Recovery of anal sphincter function following transabdominal repair of rectal prolapse: Cause of improved continence?

Twenty-eight patients with complete rectal prolapse underwent anorectal manometry before and 6 months and 1–2 years after abdominal rectopexy and sigmoid resection in a study of the mechanisms responsible for postoperatively improved anal continence. Preoperatively, 22 patients reported defective anal control. Seven patients (all with minor incontinence) regained normal control and eight other patients achieved improved continence after surgery. Anal resting, squeeze, and voluntary contraction pressures were significantly lower for defective than for normal control, with a significant rise in these pressures at 6 months after the operation, except for those incontinent patients in whom continence was not improved. No further pressure rise was seen later. Improvement of continence was not accompanied by changes in rectal sensation or reflexive functions of the internal anal sphincter. These results suggest that recovery of the resting and voluntary contraction functions of the sphincter muscles was the cause of continence improvement observed after surgery. Anal manometry was unable to predict outcome of function. Therefore, supplementary procedures for restoration of continence are not advisable, although patients with only minor incontinence are likely to regain full continence after rectopexy alone.     

Ripstein procedure is an effective treatment for rectal prolapse without constipation

The operation of choice for complete rectal prolapse is controversial. We reviewed 169 patients undergoing 185 surgical procedures for rectal prolapse over a 27-year period. The most common surgical procedure employed was the Ripstein procedure (n=142) and is the focus of this report. Other surgical procedures used included resection rectopexy (n=18), anterior resection (n=7), Altemeier's (n=9), Delorme's (n=2), and anal encirclement (n=7). The median age was 59 years (range, 12–94 years), and the female-to-male ratio was 51. The incidence of fecal incontinence, solitary rectal ulcer syndrome, and prior surgery elsewhere for rectal prolapse was 40 percent, 12 percent, and 19 percent, respectively. Operative mortality was 0.6 percent; morbidity was 16 percent. Median follow-up was 4.2 years (range, 1–15 years). Complete recurrence of prolapse after the Ripstein procedure was 8 percent; one-third of these patients recurred 3 to 14 years after surgery. Fecal incontinence improved after the Ripstein procedure or resection rectopexy in about half the patients. Persistence of prior constipation was more common after the Ripstein procedure than after resection rectopexy (57 percent vs. 17 percent;P=0.03, chi-squared). Fifteen patients developed constipation for the first time after the Ripstein procedure. About one in three patients, irrespective of surgical procedures, remained dissatisfied with the final outcome despite anatomic correction of the prolapse. The Ripstein procedure has proven to be a safe procedure with good anatomic repair of the prolapse and may improve continence. In the presence of constipation, procedures other than the Ripstein procedure may be preferable.Read in part at the meeting of The American Society of Colon and Rectal Surgeons, San Francisco, California, June 7 to 12, 1992.     

The effect of posterior rectopexy on fecal continence

Twenty-three patients with rectal prolapse or intussusception were studied to specifically focus on the effect of posterior rectopexy on fecal continence, anal pressure, and rectal capacity. Before operation, five patients were fully continent (A), 10 were continent for solid stools (B) and eight patients were fully incontinent (C). Group A remained fully continent; continence was regained nine times in group B and in group C, three patients regained full continence, two became continent for solid stools, three patients remained incontinent. Other symptoms such as constipation, false urgency, and a feeling of incomplete evacuation were not beneficially influenced by rectopexy. The patients' continence status was correlated to anorectal manometry and rectal capacity measurement. In group B, incremental pressure (P=squeeze — basal P) increased significantly (P<0.02) as well as incremental volume (V=maximum tolerated volume-volume of first sensation) (P<0.05). We conclude that, by an increase of incremental anal pressure and incremental rectal volume, posterior rectopexy offers an 83 percent chance of regaining full continence, or a major improvement, and a 17 percent chance of stabilization of fecal incontinence.Read at the spring meeting of the British Society of Gastroenterology, Bradford, April 1989.     

