes in the transverse and sigmoid colon in 22 healthy subjects. See Fig 2 and Table Ifor summary and statistical analysis. Data shown as mean (SEM).

Pulse 8(bpm) 70

(Mm Hg) 3020

Figure 4: Effect of hypocapnic hyperventilation (shadedarea) on pulse rate (bpm) and end tidal PetC02concentration (mm Hg) in 22 healthy subjects. See Table Ifor summary and statistical analysis.

concentration and increased sensitivity tocatecholamines resulting in vascular smoothmuscle relaxation.' 30 31 Central effects ofhyperventilation include decreased sympatheticoutflow, decreased vasoconstrictor tone incapacitance vessels, -4, 30 32 and increasing cir-culating catecholamine concentrations.33 Thethird component following hyperventilationrepresents reflex baroreceptor and chemore-ceptor homeostatic responses and is charac-terised by the rapid return of the pulse rateto normal on cessation of hyperventilationconsistent with increased cardiovagal activity.Our findings of hyperventilation induced

increases in colonic tone and phasic motilityin the transverse and sigmoid colon raise anumber of important questions.

Are the effects of hyperventilation on colonic tonesignificant or artefactual?The difference in baseline barostat volumes inthe transverse and sigmoid colon reflects thegreater compliance of the first region ratherthan the effects of the intraballoon operatingpressure as the higher pressures used in the sig-moid would be expected to cause a higherintraballoon volume if the two regions hadequal compliance. The increase in colonic tonewith hyperventilation might be construed asbeing clinically insignificant or artefactual as aresult of increased abdominal wall tone.However, the data are not consistent with suchan analysis. The mean reductions in balloonvolumes recorded at the two colonic sites are atleast one third the magnitude of the responseto a 1000 kcal meal (median volume reduction25%34 35). The significant increase in trans-verse and sigmoid colonic tone, the patterns ofits onset, and its persistence during the post-hyperventilation period argue strongly againsta mechanical effect from increased abdominalwall tone or diaphragmatic movement (Fig 3).In addition, eucapnic hyperventilation in thesame subjects at similar respiratory rates didnot change colonic tone. Finally, hyperventila-tion, as performed in this study in a supineposition, has been shown not to increaseelectromyographic activity of the anteriorabdominal wall (M J Ford and L P Degen,unpublished findings).

Are the effects of hyperventilation related tohypocapnia?The phenomenon of a progressive increase in

502

on Decem

ber 12, 2020 by guest. Protected by copyright.

http://gut.bmj.com

/G

ut: first published as 10.1136/gut.37.4.499 on 1 October 1995. D

ownloaded from

Effect of hyperventilation on colonic tone

Sigmoid colon

1000 H

0 5Time(me )

Figure 5: Effect of hypocapnic hyperventilation (shaded area) on phasic contrtransverse and sigmoid colon in 22 healthy subjects. Phasic contractility is exparea under the contraction curve (A UC) relative to the prehyperventilation peshow mean AUC (SEM) (mm Hg/5 min). *p<O.O1 v before study values, t.before study values.

colonic tone from the onset of hytion, increased colonic tone assochypocapnia during the immediate pventilation period, and the absencein colonic tone during eucapnic hytion strongly suggest that the iicolonic tone is a direct conse4hypocapnia. The colonic effects of Pmay either be peripherally ormediated. Studies of vascular smo(have shown that hypocapnia exerts 1smooth muscle and neural effectin smooth muscle relaxation. 1-3vascular and colonic smooth musclesimilarly to hypocapnia, a direct oreffect of hypocapnia would be moidecrease rather than increase colonshown in our study.

Are the effects of hyperventilation neurmediated?The timing of the increase in pulse i

sistent with an initial vagal inhibitiordecreased central parasympatheticould not explain either the increinterval variability as this would be c

cause a decrease in pulse interval vor the increase in colonic tone, as i

cholinergic inhibition decreases coloThe more plausible explanation iincreases in pulse interval varia

TABLE II Effect of eucapnic and hypocapnic hyperventilation on colonic toneend tidal Pco2, pulse rate, and pulse interval variability (n= 7, mean (SEM)

Eucapnia Hypocapnia

HVT Post-HVT HVT

Mean % change in tone relative to pre-HVTTransverse colon 2-2 (4.4) -0-8 (4.0) 12-8 (3.3)*Sigmoid colon 2-3 (2.4) 0-8 (1-4) 6-9 (4-1)*

Mean change in phasic contractility relative to pre-HVT (AUC, mm Hg/5 min)Transverse colon -40 (127) 50 (123) -53 (129)Sigmoid colon 334 (300) 331 (294) 522 (150)*

End tidal Pco2 (mm Hg) 38-7 (0.9) 38-8 (1.1) 21-3 (1 0)*Pulse (bpm) 69-1 (3.4)* 63-7 (3-1) 83-0 (2 1)*Coefficient of variation of pulse

interval (%) 16-6 (3.4) 14-1 (2.2) 49-9 (10-8)*

*p<0-025 and tO025>p<0 05 compared with pre-eucapnic or hypocapnic hyperpaired analyses. Abbreviations as in Table I.

