Effect of Dietary Fiber on Intraluminal Pressure and Myoelectrical Activity of Left Colon in Monkeys

Transcription

1 GASTROENTEROLOGY 77:70-74, 1979 Effect of Dietary Fiber on Intraluminal Pressure and Myoelectrical Activity of Left Colon in Monkeys A. J. M. BRODRIBB, ROBERT E. CONDON, VERNE COWLES, and JEROME J. DeCOSSE The Surgical Service, Veterans Administrative Center, Wood, Wisconsin and the Vince Lombardi Colon Clinic, Division of Surgery, The Medical College of Wisconsin, Milwaukee, Wisconsin The effects of varying intake of dietary fiber content on left colon intraluminal pressure, electrical spike discharge activity, and contractions of circular muscle were measured in stump-tailed monkeys. Animals were preconditioned to sit in restraint chairs and then had the sigmoid colon instrumented with an extraluminal strain gage force transducer, a bipolar serosal electrode and a Thomas cannula to provide access for determining luminal pressure with a perfused, open-tip catheter. Diets containing 0, 5, 10, 15, and 20 g of dietary fiber daily were fed in varying sequences for periods of 3 wk Observations were made twice during the last week of each diet. Fecal weight increased in animals on a high fibe.r diet, but changing dietary. fiber content did not alter the duration of spike discharge activity and produced only a minimal change in the contraction frequency of colon circular muscle. Colon intraluminal pressure was increased significantly by feeding lowfiber diets, but intraluminal pressure responses did not always correlate with myoelectrical activity of the colon circular muscle. Dietary fiber intake is believed to influence colonic function, and lack of dietary fiber has been suggested as an etiologic factor in diverticular disease! Patients with symptomatic diverticular disease have a lower dietary fiber intake than age- and sexmatched controls.2 Symptoms in such patients are relieved when fiber intake is increased. 3.' The high colon intraluminal pressure found in patients with Received July 17, Accepted February 8, Address requests for reprints to: Dr. Robert E. Condon, Surgical Service (112), Veterans Administration Center, Wood, Wisconsin This study was supported in part by Veterans Administration Research Funds and by the Carl E. Mueller Fund by the American Gastroenterological Association /79/ $02.00 diverticular disease s. 6 is reduced when cereal fiber is added to the diet. 3 7 In view of these suggestive clinical findings, we studied the relationships between dietary fiber intake and left colonic function in a subhuman primate. The interrelationships of colonic smooth muscle electrical spike discharges, circular muscle contractions, and associated intraluminal pressure were measured under controlled conditions in stump-tailed monkeys fed five different levels of fiber. Methods Six stumptailed monkeys (Macaca arctoides), weighing 3-5 kg, were conditioned to sit continuously in restraint chairs. This enabled collection of fecal specimens and prevented the animals from interfering with their instrumentation. Recordings also could be made without interference with the stereotyped daily routine of the animal. Because these animals are easily conditioned, care was taken to maintain the same daily routine throughout the study. After conditioning to the laboratory was completed, a laparotomy was carried out on each monkey through a midline incision under intramuscular phencyclidine anesthesia. The silicone-coated bipolar silver electrode had a 2- mm gap and was mollnted on a 3 X 5 X 8-mm silicone rubber suturing plate. The 2.5 X 5 mm strain gage was bonded to a copper-beryllium shim and coated with silicone rubber producing a 15 X 5 X 3 p ~ force m transducf)r. 8 The electrode plate was sutured to colon circular muscle with the two electrode tips oriented transversely. The force transducer also was oriented transversely and sutured under moderate tension to adjacent teniae of the same bowel segment as the electrode, 7.5 cm above the peritoneal reflection. The sensitivity of the strain gage was determined 2 wk after implantation and before beginning any experiments. The maximal response to subcutaneous injection of 1 mg urecholine was set equal to full scale deflection on the paper chart of the strain gage signal. Thereafter, the sensitiv-

