SEARCH

SEARCH BY CITATION

Keywords:

  • gastrocolonic response;
  • ileocecal junction;
  • ileocecal resection

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflict of interest
  9. References

Background  To investigate neural regulation at the ileocecal junction (ICJ) and motility changes after ileocecal resection (ICR). Previous studies showed normal basal motility at the ICJ directly by force transducers in dogs, but these observations were limited to normal contractile activity.

Methods  Continuous strain gauge recordings of stomach, terminal ileum, ileocecal sphincter (ICS), and colon were performed in dogs. The dogs were divided into four groups, namely control (CONT), extrinsic denervation at ICJ (ED), intrinsic denervation at ICJ (ID), and ICR groups. Colonic activity was recorded 2 h before a meal, in the early postprandial period (first 2 h), and in the late postprandial period (4–6 h after a meal). The meal lasted 5 min.

Key Results  Motility index was significantly increased at the ICS (= 0.0056) and proximal colon (= 0.0059) after feeding. However, such changes were not observed in the ED and ID groups. The amplitude of contractions at proximal colon in the interdigestive state was significantly decreased by ED. In the ID and ICR groups, the numbers of nonmigrating contractions were significantly decreased (< 0.05), and colonic migrating motor complex (CMMC) ratio was significantly higher than that of the CONT group (< 0.001). The dogs in these two groups had diarrhea.

Conclusions & Inferences  Gastrocolonic response at the ICJ may require both intrinsic and extrinsic innervation. When ID was performed, CMMC ratio increased. As a result, intraluminal water absorption may have decreased. ID may be one of the causes of diarrhea after ICR.


Abbreviations:
ED

extrinsic denervation

ID

intrinsic denervation

ICJ

ileocecal junction

ICS

ileocecal sphincter

GCR

gastrocolonic response

CMMCs

colonic migrating motor complexes

CNMCs

colonic non-migrating motor complexes

MI

motility index

IDS

interdigestive state (2 h before a meal)

EPS

early postprandial state (2 h after a meal)

LPS

late postprandial state (4–6 h after a meal)

EPR

early postprandial response

LPR

late postprandial response

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflict of interest
  9. References

Little is known of the functional importance and control mechanisms of the ileocecal junction (ICJ). This region is presumed to regulate the flow of chyme entering the colon and prevent coloileal reflux.1–4 Dinning et al.4 showed that cecal distension caused significant and immediate increase in ICJ tone in humans,4 a phenomenon also observable in many other species.5–8 Several investigators studied ICJ motility by manometric and electromyogram procedures.4–7,9–11 However, these procedures were indirect contractile examinations. Quigley et al.12 studied normal basal motility at the ICJ directly by force transducers in dogs, but these observations were limited to normal contractile activity.

Colonic motility is enhanced by ingestion of meals, an effect known as the gastrocolonic response (GCR).13 The GCR to meal ingestion consists of a pancolonic increase in contractions, which serve to propel the colonic contents after eating.14,15 The GCR exhibits three phases: immediate postprandial response (0–5 min after meal ingestion), early postprandial response (0–2 h after ingestion), and late postprandial response (2–8 h after ingestion).13 There is an almost immediate increase in ICJ activity by feeding in humans.4 Early GCR may occur via extrinsic nerves; however, Shibata et al.16 showed that early and late GCR is inhibited by intrinsic denervation (ID; loop intestine at the middle colon). This investigated segment in their model had no luminal contents. Diversion of chyme from the terminal ileum prevents the increased-activity period in the colon after feeding.17 Shibata et al.16 explained that their inhibition of the late GCR was caused by no luminal contents. The ICJ is a gate across the small intestine and large intestine. This region may play an important role in regulation of GCR.

Contractions of the colon are controlled by three primary mechanisms such as myogenic, chemical, and neural actions.14,15 The nerves that control colonic motor activity can be divided into CNS, peripheral autonomic nervous system (ANS), and enteric nervous system (ENS).14,15 When performing colon cancer surgery, extrinsic and intrinsic nerves must be cut. However, the effect of denervation at the proximal colon is unknown.