*.

colonic tone resulted from simultaneousdecreases in both central sympathetic andparasympathetic outflow. The increase inphasic colonic contractility, which was onlyseen in the post-hyperventilation period mayrepresent increased parasympathetic activityassociated with the ensueing homeostaticresponses, which are known to return the pulserate to normal by increased cardiovagal tone.

Do colonic responses to hyperventilation reflectcardiovascular responses?

........ The activities and sensitivities of functional15 20 subsets of sympathetic neurones subserving

vasomotor and motility functions are known to^actility in the be under independent central control and haveressed as the different propensities for respiratory modula-nod. Data tion.2 3 12 13 Indeed it has been suggested that.p<OO.05 v respiration does not influence the activity

of most motility regulating sympatheticneurones.12 13 This conclusion was based on

rperventila- studies using single and multi-unit nerveiated with recordings in anaesthetised, paralysed, and,ost-hyper- artificially ventilated cats'3 and may be of lessof change direct relevance to the normal physiology per-

,perventila- taining in conscious humans. These species'ncrease in differences may reflect the variation in experi-quence of mental design such as the levels of hypocapnialypocapnia and the potential interference with the

centrally response of the reticular activating system andoth muscle higher centres by the anaesthesia in animalboth direct studies. None the less, such findings raise thes resulting intriguing possibility that changes in centralThus, if sympathetic activity might evoke different

responded effects on colonic and vascular smooth muscle.peripheral In contrast, we believe that as colonic tone isre likely to demonstrably increased by respiratory stimula-iic tone, as tion, it is more plausible that similar central

neural responses activate changes in bothcardiac and colonic function.

Additional evidence for the effects of hyper-ally ventilation on colonic tone is provided by

studies of the role of hypocapnia in the mech-rate is con- anisms controlling hyperventilation inducedi; however, cerebral vasoconstriction and changes in theic outflow electroencephalogram.4-6 In the absence ofased pulse hypocapnia, hyperventilation in humans~xpected to produces neither electroencephalogram norrariability27 cerebral circulatory changes.4 Furthermore,muscarinic after selective alpha1 adrenergic receptor antag-nic tone.18 onism with thymoxamine or after disruption ofis that the cervical sympathetic innervation, as in tetra-bility and plegia, hyperventilation induced cerebral vaso-

constriction is dramatically diminished, showingandmo

that an intact sympathetic nervous system isand also essential in the normal response to hypo-

capnia.5 6 It seems possible therefore, that theincrease in colonic tone is attributable to

Post-HVT decreased central sympathetic outflow andinvolves sympathetic pathways in the reticular

10.3 (3.8)t activating system, hypothalamic centres, and5-2 (2.8) locus ceruleus, an effect that is critically depend-

100 (111) ent on the presence of hypocapnia and is poten-682 (177)* tially inhibited by anaesthetic agents used in30.0 (1.3)*65.9 (37) experimental animals. Inhibition of sympathetic

activity may increase colonic motility by a direct47-4 (11.4)* effect on myenteric neurones36 or indirectly, by

ventilation by modulating bulbospinal pathways therebyincreasing sacral parasympathetic activity.37-39

2000 rTransverse colon

2000 r

c

EW._-

C.

r_,

1000 H

0 5 10 15 20

503

-c

on Decem

ber 12, 2020 by guest. Protected by copyright.

http://gut.bmj.com

/G

ut: first published as 10.1136/gut.37.4.499 on 1 October 1995. D

ownloaded from

504 Ford, Caniilleri, Hanson, Wiste, Joyner

In summary, we have shown that humancolonic tone is increased by hypocapnic hyper-ventilation. Our findings differ from those ofMaxton et al 16 who studied the effect of hyper-ventilation on rectal motility in patients withIBS. Discrepancies in the two studies mayreflect several differences in experimentaldesign (bowel preparation, evaluation ofsigmoid and transverse colon in our study) orthe human subjects studied, and we need topursue further studies in IBS. The similarityof regional colonic responses, however, andthe timing of concomitant cardiovascularresponses suggest that a central reduction inautonomic nervous control may be the signifi-cant factor, liberating the colon from thenormal restraint of a centrally mediated'sympathetic brake'.36 Our findings raise thepossibility that hyperventilation in IBS patientsmay influence both colonic motor and sensoryfunction. Thus, it may be hypothesised thatincreases in colonic tone lower the set point atwhich tension receptors within the colonic wallare stimulated and contribute to the height-ened visceral perception in some patients withIBS.40A2 The methods established in thisstudy allow us to directly test this hypothesis inpatients with IBS in future studies and toevaluate the effect of the different levels ofhypocapnia associated with hyperventilation inthose patients.We wish to thank Messrs Richard L Tucker and George MThomforde for technical support and Mrs Cinds Stanislav forsecretarial assistance.

