Right and left colonic transit after eating assessed by a dual isotopic technique in healthy humans

GASTROENTEROLOGY 1992:103:80-85

Right and Left Colonic Transit After Eating Assessed by a Dual Isotopic Technique in Healthy Humans LAURENCE PICON, MARC JEAN-CLAUDE RAMBAUD,

LaMANN, BERNARD JEAN-DIDIER RAIN,

FLOUR@

and RAYMOND JIAN

Service de Gastroent&ologie et Unite de Recherche sur les Fonctions Intestinales et la Nutrition, Hapital Saint-Lazare; and Service de Gastroent&ologie et Service de Medecine NuclBaire, HBpital Saint-Louis, Paris, France

Propulsion of colonic contents after eating in the whole colon was studied in 15 volunteers by scintigraphy with injection of ‘*‘In-diethylenetriaminepentaacetic acid and gamTc-sulfur colloid into the colon through a nasogastric tube. The radionuclide was injected into the cecoascending colon (n = 7), the hepatic flexure (n = 6), the splenic flexure (n = 9),and the descending colon (n = 4). Changes of activity in the regions distal from and proximal to the injection points were determined before and after a 1606-kcal meal. Isotopic movements were also analyzed when a simultaneous injection of the two markers in the right and left parts of the colon was achieved (n = 11). During fasting, no significant change of activity was seen. After eating, radioactivity injected into the cecoascending and the hepatic flexure was transferred distally (P < 0.01and P = 0.07); radioactivity injected into the splenic flexure was transferred both distally (P = 0.07) and proximally (P< 0.02); and no significant change of activity was seen proximally from or distally to the descending colon. Both antegrade and retrograde isotopic movements increased after eating (P < O.Ol), but the number of antegrade movements was significantly greater (P< 0.05). This study confirms the colonic propulsive effect of eating and shows that this response is different in the right and left parts of the colon.

It

is well known that colonic motor activity increases after eating in humans’-4; however, the consequences on movements of colonic contents in the whole colon are still incompletely documented. Colonic scintigraphy is a physiological and quantitative method for evaluating the transfer of colonic contents in humans.5 Studies using colonic instillation of isotopic materials into the cecum by a nasogastric tube5 or using capsules with colonic disruption6 have provided information about the transit of colonic contents in the right part of the colon; how-

ever, the effects of a meal were not accurately described. A recent study using instillation of a radiolabeled marker at the splenic flexure showed antegrade and retrograde movements occurring after the mea1,7 but this study was limited to the left part of the colon and was performed after colonic preparation for retrograde insertion of the probe by colonoscopy * The aim of the current study was to evaluate the effects of a meal on the propulsion of colonic contents in the whole and unprepared colon in healthy humans and to show regional differences of intraluminal transit. For this purpose we used a simultaneous injection of two radiolabeled markers into the right and left parts of the colon. Materials

and Methods

Subjects Fifteen healthy volunteers [9 men, 6 women; mean age, 25.9 years (range, 23-34)] without history of gastrointestinal symptoms or gastrointestinal surgery, who had received no medication during the last 2 months, were studied. They all had a normal whole transit time of radio-opaque markers (~67 hours’). All subjects ingested regularly a normal western diet containing lo-20 g of dietary fiber/day (average amount, measured for 1 week). Written informed consent was obtained from each subject before participating in the study, which was approved by the local ethics committee.