2 July 1979 FIBER EFFECT ON COLON PRESSURE AND MYOELECTRICAL ACTIVITY 71.. ELECTRODE ~ STRAIN GAGE Figure 1. A diagram of instrumentation of the experimental animals. ity factor of each gage in each animal was constant throughout these experiments. A nylon Thomas cannula was inserted into the sigmoid colon 7.5 cm proximal to the electrode and strain gage and brought through the abdominal wall. Measurements of intraluminal pressure were made using a continuously perfused open-ended polyethylene catheter (ID 1.0 mm) having a compliance of 170 nl/mm Hg placed through the Thomas cannula with the tip 2.5 em above the electrode and strain gage (Figure 1). The Thomas cannula was kept sealed with a rubber bung when pressures were not being recorded, and a similar seal was used aro"und the catheter during recording sessions. The monkeys tolerated the laparotomy well. Colon contractile function usually returned to normal within 3 days in these animals. 9 At least 2 wk elapsed before any recordings were made. The monkeys were supplied with diets combining varying proportions of no fiber pellets and a commercial bran cereal (Table 1). The specially prepared, no-fiber pellets were baked using a mixture of 40% casein, 26% sucrose, 15% reagent-quality soluble starch, 10% corn oil, and 9% minerals and vitamins recommended for primates. The bran cereal (All-Bran, Kellogg) contained approximately 27% dietary fiber, 30% starch and related carbohydrates, 14% sugar, 11% protein, and 4% fat. The cereal component was supplied in quantities of 0, 20, 40, 60, or 80 g of crude fiber which is equivalent to 0, 5, 10, 15, or 20 g of dietary fiber daily including an allowance of 10% for wastage. The remainder of the diet was made up with no-fiber diet pellets to provide a total of 140 g total diet by weight each day. The daily dietary requirements lo Table 1. Experimental Diet Composition (Intake/Day) Dietary Total Protein Carbohydrate Fat fiber (g) calories (g) (g) (g) of m o ~ kare e y~ g/kg s protein, 1.2 g/kg fat and 100 Cal/kg. The diets supplied exceeded these minimal requirements in.all animals. All animals maintained a stable body weight throughout these experiments. Each animal was fed one dry meal daily, at 12 noon, and was supplied with water ad libitum. The dietary regimens all were readily accepted by the monkeys; meals were consumed in less than 1 hr by all animals. Consumption of meals was verified visually in each instance. Each animal was given each diet for a 21-day period. One pair of monkeys started with the zero fiber diet which was increased by 5 g of fiber content every 3 wk. A second pair started with 20 g fiber intake which was reduced by 5 g every 3 wk. The third pair started with 15 g fiber, followed by 20 g, 0 g, 5 g, and 10 g. The animals were studied at least twice during the third week on each of the five diets. The recording apparatus was kept out of sight. The catheter was connected to a Statham P23Dc pressure transducer and was irrigated, using a Harvard pump, at 0.02 ml/min. Intraluminal pressure, electrical activity, and contractions of the colonic circular muscle were recorded simultaneously with a Beckman 8- channel pen recorder (Beckman Instruments, Inc., Fullerton, Calif.). A time constant of 0.3 sec was used in recording these experiments; the recorder amplifier filters frequencies outside Hz. Recording was started at 11:00 AM, 1 hr before feeding. The animals were then provided with the dry meal, and recording continued for a further 3 hr. Recordings were divided into 3D-min periods for analysis. Electrical activity was found to consist of very active periods of spike bursts alternating with quiet periods of relative electrical silence. The duration of periods of electrical activity was random and not suggestive of an interdigestive myoelectrical complex. The percentage of time during which electrical spike discharges occurred was recorded. Contractile activity was recorded as the number of contractions/3d min that produced at least a 5-mm deflection of the recording pen. While smaller deflections also may represent colonic muscular contractions, they were difficult to differentiate from respiratory and voluntary movements of the monkey. Intraluminal pressure was analyzed as an index calculated as the product of the mean wave height, the mean wave duration, and the number of waves/3d-min recording period. During the third week of feeding each diet, all feces were collected from each monkey. Three fresh specimens of feces were c o l l ewithin c ~ e d afew minutes of defecation and the water content measured by ~ e i g before h i n g and after drying to constant weight. All collected stools were dried in an oven at 100 C for a minimum of 24 hr. Mean daily stool weight was recorded, and mean daily wet weight of feces was calculated. The effects of diet and time relative to feeding were assessed by three-way analysis of variance and, when further indicated, a paired comparison t-test. Results presented in figures depict means plus or minus the standard error, where the standard error is based on all measurements obtained at the indicated point in time.