Little is known of the motility changes after colectomy. Ileocecal resection (ICR) is often indicated in patients with benign diseases such as Crohn’s disease and appendicitis. However, ICR may cause an unwanted effect of diarrhea.18–21

This study aimed to investigate the effects of denervation at ICJ and motility changes at the distal colon after ICR.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflict of interest
  9. References

Preparation of animals

Experiments were completed in 16 healthy conscious dogs of either sex, each weighing 8–11 kg. The procedures were approved by the Review Committee on Animal Use of Gunma University, Maebashi, Japan. Overnight-fasted dogs were anesthetized by single intravenous injection of thiopental sodium (20 mg/kg body weight, Ravonal; Tanabe Pharmaceutical, Osaka, Japan), and general anesthesia was maintained by intratracheal inhalation of halothane (Fluothane; Takeda Chemical Industries, Osaka, Japan) and oxygen. A silastic tube (Silastic 602–205; Dow Corning, Midland, MI, USA) was inserted into the superior vena cava through a branch of the right internal jugular vein (jugular tube). The abdominal cavity was opened, and eight force transducers22 were implanted on the serosal surfaces of the gastric antrum, terminal ileum (5 and 15 cm proximal to ileocecal sphincter) (I1/I2), ileocecal sphincter (ICS), and colon (C1–C4); C1 was placed 5 cm distal to the ICS, C4 5 cm proximal to the peritoneal reflection (Fig. 1), and C2 and C3 between C1 and C4 at equal distances from each other. ICS was identified by inspection and palpation. The dogs were divided into four groups, namely control (CONT), extrinsic denervation (ED), ID, and ileocecal resection (ICR). Wires from each force transducer were tunneled subcutaneously to the dorsum and connected to an eight-channel telemeter (GTS-800; Star Medical, Tokyo, Japan); gastrointestinal and colonic contractile activities were thereby continuously recorded on a computer (Adif1412.dill; Star Medical).

image

Figure 1.  Scheme of dog models and location of force transducers (C1-4, I1-2 and A) implanted in canine gastrointestinal tract and colon. Force transducers in the colon were implanted on the serosal surfaces of the gastric antrum (A), terminal ileum (5 and 15 cm proximal to ileocecal sphincter [ICS]; I1-I2), ICS, and colon (C1–C4); C1 was placed 5 cm distal to the ICS and C4 5 cm proximal to the peritoneal reflection. C2 and C3 were implanted between C1 and C4 at equal distances from each other. The double lines show resection (double dotted line) and transaction line (double solid line). The ascending colon and terminal ileum are innervated with extrinsic nerve fibers (single dotted line) along the ileocecal artery. Intrinsic nerve fibers (single solid line) in the colon contain a highly organized and complex network of neurons. The myenteric and submucosal plexuses are the two major plexuses in the colon. The myenteric plexus exists between the longitudinal and circular muscle layers, and the submucosal plexus exists in the submucosa.14,15

Download figure to PowerPoint

The first group (CONT dogs 1–4) underwent force transducer implantation alone as control (Fig. 1). In the second group, (ED dogs 5–8) extrinsic autonomic denervation of the ileocecal site (I2–C1) was performed. The ileocolonic artery (ICA) was dissected at its origin. Descending fibers from around the ICA and ICA were dissected with this procedure. In the third group (ID dogs 9–12), transection of the intestine at points between I1 and I2, C1 and C2 was performed, followed by end-to-end anastomosis (Fig. 1). In this procedure, ID of ileocecal site (I2–C1) was achieved. In the fourth group, (ICR dogs 13–16) ICR was performed. The ileum and colon were transected close to the ICJ. The last part of the ileum and cecum and the first part of the colon were removed, and the ileum was rejoined to the colon. In this procedure, resection of the ileocecal site (I2–C1) was achieved.

After the operation, the dogs were housed in individual experimental cages. The dogs in each group were fasted for 2 days after this procedure and maintained by intravenous infusion of LactecG (Otsuka Pharmaceutical, Tokyo, Japan) at a daily volume of 500 mL. Cefmetazole (1 g) was administered intravenously, once preoperatively and once on the first day after operation. The dogs were allowed to recover for ≥10 days. They were fed normal dog food (20 g kg−1; Funabashi Farm, Funabashi, Japan) once daily and given water ad libitum.

