effects truncal vagotomy to stimulation and in

Gut, 1987, 28, 1569-1576

Effects of partial truncal vagotomy on intragastricpressure responses to vagal stimulation and gastricdistension in ferretsS A ASALA*, A J BOWER, AND I N C LAWES

From the Department ofAnatomy and Cell Biology, University of Sheffield, Sheffield

SUMMARY Changes in intragastric pressure after dorsal truncal vagotomy, investigated bystimulation of the surviving vagal branches and by step inflation of the stomach, were divided intoan early phase lasting five days, and a late phase continuing for at least three months. During theearly phase the amplitude of vagal evoked contraction was diminished but the resting pressure andthe response to gastric inflation were increased. After the fifth day vagal evoked contractionsdoubled in amplitude but the resting pressure and the response to step inflation of the stomachreturned to control levels. Ventral vagotomy did not produce any substantial changes. Alterationsto gastric and body weight, or to the relation between resting pressure and evoked contraction andrelaxation were excluded as causes of the enhanced vagal effectiveness. Sprouting of axons intodenervated territory occurred too late to explain the changes, but an increase in synaptic densitywithin the innervated territory has not been ruled out.

Changes in the innervation of gastric smooth muscleafter partial vagotomy merit investigation becausemany of the complications of vagotomy relate toalterations in motility. Selective or truncal vagotomywith drainage causes more complications relatedto motility than highly selective vagotomy.'`Recurrence rates of ulceration after highly selectivevagotomy rise considerably, however, with length offollow up' which implies that there is an elementof plasticity involved in secretomotor innervation.Surgery for peptic ulceration is therefore a com-promise between avoiding recurrent ulceration andrisking disorders of motility.Changes in motiity after various forms of vagotomy

include alteration in the rate of gastric emptying.'reduction of phase III of cyclic motor activity89 andirregularity of pacemaker potentials."' The effects ofvagotomy on motility alter with time""'3 and many

Address for correspondence: I N C Lawcs, Dcpartmcnt of Anatomy,University of Sheffield, Western Bank, Shefficld Sl(O 2TN.

Received for publication 8 May 1987.

* Present address: Dept. of Human Anatomy, Faculty of Mcdicine,Ahmadu Bello University, Zania, Nigeria

1569

of the effects are particularly evident in the first fewpostoperative months. The present investigation wasdesigned to characterise plasticity in motor innerva-tion, particularly with respect to which vagal trunk isinvolved, the time course of such changes and thepossible underlying mechanisms.The recovery of vagal evoked contractions of the

ferret stomach after partial truncal vagotomy mayinvolve changes in the target organ, increased effec-tiveness of the surviving extrinsic innervation, orboth. Resting pressure partly determines the ampli-tude of vagal evoked contractions.'4 This may explainwhy larger stomachs are associated with greaterevoked contractions"' which, in turn, may explain thecorrelation of body weight with the amplitude ofevoked contractions.'5 If vagotomy altered any ofthese factors then the amplitude of evoked contrac-tions could have increased as a consequence ofchanges in the target organ alone, rather than in thesurviving vagal branches. Thus if vagotomy caused anincrease in gastric weight or an increase in restinggastric pressure, then an increase in vagal evokedcontraction could be explained without recourse toneuronal plasticity. For the purposes of this paper,neuronal plasticity is defined as any longterm adap-

on Novem

ber 26, 2021 by guest. Protected by copyright.

http://gut.bmj.com

/G

ut: first published as 10.1136/gut.28.12.1569 on 1 Decem

ber 1987. Dow

nloaded from

http://gut.bmj.com/

Asala, Bower, and Lawes

tive change in a nerve which, in the present context,compensates for loss of function.

Differential loss of relaxatory influences" and theredundancy known to exist in the vagal innervation ofthe stomach'" are additional potential mechanismswhich have to be excluded before functional recoverycan be attributed to plasticity in the vagus. Aprevious study" indicated that sectioning the leftcervical vagus induced an increase in responsesevoked by the surviving right vagus, but not viceversa, suggesting the possibility that the left vagus ismore potent than the right. Their apparent equival-ence may be the result of smooth muscle limitationsat suprathreshold levels of stimulation.'4 This possi-bility was investigated.