This study was supported in part by General ClinicalResearch Center grant no RR00585 from the NationalInstitutes of Health. Dr Ford was supported by awards from theWellington Foundation, Myre Sim Fund, and the FulbrightCommission.

1 Jung RC, Walsh JA, Hyman C. Response of human forcarmmuscle blood vessels to hyperventilation. Cardiovasc Res1971; 5: 347-52.

2 Delius W, Hagbarth KE, Hongell A, Wallin BG.Manoeuvres affecting sympathetic outflow in humanmuscle nerves. Acta Physiol Scaiid 1972; 84: 82-94.

3 Vissing SF, Scherrer U, Victor RG. Stimulation of skin sym-pathetic nerve discharge by central command. Differentialcontrol of sympathetic outflow to skin and skeletal muscleduring static exercise. Circ Res 199 1; 69: 228-38.

4 Patel VM, Maulsby RL. How hyperventilation alters theelectroencephalogram: a review of controversial view-points emphasizing neurophysiological mechanisms._Clinl Neurophysiology 1987; 4: 101-20.

5 Corbett JL, Eidelman BH, Debarge 0. Modification of cen-tral vasoconstriction with hyperventilation in normal manby thymoxamine. Lanicet 1972; ii; 461-3.

6 Eidelman BH, Debarge 0, Corbett JL, Frankel H. Absenceof cerebral vasoconstriction with hyperventilation intetraplegic man. Evidence for neurogenic control of cere-bral circulation. Lanicet 1972; ii: 457-60.

7 Gorman JM, Liebowitz MR, Fryer AJ, Stein J. A neuro-anatomical hypothesis for panic disorder. Ant .7 Psychiao-1989; 146: 148-61.

8 NMcKell TE, Sullivan AJ. The hyperventilation syndrome ingastroenterology. Gastroenterology 1947; 9: 6-18.

9 Chambers JB, Bland JM. Hyperventilation and irritablebowel syndrome. Lanicet 1986; i: 221.

10 Nyhlin H, Ford MJ, Eastwood J, Smith JH, Nicol EF, EltonRA, Eastwood MA. Non-alimentary aspects of theirritable bowel syndrome. .7 P.svchosom;z Res 1993; 37:155-62.

11 Camilleri M, Neri M. Motility disorders and stress: signifi-cance to the irritable bowel syndrome. Dig Dis Sci 1989;34: 1777-86.

12 Janig W, McLachlan EM. Specialized functional pathwaysare the building blocks of the autonomic nervous system.7Auron1 Nerv Syst 1992; 41: 3-14.

13 Boczek-Funcke A, Halber HJ, Janig W, Michaelis M.Respiratory modulation of the activity in sympathetic

neurones supplying muscle, skin, and pelvic organs in thecat. _] Phvsiol 1992; 449: 333-61.

14 Valori RMv, Cole E, Howard R, Cockel R. Hyperventilation(HV) increases the amplitude of esophageal contractionsin healthy volunteers: a controlled study. Guit 1988; 29:A1447.

15 Danielopolu D, Savulesco A. Les effets de l'hvperventila-tion pulmonaire sur la motilite gastrique. W7 de Psio/ ct (/bI'atlrol Geizcer 1928; 16: 225-31.

16 Maxton DG, Prior A, Whorwell PJ. Effect of hyperventila-tion on distal colonic motility and rectal sensitivity in irri-table bowel syndronmc. Digestion 1991; 48: 70-4.

17 Camilleri M, Fealey RD. Idiopathic autonomic denervationin eight patients presenting with functional gastrointesti-nal diseasc. A causal association? Dig Dis Sci 1990; 35:609-16.

18 Bharucha AE, Camilleri M, Low PA, Zinsmeister AR.Autonomic dysfunction in gastrointestinal motility disor-ders. Giut 1993; 34: 397-401.

19 Aggarwal A, Cutts TF, Abell TL, Cardoso S, Familoni B,Bremer J, et al. Predominant symptoms in irritable bowelsyndrome correlate with specific autonomic nervoussystem abnormalities. Gastroenterology 1994; 106: 945-50.

