The gastrocolic response: Evidence for a neural mechanism

GASTROENTEROLOGY

The Gastrocolic Response: Neural Mechanism

Evidence

77:1235-1240.1979

for a

W. J. SNAPE, Jr., S. H. WRIGHT, W. M. BATTLE, and S. COHEN Gastrointestinal Section, Department Philadelphia, Pennsylvania _

of Medicine, Hospital of the University

The aim of this study is to determine the effect of anticholinergic therapy on the gastrocolonic response to a standard meal or its major constituent fat. A rapid increase in rectosigmoidal spike activity occurs after ingesting the standard meal or the fat meal (P < 0.02). Distal colonic motility returns to fasting levels 50 mm after both meals. There is no further increase in spike activity after the 1000-caJorie meal, but spike activity increases again 70 min after ingesting the fat (P < 0.02). The anticholinergic drug, clidinium bromide, inhibits the early increase in spike activity after both meals. However, the anticholinergic has no effect on the delayed peak of activity following the ingestion of fat. This study suggests that (a) the early gastrocolic response to a standard meal and a fat meal is cholinergically mediated and @) the late increase in rectosigmoidal motility occurs only after fat ingestion and may be controlled by other neural mediators or possibly the gastrointestinal hormones.

Previous studies showed an increase in colonic spike and motor activity after eating a “standard” American lunch.’ This gastrocolonic response to eating may be mediated by gastrointestinal hormonal stimulation of colonic smooth muscle or by neural re-

Received October 2,1976. Accepted June 21,1979. Address requests for reprints to: William J. Snape, Jr., M.D., Gastrointestinal Section, Hospital of the University of Pennsylvania, 36th and Spruce Streets, Philadelphia, Pennsylvania 19164. This work was supported in part by Research Grant R01 AM 19379 and Research Grant NIH 5 MO1 RR66646 to the Clinical Research Center of the University of Pennsylvania from the National Institutes of Health and by a grant from Ileitis and Colitis Foundation. These data were presented at the Annual Meeting of the American Gastroenterological Study in Las Vegas, Nevada, May 24, 1976. The authors thank Ms. Patricia Rosso for technical assistance and Ms. Marie Simmons for secretarial assistance. 0 1979 by the American Gastroenterological Association ~165085/79/121235-~~2.~

of Pennsylvania,

flexes that originate in the upper part of the gastrointestinal tract.le3 The gastrointestinal hormones, gastrin or cholecystokinin, stimulate colonic motility when infused in concentrations that mimic physiologic blood levels of the hormones.’ The gastrointestinal hormones may be important in the pathogenesis of symptoms in patients with colonic motility disorders.4-s Abnormal colonic contractility occurs following the intravenous administration of cholecystokinin and pentagastrin in patients with the irritable bowel syndrome.4 Harvey and Read reported that cholecystokinin release after a meal or after magnesium citrate administration may exacerbate symptoms of the irritable bowel syndrome.5.6 The cholinergic nerves also stimulate colonic smooth muscle.7-g Following eating, multiple neural reflexes arcs are initiated throughout the gastrointesin colonic continal tract.3.‘0 The abnormality tractility induced by a meal in patients with the irritable bowel syndrome can be inhibited by pretreatment with an anticholinergic.” Thus, the cholinergic nervous system also seems to play a role in the pathogenesis of the irritable bowel syndrome. The present studies were performed to determine the role of the cholinergic nervous system in the genesis of the gastrocolic response after a 1000-calorie meal or after the primary component of this meal, 600 calories of fat.

Methods Studies were performed on 14 normal subjects of both sexes, ages 20-35 yr. The subjects had no history of gastrointestinal disease or previous abdominal surgery, and were taking no medications. All subjects fasted at least 12 hr before each study. Informed consent was obtained from each subject. Studies were approved by the Committee on Studies Involving Human Beings at the University of Pennsylvania. All subjects underwent sigmoidoscopy without air insufflation, enemas, or cathartics. Two bipolar silver-silver

1236

SNAPE ET AL.