In the ferret,"';" the dorsal trunk is distributed tothe dorsal surface of the corpus and both surfaces ofthe antrum. The ventral trunk reaches only theventral surface of the stomach, principally the corpus.The number of efferent axons is the same in thedorsal and ventral trunks and there are also equalnumbers of efferents in the left and right cervicalnerves. Equal numbers of neurones in the dorsalmotor nuclei project to the stomach," but more in theleft nucleus project to the corporoantral junctionwhere contractions begin.The aims of the present investigation were: (i) to

determine the time course of recovery of evokedgastric contractions after partial vagotomy; (ii) toinvestigate any changes in the intrinsic motor res-ponses of the stomach; (iii) to examine the possibilitythat changes in body weight, gastric weight, or restingintragastric pressure might contribute to the increasein vagal evoked contractions; (iv) to investigatepossible changes in vagal evoked relaxation, and inparticular to determine whether a selective loss ofinhibition explained enhancement of contraction.

Gastric emptying and rhythmic contractions arecomplex phenomena which depend on a number offactors, any of which might have been altered inopposing directions by vagotomy. For this reason theless complicated contractions evoked by gastric in-flation and vagal stimulation were investigated in thisstudy.

Methods

ANIMALSSeventy seven ferrets of either sex weighing 415-1600g were used. Animals for vagotomy were anaesthe-tised with pentobarbitone (60 mg/kg ip). The dorsalor the ventral abdominal trunk of the vagus wasligated below the diaphragm with two silk ligatures1 cm apart. The intervening nerve was removed andlater examined histologically. The abdominal woundwas closed in layers and the animals were allowed to

recover. The postoperative diet was water on thefirst day, milk on the second, and meat with fluidthereafter.Animals were randomly allocated to one survival

group, the groups being allowed to survive for 3, 5, 7,21, 28 or 84 days. At the end of the survival period theanimals were anaesthetised with urethane (1-5 g/kg50% w/v in 154 mM-NaCl ip). A tracheostomy wasdone in all animals. The right external jugular veinwas cannulated for the administration of drugs. Thepylorus was occluded by a ligature. The stomach wasintubated via the oesophagus with a firm plastic tube5 mm external diameter, inside which was a secondtube 2 mm external diameter. The outer tube wasused to inflate the stomach rapidly with 20 ml154 mM-saline at 380C. Intragastric pressure wasmonitored via the inner tube connected to a pressuretransducer (Bioscience model 8138), the output fromwhich was displayed on a chart recorder (Bryans28000).The cervical vagi were divided and the peripheral

cut ends were stimulated (Devices Isolated Stimulatortimed by a Digitimer D4030) using parameters givingmaximal responses" (10 Hz, 10 s trains, 0(5 mspulses, voltages as in the results) delivered via bipolarsilver electrodes immersed in liquid paraffin at 39°C.Care was taken to shield the electrodes with plasticcovers.The sections of tissue removed at vagotomy were

examined histologically and were shown to be neural.At the end of the experiment, the cervical vagi werestimulated after division of the surviving vagal trunk.In no case was there any evoked contraction, con-firming that the previously vagotomised trunk hadbeen totally divided as far as motor axons wereconcerned.

In all cases the pressure was measured with thetransducer placed at the same height as the supineanimals' vertebral column, or a correction wasapplied to give the equivalent pressure.

STATISTICAL ANAL YSIS

The data were subjected to factor analysis, regressionanalysis, and paired (within a group of animals) orunpaired (between groups of animals) t tests. Resultsare expressed as means±SE of the mean.

Results

VAGAL EVOKED CONTRACTIONSDorsal vagotomyStimulation of the cervical vagi (20 V, 10 Hz, 0(5 mspulses, 10 s duration) caused a brief contraction ofthe stomach followed by a relaxation lasting severalminutes.The results of stimulating the left and/or right

1570

on Novem


http://gut.bmj.com

/G


ber 1987. Dow

nloaded from

http://gut.bmj.com/

Gastric motility after truncal vagotomy

Table 1 Vagal evoked contractions after dorsal vagotomy

Source of Sumn of Meanvariance squarex if square F P

Nerves 38X1 2 19'0 10( nsDays 2213 8 6 369 0 20(1 <(0(1Nervexday 62-2 12 5 2 0(3 nsWithingroup 1931 4 105 18 4Total 4245-5

Five animals per group. Nerves: left. right or both cervical evokedresponses; day: postoperative survival for 0, 3, 5, 7, 21 or 28 days;nerve x day: interaction between ncrves and postoperative survival;within group: variancc within cach experimental group.

cervical vagi were subjected to factor analysis (Table1), excluding the group which survived 84 daysbecause it had only four members. Time had a highlysignificant effect on the amplitude of evoked res-ponse, but it did not matter which nerve was stimula-ted - that is, stimulation of the left, right or bothcervical nerves evoked the same amplitudes of res-ponse. Because the identity of the stimulated nervemade no difference, subsequent analysis was con-fined to left cervical evoked responses.