Colonic Intubation A tube assembly, which consisted of a 4-m long eight-lumen polyvinyl system with a balloon tip, was introduced by the nose. The tube assembly had an outer diameter of 6 mm, and each lumen had an inner diameter of 0.5 mm. The tip consisted of a mercury-weighted latex balloon that could be filled with or emptied of mercury and/ or air. The six proximal lumens of the tube were intended 0 1992 by the American Gastroenterological 0016-5065/92/.$3.00

Association

RIGHT AND LEFT COLONIC TRANSIT

July 1992

for radionuclide instillation. Their openings were 5,10,15, 20,40,and 50 cm from the tip of the tube. During the 36 hours required for tube progression (range, 18-40 hours), to standardize intestinal contents, the subjects were maintained on a controlled diet containing 13-15 g of dietary fiber/day. When the tube was fluoroscopically confirmed to be in the colon, two lumen openings were chosen to inject a bolus of 150 FCi of “‘Indiethylenetriaminepentaacetic acid (“‘In-DTPA) mixed with .Z mL of saline into the right part of the colon (cecoascending colon or hepatic flexure) and a bolus of 800 pCi of ggmTc-sulfur colloid (ggmTc-SC) mixed with 2 mL of saline into the left part (splenic flexure or descending colon). A flush of 5 mL of saline was used to ensure that no radionuclide remained in the tube. All radionuclide injections were performed between 10 and 11 AM. After injection of the isotopic markers, the tube was cut under the nose and then, after completion of the scintigraphic study, expelled through the anus.

Site of Injection Our objective was to obtain a simultaneous injection into the right (cecoascending or hepatic flexure) and left (splenic flexure or descending colon) parts of the colon. The injection region of interest (ROI) was defined on the first image taken immediately after injection of the two markers, as the ROI containing >50% of the total colonic activity. Table 1 gives the indium and technetium injection ROIs for each subject. Injection ROI was the cecoastending colon in 7 subjects, the hepatic flexure in 6, the transverse colon in 2, the splenic flexure in 9, the descending colon in 4, and the sigmoid colon in 1. The position of the probe allowed a simultaneous injection of the two radionuclides into the right and left parts of the colon in 11 of the 15 subjects. Because of the small number of experiment, data concerning the transverse and the rectosigmoid colon were not analyzed. Technetium injection was not performed in one subject for technical reasons.

Scintigraphic

Subjects Subject la 2 3a 4a 5” 6 7O 8 go 1oa 11 12” 13” 14” 15”

and Technetium

Injection

ROIs in the 15

Studied Indium

Technetium

Hepatic flexure Hepatic flexure Hepatic flexure Hepatic flexure Cecoascending Splenic flexure Hepatic flexure Cecoascending Hepatic flexure Cecoascending Transverse Cecoascending Cecoascending Cecoascending Cecoascending

Splenic flexure Transverse Descending Descending Splenic flexure Rectosigmoid Splenic flexure Not done Splenic flexure Descending Splenic flexure Descending Splenic flexure Splenic flexure Splenic flexure

“Subjects in whom a simultaneous injection of the two radionuelides in the right and left parts of the colon was obtained.

Imaging

Immediately after their injection, both markers were detected simultaneously using specific windows of a gamma camera. The two marker counts were recorded by taking l-minute anterior and posterior abdominal images every 15 minutes for 6 hours. An external technetium Tc 99m point source was taped on the lower right thorax to allow accurate repositioning of the subject. After a 3-hour fasting period, a 1000-kcal meal (20% protein, 50% carbohydrate, and 30% lipid) used in our laboratory to stimulate human colonic motility9 was given, and data were collected during 3 hours. All images were taken with the subjects in the erect position, and they were sitting between two scans. Quantification

ofActivity

All images were visualized to generate seven ROIs according to the following anatomic subdivision: cecoastending, hepatic flexure, transverse, splenic flexure, descending, rectosigmoid, and the whole colon. After drawing the outline of the whole colonic activity, we used the following criteria to define the six other ROIs: (a) the separation between the cecoascending colon and the hepatic flexure was made by drawing an horizontal line originating from the internal angle of the hepatic flexure; (b) the separation between the transverse colon and the splenic flexure was made by drawing a vertical line originating from the internal angle of the splenic flexure; and (c) except for the rectosigmoid ROI, we drew the other separations between ROIs to obtain areas of similar size. Anterior and posterior geometrical means were calculated to take into account heterogeneity of tissue attenuation. All counts were corrected for background activity, scatter of indium In 111 Compton activity into the technetium Tc 99m window, and physical decay of isotopes. The activity in each ROI was expressed as a percentage of total colonic activity on each image. Scintigraphic