250

GASTROENTEROLOGY Vol. 77,No. 6

-

n=36

p(

0

20

40 Spike

Figure

1.

60

60

Potent~ols / 30 VlWOl Counting

01

100 m,n

Comparison of electronic counting of spike potentials and visual counting of spike potentials during identical 30-min test periods. Regression analysis shows a good correlation (r = 0.90) (P < 0.01). Thirty-six time periods were compared in 9 patients during fasting and after ingestion of the various meals.

chloride wire electrodes were attached to the colonic mucosa 5-25 cm from the anus under direct vision through the sigmoidoscope. The electrode used in these studies has been previously describedeg Each bipolar electrode was connected to a junction box containing a l/16 A fuse. The junction box was in turn connected to a rectilinear recorder (Beckman R411, Beckman Instruments, Inc., Fullerton, Calif.) through an AC coupler (9806A) with a time constant of 1.0 set (0.16 Hz) and a 22-Hz filter. All subjects were grounded through an EKG surface electrode attached to the right leg. Intraluminal pressure was measured at the same level as each recording electrode. Pressure was measured using polyvinyl catheters which were continuously perfused at 4 cc/min with distilled water using an infusion pump (Harvard 2215, Harvard Apparatus Company, Millis, M~ss.).~ Pressure was transmitted to transducers (Statham P321A, Statham Instruments, Oxnard, Calif.) and recorded simultaneously with the myoelectrical activity. Respirations were monitored by a pneumograph belt placed around the chest and connected to a transducer. Basal myoelectrical and intraluminal pressure recordings were begun 30 min after the removal of the sigmoidoscope. After a 30-min basal recording period, the subjects were fed a meal. Each subject ate the meal in less than 5 min. The 1000-calorie meal consisted of the following: a roast beef sandwich and a milkshake with whole milk and vanilla ice cream.’ The pure fat meal was designed to ap-

proximate the caloric content of fat in the lOOO-calorie meal. The pure fat meal consisted of 75 cc of olive oil (Progresso), providing a total of 600 calories of fat. Normal saline was added to fat meal to maintain the volume constant at 450 cc. Five milligrams of clidinium bromide or placebo was administered 90 min before eating the lOOOcalorie meal or the 600-calorie fat meal. Subjects were pretreated with the placebo or active drug in random order. Three subjects who were fed the whole meal had only one study of the drug-placebo pair. Four subjects were pretreated with 5 mg of clidinium 20 min before eating the 600-calorie meal. Myoelectrical activity and intraluminal pressure were recorded for 90 min after the consumption of each meal. Each myoelectrical recording was evaluated for slow waves and spike potentials by two investigators in a blinded manner. Slow waves appeared as regular, cyclical changes in electrical potential that were greater than 0.02 mV in amplitude. Spike potentials appeared as rapid fluctuations in electrical potential that were not associated with simultaneous movement artifact on the respiration or pressure recordings. The rapid deflections were counted only if they were greater than 0.02 mV in amplitude. Spike potentials were counted visually for each IO-min period. Only one spike potential was counted in a slow wave cycle regardless of the number of rapid deflections during that period. Therefore, no more than one spike potential could be counted in a 7.5-set period since this was the shortest slow wave cycle identified. This method of analysis gives equal weight to slow wave cycles that had one spike potential with those cycles that had multiple spike potentials. Thus, the data is an estimation of the total number of spike potentials or an expression of the number of slow wave cycles with accompanying spike potentials that are present during the recording periods. In 9 subjects, the results of visual analysis of spike activity were compared with results obtained with an automated electronic digital counter. The myoelectrical impulses were simultaneously recorded on % inch analogue magnetic tape (Scotch 3M 207, Minnesota Mining and Manufacturing Co., St. Paul, Minn.) by an 8 channel online FM recorder (Vetter Co., Rebersburg, Penn.) at a top speed of 15/16 inches per second. The tape recorder was flutter compensated to allow replay at a faster speed. Each myoelectrical recording was analyzed by replaying the magnetic tape at 15 inches per second (16 times faster than actual time) through an electronic digital counter which contained lo-pole Butterworth filters (Med Associates, East Fairfield, Vt. and Bloom Associates, Narbeth, Penn.). The filter bandwidth for analysis of the spike potentials was 80-208 Hz, which corresponds to 5-13 Hz in real time. The band-pass filters removed all slow wave activity and allowed only spike activity to pass through. The digital counter was set to record every pulse of electrical activity that had an amplitude of 0.02 mV. The response rate of the counter was 350 Hz, which was greater than the maximum frequency of our signal. Figure 1 shows the comparison between electronic and visual analysis of spike activity. Spike activity analyzed by both methods, is plotted for 30 min periods. Spike activity analyzed by both methods correlated