Responses evoked at different intervals aftervagotomy were compared with each other using anunpaired t test. In the control group of six ferretsstimulation of the left cervical vagus evoked contrac-tions of 2-39±0 25 kPa (24.4±+25 cmH2O, Fig. 1).After division of the dorsal trunk (day 1 responseevoked via the ventral trunk) the amplitude fell to1 22±0*39 kPa (12.4±4.0 cmH2O, p<0.005), similarto the level obtained by stimulating the ventral trunk

27-_3I 24-

u 21

c

.o 18-

C 15-

c 12-

9-w

0

iQ> 6-

3a

Table 2 Vagal evoked contractionis after ventral vagotomy

Source of Sumn of Meantvariance squares (if square F P

Nerves 6-7 2 3 3 0(3 nsDay 1815 5 36 3 3.1 <0(05Nervexday 46 2 10O 4.6 0(4 nsWithingroup 849 5 72 12-0Total 1083-9

Four animals pcr group. Nerves: left, right or both cervical evokedresponscs: day: postopcrative survival for 0, 3, 5, 7, 21 or 28 days;nerve xday: interaction between nerves and postopcrative survival:within group: variance within eaich cxperimentail group.

before dorsal vagotomy. The amplitude on the thirdday (1 35±0 25 kPa, 13 8±2 5 cmH2O) was lowerthan the control response (p<002) and was notsignificantly different from the day 1 response. By thefifth day, however, the amplitude (2.01±0-23 kPa,20 5±2-3 cmH2O) had returned to control levelsand had increased significantly above day 1 levels(p<0-01); on all subsequent days the responsesremained significantly above the day 1 levels(p<0005). As a check to ensure that the recoverywas sustained, the responses on day 84 were com-pared with the greatest responses (day 21) and foundnot to be significantly different.Thus dorsal vagotomy halved the amplitude of the

response evoked by stimulation of the cervical vagus,and the response remained low for three days. By thefifth day responses showed a substantial enhance-ment to prevagotomy levels, where they remainedfor at least three months.

Ventral vagotomyThe responses to stimulation of the left cervical vagusin control animals and at various times after ventralvagotomy are shown in Figure 2. Factor analysis(Table 2) indicated that there was no significantdifference in the responses to stimulation of the left,

271

C O 3 57 21 28Days after dorsal vagotomy

c

a4

2

C21

U0

b I 18'

O Ea, u 15

84 u 12a.

Fig. 1 Gastric contractions evoked by stimulation ofthe leftcervical vagus after dorsal vagotomy. On thefirstpostoperative day contractions were reduced to the amplitudeofthose evoked by stimulating the surviving ventral trunk. Bythefifth postoperative day, however, they had increased toprevagotomy values, equal to those previously evoked bystimulating the dorsaltrunk. (10 cmH,0=0-098 kPa)

C0 7 21 28 84Days after ventral vagotomy

Fig. 2 Gastric contractions evoked by stimulation ofthe leftcervical vagus after ventral truncal vagotomy. There was littlechange immediately after ventral vagotomy, and minimalincreases above the immediate postvagotomy levels werefound after2l and 84days. (10 cmHO-=0098kPa)

I

1571

O L

II imI

on Novem


http://gut.bmj.com

/G


ber 1987. Dow

nloaded from

http://gut.bmj.com/


right or both cervical nerves. Thus the maximalcontributions of the left and right cervical vagi to thedorsal trunk were equal.The control amplitude of contraction (2-24±0419

kPa, 22-9±1 9 cmH2O) was not significantly differ-ent from any of the amplitudes after vagotomy. Theresponses immediately after vagotomy (day 1 res-ponses evoked via the dorsal trunk, 1.87+0-11 kPa,19 1± 11 cmH2O) were significantly lower than theresponses on day 21 (2-33±0.12 kPa, 23-8±1-2cmH2O, p<0.025) and on day 84 (2.26±0.14 kPa,23 1±14 cmH2O, p<0.05) but these differenceswere small and unlikely to be of great physiologicalimportance. This shows that surgery per se, even ofthe ventral trunk, does not cause changes of themagnitude seen after dorsal vagotomy. Becausethere were no substantial changes after ventralvagotomy, the ventral groups were not examinedfurther.