Table 1. Indium

81

Data Analysis

Data were analyzed in two ways: radionuclide transfers according to the injection ROI and isotopic movements in the whole colon. For each injection ROI, two time-activity curves were created, which represented the changes of activity in all the ROIs proximal to and distal from the injection ROI, respectively. Isotopic movements in the whole colon were also analyzed when a simultaneous injection of the two radiolabeled markers in the right and left parts of the colon was obtained. Isotopic movements were defined as an increase or decrease of >lO% of the ROI activity, on at least two consecutive images. Direction (antegrade or retrograde), time of onset (fasting or postprandial), and number of ROIs covered were studied for each movement. Dosimetry Radiation

doses for each scintigraphic

study were

0.037 and 0.966 rad for total body and lower large intestine

(critical organ), respectively.

82 PICON ET AL.

GASTROENTEROLOGYVol.103,No.1

Statistical Analysis

Discussion

Analysis of variance (ANOVA) was used to compare changes of activity on the time-activity curves. Paired t test were used to compare isotopic movements in the whole colon. Data are given as mean ? SEM. The level of significance accepted was P < 0.~15.

Results Radionuclide

Transfer From the Injection ROI

Figure 1 shows time-activity curves related to the four injection ROIs before and after the meal. During fasting, no significant radioactivity change was seen proximally to or distally from the four injection ROIs. After eating, radioactivity significantly increased distally from the cecoascending colon (P < 0.01) and increased without reaching a significant level distally from the hepatic and splenic flexures (P = 0.07). In contrast, an increase of radioactivity proximally to the injection ROI was observed only at the splenic flexure (P < 0.02). In the descending colon, no significant radioactivity change was seen proximally or distally after eating. No reflux into the ileum was observed, and no subject had stools during the study. Isotopic Movements

in the Whole Colon

Figure 2 shows examples of isotopic movements occurring before (Figure 2B) and after the meal (Figure 2A and B). Table 2 indicates the number of isotopic movements according to their time of onset, direction, and number of ROIs covered. During the 3 hour-fasting period, the total number of isotopic movements per subject was 1.54 f 0.31. The number of antegrade and retrograde isotopic movements was similar, and most of these isotopic movements covered less than three ROIs. After eating, both antegrade and retrograde isotopic movements increased significantly (P < 0.01 and P < 0.05, respectively), but the number of antegrade isotopic movements was significantly greater than the number of retrograde isotopic movements (P < 0.05). The number of isotopic movements covering either less than three ROIs or three ROIs or more were both significantly greater after eating than during fasting (P < 0.05 and P < 0.01, respectively). Two subjects had a rapid antegrade isotopic movement that resulted in most of the activity being transferred from the cecoascending to the rectosigmoid colon (mass movement), occurring 1 and 2 hours after the meal, respectively.