December 1979

A.

NEURAL CONTROL OF GASTROCOLIC

RESPONSE

1237

MYOELECTRlCdL

O.lrnV

I IOmmliq

FIESIIRATIONS

+

I MINUTE __I +

I MINUTE---l

Figure 2. Myoelectrical activity, intraluminal pressure and respirations in a normal subject during fasting (A) and 20 min after ingesting the fat meal (B). A. During fasting slow wave activity is present at a frequency of 6.2 cycles/min and there is no superimposed spike activity. There are no contractions present on the intraluminal pressure tracing. B. Following the meal, spike potentials are superimposed on the slow waves. There is a concomitant increase in colonic contractio’ns shown by changes in intraluminal pressure. closely (r = 0.90). More spike potentials were counted by the electronic method, shown by the slope 2.45. These results are expected since the electronic counter measures all spike potentials within a test period, whereas visually only one spike potential was counted per slow wave cycle.

The pattern of responses after the experiments was identical for both methods of analysis. Intraluminal pressure was evaluated by calculating a motility index (i.e., product of the mean amplitude of the pressure waves multiplied by the sum of duration of each

pressure wave) for each lo-min period.* Statistical analysis was made using the paired and unpaired Student’s t-test.

Results Figure 2 shows myoelectrical and intraluminal pressure tracings recorded during fasting (A) and after eating the 600-calorie fat meal (B). Dur-

Time (min)

Figure

3.

Colonic spike activity during fasting and after ingestion of a 1606-calorie meal. The studies were performed after pretreatment with either a placebo or clidinium bromide.

ing fasting (A) there are no spike potentials superimposed on the slow wave activity. Similarly there is no increase in distal colonic intraluminal pressure. Twenty minutes after eiting, spike activity is increased and it is superimposed upon the slow wave activity. The intraluminal Pressure tracing shows a concomitant increase coin&ding with the increase in colonic spike activity. Figure 3 shows the distal colonic spike response to a lOOO-calorie meal with and without pretreatment with the anticholinergic. Spike activity was stable during the fasting period with 2.1 f 0.1 spike potentials present in a lo-min ,period. Spike activity remained constant through&t the entire fasting period. In the normal subjects, spike activity increased rapidly following eating, reaching a peak of 10.6 + 1.9 spike potentials/lo min (P < 0.01) 30 min after the meal. Spike activity returned to fasting levels by 50 min after eating. Following pretreatment with the anticholinergic drug, there was no significant increase in rectosigmoidal spike activity after eating the lOOO-calorie meal (P > 0.05). Thus, the increase in spike activity that occurred following ingestion of a 1000-calorie meal was completely abolished by pretreatment with the anticholinergic. Figure 4 shows the colonic spike response to 600 calories of fat. The response is measured with and without treatment with clidinium bromide 90 min before the meal. In the absence of pretreatment with the anticholinergic, rectosigmoidal spike activity rapidly increased after ingestion of fat. The peak spike response occurred within the first 10 min when spike activity increased to 14.2 f 3.6 spike potentials/lo min. Spike activity remained significantly elevated for 40 min after eating (P < 0.01). The spike activity returned tn fasting levels fnr the

1238

SNAPE

ET AL.