Body and gastric weightsTo confirm that the allocation of ferrets to differentsurvival groups had been randomly allocated success-fully, the body weights of the ferrets which hadreceived dorsal vagotomies were subjected toanalysis of variance. Allocation to different survivaltimes had no significant effect (F=0-963, df=6 and21) indicating the the groups were comparable.

Analysis of variance of the stomach weights of theanimals which had received a dorsal vagotomy in-dicated that there was no significant differencebetween the groups (F= 1-669, df=6 and 21).

Comparison of cervical vagiThe cervical vagi were stimulated at various intensi-ties, from near threshold (4 V) to maximal levels (16V). The responses to stimulation of either cervicalvagus were equal at 8, 12, and 16 V. At 4 V, however,the left vagus evoked responses 50-100% greaterthan those obtained from the right (p<002).

RESPONSES TO STEP INFLATION

Responses to inflation of the stomach after division ofboth the cervical vagi must reflect intrinsic reflexes ormyogenic contraction of the smooth muscle. Changesin either of these non-vagal mechanisms were testedcollectively by giving a step inflation of the stomach.Inflation of the stomach with 20 ml warm saline inunder 2 s caused an immediate increase in pressurewhich fell to a plateau within five minutes.The peak pressure was measured in control ferrets

and at various times after dorsal vagotomy (Fig. 3).All groups had a bilateral cervical vagotomy an hourbefore step inflation of the stomach. After dorsalvagotomy peak pressures rose by more than 30% ofthe control values (2 18±0 17 kPa, 22-2+±17 cmH2O

28-

24-

E 20-w

,, 16-

a 12-a,L -0 4

4-1

0 CO 3 57~~212 8IC 0 3 5 7 21 28 84

Days after dorsal vagotomy

Fig. 3 Peak pressures in response to20 ml step inflationafter dorsal vagotomy. For thefirst three days there was anincrease in the pressure obtained, but by thefifth day thepeakpressure had returned to control levels. These changes are theinverse ofthose affecting vagal evoked contractions (Fig. 1).(10 cmrHO=0098 kPa)

v 1-63+0-11 kPa, 16-6+1-1 cmH2O, p<0.01). Onday 3 the peak pressures were still significantly raised(2-35±0-25 kPa, 24-0+2-6 cmH2O, p<0025). By thefifth day, however, the peak pressures were no longersignificantly above control levels, and on subsequentdays they returned to the control values.Enhanced contraction was temporary, therefore,

disappearing at the same time (day 5) as the survivingventral trunk became more effective.

RESTING INTRAGASTRIC PRESSURE

The resting pressure was measured 30 minutes afterinflation of the stomach with 20 ml warm saline (Fig.4). After dorsal vagotomy the resting pressure in-

9

I

E_

6

._ o

a;aL.

u

:- 3.

CO 3 57 2128 84Days after dorsal vagotomy

Fig. 4 Resting pressures with an intragastric volume of20ml after dorsal vagotomy. Resting pressure was significantlyraisedfor thefirst three days after dorsal vagotomy, but itreturned to control levels afterfive days. There was nosignificant reduction below control levels. (10 cmH,O=0-098 kPa)

n, . .

1572

0

II

4 4 0

-L

i

on Novem


http://gut.bmj.com

/G


ber 1987. Dow

nloaded from

http://gut.bmj.com/

Gastric motility after truncal vagotomy

creased, reaching its highest level three days later(0-76±0-02 kPa, 7-8±0-2 cmH2O), when it wassignificantly higher than control levels (0.61±0.04kPa, 6-2+04cmH20, p<0005). The resting pressurethen fell, and by the fifth day it was no longersignificantly raised. The pressure on day 3 wassignificantly higher than on day 7 and on all later days(p<0.05).Thus dorsal vagotomy had a similar effect on

resting pressure as on peak pressure: the increase inresting pressures lasted only until the survivingventral branch became more effective in evokinggastric contractions (Fig. 1 v Fig. 4). There was noevidence that the resting pressure fell below controllevels.

