Electrophysiological study of the rectosigmoid canal: evidence of a rectosigmoid sphincter

Abstract

Previous studies strongly suggest the presence of a sphincter at the rectosigmoid junction,an area with a mean length of 2.8 cm in adults, called the rectosigmoid canal (RSC). To find supporting evidence of a sphincteric function for the RSC, two recording electrodes were applied to each of the sigmoid colon (SC), RSC and rectum (R) in 11 subjects during operative repair of huge incisional hernias. The RSC, SC and R were individually stimulated by a further electrode and their pressures monitored by a three-channel microtip catheter. The variables of the slow waves or pacesetter potentials, recorded at rest from the RSC and R, were significantly higher than those of the SC. While the frequency and conduction velocity of pacesetter potentials of the RSC and R were similar, the potential pacesetter amplitude of the RSC was significantly higher. The increase of the electrical activity and pressure upon electrostimulation was significantly greater in the RSC than that of the SC or R. Electrostimulation led to an increase in pressure of all three areas, the RSC increase being significantly the greatest. The greater increase of the electrical activity and pressure of the rectosigmoid canal upon electrostimulation, compared to that of the SC or R, strongly supports the presence of a rectosigmoid sphincter.

Introduction

The concept of a rectosigmoid sphincter was initially proposed by O’Beirne and later supported by Mayo (Goligher & Duthie 1984), Ballantyne (1986) and Longo et al. (1989). However, other investigators have denied the existence of such a sphincter (Stoss, 1990). Stoss (1990) reported that the rectosigmoid junction is not a true sphincter though it may be regarded as a functional one. Ballantyne (1986) and Wadhwa et al. (1996) stated that approximately 40–50% of normal men have a high pressure zone at the rectosigmoid junction.

We have recently demonstrated that the rectosigmoid junction is not just a junction but an area with a mean length of 2.8 cm in the adult and 0.7 cm in the neonate (Shafik et al. 1999), and we designated it the rectosigmoid canal (RSC). The RSC pressure was higher than that in the sigmoid colon (SC) or rectum (R)(Shafik, 1996, 1999; Wadhwa et al. 1996). Furthermore, its response by relaxation or contraction upon SC or rectal contraction, respectively, suggested the presence of a physiological sphincter (Shafik, 1996, 1999).

The stools in the SC, on reaching a certain volume, stimulate the mechanoreceptors which evoke the sigmoidorectal junction inhibitory reflex (Shafik, 1996), with a resulting delivery of the SC contents to the R. During passage of the fecal matter to the R, RSC distension initiates the rectosigmoid–rectal reflex, which effects rectal contraction (Shafik, 1999), which in turn evokes the rectoanal inhibitory reflex (Denny-Brown & Robertson, 1935; Corman, 1998).

Recent studies strongly suggest the presence of an anatomical sphincter at the RSC (Shafik et al. 1999). The RSC musculosa was greatly thickened compared to that of the SC or R. The mucosa was thrown into multiple folds forming a mucosal rosette.

The aforementioned data support the existence of a sphincter at the RSC which supposedly regulates the passage of stools from the SC to the R.

Moreover, electrophysiologic studies (Shafik et al. 2001) have demonstrated that, although the frequency and conduction velocity of the electric waves of the RSC and R were similar, the wave amplitude of the RSC was greater than that of both the R and the SC due probably to the thicker musculosa of the RSC (Shafik et al. 1999). The greater RSC wave amplitude was suggested to be electrophysiological evidence of the sphincteric function of the RSC.

In view of the above-mentioned data, which strongly suggest the presence of a sphincter at the RSC, we hypothesized that RSC electrostimulation would produce an increase of the RSC electric activity and pressure greater than that of the SC and R and this could be further evidence of the sphincteric function of the RSC. The current communication investigated this hypothesis.

Materials and methods

Subjects

The study comprised 14 subjects: 11 women and three men with a mean (± SD) age of 43.6 ± 12.3 years (range 27–56). They gave their informed consent before participating in the study. The 11 women were due to be operated upon for huge abdominal incisional hernias after Caesarean section, while the three men were due to undergo laparotomy for complicated acute appendicitis. None of the subjects had any gastrointestinal complaint. Laboratory investigations and barium enema studies were normal. The study was approved by our Faculty Review Board and Ethics Committee.