GASTROENTEROLOGY

10

16 i

T

Time

hn)

Figure 4. Colonic spike activity during fasting and after ingestion of a MM-calorie meal of fat (olive oil). The studies were performed after pretreatment with either a placebo or clidinium bromide.

Vol. 77, No. 6

the increase in rectosigmoidal motility following the meal. The intraluminal pressure stayed at fasting levels during the entire postprandial period. After the ingestion of fat, two distinct increases in intraluminal pressure were observed that were similar to the changes observed in the spike activity. The early peak increase in the motility index occurred 20 min after eating the fat meal (1840 f 345/10 min). A second increase in intraluminal pressure occurred 80 min after eating the meal (975 + 310/10 min). The anticholinergic inhibited the early increase in distal colonic motility. The late increase remained intact following administration of clidinium 90 min before the meal (1012 f 359/10 min) and 20 min before the meal (959 + 194/10 min). Discussion

The gastrocolonic response after eating connext 30 min. A second peak of the spike activity then sists of an increase in distal colonic motor activity occurred from 70 to 90 min after the olive oil meal (P that is related to the ingestion of food. The gastroco< 0.02). Thus, the distal colonic spike response to fat ionic response may be due to (1) the stimulation of differed from the response to the 1000-calorie whole neural receptors or (2) the release of gastrointestinal meal, since there were two peaks of spike activity hormones. These studies suggest that both mechafollowing fat compared with the single peak follownisms may be operative, depending on the composiing the whole meal. tion of the meal. Following pretreatment with clidinium bromide, Only one spike potential was counted per slow the early increase in colonic spike activity after the wave cycle because of the difficulty in accurately de600-calorie olive oil meal was inhibited (P > 0.05). termining the number of individual spikes present in However, spike activity from 70 to 90 min postprandially was unaltered. Spike activity increased to a maximum of 8.4+ 1.6spike potentials/lo min, (P < 0.01) 90 min after eating the fat. Thus, administration of the anticholinergic inhibited the early re** p<.Ol ** sponse to fat ingestion, but not the late response. T Clidinium was administered to 4 subjects 20 min 2400 before the meal to attain adequate blood levels of the drug from 70 to 90 min after eating. In these pa2000 tients the basal spike response was 2.2 of: 0.3 SP/lO min. The early peak of spike activity was again in1000 Cal meal ----0 meal ond Cllduvum hibited, but the distal colonic spike response for the 1600 period 70-90 min after eating fat had increased spike activity (8.9 f 2.8 SP/lO min) (P < 0.02). Therefore 1200 the late increase in colonic spike activity occurred following the ingestion of the fat meal despite adequate serum levels of the anticholinergic. 800 Rectosigmoidal intraluminal pressure increased concomitantly with the changes in spike activity. 400 Figure 5 shows the effect of eating the 1000-calorie meal on colonic intraluminal pressure, expressed as 0’ ’ the motility index. An early increase in intraluminal 60 I30 I00 0 20 40 -20 pressure occurred after the lOOO-calorie meal. The Tmw ( minutes) peak increase in rectosigmoidal motility occurred Figure 5. Colonic intraluminal pressure expressed as a motility during the first 10 min after eating. The motility deindex during fasting and after the ingestion of a 106~ creased to fasting levels by 50 min after the meal. calorie meal. The studies were performed after prePretreatment with the anticholinergic drug inhibited treatment with either a placebo or clidinium bromide. ??