EVOKED CONTRACTIONS AND RESTINGPRESSUREWithin hours of vagotomy'4 the regression of evokedcontractions (Y) on resting pressure (X) was:Y=2*61-1-74xX, n=64, p<0-001 (kPa)Y=26-67-1-74xX, n=64, p<0001 (cmH2O)

Evoked contractions in 10 of 12 (83%) of the ferretsin the 0-3 day groups were less than 1-5 cmH2Oabove the pressure predicted by this equation,whereas 21 of 28 (75%) of the evoked contractions inthe 5-84 day groups were more than 0-15 kPa (1-5cmH2O) above the predicted pressure (X=11i697,df=1, p<0001). In short, for a given resting pres-sure, a greater evoked contraction was obtained afterday 5 than before. This is further evidence that

0-

o -1.I

C

o -2-

-3-

4 1 I

0

00o0\o o

0

0

00 0

0

24 6 8 ~~~~~~~~10Resting pressure (cm H20)

Fig. 5 Relation ofvagal evoked relaxation to restingpressure more than one day after dorsal vagotomy(continuous line) . For comparison the regression lineobtained on the day ofvagotomy is included (broken line).There is no evidence ofreduced relaxation at a given pressurein the vagotomised group. (JO cmH,O=0-098 kPa)

changes in resting pressure did not account for thedifferences in evoked contractions before and afterthe fourth day.

VAGAL EVOKED RELAXATIONIt might be postulated that an increase in vagalevoked contraction was the result of a selective loss ofvagal inhibition, leaving excitatory effects unchal-lenged. Immediately after 10 s of vagal stimulationthere is a contraction followed by a period ofrelaxation, the latter being an indication of vagallydependent inhibitory processes.'3

Vagal evoked relaxation takes over 10 min torecover even after brief periods of stimulation, soonly the first relaxatory response was analysed ineach ferret. The relaxation obtained 1-84 days afterdorsal vagotomy is plotted versus resting pressure inFigure 5. As the resting pressure increased the degreeof relaxation also increased. The regression equationwas:Y=0 08-0 40xX, n=40, p<0-001 (kPa)Y=0-77-0*40xX, n=40, p<0-001 (cmH2O)

where Y is the degree of relaxation and X is theresting pressure. For comparison the regression lineobtained from a control group (one hour aftervagotomy) in a previous experiment is also shown inFigure 5. It is evident that the slopes of the two linesare very similar. The degree of relaxation obtainedafter dorsal vagotomy was marginally greater than inthe control group.Thus there is no evidence for a selective loss of

inhibition after dorsal vagotomy nor for an alterationof the relation between vagal evoked relaxation andresting pressure.

Discussion

Two processes appear to affect vagal evoked gastriccontractions after partial truncal vagotomy: an earlyprocess enhances the contraction of'the stomach inresponse to a step inflation or a static intragastricvolume, while a later process coming into effect asthe first process wanes, enhances the effectiveness ofsurviving vagal branches. Enhancement of vagalevoked contractions was not explained by changes inthe weights of ferrets or of their stomachs, by adecrease in resting pressure, nor by a differential lossof vagal inhibition.

EARLY CHANGESEither an increased ability of muscle to contract or anenhanced excitability of the intrinsic reflexes wereindicated by the increases in resting and inflationevoked pressure. The changes in peak (Fig. 3) andvagal evoked (Fig. 1) pressures were striking mirrorimages of each other, suggesting that the increased

1573

-/

on Novem


http://gut.bmj.com

/G


ber 1987. Dow

nloaded from

http://gut.bmj.com/


effectiveness of the surviving nerve might have beena signal for the disappearance of peripheral compen-sation, analogous to the loss of extrajunctional recep-tors after reinnervation of skeletal muscle."4 Changesin the muscle or the intrinsic reflexes did not,therefore, account for the eventual increase in effec-tiveness of the surviving vagal branches.These early changes develop at the same time as

the hypergastrinaemia which follows vagotomy indogs"5 and man."6 Although gastrin is known to affectthe gastric pacemaker and the generation of actionpotentials,"7 8 no link with the early changes in thepresent experiments has been established.

LATER CHANGESWith the exclusion of non-vagal factors and ofdifferential loss of vagal inhibition, the increase ineffectiveness of the surviving vagal branches must becaused by changes in the branches themselves or inthe distribution of their activity. Possible mech-anisms include neuronal sprouting, which couldincrease the area occupied by preganglionic neurones;increased synaptic density without change in territory;or redistribution of postganglionic neurones.