20 Talley NJ, Phillips SF, Wiltgen CM, Zinsmeister AR,Melton LJ. Assessment of functional gastrointestinaldisease: the bowel disease questionnaire. Mayo Clin1 Proc1990; 65: 1456-79.

21 Zigmond A, Snaith R. 'The hospital anxiety depressionscale. Acta P.svchiatr Scanid 1983; 67: 361-70.

22 Steadman CJ, Phillips SF, Camilleri M, Haddad AC,Hanson RB. Variation of muscle tone in the human colon.Gastroenterology 1991; 101: 373-81.

23 von der Ohe MR, Hanson RB, Camilleri MI. Comparison of'simultaneous recordings of human colonic contractionsby manometry and a barostat. \Nenr-ogastroe?terro/ogy and/Mortil/ity 1994; 6: 213-22.

24 Rosenthal RR. Simplified eucapnic voluntary hyperventila-tion challenge. _7 Allergy Cli/u Ininuinol 1984; 73: 676-9.

25 Vansteenkiste J, Rochette F, Demedts M. Diagnostic tests ofhyperventilation syndrome. Eur Respir _f 1991; 4: 393-9.

26 Phan CQ, Tremper KK, Lee SE, Barker SJ. Noninvasivemonitoring of carbon dioxide: a comparison of the partialpressure of transcutaneous and end-tidal carbon dioxidewith the partial pressure of arterial carbon dioxide. 7 C/inMonlitorintg 1987; 3: 149-54.

27 Ewing DJ, Dorsev DQ, Bellavere F, Clarke BF. Cardiacautonomic neuropathy in diabetes: comparison of'measures of R-R interval variation. Diabetol/ogia 1981; 21:18-24.

28 Goodwin GM, McCloskcy DI, Mitchell JH. Cardiovascularand respiratory responses to changes in central commandduring isometric exercise at constant muscle tension. 7Pv/2i()1 1972; 226: 173-90.

29 Browse NL, Hardwick PJ. The deep breath-venoconstric-tion reflex. Clin1 Sci 1969; 37: 125-35.

30 Onrot J, Bernard GR, Biaggioni I, Hollister AS, RobertsonD. Direct vasodilator effect of hyperventilation-inducedhypocarbia in autonomic failure patients. Ani .7 Med Sci1991; 301: 305-9.

31 Baisset A, Cotonat J, Montastruc P. Influence de l'alcalosegazeuse hypocapnique sur les effets beta des svmpathe-tiques. Path Biol (Pari.s) 1969; 17: 21-9.

32 Uvrnas B. Cholinergic vasodilator nerves. Feel Proc 1967; 25:1618-22.

33 Chauhan A, Mullins PA, Taylor G, Petch MC, SchofieldPM. Effect of hyperventilation and mental stress oncoronary blood flow in syndrome X. Br Heart7 1993; 69:516-24.

34 von der Ohe M, Camilleri M, Kvols LK, Thomforde GM.Motor dysfunction of the small bowel and colon inpatients with the carcinoid syndrome and diarrhea. NEnigl.7 Med 1993; 329: 1073-8.

35 von der Ohe M, Hanson RB, Camilleri NI. Serotonergicmediation of postprandial colonic tone and phasicresponses in humans. Guit 1994; 35: 536-41.

36 Gillis RA, Dias Souza J, Hicks KA, Mangel AW, PaganiFD, Hamilton BL, er al. Inhibitory control of proximalcolonic motilitv by the sympathetic nervous system. Ain, 7Physiol 1987; i53: G531-9.

37 Coote JH, Macleod VH. The influence of gulbospinalmonoaminergic pathways on sympathetic nerve activity.-.JPhYsiol 1972; 224: 74-5.

38 Loewv AD. Forebrain nuclei involved in autonomic control.Prog Brain1 Res 1991; 87: 253-68.

39 Loewv AD. Descending pathways to sympathetic andparasympathetic preganglionic neurons. JAuton Nerv Syst1981; 3: 265-75.

40 Ritchie J. Pain from distension of the pelvic colon by inflat-ing a balloon in the irritable bowel syndrome. Gutt 1973;14: 125-32.

41 Whitehead WE, Holtkotter B, Enck P, Hoelzl R, HolmesKD, Anthony J, et al. Tolerance for rectosigmoid disten-sion in irritable bowel syndrome. Gastroeniterology 1990;98: 1187-92.

42 Cook IJ, Van Eeden A, Collins SM. Patients with irritablebowel syndrome have greater pain tolerance than normalsubjects. Gastro>enrer*o/og 1 987; 93: 727-33.

on Decem

ber 12, 2020 by guest. Protected by copyright.

http://gut.bmj.com

/G

ut: first published as 10.1136/gut.37.4.499 on 1 October 1995. D

ownloaded from