In this study, we used a scintigraphic technique with simultaneous intracolonic injection of two isotopic markers into the right (cecoascending colon or hepatic flexure) or the left (splenic flexure or descending colon) parts of the colon to evaluate the transfers of radionuclides 3 hours before and 3 hours after a test meal. In the right part of the colon, there was no significant transfer of radionuclides during fasting, whereas eating led to an antegrade transfer. In contrast, we did not observe any significant postprandial retrograde transfer either from the hepatic flexure to the more proximal part of the colon or from the cecoascending to the ileum. This antegrade propulsion of the right colonic contents after eating has been previously noted in humans5v6; however, in these studies it was difficult to identify the relationship between ingestion of the meals and cecoascending colonic emptying. When the isotopic marker was injected at the splenic flexure, no transfer occurred during fasting, whereas eating led to both antegrade and retrograde transfers. This result is in agreement with a previous study7 in which radionuclides were injected at the splenic flexure by means of a probe introduced by colonoscopy after colonic preparation. Retrograde transfers have also been described in studies using instillation of radionuclides in the right part of the human colon5; however, under these conditions, the amount of radionuelides reaching the splenic flexure is probably not sufficient to detect accurately all the transfers of activity from this area. When the radionuclides were injected in the descending colon, no significant transfer to the splenic flexure or to the rectosigmoid was seen either before or after the meal. Individual analysis of the time-activity curves concerning this ROI did not show any retrograde transfer from the descending colon to the splenic flexure. In contrast, antegrade transfers of radionuclides to the rectosigmoid were often noted but were negated by subsequent opposite transfers, so that the mean time-activity curve was not significantly affected. Time-activity curves represent a summation of different and sometimes opposite movements. In an attempt to describe colonic propulsion more precisely, we evaluated the individual movements of radionuclides in the whole colon. A similar approach had been previously reported by Ritchie et aLlOusing barium sulfate and iterative X-ray examinations, but this method was not quantitative and led to radiation hazards. Movements in the whole colon could be accurately characterized when enough radionuclide material was present in all parts of the colon. Our technique with simultaneous injection into the right

RIGHT AND LEFT COLONIC

July 1992

100 d) !! 80 t ; 60

‘CI

TRANSIT

83

C

-A

Meal

T-r

T

4 5 6 Time after injection (hours)

5 6 2 3 4 1 Time after injection (hours) 100 80

80

Meal

-40

-40

lL t.l~l.l.ltl’li

0

1

2

3

4

5

6

Time after injection (hours)

0

5 6 1 2 3 4 Time after injection (hours)

Figure 1. Radionuclide transfers according to the four injection ROIs. Solidlines represent the changes of colonic activity in the ROIs distal from the injection ROIs and dotted lines the changes of colonic radioactivity in the ROIs proximal to the injection ROIs. (A) Injection into the cecoascending colon (n = 7). Solid line shows an antegrade transfer of activity toward the more distal ROIs after the meal (P < 0.01). There is no curve representing change of activity in the proximal ROIs, because no reflux into the ileum was observed. (B) Injection into the hepatic flexure (n = 6). After the meal, solid line shows an antegrade transfer of activity toward more distal ROIs (P = 6.07), whereas dotted line shows an antegrade transfer of activity from the cecoascending colon. (C) Injection into the splenic flexure (n = 9). After the meal, solid line shows an antegrade transfer of activity toward more distal ROIs (P = 0.07) and dotted line a retrograde transfer of activity toward more proximal ROIs (P < 0.02). (D) Injection into the descending colon (n = 4). No significant change of activity was seen.

and left parts of the colon enabled us to reach this methodological requisite. Movements were defined as a change of at least 10% of a ROI activity on more than two consecutive images.” It may be argued that the 15-minute interval between images may be inadequate to detect all the individual motor events and that continuous detection would be the sole method

permitting evaluation of all colonic movements. However, it would be technically difficult and somewhat unphysiological for a 6-hour study in humans. Although no significant change on time-activity curves was noticed during fasting, we found antegrade and retrograde movements, usually limited to short distances. These findings are in agreement

84

PICON ET AL.

GASTROENTEROLOGY

Vol. 103, No. 1

Figure 2. (A) Colonic scintigraphy in a subject after “‘In-DTPA injection into the hepatic flexure. No movement of the tracer was seen during the first 3 hours corresponding to fasting (top lefi, image taken z hours after injection). The meal was given 3 hours after the injection, and a transfer of activity from the hepatic flexure to the distal ROIs was observed 30 minutes later (top right), reaching the sigmoid on the last images (bottom). (B) Colonic scintigraphy in another subject after g9mTc-SCinjection into the splenic flexure. During fasting, a fraction of activity moved toward the descending colon (top Jeff, image taken 2 hours after injection). The meal was given 3 hours after the injection, and a simultaneous transfer of activity was observed 30 minutes later from the splenic flexure to the sigmoid and the transverse (fop right and bottom).