Autoradiographic studies,'" done in conjunctionwith the present study, showed that territorial gainsby sprouting did occur, but too late (Al-Muhtaseb,personal communication) and in the wrong trunk toaccount for the changes in contractility recorded inthe present work, although they may be related tolater changes in secretion.While sprouting into neighbouring territory was

too late, an early increase in synaptic density withinthe original territory of the surviving branches is notexcluded. Increases in synaptic density29 and theformation of functional synapses3';32 occur earlyenough to account for the enhanced effectiveness ofthe ventral vagus found on the fifth postoperativeday, although there is no positive evidence of this.An alternative explanation of the results is that

after dorsal vagotomy, postganglionic neuronesspread into the denervated territory, distributingactivity from the ventral vagus.33

FACTORS GOVERNING THE PHYSIOLOGICALCHANGESThe fact that dorsal vagotomy caused a substantialincrease in the effectiveness of the ventral trunk andnot vice versa requires explanation, as does theprevious finding'7 that left, but not right, cervicalvagotomy caused a similar increase.

Differences between branches of the vagusThere has been disagreement over whether the leftcervical vagus has gastric effects greater than, orequal to, those of the right in dogs, cats, rabbits,4

rats,` cats"' and ferrets." These discrepancies may beresolved by the present finding in the ferret thatwhereas the cervical vagi evoked equal responses atall suprathreshold levels of stimulation (when smoothmuscle imposed a ceiling), at threshold levels the leftcervical vagus evoked greater contractions. Thosewho found a greater potency of the left vagus mayhave been using lower intensities of stimulation.Sectioning the dorsal trunk or left cervical nervewould cause a greater loss of excitatory activity at low(presumably more physiological) intensities of stimu-lation than sectioning the ventral trunk or rightcervical nerves. Loss of activity is a stimulus tosprouting."'The difference in the ability of the ventral and

dorsal trunks to evoke gastric contractions relates tothe difference in the distribution of these nerves inthe ferret.'""' Nerve evoked gastric activity spreadsfrom one part of the stomach to another only whenthere is a neural connection present.8

Degeneraton as a stimulusDegenerating nerves induce growth of neurites inneighbouring axons. 9- l8 Degeneration of affer-ents,39 or loss of an inhibitory factor4" also inducesprouting in efferent nerves. The dorsal trunk hasconsiderably more afferents than the ventral'8 andthis may explain why sectioning the dorsal trunkcaused a substantial increase in ventral evoked res-ponses but not vice versa. A similar argument hasbeen applied to left versus right cervical vagotomy inferrets. '7

Functional overlap and peripheral ceilingsIn intact ferrets stimulation of the dorsal vagus gives aresponse as great as that obtained from stimulation ofboth abdominal vagi together,'7 suggesting that theperipheral mechanism imposes a ceiling on theamplitude of evoked contraction'4 which the normaldorsal trunk is capable of reaching independently.Thus plastic changes in the dorsal vagus after ventralvagotomy would not be expected to exceed theceiling. The normal ventral vagus, on the other hand,cannot usually evoke a maximal contraction.Plasticity occurring in the ventral vagus, therefore,may allow this branch fuller access to the peripheralmechanism and permit visible changes in evokedcontraction.Changes in motility in the first few postoperative

days after partial vagotomy involve compensatorymechanisms confined to smooth muscle or its in-trinsic plexuses. By the fifth postoperative day,however, surviving branches of the vagus becomemore effective and the non-vagal compensatorymechanisms disappear. Sprouting of axons intoneighbouring territory occurs at three months, too

1574

on Novem


http://gut.bmj.com

/G


ber 1987. Dow

nloaded from

http://gut.bmj.com/

Gastric motility after truncal vagotomy 1575

late to account for the enhanced effectiveness of thesurviving vagus. An increase in synaptic densitywithin the original territory must therefore be con-sidered as an explanation. The extent to which thesechanges apply to man remains to be determined. Ifearly changes in synaptic density do occur, thenmethods which prevent this4'4" may be able to con-tribute to the reduction of recurrent ulceration,alongside developments in surgical technique whichcombine the advantages of truncal and parietal cellvagotomy."

We would like to thank P L R Andrews who madehelpful comments on an earlier draft of this paper.