After all experiments were completed, the dogs were sacrificed by an overdose of potassium chloride. Specimens of ICS were then fixed in 10% formalin and stained with hematoxylin and eosin. Proper placement of transducers was confirmed.

Recording of contractile activity

The dogs were fasted overnight before each experiment. After the interdigestive motor complex had been recorded at the site of antrum, ≥2-h contractile activity was recorded in interdigestive state. Then, dogs were fed and contractile activity was recorded for 2-h periods (0–2 h and 4–6 h) in the early and late postprandial states. Three feeding experiments were performed in each dog.

Data analysis

Colonic contractile activity was analyzed in the interdigestive states (2 h before a meal), early postprandial states (2 h after a meal), and late postprandial states (4–6 h after a meal) at each site from ICS to C4. The recorded mechanical activities were analyzed using software for analysis of gastrointestinal motility (8STAR, Star Medical).

Colonic contractile activity consisted of a burst of contractions lasting ≥1 min. Two contractile states were separately accepted when there was an intermittent quiescent state of ≥2 min. Bursts of colonic contractions that migrated over three recording sites (C2–C4) were considered colonic migrating motor complexes (CMMCs); all other contractile activities were considered colonic non-migrating motor complexes (CNMCs). This distinction followed criteria defined by Sarna et al.23

The following parameters were measured: no. CMMCs and CNMCs; total count of CMMCs and CNMCs; mean duration of colonic contractile activity; time between the start and end points of a burst of contractions at each site; mean cycle duration of colonic contractile activity; time between the start of two consecutive contractions; mean amplitude of contraction; amplitude peak of the individual contraction; %CMMC (the ratio of CMMC contractions to total contractions on each recording day); mean motility index (MI) of colonic contractile activity; and average MI of colonic contractile activity at each site from C1 to C4. The MI was defined as the integrated area between the baseline (zero level) and the contractile wave expressed in motor units. Fecal moistness was also assessed.

Statistical analysis

The results are expressed as mean ± SE. Fisher’s protected least significant difference test was used to test the significance of differences among groups. < 0.05 was considered statistically significant.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflict of interest
  9. References

Placement of transducers

All dogs remained healthy during the experiments periods. Histologic examination of each dog at ICS was performed so as to confirm that transducers were placed correctly.

Characteristics of gastric and ileal motility

The migrating motor complexes occurred at regular intervals of 100–120 min in the stomach and migrated through the ileum in the interdigestive state. This pattern was disrupted by feeding. Gastric motility showed no significant change by denervation and resection at the ICJ.

Spontaneous contraction of proximal colon

Three types of contraction appeared at the ICS (Fig. 2A) such as irregular random phasic waves, rhythmic bursts of contractions corresponding to phase III of the migrating motor complex, and single, propagated, high-amplitude waves. These contractions were consistent with the findings of Quigley et al.12 Contractions of each type were counted over 6 h before feeding (interdigestive state). In the ID model, counts of such high-amplitude waves significantly decreased (= 0.019) (Fig. 2B).

image

Figure 2.  (A) Three types of contraction appeared at the ICS: (i) irregular random phasic waves; (ii) rhythmic bursts of contractions corresponding to phase III of the migrating motor complex; and (iii) single, propagated, high-amplitude waves. (B) Contractions of each type were compared over 6 h before feeding. In the ID model, such high-amplitude waves were significantly decreased.

Download figure to PowerPoint

Two types of contraction were identified at C1: short-duration contractions with an underlying long-duration contraction (type A); and short-duration contractions without an underlying long-duration contraction (type B) (Fig. 3A). Type B contractions resembled the wave pattern of the small intestine, but these contractions were regardless of phase III of the migrating motor complex. Type B contractions were observed only at C1 and were not seen in the ED group (Fig. 3B).

image

Figure 3.  (A) Colonic contractile activities in control group after ingestion of a meal. Two types of contraction were identified at C1: (i) short-duration contractions with underlying long-duration contraction (type A); and (ii) short-duration contractions without underlying long-duration contraction (type B). (B) Effects of extrinsic denervation (ED) and intrinsic denervation (ID). All contractions were recorded in early postprandial state. Type B contractions disappeared in ED group. Amplitude of contractions at ED segment was decreased.

