[ Next ] [ Prev ] [ Abs ] [ Chi ] [ TOC ] [ Home ]
Chin Med J (Taipei) 1997;60:252-8.
Division of Colorectal Surgery, Department of Surgery, Veterans General Hospital-Taipei; and National Yang-Ming University School of Medicine, Taipei, Taiwan, R.O.C.
Background. Anorectal dysfunction may occur following sphincter-saving resection for rectal carcinoma. The dysfunction may present clinically with increased stool frequency and varying degrees of fecal incontinence. It is postulated that these presentations come about from reduced neorectal capacity, as well as internal anal sphincter injury, during transanal instru-mentation or through damage to the nerve supply in the course of rectal dissection. The purpose of this study was to assess the functional results of low anterior resection (LAR), and the relative importance of each mechanism.
Methods. Thirty-one patients were included in this study, eighteen of whom had standard LAR for rectal carcinoma. Bowel continuity was reestablished by handsewn (HS) 2-layer suture in 10 patients and stapled EEA (U.S. Surgical Corporation) anastomosis in the other 8 patients. Thirteen patients who had received abdominal surgery other than LAR were the control group. Anorectal manometry was performed preoperatively and one week, then six months post-operatively. Clinical assessment was done pre-operatively and six months post-operatively.
Results. Resting anal pressure was significantly reduced in both HS and EEA groups post-operatively (HS: 69.4 +/- 14.8 mmHg, median: 71.5 mmHg vs. 43.7+/- 16.2 mmHg, median: 48.5 mmHg, 95% confidence interval of mean difference: 6.4-24.6 mmHg; EEA: 51.3 +/- 14.6 mmHg, median: 48.0 mmHg vs. 38.8 +/- 16.6 mmHg, median: 41.5 mmHg, 95% confidence interval of mean difference: 5.6-49.8 mmHg), partial recovery was noted in the HS group six months later. The squeeze pressure and functional length of the anal canal showed no difference pre- and post-operatively. Rectoanal inhibitory reflex was present in 90% of the patients preoperatively, but in only 70% of the HS, and 38% of the EEA group, six months post-operatively. Clinically, increased bowel frequency and varying degrees of incontinence were experienced postoperatively. There was no difference in bowel frequency between the two groups, but worse continence grade was seen in the EEA group.
Conclusions. LAR for rectal carcinoma resulted in impaired anorectal function. Handsewn anastomosis seemed to have a better functional outcome than EEA stapled anastomosis.
[Chin Med J (Taipei) 1997;60:252-8.]
Keywords: low anterior resection, resting anal pressure, rectoanal inhibitory reflex
Received: October 2, 1996.
Accepted: September 2, 1997.
Address reprint requests to: Jen-Kou Lin, Division of Colorectal Surgery, Department of Surgery, Veterans General Hospital-Taipei, No. 201, Sec. 2, Shih-Pai Road, Taipei, Taiwan, R.O.C.
It is widely accepted that carcinoma of the rectum can be treated by low anterior resection (LAR). The perioperative morbidity is low and the oncologic safety in terms of local recurrence rates and tumor-free survival are equal to that of abdominoperineal combined resection [1-3]. With the introduction of circular stapling devices, even lower anastomosis could be performed; hence now a vast majority of rectal cancers are treated with low anterior resection. However, unsatisfactory anorectal function does occur following LAR, and may present with increased stool frequency or varying degrees of incontinence [4,5]. The pathophysiology of these functional disturbances is still controversial. Reduced neorectal capacity as well as pelvic nerve injury during dissection of the rectum, and direct sphincter damage during transanal introduction of stapling devices have been implicated [6,7].
The purpose of this study was to assess the functional results of this operation, and the relative contribution of each mechanism to anal sphincter impairment by comparing handsewn anastomosis with stapling EEA anastomosis.
Eighteen patients who underwent low anterior resection for rectal carcinoma were studied and compared with 13 controls who received abdominal colectomy other than LAR (Table 1). The LAR group was divided into two subgroups, according to the method of anastomosis (handsewn in 10 patients or stapled EEA anastomosis in 8 patients). The mean age of each group did not differ. The distance of anastomosis from anal verge, as measured by proctoscopy, was 5.5 +/- 0.5 cm in the handsewn (HS) group (range:5-7 cm, median:6.0 cm), and 6.0 +/- 0.0 cm in the EEA group (range:5-8 cm, median:6.5 cm), and was of no significant difference. The types of surgery in the control group were right hemicolectomy (RH) in two patients, left hemicolectomy (LH) in four and anterior resection (AR) in seven. Low anterior resection was a standard procedure which should include mobilization of the rectum to the pelvic floor, high ligation of the inferior mesenteric vessels, division of lateral ligaments, total excision of the mesorectum with identification and preservation of pelvic nerves in all cases. Anastomosis was then done by handsewn method, using two-layer sutures, or by stapled anastomosis by EEA (31 mm). No temporary defunctioning colostomy was performed in any case.