Methods

The tests were performed during repair of the incisional hernias. The type of anaesthesia administered was the same in all the patients, i.e. 5% halothane/95% oxygen. We did not give intra-operative fluids during the study period. Electrical activity of the SC, RSC and R was recorded by applying at each site two monopolar silver-silver chloride electrodes (Smith Kline Beckman, Los Angeles, CA, USA) fixed to the serosa by electrode gel and separated by a distance of approximately 2 cm (Fig. 1). Each electrode had a diameter of 0.8 mm and was covered by an insulating vinyl sheath sparing its tip. The electrodes were serially fixed to the lower part of the SC, RSC and upper part of the R. They were attached to a metal cannula containing a six-pin socket. The insulated wire leads were fixed to the sockets in the cannula and connected to a Brush Mark 200 rectilinear pen recorder. The electric activity including the frequency, amplitude and velocity of conduction of the waves was recorded from the six electrodes at rest and during electrostimulation.

Fig. 1.

Sites of application of the electrodes.

The above electrodes were recording ones. The SC, RSC and R were individually stimulated by another silver-silver chloride electrode. The stimulation parameters were set at a frequency of 20 Hz and a pulse width of 0.2 ms.

Manometric measurements

The pressure of the SC, RSC and R was measured at rest and during electrostimulation using a three-channel microtip catheter (Wiest Urocompact, Irvine, CA, USA). The latter was introduced per anum so that the distal transducer lay in the SC, the second one in the RSC and the third one in the R. The RSC was identified by the high-pressure zone situated between the SC and R; one transducer was placed at this zone, the other one proximal to it in the SC and the third distally in the R. The proper positioning of the transducers was ascertained at operation by palpation of the transducer at the desired location.

To ensure reproducibility of the results, the recordings were repeated two to three times in the individual subject with a lapse of 1 min between each recording, and the mean value was calculated. The third recording was performed when the second one was poorly registered. The mean duration of all the measurements was 25.3 ± 2.6 min (range 22–37). Measuring time in the first group of patients was longer but diminished gradually in the subsequent patients due to growing experience. The results were analysed statistically using Student's t-test and values were given as the mean ± standard deviation (SD). Differences assumed significance at P > 0.05.

Results

No adverse side-effects were encountered during or after performing the tests and all studies were completed.

Electrical waves were recorded from the SC, RSC and R. The slow waves or pacesetter potentials (PPs) registered from the SC were monophasic with a large negative deflection (Fig. 2), while the PPs of the RSC and R were triphasic with a small positive, large negative and another small positive deflection (Fig. 3). The PP variables of the RSC and R were significantly higher than those of the SC (Figs 2 and 3). While the frequency and conduction velocity of the PPs recorded from the R and RSC were similar, the PP amplitude of the RSC was significantly higher (P < 0.05, Fig. 3).

Fig. 2.

Electrical activity recorded from the sigmoid colon showing monophasic pacesetter potentials followed randomly by action potentials.

Fig. 3.

Electrical activity of the same subject as in Fig. 2. (a) Rectosigmoid canal showing triphasic pacesetter potentials followed randomly by action potentials, and (b) rectum showing triphasic pacesetter potentials followed randomly by action potentials of the same frequency and conduction velocity but of a lower amplitude than that of the rectosigmoid canal. The electrical activity parameters of the RSC and R were higher than those recorded from the SC.

The PPs were followed or superimposed by fast activity spikes or action potentials (APs), which were recorded from the six electrodes as multiple negative deflections (Figs 2 and 3). They occurred randomly and did not follow each PP. They had, however, similar frequencies when recorded from the two electrodes of each of the SC, RSC and R (Figs 2 and 3). The AP frequency from the RSC was the same as that recorded from the R (Figs 2 and 3).

Effects of electrostimulation on the EMG activity of the sigmoid colon, rectosigmoid canal and rectum

The PP frequency, amplitude and conduction velocity of SC, RSC and R using the same stimulation parameters showed significant increase during electrostimulation (Fig. 4). However, the RSC parameters were significantly higher than those of the SC or R (Fig. 4). Furthermore, while the PP frequency and conduction velocity of the RSC and R were similar before electrostimulation, and only the amplitude was significantly higher in the RSC (Fig. 3), during electrostimulation all the RSC parameters were significantly higher than those of the R (Fig. 4).