Download figure to PowerPoint

Amplitude of contractions

Amplitude of contractions at C1 was higher than at other sites (C2–C4) in the control group. In ED model vs control, amplitude of contraction at C1 was significantly inhibited during all states (18.8 ± 2.9 vs 52.1 ± 5.7, 18.7 ± 2.7 vs 39.2 ± 2.0, and 16.39 ± 2.0 vs 32.3 ± 2.3, respectively; Figs 3B and 4). In addition, amplitude of contraction at ICS and C2 in the interdigestive state was significantly inhibited by ED. In ID group, amplitude was not altered.

image

Figure 4.  Mean amplitude of contractions in interdigestive and postprandial state. Mean amplitude of contraction at C1 was significantly decreased by extrinsic denervation (ED) in early (= 0.0002) and late postprandial state (< 0.0001). In interdigestive state, mean amplitude of contraction at proximal colon (ICS, C1, and C2) was significantly decreased by ED (P 0.049, <0.0001, and 0.0038, respectively). In the ED model, mean amplitude of contraction at C4 was significantly increased (= 0.0029). *< 0.05 vs control.

Download figure to PowerPoint

Early and late postprandial response

All dogs in each group ate the test meal within 5 min. The difference with the time before a meal of MI was compared (Fig. 5). In the early postprandial response, MI only at the proximal colon (ICS and C1) significantly increased in the control model (356.1 ± 71.7 vs 116.7 ± 21.5 and 545.2 ± 79.5 vs 257.4 ± 43.4, respectively). In the late postprandial response, MI significantly increased at almost every site in the control model.

image

Figure 5.  Early (EPR) and late postprandial response (LPR). Alteration of the motility index (MI) (g min) at all sites by the ingestion of meal. In the EPR, MI at only the proximal colon (ICS and C1) significantly increased in the control model. In the LPR, MI significantly increased at almost every site in the control model. In both the ED and ID models, the EPR and LPR disappeared at ICS and C1. In the ICR model, the EPR and LPR appeared at the distal residual intestine (C2–C4). Values are mean ± SE. §< 0.05 vs interdigestive state. *< 0.05 vs control. IDS: interdigestive state.

Download figure to PowerPoint

In both the ED and ID models, the early and late postprandial responses disappeared at ICS and C1 (Fig. 5). However, the early postprandial response at C1 in the ID group was not significantly different. In the ICR model, early and late postprandial responses appeared at the distal residual intestine (C2–C4).

Rate of CMMCs (%CMMC) and fecal moistness

In ID and ICR groups no. CNMC was significantly decreased (3.9 ± 0.7 and 4.0 ± 0.6 vs 7.3 ± 1.2) and %CMMC was significantly greater than the control group (74.7% ± 2.1% and 75.2% ± 1.0%vs 56.5% ± 1.3%; Table 1). Fecal moistness in these two groups was significantly greater than that of the controls (57.3% ± 3.3% and 60.5% ± 2.1%vs 48.2% ± 1.0%; Table 1). Dogs in the ID and ICR models had diarrhea. The %CMMC was significantly correlated with fecal moistness (r = 0.60, = 0.038).

Table 1.   CMMC, CNMC, and fecal moistness
 CMMC (times)CNMC (times)%CMMCFecal moistness (%)
  1. CMMC, colonic migrating motor complexes; CNMC, colonic non-migrating motor complexes. %CMMC, CMMC/CNMC + CMMC. Values are mean ± SE.

  2. *< 0.05 vs control.

CONT9.3 ± 1.97.3 ± 1.556.0 ± 1.948.2 ± 1.0
ED9.2 ± 1.77.3 ± 1.355.7 ± 2.145.8 ± 1.1
ID11.4 ± 2.33.9 ± 0.8*74.6 ± 3.5*57.3 ± 3.3*
ICR11.3 ± 2.24.0 ± 0.7*73.9 ± 1.0*60.5 ± 2.1*

Duration of contractile activity

Feeding exerted no effect on mean duration of contractile activity in the control group (Fig. 6). In ID group, mean duration of contractile activity significantly decreased at various sites. In ED group, this duration significantly decreased only at ICS in early postprandial state. On the other hand, these activities in the ICR group tended to be longer than other groups.

image

Figure 6.  Mean duration (min) of contractile activity in interdigestive state (IDS), early postprandial state (EPS), and late postprandial state (LPS). In the control group, there were no effects of meal ingestion. In the ID group, the mean duration of contractile activity significantly decreased at various sites. In the ED group, this duration significantly decreased only at the ICS in EPS. On the other hand, these activities in the ICR group tended to be longer than those of other groups. Values are mean ± SE. §< 0.05 vs IDS. *< 0.05 vs control.