Manometry was performed the day before surgery, one week and six months after surgery. The manometric instruments consisted of a catheter, infusion pump, transducers and polygraph recorder. The catheter was a low compliance 5-lumen catheter assembly consisting of a large central catheter (4.5 mm in external diameter with lumen 2.4 mm in diameter) with four pressure-recording catheters on four sides 90 degree apart. A central distending balloon was attached to the distal end of the central catheter. The catheter was connected to a hydraulic infusion system with continuous infusion of normal saline with perfusion rate 0.5 ml/min at 15 psi (Medical Specialties Inc.); the perfusion system then was attached to a Gould/Brush transducer preamplifier (Model 13-4615-50) and recorders (Gould Inc., Instrument System Division, Cleveland, OH). The patient was placed in the left lateral position with hip and knees flexed; no bowel preparation was given. The anal pressure was recorded by l-cm station pull-through technique from 6 cm to 0 cm. The average value of the highest pressure in four quadrants when the external sphincter muscle was at rest was considered as maximal resting pressure (MRP). Then the catheter was repositioned so that the sideholes stay at the level of maximal resting pressure. Patients were requested to exert three consecutive maximal squeeze efforts, and the average of those pressures was considered as maximal contraction pressure (MCP). Squeeze pressure (SP), which represented contraction of the external anal sphincter, was the difference between MCP and MRP. The average length of the high pressure zone was considered as the functional length (FL). The rectoanal inhibitory reflex (RAIR) was assessed by inflation of the distending balloon with 40 ml of air, while recording anal resting pressure, a reproducible drop of 20% or more in resting pressure, was defined as a positive reflex. The rectal capacity was measured by inflation of the balloon with air, the volume of perception of threshold for first sensation (V1st), urgency (Vurgency) and maximal tolerable volume (MTV) were recorded.
Clinical evaluation was performed preoperatively and six months postoperatively, and assessed by two parameters: bowel frequency in 24 hours, quality of fecal continence [Grade 1-5; Grade 1: perfect continence; Grade 2: incontinence of flatus; Grade 3: occasional minor soiling; Grade 4: frequent major soiling; Grade 5: incontinence (require colostomy)] [6]. Since frequent bowel movements might be experienced in low rectal carcinoma from tenesmus sensation, tumor bleeding and partial obstruction by the tumor, the number of bowel movement prior to the symptoms was recorded as pre-operative stool frequency.
Results were expressed as mean +/- standard deviation. Statistical differences between pre-operative and postoperative means, including MRP, MCP, SP, FL, rectal capacity (V1st, Vurgency, MTV) and clinical data (stool frequency and continence grading) were determined by Mann-Whitney U test. The changes in RAIR were compared by chi-square test. A p-value less than 0.05 was considered statistically significant.
MRP was significantly decreased in both HS and EEA group at one week postoperatively (69.4 +/- 14.8 mmHg, median: 71.5 mmHg vs. 43.7 +/- 16.2 mmHg, median: 48.5 mmHg in HS group, p < 0.05; 51.3 +/- 14.6 mmHg, median: 48.0 mmHg vs. 38.8 +/- 16.6 mmHg, median: 41.5 mmHg in EEA group, p < 0.05). The 95% confidence interval (C.l.) of mean difference were 5.6-45.8 mmHg for HS group, and 6.4-24.6 mmHg for EEA group. No significant change was noted in the control group (Table 2). A tendency of recovery of the MRP was observed in the HS group six months postoperatively. This was not seen in the EEA group (Figure 1). In three groups, the squeeze pressure and functional length showed no significant change pre- and postoperatively.
Rectoanal inhibitory reflex (RAIR) was present in about 90% of the patients preoperatively. Recovery of the reflex at six months was seen in 70% of HS and 38% of the EEA group, a marked difference though not significant statistically (p = 0.4) (Table 3).
There was no difference in the control and HS group pre- and post-operatively, yet significant decrease of volume of first sensation, urgency and maximal tolerable volume was observed in the EEA group (Vlst: 105 +/- 14.6 ml, median: 105 ml vs. 66.0 +/- 20.7 ml, median: 70 ml, 95% C.I. of mean difference: 9.2-68.8 ml. Vurgency: 158.3 +/- 27.9 ml, median: 150 ml vs. 110.0 +/- 36.1 ml, median: 100, 95% C.I. of mean difference: 6.1-69.9 ml. MTV: 201.7 +/- 34.3 ml, median: 185 ml vs. 162.0 +/- 38.9 ml, median: 150 ml, 95% C.I. Of mean difference: 5.4-54.1 ml) (Table 4).
Bowel movements per day significantly increased after LAR (HS:1.3 +/- 0.5/day, median:l.0/day vs. 3.0 +/- 1.2/day, median:3.0/day, 95% C.I. of mean difference:-3.9~-0.1; EEA: 1.6 +/- 1.5/day, median: 1.0/day vs. 3.3 +/- 2.5/day, median:2.0, 95% C.I. of mean difference:-5.8~-0.1), but no significant difference was seen between the HS and EEA group (Table 3).