December 1979

a burst. The magnitude of the postprandial spike response is underestimated by this method of analysis because spike activity occurs more often as bursts following eating than during control periods, However, there was a high degree of correlation between the visual and the electronic methods of spike analysis. The pattern of the spike response after stimulation was identical with both methods of analysis. Thus, visual analysis appears to be a good method for estimating the spike response to various stimuli. Previous studies showed that gastrin or octapeptide of cholecystokinin are capable of stimulating colonic motility at serum concentrations that are possibly physiologic. However, following a meal the levels of serum gastrin had a minimal correlation with colonic motility from 45 to 90 min after eating.’ These studies were performed to determine a possible role of the cholinergic nervous system in the genesis of the gastrocolonic response. The increase in rectosigmoidal spike activity occurred rapidly after the lOOO-calorie meal. The rapidity of this response suggests neural mediation, since serum gastrin and cholecystokinin do not reach peak serum concentrations until 45-60 min after eating.‘,‘2.‘3 Further evidence of the importance of cholinergic nerves in the gastrocolonic response is the inhibition of the colonic response to the lOOO-calorie meal by pretreatment with the anticholinergic agent, clidinium bromide. Therefore, the cholinergic nervous system seems to be a major contributor to the gastrocolic response after the WOO-calorie meal. However, previous studies on the lower esophageal sphincter have suggested that the stimulatory effect of the gastrointestinal hormones gastrin or motilin may be mediated through receptors present on cholinergic neurons.‘4”5 Some gastrointestinal hormones are capable of releasing acetylcholine from the myenteric plexus.” Not only may a cholinergic blockade inhibit action of the hormones at the neuromuscular junction, but the release of the hormones may be affected. Cholecystokinin release is inhibited by the administration of an anticholinergic.” By contrast, postprandial serum levels of gastrin are greater following pretreatment with Therefore the anticholinergic anticholinergics.‘” agent may be inhibiting the gastrocolonic response (1) solely through the cholinergic nervous system, (2) by blocking the neural effector of hormonal mediation, or (3) by altering the rate of secretion of the gastrointestinal hormones. The gastrocolonic response after fat was different from the response after the whole meal. Following ingestion of the olive oil meal there was also a rapid increase in colonic spike activity similar to the lOOOcalorie meal. In addition, there was a late increase in distal colonic motility. The early increase in colonic

NEURAL

CONTROL

OF GASTROCOLIC

RESPONSE

1239

motility following the ingestion of olive oil was inhibited by the anticholinergic. Thus, the early response following the fatty meal appears cholinergitally mediated, similar to the response following the UKGcalorie meal. However, the late peak in the gastrocolonic response was not inhibited by the anticholinergic. The lack of inhibition could be due to either inadequate blood levels of the drug or absense of a cholinergic mediation for the late peak of activity. There are no serum assays for the drug, clidinium bromide. Therefore, evidence for the duration of action of the drug must be pharmacologic. Meal stimulated gastric acid secretion can be inhibited for 4 hr by 5 mg of clidinium bromide.‘g Thus, adequate biologic activity of the drug is probably present during this study. Furthermore, the anticholinergic was administered only 20 min before eating so that adequate serum levels of the drug would be present during the period of the late peak in colonic activity stimulated by fat ingestion. The late increase in coionic spike activity was not altered despite probably adequate anticholinergic activity. These studies suggest that the late increase in colonic motility may be secondary to direct gastrointestinal hormonal stimulation of the colon. Previous studies have shown that gastrin and cholecystokinin do stimulate colonic motility.‘~‘~2” Cholecystokinin might be responsible for the late peak in colonic motility after eating a meal composed of fat, for fat is a potent stimulus to cholecystokinin release.13 However, cholecystokinin release supposedly can be inhibited by anticholinergics.” Therefore, there is no direct evidence which hormone is involved. Possibly other candidate hormones such as substance P, or motilin are also involved. The gastrocolic response that occurs rapidly after a meal appears to be predominantly neurally mediated. However, ingestion of fat stimulates a late gastrocolonic response. The mechanism for this response is unclear References 1. Snape WJ Jr, Matarazzo SA, Cohen S: Effect of eating and gastrointestinal hormones on human colonic myoelectric and motor activity. Gastroenterology 75:373-378,1978 2. Connell AM, Logan CJH: The role of gastrin in gastroileocolic responses. Am J Dig Dis 12277~284,1967 3. Sharma KN, Nasset ES: Electrical activity in mesenteric nerves after perfusion of gut lumen. Am J Physiol 202:725730.1962 4. Snape WJ Jr. Carlson GM, Matarazzo SA, et al: Evidence that abnormal myoelectrical activity produces colonic motor dysfunction in the irritable bowel syndrome. Gastroenterology 72:383-387.1977 5. Harvey RF, Read AE: Effect of cholecystokinin on colonic