Pudendal nerve terminal motor latency influences surgical outcome in treatment of rectal prolapse

Purpose: This study was undertaken to document the effect of pudendal nerve function on anal incontinence after repair of rectal prolapse. METHODS: Patients with full rectal prolapse (n=24) were prospectively evaluated by anal manometry and pudendal nerve terminal motor latency (PNTML) before and after surgical correction of rectal prolapse (low anterior resection (LAR; n=13) and retrorectal sacral fixation (RSF; n=11)). RESULTS: Prolapse was corrected in all patients; there were no recurrences during a mean 25-month follow-up. Postoperative PNTML was prolonged bilaterally (>2.2 ms) in six patients (3 LAR; 3 RSF); five patients were incontinent (83 percent). PNTML was prolonged unilaterally in eight patients (4 LAR; 4 RSF); three patients were incontinent (38 percent). PNTML was normal in five patients (3 LAR; 2 RSF); one was incontinent (20 percent). Postoperative squeeze pressures were significantly higher for patients with normal PNTML than for those with bilateral abnormal PNTML (145 vs.66.5 mmHg; P =0.0151). Patients with unilateral abnormal PNTML had higher postoperative squeeze pressures than those with bilateral abnormal PNTML, but the difference was not significant (94.8 vs.66.5 mmHg; P=0.3182). The surgical procedure did not affect postoperative sphincter function or PNTML. CONCLUSION: Injury to the pudendal nerve contributes to postoperative incontinence after repair of rectal prolapse. Status of anal continence after surgical correction of rectal prolapse can be predicted by postoperative measurement of PNTML.Read at the meeting of The American Society of Colon and Rectal Surgeons, Seattle, Washington, June 9 to 14, 1996.     

Long-term results of suture rectopexy in patients with fecal incontinence associated with incomplete rectal prolapse

Suture rectopexy is the recommended therapy for complete rectal prolapse that is associated with fecal incontinence. It has been suggested that correction of an incomplete rectal prolapse is also worthwhile for patients with fecal incontinence. PURPOSE: Aims of this study were 1) to evaluate the clinical outcome of suture rectopexy in a consecutive series of patients with incomplete rectal prolapse associated with fecal incontinence, and 2) to compare these results with those obtained from patients with complete rectal prolapse. METHODS: Between 1979 and 1994, suture rectopexy was performed in 13 incontinent patients (3 males; median age, 65 (range, 45–77) years) with incomplete rectal prolapse (Group I) and in 24 incontinent patients (21 females; median age, 71 (range, 24–86) years) with complete rectal prolapse (Group II). RESULTS: After a median follow-up of 67 months, continence was restored in 5 of 13 (38 percent) patients with incomplete rectal prolapse and in 16 of 24 (67 percent) patients with complete rectal prolapse. In both groups, all male patients became continent. CONCLUSIONS: For the majority of incontinent patients with incomplete rectal prolapse, a suture rectopexy is not beneficial. The clinical outcome of this procedure is only good in incontinent patients with complete rectal prolapse. Based on these data, it is questionable whether incomplete rectal prolapse plays a causative role in fecal incontinence.Read at the meeting of the International Society of University Colon and Rectal Surgeons, Lisboa, Portugal, April 14 to 18, 1996.No reprints are available.     

Perineal excision of rectal procidentia in elderly high-risk patients

PURPOSE: This report presents a ten-year experience with perineal excision and posterior levator ani repair in elderly, high-risk patients with complete rectal procidentia. METHODS: Seventy-two patients with rectal prolapse were treated with perineal excision. Nine presented with acute incarcerated rectal prolapse. Mortality, morbidity, recurrence rates, and improvement of anal continence were assessed. RESULTS: Recurrence rate was 5.5 percent. Improvement in anal continence was seen in 66.7 percent of patients. Morbidity and mortality was low. CONCLUSIONS: Perineal excision of rectal prolapse is safe and has a low recurrence rate. Posterior levator ani repair seems to improve anal continence.Poster presentation at the meeting of The American Society of Colon and Rectal Surgeons, Chicago, Illinois, May 2 to 7, 1993.     

Treatment of rectal prolapse in the elderly by perineal rectosigmoidectomy

The results and complications of perineal rectosigmoidectomy for complete rectal prolapse in 114 patients have been reviewed. Most patients were elderly and high risk by virtue of other concurrent medical conditions. Fourteen patients (12 percent) developed significant postoperative complications. Hospital stay was short (median, four days). Ten patients were lost to follow-up. The remaining 104 patients were followed for 3 to 90 months. Eleven patients (10 percent) developed recurrent fullthickness rectal prolapse; six of them underwent repeat perineal rectosigmoidectomy. Sixty-seven patients had fecal incontinence prior to surgery. Eleven patients underwent concomitant levatoroplasty; 10 of them either improved or regained full continence of feces postoperatively. Twenty-six of the 56 patients who underwent perineal rectosigmoidectomy alone improved or regained full continence. Rectal prolapse can be successfully treated by perineal rectosigmoidectomy in elderly, highrisk patients with minimal morbidity. Levatoroplasty dramatically improves fecal incontinence occurring in association with rectal prolapse.Read at the meeting of The American Society of Colon and Rectal Surgeons, Boston, Massachusetts, May 12 to 17, 1991.     