Worse continence grading post-operatively was observed in both groups, but was more evident in the EEA group (Table 5).
Tonic contraction of the internal anal sphincter comprises more than 80% of the anal resting pressure which is significantly reduced following low anterior resection both in HS and EEA groups [6,8]. The mechanism of this decrease is discussed; both direct or indirect injury to the innervation of internal anal sphincter have been implicated [6,9]. Injury to the extrinsic innervation during rectal mobilization, division of the inferior mesenteric artery, and lateral ligaments, or damage to the intrinsic innervation during rectal transection may interrupt the autonomic innervation of the sphincter. Since the sympathetic element of the autonomic nerve system is stimulatory to the sphincter, its injury is the most likely explanation for sphincter dysfunction following handsewn anastomosis where no perianal manipulation was performed [6]. In EEA anastomosis, a stretching effect of the stapling devices leads to additional injury of the internal anal sphincter [9]. In this study, MRP dropped significantly after LAR with tendency to recover in HS group six months later. This recovery was thought to result from regeneration of the nerve damaged during surgery, whereas in the EEA group, no recovery of the MRP was seen. This finding might indicate that the direct stretching injury of the internal sphincter during transanal introduction of stapling devices is not reversible, at least six months later.
Reports on squeeze pressure after LAR vary. Some authors demonstrated fall of SP following LAR [4,9], but others stated that LAR had no effect on SP [10]. Squeeze pressure is produced by voluntary contraction of the external anal sphincter, innervated by the pudendal nerve which is unlikely to be hurt during surgical dissection [7,11]. In this study, SP showed no significant change in either the control or LAR group. This finding suggests that transanal introduction of the EEA stapler left no damage to the external anal sphincter and, as mentioned above, no injury to the pudendal nerve during the dissection. Hence no change of SP could be found.
Rectoanal inhibitory reflex allows a sample of rectal contents into the upper anal canal where contact with the sensitive transitional mucosa enabling differentiation between flatus and feces, and allows effective discrimination, so that the subject can pass flatus safely or close the anus voluntarily in the face of stool. Loss of this reflex may contribute to imperfection of continence. This reflex is mediated primarily through an intramural neuronal plexus and is usually present in normal persons. Some of the patients did not have RAIR before operation, possibly the influence of the tumor in the rectum, as also reported by some authors [5,9]. The reflex remains intact following full rectal mobilization but is abolished by transection of the rectum, as occurs during LAR [12]. The reflex may recover after regeneration of the nerve fibers across the anastomosis as shown by Suzuki et al [13] and Williams et al. [14] in handsewn anastomosis, and also by O'Riodain et al. [l2] in stapled anastomosis. RAIR was present in 70% of the HS group and only 38% of the EEA group six months postoperatively. The return of the reflex, as stated above, was initiated by regeneration of the myenteric plexus across the anastomosis, very difficult in the inverted EEA anastomosis. Hence more time might be needed to regain the reflex. O'Riordain has stated that one to two years are needed for the reflex to recover [l5].
Rectal capacity examination showed no significant difference in pre- and postoperative perception of thresholds for first sensation, urgency and maximal tolerable volume in the control and HS group, whereas in the EEA group, the Vlst, Vurgency, and MTV decreased significantly six months postoperatively.
Impaired anorectal function after LAR may presented clinically by increased stool frequency and varying degrees of incontinence. It is believed that frequent bowel movements are the result of reduced neorectal capacity, and that imperfect continence is related to decreased anal resting pressure and absence of RAIR [9]. In this study, increased bowel frequency was noted in both groups post-operatively, a finding not compatible with the manometric studies, since no significant change of rectal capacity was seen in the HS group. It is thought that not only rectal capacity, but also other factors such as colonic transit, consistence of the bowel content as stated by Vassilakis [9], may influence bowel frequency. Worsened continence grading was noted in both groups, yet was more evident in the EEA group, which was compatible with the manometric findings (drop of MRP with no recovery in EEA group; delay return of RAIR in EEA group). These findings seemed to suggest that handsewn anastomosis may carry a better functional outcome than the EEA anastomosis.
In conclusion, anorectal dysfunction is inevitable after LAR, and may present with varying degrees of fecal incontinence and increased bowel frequency. The former is the result of impaired internal anal sphincter activity, possibly because of damage to the innervation and of direct stretching. The latter is thought to be the result of reduced neorectal capacity and increased colonic transit. This study shows that there is some influence on anal continence according to the method of anastomosis, better functional outcome was anticipated with handsewn anastomosis, because of earlier recovery of resting pressure and RAIR. Although significant decrease of rectal capacity was noted following EEA anastomosis, bowel movement frequency was not affected by the way anastomosis was performed.
Wide difference in the functional outcome seemed to lie with the surgeon rather than with the operation [14]. Although handsewn anastomosis seemed to have better functional outcome, stapled anastomosis should not be negated in difficult anastomosis. With meticulous dissection and skillful anastomosis defecatory disorders can be minimized.
Copyright: 1997, Chinese Medical Association (Taipei)