Download figure to PowerPoint

Cycle duration of contractile activity

In control group, cycle duration of contractile activity in the postprandial state was shorter than in the interdigestive state (Fig. 7). In ID and ICR groups, cycle duration of contractile activity at the distal colon in the late postprandial state was significantly longer than control group (Fig. 7). Cycle duration in ED group varied.

image

Figure 7.  Mean cycle duration (min) of contractile activity in interdigestive state (IDS), early postprandial state (EPS), and late postprandial state (LPS). In the control group, the cycle duration of contractile activity in the postprandial state was shorter than that in the IDS. In the ID and ICR groups, the cycle duration of contractile activity at the distal colon in the LPS was significantly longer than that in the control. Cycle duration in the ED group varied. Values are mean ± SE. §< 0.05 vs IDS. *< 0.05 vs control.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflict of interest
  9. References

We investigated colonic motility patterns in denervation models. This is the first study to investigate neural regulation at ICJ by force transducer in denervation models. The GCR disappeared in the ID and ED models. When ID at proximal colon was performed, %CMMC and fecal moistness were increased at distal colon. In the ED model, amplitude of contractions at C1 (denervation segment) was significantly decreased in interdigestive and postprandial states.

Quigley et al.12 investigated motor activity at the normal ICS directly by force transducer. Consistent with their findings, we observed three types of contraction appearing at the ICS: irregular random phasic waves; rhythmic bursts of contractions corresponding to phase III of the migrating motor complex; and single, propagated, high-amplitude waves. Quigley et al.12 showed that high-amplitude waves appeared on average every 115.5 min in the interdigestive state. In our investigation, high-amplitude waves appeared 2.78 ± 0.76 times over 6 h before feeding, in close accordance with previous findings. High-amplitude wave count significantly decreased in the ID model. Sarna et al.24 showed that arrival of phase III activity in the terminal 25-cm length of ileum triggered a cecal giant migrating contraction. Possibly, this trigger may propagate via intrinsic nerves. Otterson et al.25 showed that intrinsic denervation at the terminal ileum caused an increase in the ileal giant migrating contraction proximal to the denervation point. However, the giant migrating contraction that propagated across the ICJ decreased. We observed a giant migrating contraction at the ICS (distal to the denervation point). Our results were similar to those of Otterson et al.25

Sarna et al.13 reported that GCR comprises three phases such as immediate (0–5 min after ingestion of a meal), early (0–2 h after ingestion), and late phase (2–8 h after ingestion). In our study, GCR both in early and late states at the proximal colon was significantly inhibited not only in ED group but also in ID group. Precise mechanisms of the immediate and early postprandial responses are not well understood, although both neural and hormonal influences have been suggested.13 Rosso et al.26 revealed the presence of sympathetic fibers distributed at the ICJ and encircling the myenteric plexus and submucosal plexus in an immunohistochemical study of the horse. In our ED models, these autonomic fibers were cut, and this extrinsic denervation caused a reduction of GCR. The extrinsic innervation may be needed for GCR. Shibata et al.16 showed that early GCR was reduced in the innervated loop (ID) at middle colon, and late GCR was significantly inhibited. In late GCR, our results and those of Shibata et al.16 were similar. However, they suggested that inhibition of late GCR was caused by diversion of chyme to the loop.17 In our models, colonic continuity was restored by end-to-end anastomosis, and intraluminal contents always presented at the denervation segment. One reason for this inhibitory effect is inflow disorder at the anastomotic region. Previous studies have shown that orocecal transit time in dogs is 2–4 h.27 In the present study, late GCR was investigated from 4 h after feeding. We consider that inflow disorder had little effect. Dickson et al.28 suggested that serotonin released from enterochromaffin cells activates the internuncial neuronin intrinsic nervous network, causing CMMC. In the ID model, intrinsic nervous network normally sensing intraluminal contents may have failed. Therefore, intrinsic innervation at ICJ may play some role in regulation of late GCR. The inhibitory effects in our ED model accord well with previous findings.16