1240

6.

7.

8.

9.

10.

11.

12.

13.

SNAPE

ET AL.

motility and symptoms in patients with the irritable bowel syndrome. Lancet l:l-3,1973 Harvey RF, Read AE: Effects of oral magnesium sulphate on colonic motility in patients with the irritable bowel syndrome. Gut 14:983-987.1973 Devroede G, Lamarche J: Functional importance of extrinsic parasympathetic innervation to the distal colon and rectum in man. Gastroenterology 86:273-2891974 Bennett A, Stockley, HC: The intrinsic innervation of the human alimentary tract and its relation to function. Gut 18:443453.1975 Snape WJ Jr, Carlson GM, Cohen S: Human colonic myoelectric activity in response to prostigmine and the gastrointestinal hormones. Am J Dig Dis 22:881-887,1977 Kosterlitz HW: Intrinsic and extrinsic nervous control of motility of the stomach and the intestines. In: Handbook of Physiology, Section 8, Alimentary Canal, Vol. 4, Motility. Edited by CF Code. Washington, D.C., American Physiological Society, p 2147-2171.1968 Sullivan MA, Cohen S, Snape WJ Jr: Colonic myoelectrical activity in irritable bowel syndrome--effect of eating and anticholinergics. N Engl J Med 298:878-883,1978 Dockray GJ, Taylor IL: Heptadecopeptide gastrin: measurement in blood by specific radioimmunoassay. Gastroenterology 71:971-977,1976 Rayford PL, Fender HR, Ramus NI, et al: Release and half-life

GASTROENTEROLOGY

Vol. 77, No. 6

of CCK in man from Symposium Gastrointestinal Hormones. Edited by James C. Thompson. Austin, Texas, University of Texas Press, 301-318,1975 14 Lipshutz WH, Tuch AF, Cohen S: A comparison of the site of action of gastrin I on lower esophageal sphincter and antral circular smooth muscle. Gastroenterology 61:454-4891971 AJ, Bowes KL, Zwick R, et al: Effect of motilin on 15. Meissner the lower esophageal sphincter. Gut 17:925-932,1976 16. Vizi SE, Bertaccini G, Impicciatore M. et al: Evidence that acetylcholine released by gastrin and related polypeptides contribute to their effect on gastrointestinal motility. Gastroenterology w268-277,1973 17. Konturek SJ, Tasler J, Oblulowicz W: Effect of atropine on pancreatic responses to endogenous and exogenous cholecystokinin. Am J Dig Dis 17:911-917,1972 18. Walsh JH, Yalow RS, Benson SA: The effect of atropine on plasma gastrin response to feeding. Gastroenterology Bo:I621.1971 19. Peterson WL, Fordtran JS: Reduction of gastrin activity. In: Gastrointestinal Disease. Edited by MH Sleisenger, JS Fordtran. Philadelphia, WB Saunders Co., Chapter 52, p 891-913, 1978 20. Dinoso VP, Meshkinpour H, Lorber SH, et al: Motor responses of the sigmoid colon and rectum to exogenous cholecystokinin and secretin. Gastroenterology 65:438-444,1973