Perineal rectosigmoidectomy in the elderly

Between April 1989 and October 1991, 20 consecutive patients underwent perineal rectosigmoidectomy and coloanal anastomosis for full-thickness rectal prolapse. These 16 females and 4 males, with a mean age of 82 (range, 68–101) years, were evaluated by detailed functional assessment and physiologic testing. A grading scale from 0 to 24 was based upon the frequency and type of incontinence, 0 representing full continence. The mean preoperative continence score was 14.5, while the mean postoperative continence score was 8.4. The mean length of resected rectosigmoid was 23 cm. There was one postoperative death and one significant complication, a postoperative pelvic hematoma that required reoperation. There were no full-thickness recurrences at a mean follow-up of 26 months. Six of the 10 patients who underwent preoperative pudendal nerve terminal motor latency (PNTML) testing had evidence of severe neuropathy (latencies greater than 2.5 milliseconds). Prolonged PNTML, however, was not shown to be an accurate predictor of postoperative incontinence because four of the six patients with neuropathy regained excellent to good control. In conclusion, perineal rectosigmoidectomy is a safe operation for the treatment of full-thickness rectal prolapse in the elderly patient. Improved postoperative continence was noted in 90 percent of patients, with improvement seen even in those patients with severe pudendal neuropathy.Read at the meeting of The American Society of Colon and Rectal Surgeons, San Francisco, California, June 7 to 12, 1992.     

Anorectal dysfunction and rectal prolapse in progressive systemic sclerosis

Our aim was to characterize the clinical spectrum of anorectal dysfunction among eight patients with progressive systemic sclerosis (PSS) who presented with altered bowel movements with or without fecal incontinence. The anorectum was assessed by physical examination, proctosigmoidoscopy, and anorectal manometry. There was concomitant involvement of the other regions of the digestive tract in all patients as determined by barium studies, endoscopy, or manometry: eight esophageal, three gastric, four small bowel, and two colonic. Seven patients had fecal incontinence, and four also had second-degree complete rectal prolapse. Abnormal anorectal function, particularly abnormal anal sphincter resting pressures, were detected in all patients; anal sphincter pressures were lower in those with rectal prolapse. Rectal capacity and wall compliance were impaired in seven of seven patients. Successful surgical correction of prolapse in three patients resulted in restoration of incontinence for six months and seven years in two of the three patients. We conclude that rectal dysfunction and weakness of the anal sphincters are important factors contributing, respectively, to altered bowel movements and fecal incontinence in patients with gastrointestinal involvement by PSS. Rectal prolapse worsens anal sphincter dysfunction and should be sought routinely as it is a treatable factor aggravating fecal incontinence in patients with PSS.This work was presented in part at the Annual Meeting of the American Gastroenterological Association, May 1992, and appears in abstract form in Gastroenterology 1992;102:A473.Supported in part by the General Clinical Research Center Grant 00585 from the National Institutes of Health.     

Continence is improved after the ripstein rectopexy

PURPOSE: This study was undertaken to evaluate anal manometric changes after Ripstein's operation for rectal prolapse and rectal intussusception and to study the clinical outcome following the operation, with special reference to anal incontinence. METHODS: Forty-two patients with rectal prolapse or rectal intussusception were subjected to anorectal manometry preoperatively and seven days and six months postoperatively. A detailed history was obtained from each patient preoperatively and six months postoperatively. RESULTS: Preoperatively, patients with rectal intussusception had higher maximum resting pressure (MRP) (52±23 mmHg) than patients with rectal prolapse (34±20 mmHg;P <0.01). In the group of patients with rectal prolapse, there was a postoperative increase in MRP after six months (P <0.001) but not after seven days. Maximum squeeze pressure (MSP) did not increase. Neither MRP nor MSP increased postoperatively in patients with internal rectal procidentia. Continence was improved postoperatively both in patients with rectal prolapse (P <0.01) and rectal intussusception (P <0.01). There was no postoperative increase in rectal emptying difficulties. CONCLUSION: Ripstein's operation often improved anal continence in patients with rectal prolapse and rectal intussusception. This improvement was accompanied by increased MRP in patients with rectal prolapse, indicating recovery of internal anal sphincter function. No postoperative increase in MRP was found in patients with rectal intussusception. This suggests an alternate mechanism of improvement in patients with rectal intussusception.Supported by grants from Kjell and Märta Beijers Stiftelse and Marianne and Marcus Wallenbergs Stiftelse.     