References

1 Donahue PE, Bombeck CT, Condon RE, Nyhus LM.Proximal gastric vagotomy versus selective vagotomywith antrectomy: results of a prospective, randomizedclinical trial after four to twelve years. Surgery 1984; 96:585-90.

2 Hoffman J, Jensen H-E, Schulze S, Poulsen PE,Christiansen J. Prospective controlled vagotomy trialfor duodenal ulcer: results after five years. Br J Surg1984; 74: 582-5.

3 Koffman CG, Hay DJ, Ganguli PC, et al. A prospectiverandomized trial of vagotomy in chronic duodenalulceration. Four year follow up. Br J Surg 1983; 70:342-5.

4 Stoddard CJ, Johnson AG, Duthie HL. The four toeight year results of the Sheffield trial of electiveduodenal ulcer surgery - highly selective or truncalvagotomy? Br J Surg 1984; 71: 779-82.

5 Adami H-0, Enander L-K, Enshog L, Ingvar C, Ryd-berg B. Recurrences 1 to 10 years after highly selectivevagotomy in prepyloric and duodenal ulcer disease. AnnSurg 1984; 199: 393-9.

6 Andersen D, Amdrup E, Hostrup H, Sorensen FH. TheAarhus County vagotomy trial: trends in the problem ofrecurrent ulcer after parietal cell vagotomy and selectivegastric vagotomy with drainage. World J Surg 1982; 6:86-92.

7 Gulsrud PO, Taylor IL, Watts HD, Cohen MB,Elashoff J, Meyer JF. How gastric emptying of carbo-hydrate affects glucose tolerance and symptoms aftertruncal vagotomy with pyloroplasty. Gastroenterology1980; 78: 1463-71.

8 Gleysteen JL, Sarna SK, Myrvik AL. Canine cyclicmotor activity of stomach and small bowel: the vagus isnot the governor. Gastroenterology 1985; 88: 1926-31.

9 Mayer EA, Thomson JB, Jehn D, et al. Gastric empty-ing and sieving of solid food and pancreatic and biliarysecretions after solid meals in patients with nonresectiveulcer surgery. Gastroenterology 1984; 87: 1264-71.

10 Bortolotti M, Labo G, Serantoni C, Ciani P. Effect ofhighly selective vagotomy on gastric motor activity ofduodenal ulcer patients. Digestion 1978; 17: 108-20.

11 Cowley DJ, Vernon P, Jones T, Glass HI, Cox AG.Gastric emptying of solid meals after truncal vagotomy

and pyloroplasty in human subjects. Gut 1972; 13:176-81.

12 Donovan IA, Clarke RJ, Gunn IF, Alexander-WilliamsJ. A comparison of gastric emptying at 3 and 12 monthsafter proximal gastric or selective vagotomy withoutpyloroplasty. Br J Surg 1974; 61: 889-92.

13 Gleysteen JJ, Kalbfleisch JH. Progression of changes ingastric emptying of hypertonic liquids after proximalgastric vagotomy. Dig Dis Sci 1981; 26: 119-23.

14 Andrews PLR, Lawes INC. Gastric tone modifies theresponses to extrinsic neural stimuli in the anaesthetisedferret. J Physiol 1985; 366: 1-16.

15 Andrews PLR, Lawes INC. The role of vagal andintramural inhibitory reflexes in the regulation of intra-gastric pressure in the ferret. J Physiol 1982; 326:435-51.

16 Andrews PLR, Grundy D, Lawes INC. The role of thevagus and splanchnic nerves in the regulation of intra-gastric pressure in the ferret. J Physiol 1980; 307:401-11.

17 Andrews PLR, Lawes INC, Bower AJ. Peripheralfunctional organisation of vagally evoked gastric motorresponses in the ferret. Gut 1980; 21: 811-7.

18 Asala SA, Bower AJ. An electron microscopic study ofthe vagus nerve composition in the ferret. Anat Embryol1986; 175: 247-53.

19 Asala SA, Bower AJ. Vagal innervation of the gastricwall - an autoradiographic study. J Anat 1984; 138: 589.

20 Mackay TW, Andrews PLR. A comparative study of thevagal innervation of the stomach in man and the ferret. JAnat 1983; 136: 449-81.

21 Odekunle A, Bower AJ. The brainstem localization ofthe vagal gastric efferent neurons in the ferret; ahorseradish peroxidase neurohistochemical study. JAnat 1983; 138: 589.