with manometric and electromyographic studies showing fasting motor activity.‘,” The significance of such fasting movements without net propulsive effects remains uncertain. They may be responsible for the mixing of colonic contents. After eating, we found an intensification of both retrograde and antegrade movements associated with changes on timeactivity curves. In particular, antegrade movements covering more than two ROIs obviously increased, and in two subjects, a mass movement was observed. As suggested by Narducci et al.,’ these antegrade movements could be an equivalent of the high-amplitude propagated contractions recorded in manometric. In contrast, motor events responsible for the retrograde movements are not yet established. In electromyographic studies, retropropagated long spike bursts have been reported,‘*” but the manometric equivalent has not yet been described in normal subjects. Moreno-Osset et ale7 studied correlation between colonic pressure change and propulsion of radionuclides injected at the splenic flexure. They suggested that antegrade and retrograde movements starting at the splenic flexure were caused by a positive pressure gradient between this high-pressure area and other parts of the colon. In our opinion, individual movement approach will be necessary in fur-

ther studies to establish more precisely correlations between colonic motor events and propulsion. The results of the current study suggest that the cecoascending colon and hepatic flexure act as a storage area during fastingI and that eating induces Table 2.

Number of Isotopic Movements per 3 Hours According to Their Time of Onset, Direction, and Number of ROIs Covered Onset of movement Fasting

Direction of movements Total Antegrade Retrograde No. of movements covering <3 ROIs Antegrade Retrograde No. of movements covering r3 ROIs Antegrade Retrograde

Postprandial

0.73 f 0.24

3.82 f 0.35” 2.27 f 0.27’ 1.54 + 0.21b,C

0.54 f 0.19 0.64 f 0.23

1.27 k 0.32b 1.27 + 0.23b

0.27 + 0.13 0.09 f 0.09

1.00 f 0.26’ 0.27 + 0.13”.d

1.54 f 0.31

0.62 f 0.23

NOTE. Results are expressed as mean + SEM. “P < 0.01, bP < 0.05 fasting vs. postprandial. “P < 0.01, dP i 0.05 antegrade vs. retrograde.

RIGHT AND LEFT COLONIC TRANSIT

July 1992

their emptying. In contrast, in the transverse colon, postprandial antegrade and retrograde movements coming from the right colon and the splenic flexure may be responsible for the mixing of intraluminal contents. Movements in the descending and rectosigmoid colon are more limited, and no obvious propulsive response to the meal is seen in this area. This concept of colonic functional division is in agreement with several observations: [a) embryologic origins and blood supply of the right and left parts of the colon are different14; (b) the right colon absorbs more water and electrolytes than the left15s16;and (c) electromyographic studies provided evidence for differences between the sigmoid area and the other parts of the colon.” These regional differences in colonic transit have to be taken into account for studying patients with colonic motor abnormalities. In chronic idiopathic constipation, two different pathological conditions have been reported3,‘7,‘8; in some patients, a decrease of antegrade movements coming from the cecoastending colon may be implicated, whereas in others, an intensification of retrograde movements coming from the left colon (splenic flexure, descending or sigmoid colon) may be present. In functional diarrhea, an exacerbation of the colonic response to eating could be involved,13 possibly associated with a failure of a left colonic “brake.” Likewise, the effects of drugs on colonic propulsion deserve studying in different parts of the colon. For example, a drug that simultaneously increases both antegrade and retrograde movements in the proximal and distal part of the human colon may in fact have no propulsive effect on the colon. References 1.

2.

3.

4.

5.