3 72 BRODRIBB ET AL. GASTROENTEROLOGY Vol. 77, No.1 LEFT COLON (ELECTRICAL) 9 FIBER/d FEEDING LEFT COLON IMOTOR) iii, i ' Figure 2. Duration of electrical spike discharge activity (mean percent time spikes present ± SE) in monkey colon before and after feeding diets of varying fiber content. Feeding occurred at time zero. The left colon exhibits spike discharge activity about 35% of the time. There is no relationship between dietary fiber consumption and mean duration of colonic spike discharge activity. Results There were no differences in responses related to the order in which the various diets were fed. Duration of spike burst activity in the left colon of the 6 monkeys over the 4 hr recording period is shown in Figure 2. There was no difference in recorded activity with the different diets. Left colon electrical activity was not stimulated by eating. Mean contractile frequency in the left colon of the six monkeys is shown in Figure 3. Muscular contractions always were associated with electrical spike discharges. There was not a marked change in contractile frequency with the diets of differing fiber intake, although the no fiber diet was associated with more activity than higher-fiber diets. Contractile frequency was not stimulated by eating. Intraluminal pressure showed a difference with the different dietary regimes (Figure 4). On a high-fiber diet, the pressure index was low, but as fiber intake decreased below 10 g/day, the index increased. There was no significant increase in pressure with eating. Contractions of circular muscle and increases in intraluminal pressure did not have any regular relationship to each other. Increases in intraluminal pressure associated with contractions of circular muscle were present during portions of recorded activity in all animals, but such association was observed during less than half of all records and was not related to feeding or dietary fiber intake (Figure 5).,0..L..-+-T""-,...-,..---,, ,-..., * DIFFERENCE FROM 20 g/d DIET P<O.02 Figure 3. Contractions of colon circular muscle (mean number of contractions equal or exceeding 5 mm deflection per 30 min ± SE) in monkeys before and after feeding diets of varying fiber content. Muscle contractions always were associated with an electrical spike discharge. Feeding occurred at time zero. Contractions of the sigmoid colon occurred about once per minute during these experiments and were not influenced by dietary fiber consumption. Mean colon contraction frequency in animals consuming no dietary fiber was significantly increased just before and after feeding but not at other times. Fecal Weight Mean daily dry fecal weight and mean percent water content of feces on the different diets are displayed in Table 2. There was a difference in dry stool weight, but no difference in percent water content with the different fiber intakes. Discussion Long-term investigation of colonic motor function using chronically implanted strain gages is :i " 200 ;!: w a: :::> 160 (J) 0:3 g: LEFT COLON (INTRALUMINAL PRESSURE) 4 0 ~ - ~ - - r - - r - - r - - r * DIFFERENCE FROM 20 g/d DIET P<O,Q2 Figure 4. Colon intraluminal pressure index (mean ± SE) before and after feeding diets of varying fiber content. Intraluminal sigmoid colon pressure is significantly greater in animals consuming low (5 g) or no fiber diets than in animals consuming 10 g or more of dietary fiber daily.