What affects continence after anterior resection of the rectum?

Functional results after anterior rectal resections are commonly considered satisfactory but variable percentages of postoperative incontinence are often reported. Continence was evaluated after 20 low anterior resections (LAR) and 13 high anterior resections (HAR) by means of clinical assessment, anorectal manometry, and evacuation proctography. Whereas all HAR patients had perfect continence, 10 patients (50 percent of the LAR group had occasional episodes of soiling from liquid feces, 5 patients (25 percent had frequent soiling or occasional incontinence for solid feces, and 1 patient (5 percent had frequent solid stool loss requiring surgical treatment. Anal canal resting pressure at 3 and 4 cm from the anal verge was significantly lower in the LAR group (P<0.02 and P<0.05, respectively) than in the HAR group. However, the maximum voluntary contraction did not differ between the two groups. Rectoanal inhibitory reflex was found to be present in 17 of the 20 patients with LAR and in all patients with HAR. The volume at which the anal sphincter is continuously inhibited was significantly reduced in the LAR group (P<0.001). Also, the conscious rectal sensibility volumes were found to be significantly reduced for threshold, constant, and maximum tolerated volume. Threshold volume for internal sphincter relaxation was lower than the threshold volume for rectal sensation in some patients with LAR. This could allow postoperative fecal soiling. Rectal compliance was decreased (P<0.001) in the LAR group. Evacuation proctography, performed in six LAR patients affected by major soiling or solid stool loss, revealed an abnormal obtuse anorectal angle and pathologic lowering of the perineum at rest and during defecation. The concomitance of internal anal sphincter impairment, reduction in rectal compliance, and previous pelvic floor muscle damage are postulated as cause affecting continence in patients who underwent LAR.Read at the Congress on Colo-Rectal Disease Milan, Italy, June 29–30, 1989.     

Results of colectomy for severe slow transit constipation

PURPOSE: This study assesses the outcome of a standardized operation performed by two surgeons for severe idiopathic slow transit constipation that was resistant to laxative treatment. METHODS: Fifty-nine consecutive patients, 4 men and 55 women, with a mean age of 42.3 years, underwent colectomy with ileorectal anastomosis. Slow colonic transit was demonstrated in each case. Fifty-two patients were available for follow-up, with median time to follow-up being 42 (range, 3–81) months. RESULTS: Median bowel frequency was 4 per 24 hours. Sixty-nine percent had four or less bowel movements daily. Ten percent used antidiarrheal medication regularly. One patient had a stoma for recurrent severe constipation. Mean continence score was 1.8 (on a scale of 0–20); six patients were incontinent, and four of these six had normal preoperative anal manometry. Fourteen patients (27 percent) had difficulty with rectal evacuation. Preoperative defecating proctography was a poor predictor of postoperative evacuation difficulties. Twenty-seven patients (52 percent) had persisting abdominal pain, but there was a significant improvement in the degree of pain (P <0.00001). Forty-seven patients (90 percent) were satisfied with the outcome of the operation (and would elect to have it done again). Dissatisfied patients had recurrent constipation or diarrhea and incontinence. CONCLUSION: Colectomy with ileorectal anastomosis produces a satisfactory functional outcome in the majority of patients undergoing surgery for severe constipation with proven slow colonic transit.Supported by the Division of Surgery and the Colorectal Research Fund.Read at the meeting of the Royal Australasian College of Surgeons, Perth, Australia, May 1995.     