Frexinos J, Bueno L, Fioramonti J. Diurnal changes inmyoelectric spiking activity of the human colon. Gastroenterology 1985;88:1104-1110. Narducci F, Bassoti G, Gaburri M, Morelli A. Twenty four hour manometric recording of colonic motor activity in healthy man. Gut 1987;28:17-25. Bazzocchi G, Ellis J, Villanueva-Meyer J, Jing J, Reddy SN, Mena I. Snape WJ. Postprandial colonic transit and motor activity in chronic constipation. Gastroenterology 1990;98:686693. Snape WJ, Matarazzo SA, Cohen S. Effect of eating and gastrointestinal hormones on human colonic myoelectrical and motor activity. Gastroenterology 1978;75:373-378. Krevsky B. Malmud L, D’Ercole F, Maurer A, Fisher R. CoIonic transit scintigraphy. A physiologic approach to the

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

85

quantitative measurement of colonic transit in humans. Gastroenterology 1986;91:1102-1112. Proano M. Camilleri M, Phillips SF, Brown ML, Thomforde GM. Transit of solids through the human colon: regional quantification in the unprepared bowel. Am J Physiol 1990:258:G856-G862. Moreno-Osset E. Bazzocchi G, Lo S. Trombley B, Ristow E, Reddy SN, Villanueva-Meyer J, Fain JW, Jing J, Mena I. Snape WJ. Association between postprandial changes in colonic intraluminal pressure and transit. Gastroenterology 1989;96: 1265-1273. Metcalf AM, Phillips SF, Zinsmeister AR, McCarty RL, Beart RW. Wolff BG. Simplified assessment of segmental colonic transit. Gastroenterology 1987;92:40-47. Flourie B, Lemann M, Nicolov S, Jian R, Franchisseur C, Rambaud JC. Colonic motility: a new technique of intraluminal pressure recording in the whole and unprepared human colon (abstr). J Gastrointest Motil 1990:1:60. Ritchie JA, Truelove SC, Ardran GM, Tuckey MS. Propulsion and retropulsion of normal colonic contents. Dig Dis 1971:16:697-704. Camilleri M. Colemont LJ, Phillips SF, Brown ML, Thomford GM. Chapman N, Zinsmeister AR. Human gastric emptying and colonic filling of solids characterized by a new method. Am J Physiol 1989;257:G284-G290, Dapoigny M, Trolese JF, Bommelaer G, Tournut R. Myoelectric spiking activity of right colon, left colon and rectosigmoid of healthy humans. Dig Dis Sci 1988:33:1007-1012. Jian R. Najean Y, Bernier JJ. Measurement of intestinal progression of a meal and its residues in normal subjects and patients with functional diarrhoea by a dual isotope technique. Gut 1984;25:728-731. Youmans WB. Innervation of the gastrointestinal tract. In: Code CF. ed. Handbook of physiology. Section 6. Alimentary Canal. Washington, DC: American Physiological Society, 1968:1655-1663. Phillips SF, Devroede G. Functions of the large intestine. In: Crane RK. ed. International review ofphysiology. Gastrointestinal physiology III. Baltimore: tlniversity Park, 1979:263290. Levitan R. Fordtran JS, Burrows BA, Ingelfinger FJ. Water and salt absorption in the human colon. J Clin Invest 1962;41:1754-1759. Schang JC, Devroede G, Duguay C, Hemond M. Hebert M. Constipation par inertie colique et obstruction distale: etude electromyographique. Gastroenterol Clin Biol 1985;9:480485. Bassotti G, Gaburri M, Imbimbo BP, Rossi L, Farroni F, Pelli MA. Morelli A. Colonic mass movements in idiopathic chronic constipation. Gut 1988:29:1173-1179.

Received February 25, 1991. Accepted December 20,1991. Address requests for reprints to: Marc Lemann, M.D., Service de Gastroenterologie, Hopital Saint-Louis, 1 Avenue Claude Vellefaux, 75475 Paris, France. The authors thank E. Cheval for technical assistance. This work was presented in part at the annual meeting of the American Gastroenterological Association, San Antonio, Texas, May 1990, and has appeared in abstract form (Gastroenterology 1990:98:A380).