4 July 1979 FIBER EFFECT ON COLON PRESSURE AND MYOELECTRICAL ACTIVITY 73 Intraluminal Pressure Electrical Activity Intraluminal Pressure I n t r ~ ~ ~ C : ::..:.. ~.,... f- '1.A " 1 ~ ' ~ ~.... " ". e :., ;,,,,,, - ~,,,,,,,,,,,,,,,,,,,,,,, -,,, ~ _..., Electrical "', ~ ~, I, A e t j v : i ~ i, ~ mmhg Il0MM Il0MV Figure 5. Simultaneous recordings of sigmoid colon intraluminal pressure (mmhg), circular muscle contractions (strain gage) (mm per. deflection) and spike discharges (electrical activity) (mv). In A, spike discharges and muscle contractions are associated with simultaneous increases in intraluminal pressure. In B, irregular spike discharges are associated with muscle contractions of varying force; changes in recorded intraluminal pressure are also seen but bear no regular relationship to the myoelectrical events. In C, regularly recurring spike discharges and circular muscle contractions are not associated with alteration of intraluminal pressure. Table 2. Effect of Fiber Intake on Stool Weight and Water Content presently not feasible in humans. primates are an appropriate animal model for such studies, because their large bowel is comparable anatomically to that of humans and because they can consume diets which are not dissimilar to those of humans in terms of nutrient composition. The main variation in the several diets studied in these experiments was dietary fiber content. The other components of the diet, such as sugar and fat, were kept as constant as practicable. It was not possible to dictate exactly how much fiber was eaten by each monkey, but the bran cereal was eaten with enthusiasm, and wastage was small, so the total diet- Signif- Signif- Mean dry icance of Water icance of Dietary fecal weight differ- content differfiber (g/day ± ence (P- (percent ence (P- (g/day) SE) value)o ±SE) value)o ± ± ± ± ± ± 0.75 < ± ± ± ± 0.83 a Analysis of variance; indicates not signijicant. ary fiber content consumed closely approximated the intended quantity. Because of the varying sequence in which diets of differing fiber content were offered, minor alterations in technique or animal environment that might have influenced colonic responses during the course of these experiments should not appreciably bias the data reported. It was not possible to quantify tension in colonic circular muscle by means of the strain gage. Even though strain gage response previously had been calibrated in vitro, it cannot be assumed that the gage in vivo necessarily continues to quantitatively reflect force of contraction. The strain gage signal in vivo also may be variably influenced by the tension under which the gage has been sutured in place and the later effects of connective tissue encapsulation of the gage and of adhesions. Although not quantifiable in absolute terms, changes in strain gage resistance do directly reflect muscle contraction. Recordings taken during the hour before the monkeys ate do not represent fasting basal activity. Since they were fed only a single meal each day, the monkeys were conditioned to expect feeding and anticipated the presentation of food for at least 2 hr before eating. The lack of change in left colonic myoelectrical activity associated with feeding noted in these experiments may only reflect high anticipatory preprandial activity. The intrinsic myoelectrical activity of the instrumented segment of left colon was unaffected by a wide range of changes in dietary fiber content. This finding is quite different from our observations of the right colon which shows increased contractile and electrical activity in response to a meal" and changes in contractile activity with differing fiber intake (unpublished observations). We were not able to derive any evidence from our experiments to support the hypothesis that a high-fiber diet acts as an inhibitory influence on circular muscle activity or that a low-fiber diet has a "spasmogenic" stimulant effect on the bowel. 12 1J The fiber content of the diet did have a marked effect on in.traluminal pressure during both the preprandial and postprandial periods. Low-fiber diets were associated with higher pressure in the colon lumen. The increase in intraluminal pressure, however, was not accompanied by significant increases in electrical spike discharge activity or in the frequency of circular muscle contractions. The observation that intraluminal pressure changes did not correlate with discrete changes of intrinsic electrical activity or muscle contraction demonstrates that the pressure recorded by a perfused intraluminal catheter in the colon depends on a number of factors, of which contraction of the colon smooth muscle immediately adjacent to the cath-