22 Andrews PLR, Scratcherd T. The gastric motilitypatterns induced by direct and reflex excitation of thevagus nerves in the anaesthetized ferret. J Physiol 1980;302: 363-78.

23 Andrews PLR, Lawes INC. Characteristics of thevagally driven non-adrenergic non-cholinergic inhibitoryinnervation of ferret gastric corpus. J Physiol 1985; 363:1-20.

24 Frank E, Jansen JKS, Lomo T, Westgaard RH. Theinteraction between foreign and original motor nervesinnervating the soleus muscle of rats. J Physiol 1975;247: 725-43.

25 Hollinshead JW, Debas HT, Yamada T, Elashoff J,Osadachey B, Walsh JH. Hypergastrinaemia developswithin 24 hours of truncal vagotomy in dogs. Gastro-enterology 1985; 88: 35-40.

26 Busman DC, Munting JDK. Results of highly selectivevagotomy in a non-university teaching hospital. Br JSurg 1982; 69: 62t-4.

27 Morgan KG, Schmalz PF, Go VLW, Szursewski JH.Effect of pentagastrin, G17 and G34 on the electricaland mechanical activities of canine antral smoothmuscle. Gastroenterology 1978; 75: 304-412.

28 Kelly KA. Effect of gastrin on gastric myoelectricactivity. Am J Dig Dis 1970; 15: 399-405.

29 Cotman CW, Nieto-Sampedro M. Cell biology ofsynaptic plasticity. Science 1984; 225: 1287-94.

on Novem


http://gut.bmj.com

/G


ber 1987. Dow

nloaded from

http://gut.bmj.com/

1576 Asala, Bower, and Lawes

30 Betz WJ, Caldwell JH, Ribchester RR. Sprouting ofactive nerve terminals in partially inactive muscles of therat. J Physiol 1980; 303: 281-97.

31 Henderson CE, Hutchet M, Changeux J-P. Denervationincreases a neurite-promoting activity in extracts ofskeletal muscle. Nature 1983; 302: 609-11.

32 Holland RL, Brown MC. Postsynaptic transmissionblock can cause terminal sprouting of a motor nerve.Science 1980; 207: 649-51.

33 Brackett KA, Crocket A, Joffe SN. Sequential ultra-structural study of mucosal innervation followingparietal cell vagotomy and antrectomy. Experientia1984; 40: 850-2.

34 McCrea ED, McSwiney BA, Stopford JSB. The effecton the stomach of stimulation of the peripheral end ofthe vagus nerve. Q J Exp Physiol 1925; 15: 201-33.

35 Lewis PR, Scott JA, Navaratnam V. Localization in thedorsal motor nucleus of the vagus in the rat. JAnat 1970;107: 197-208.

36 Bauer RF. Electrically induced vagal stimulation ofgastric acid secretion. In: Brooks FP, Evers PW, eds.Nerves and the gut. New Jersey: CB Slack Inc, 1977,86-95.

37 Duchen LW, Tongue DA. The effects of implantation of

an extra nerve on axonal sprouting usually induced bybotulinum toxin in skeletal muscle of the mouse. J Anat1977; 124: 205-16.

38 Daniel EE, Sarna SK. Distribution of excitory vagalfibers in canine gastric wall to control motility. Castro-enterology 1976; 71: 608-613.

39 Brown MC, Ironton R. Sprouting and regression ofneuromuscular synapses in partially denervated mam-malian muscles. J Physiol 1978; 278: 325-48.

40 Diamond J, Cooper E, Turner C, Macintyre L.Trophic regulation of nerve sprouting. Science 1976;193: 371-7.

41 Schwartz M, Spearman N. Sprouting from chickenembryo dorsal root ganglia induced by nerve growthfactor is specifically inhibited by affinity-purified anti-ganglioside antibodies. Proc Natl Acad Sci 1982; 79:608W-3.

42 Gurney ME. Suppression of sprouting at the neuro-muscular juncton by immune sera. Nature 1984; 307:546-8.

43 Taylor TV, Gunn AA, Macleod DAD, MacLennan I.Anterior lesser curve seromyotomy and posteriortruncal vagotomy in the treatment of chronic duodenalulcer. Lancet 1982; ii: 846-9.

on Novem


http://gut.bmj.com

/G


ber 1987. Dow

nloaded from

http://gut.bmj.com/

effects truncal vagotomy to stimulation and in

Documents