When ID at proximal colon (ID and ICR models) was performed, no. of CNMC significantly decreased and %CMMC increased. Tanabe et al.29 showed that extrinsic autonomic denervation of the entire colon and ID at the proximal colon significantly increased %CMMC – in close agreement with our results – suggesting that intrinsic myenteric plexus may inhibit propagation of CMMC. In addition, in the groups in which %CMMC increased, fecal moistness increased. Nonpropagating contractions result in to-and-fro movements to mix, knead, and churn luminal contents.14,15 Sethi et al.30 showed that propagation of contractile states in the colon is a major factor in slow net distal propulsion of colonic contents and individual phasic contractions may mainly produce mixing and agitation of colonic contents. In our study, %CMMC was significantly correlated with fecal moistness (r = 0.60). The increasing %CMMC might have caused decreases of mixing and absorption of colonic contents. The intrinsic myenteric plexus may control migration of contraction and absorption of colonic contents. Previous studies showed that ICR caused diarrhea.18–21 This effect may be the major cause of diarrhea after ICR.

We observed that type B contractions disappeared in ED group. Ehrlein et al.31 described three types of contractions in rabbit proximal colon such as high-frequency repetitive contractions, low-frequency rises of baseline, and monophasic progressive contractions, which are comparable to our observations in dogs. Brierley et al.32 showed that neurokinin (NK)1 and NK2 receptor antagonism abolished long-duration contractions, but not short-duration contractions, suggesting that the former are mediated by tachykinins (TKs). The TKs secretion may be regulated via extrinsic nerves at the proximal colon. In the ED model, amplitude of contraction at C1 was significantly decreased. It was previously shown that mean amplitude of waves at the distal colon was significantly decreased in patients with spinal cord injury (SCI).33,34 The patients with spinal cord injury and those in our ED model suffered injury of both sympathetic and parasympathetic nerves. Therefore, sympathetic and/or parasympathetic nerve may be needed to preserve amplitude of contractions.

Mean duration of contractile activity was not altered by feeding; that in the ICR group tended to be longer than other groups. Sarna et al.35 simulated secretory diarrhea by perfusion of non-nutrient fluids into the colon, during which they noted increases of mean duration. As the ICR group had diarrhea, duration time might be protracted. However, mean duration time was not increased in ID group, suggesting that another reason might exist.

After meal ingestion, cycle duration of contractile activity in the control group was shorter at most sites. This effect disappeared in ED group at most sites, suggesting involvement of extrinsic nerves. The effect also disappeared in ID group at distal site of transection line, suggesting intrinsic nerve involvement in intestine. In ID and ICR groups, cycle duration of contractile activity at the distal colon in the late postprandial state was significantly longer than control. In the ID and ICR models, intrinsic nerves network normally sensing intraluminal contents may have failed.

Concerning blood flow, Shimizu et al.36 revealed that colonic motility showed no significant difference between the caudal mesenteric artery (CMA) transecting group and CMA preserving group in rats.36 In ED group in our study, due to dissection of ICA, blood supply to the proximal colon was less than in the other groups, but the influence of dissection on colonic motility seemed slight. Although it is possible to perform ED by a vessel-preserving method, it is not perfect for ED. No changes of colonic motor activity were observed in dogs undergoing autonomic denervation of the paraaortic nerves with vessel preservation.37 However, our model provided perfect ED.

In conclusion, GCR was inhibited by ID and ED at ICJ. Amplitude of contraction at proximal colon was inhibited by ED. Rate of CMMCs and fecal moistness was increased by ID. Although diarrhea after ICR may have various causes, we saw a close relationship between ID and diarrhea. Therefore, ID may be one of the causes of diarrhea after ICR.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflict of interest
  9. References

We gratefully acknowledge the contributions of Dr Hayato Yamauchi and Toshinaga Suto. We thank Mr. Arimitsu Bettou for making the jackets of dogs.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflict of interest
  9. References