Fig. 4.

Electrical activity during electrostimulation of the sigmoid colon (a), rectosigmoid canal (b) and rectum (c). An arrow marks the point of stimulation.

Effects of electrostimulation on the pressure of the sigmoid colon, rectosigmoid canal and rectum

The basal pressure of the RSC was significantly higher than that of the SC or R (Table 1). Electrostimulation increased the pressure of the SC, RSC and R, with the RSC pressure being higher than that of the SC or R (Fig. 5; Table 1). Meanwhile, the SC pressure elevation was similar to that of the R with no significant difference (Fig. 5).

Table 1.

The pressure of the sigmoid colon (SC), rectosigmoid canal (RSC) and rectum (R) before and during electrostimulation^†

	Pressure (cmH2O) Before electrostimulation		During electrostimulation
	mean	range	mean	range
SC	7.4 ± 0.9	6–8	24.2 ± 1.6*	22–28
RSC	26.8 ± 6.7	22–32	86.4 ± 7.3**	72–92
R	7.2 ± 1.4	5–9	22.8 ± 2.1*	18–26

^†Values were given as the mean ± standard deviation (SD).

^*P < 0.05;

^**P < 0.01. P-values before stimulation were compared with those during stimulation.

Fig. 5.

Pressure tracing during electrostimulation of the sigmoid colon (a), rectosigmoid canal (b) and rectum (c). An arrow marks the point of stimulation.

We found no significant differences in the pressure measurements or electrical activity recordings between the men and women. The aforementioned results were reproducible with no significant difference when the tests were repeated in the individual subject.

Discussion

The current electrophysiological study may support the evidence of a sphincteric function of the RSC. At rest, RSC electrical activity exhibited an amplitude higher than that of the SC or R. Meanwhile, the electrical wave pattern of the SC differed from that of the RSC and R, suggesting that the waves are evoked from two different pacemakers.

During electrostimulation, using the same stimulation parameters, the wave variables of all of the SC, R and RSC increased significantly, with the ones of the RSC being significantly higher than those of the SC and R. These findings presumably denote an increased motor activity of the RSC on electrostimulation compared with that of the SC or R. The increased RSC motor activity is evidenced not only by the increased electrical activity but also by the elevated RSC pressure, which recorded a significantly higher value than that of the SC or R.

The RSC pressure elevation on electrostimulation was so high that it could effect RSC closure. This finding is suggested to explain the reactive processes of the RSC and R during their filling. Thus the presence of an RSC sphincter supposedly clarifies why the colonic contents, on left colonic mass contraction, stop short of the RSC and do not proceed to the R. We demonstrated in a previous study (Shafik, 2000) that left colonic mass contraction evokes the colo-rectosigmoid reflex with a resulting RSC contraction that prevents passage of the stools from the colon directly to the rectum. Furthermore, RSC pressure reduction on SC distension, allowing passage of stools to the rectum, and RSC pressure elevation on rectal distension, thus preventing reflux of the rectal contents to the SC (Shafik, 1996), are evidence of RSC sphincteric function. It appears that stimulation of the mechanoreceptors in the SC or in the R reflexly modulates the contractile activity of the RSC through the sigmoido-rectal reflexes (Shafik, 1996, 1999).

The SC ending with the RS sphincter is suggested to resemble the rectum with the distal anal sphincters. The distal sphincter in both of the SC and the R responds reflexly to distension by relaxation. With this view in mind, we experimented with the possibility of using the SC with the distal RS sphincter to replace the excised rectum with its distal internal sphincter (Shafik & El-Sibai, 1999).

In conclusion, the greater increase of the RSC electrical activity and pressure compared to that of SC or R upon electrostimulation using the same parameters strongly supports the presence of an RS sphincter. Further studies are needed to determine whether the RS sphincter could be involved in pathological lesions that may affect the defaecation process and to investigate the implications of the current findings for pharmacological approaches to the management of diarrhoea and constipation.