5 74 BRODRIBB ET AL. GASTROENTEROLOGY Vol. 77, No.1 eter tip is only one" component. '4 Interpretation of our findings is limited by the fact that the perfused catheter tip and the force transducer were slightly displaced from each other. We believe that intraluminal catheter pressure recordings cannot be assumed always to give an accurate picture of smooth muscle activity in a discrete colon segment. An increase in fiber intake increased the daily volume of feces produced by our animals. Multiple comparison testing showed t h the a ~ 0, 5, and 10 g fiber diets and the 15 and 20 g fiber diets constitute two groups within which no significant differences were observed but between which any pairs of means were significantly different. The percent water content of the feces was not changed by fiber intake, a finding consistent with previous human studies using cereal fiber in which water content showed little change. 15 The fact that the proportion of stool water content was not influertced by the fiber content of the diet does not support the hypothesis '6 that alterations in stool bulk and transit time by high fiber diets are secondary to the capacity of fiber or its breakdown products to hold extra water in a gellike state. Stool consistency was noticeably altered by increasing fiber intake; feces became increasingly soft and plastic as the fiber content increased.' This change in the nature of the stool may affect not only the diameter of the bowel lumen but also may provide an alternative explanation for the reduction in luminal pressure with the high-fiber diet. It is possible that a high-fiber intak-e results in increased filling of the colon with soft stool which, in turn, prevents "closed chambers" from forming in the descending colon.'7 As soon as one area of the bowel starts to contract, the soft fecal material would be displaced into the next segment and pressure would be reduced. Our observation in monkeys that intraluminal pressure is influenced by changes in dietary fiber intake is consistent with the findings of Hodgson,'8 who reported that a low-fiber diet given to rabbits for 4 mo increased intraluminal pressure, particularly after administration of neostigmine. It is also consistent with the clinical observation that increasing fiber intake by patients with diverticular disease results in a reduction of intra sigmoid pressure,3.7 and supports the hypothesis that a low-fiber diet is an etiologic factor in diverticular disease of the colon and a high-fiber diet may prevent its development." References 1. Painter, Burkitt DP: Diverticular disease of the colon: a deficiency disease of Western civilization. Br Med J 2: , Brodribb AJM, Humphreys DM: Diverticular disease: three studies. Part I-Relation to other disorders and fibre intake. Br Med J 1: , Brodribb AJM, Humphreys DM: Diverticular disease: three studies. Part II-Treatment with bran. Bt Med J 1: , Brodribb AJM: Treatment of symptomatic diverticular disease with a high fibre diet. Lancet 1: , Arfwidsson S: Pathogenesis of multiple diverticula of the sigmoid colon in diverticular disease. Acta Chir Scand Suppl 342:3-68, Painter, Truelove SC: The intraluminal pressure patterns in diverticulosis of the colon. 1. Resting patterns of pressure. II. The effect of morphine. Gut 5: , Findlay JM, Smith AN, Mitchell WD, et al: Effects of unprocessed bran on colon function in normal subjects and in diverticular disease. Lancet 1: , Cowles VE, Condon RE, Schulte WJ, Woods JH, Sillin LF: A quarter wheatstone bridge strain gage force transducer for recording gut motility. Am J Dig Dis (accepted for publication) 9. Woods JH, Erickson LW, Condon RE, et al: Postoperative Ileus: a colonic problem? Surgery 84: , Whitney RA Jr., Johnson DJ, Cole WC: Laboratory Primate Handbook. New York, Academic Press, Inc., 1973, p Sillin LF, Schulte WJ, Condon RE, Woods JH, Bass P: Gastrocolic response: is it G.LP.? Proceedings of the Sixth International Symposium on Gastrointestional Motility, Edinburgh, Scotland, Spring Plumley P: Fecal fibre fortunes. Br Med J 3:305, Srivastava GS, Smith AN, Painter : Sterculia bulkforming agent with smooth muscle relaxant versus bran in diverticular disease. Br Med J 1: , Misiewicz JJ: Colonic motility. Gut 16: , Cowgill GR. Anderson WE: Laxative effects of wheat bran and "washed bran" in healthy men: comparative study. JAMA 98: , Cummings JH: Dietary fibre and colonic function. J R Soc Med (Lond) 71:81-83, Painter : Diverticular Disease of the Colon, a Deficiency Disease of Western Civilization. London, England, W. Heineman Medical Books Ltd., Hodgson WJB: An interim report on the production of colonic diverticula in the rabbit. Gut 13: , 1972