Abdominal rectopexy for complete prolapse: Prospective study evaluating changes in symptoms and anorectal function

The effect of abdominal rectopexy on bowel function is difficult to assess in retrospective studies because preoperative bowel habit cannot be determined accurately. This study examined bowel symptoms and physiologic tests of anorectal function prospectively in 23 patients before and at three months after rectopexy. Rectopexy eliminated complete prolapse in all and stopped bleeding in 16 of 18 patients. Incontinence improved significantly. Constipation (<3 bowel actions per week or straining for more than 25 percent of defecation time) was relieved in 4 of 11 affected patients but developed in 5 of the 12 who were not constipated preoperatively. Since the median bowel frequency was 21 motions per week before surgery and 17 afterward, the main determinant of constipation was straining. Abdominal pain was relieved after rectopexy in 6 of 12 patients but developed in 3 of 13 who were pain-free before surgery. Three patients (13 percent) had a first-degree relative with rectal prolapse. Perineal descent decreased significantly. Maximal anal resting pressure increased significantly, but this did not correlate significantly with improved continence. Twenty-one patients (91 percent) could expel a 50-ml balloon preoperatively; 18 of those 21 could still do so postoperatively. The two patients who could not expel the balloon preoperatively were able to do so postoperatively. This study shows that rectal prolapse is associated with profoundly abnormal defecation and abdominal pain. While abdominal rectopexy improved continence, it may improve or worsen other bowel symptoms, including constipation.Support for this study was received from the Imperial Cancer Research Fund, ICI Pharmaceuticals (SA) Ltd., the St. Mark's Research Foundation, and the Medical Research Council of South Africa.     

Surgical treatment of recurrent complete rectal prolapse

PURPOSE: Complete recurrent rectal prolapse (RRP) after initial prolapse surgery is well described. Our aim was to examine the possible causes for RRP, to learn of the operations performed most frequently, and to examine the outcome following recurrence surgery. METHODS: Patients with RRP were reviewed retrospectively from 1963 to 1993. RESULTS: A total of 24 patients (19 females) had RRP. Of these, 29 operations were performed; three patients had 2 RRP operations, and one patient had 3 RRP operations. Median age was 56 (range, 18–88) years. Median follow-up and median duration to recurrence were 6.75 (range, 0.08–17) years and 2 (range, 0.1–29) years. One patient had RRP at the end of the follow-up period. RRP occurred after 15 abdominal and 9 perineal operations. Treatment for RRP included 25 abdominal and 4 perineal operations. Causes for RRP were identified in 41 percent of cases and was most often attributable to problems with the mesh following the Ripstein procedure. Preoperative incontinence and constipation were largely unchanged by RRP operation. CONCLUSION: RRP occurred most commonly because of problems with the mesh, but no etiologic factor was found in the majority of patients. Abdominal operations were performed more frequently than perineal approaches for RRP. Elimination of prolapse can be obtained, but bowel dysfunction still remains in 60 percent of patients.     

Results of Delorme's procedure for rectal prolapse

PURPOSE: A retrospective study was undertaken to assess the results of Delorme's procedure for rectal prolapse and to determine the advantages of an innovative extended transrectal repair, which aims at performing a total pelvic floor repair. METHODS: A total of 85 patients, ranging in age from 21 to 97 years, were operated on. Sixty-five (82 percent) patients had varying degrees of fecal incontinence. Similar groups of patients were compared with regard to control of the prolapse and restoration of continence according to 1) age and medical condition and 2) operative technique: original vs.extended operation. RESULTS: Twelve patients (14 percent) developed postoperative complications. There was one perioperative death (1.2 percent). Eighty patients were followed for 6 to 136 (median, 33) months. Eleven (13.5 percent) developed recurrent full-thickness prolapse. The recurrence rate was significantly different 1) between 44 elderly and poor operative risk patients not suitable for abdominal surgery (22.5 percent) and 41 younger patients without concurrent medical conditions, electively submitted to perineal repair (5 percent) (P <0.05), and 2) between the original procedure (21 percent of 44 patients) and the modified technique (5 percent of 41 patients) (P <0.05). Forty five patients (69 percent) improved or regained full continence. No patient worsened. No residual dysfunction was induced. Restoration of continence was not influenced by selection of patients or surgical technique. CONCLUSIONS: Despite increased morbidity (22 percent; P <0.05), advantages of the modified technique were 1) over the original procedure, a reduced recurrence rate, 2) over perineal proctectomy, the absence of coloanal anastomosis and better functional outcome, and 3) over abdominal rectopexy, a less aggressive approach without disturbing effects on bowel habits.     

Delorme's procedure for complete rectal prolapse in severely debilitated patients

PURPOSE: This study was designed to assess the results of a minimally invasive surgical procedure for the correction of complete rectal prolapse in a poor surgical risk group. METHODS: Over a ten-year period, 40 patients underwent 41 Delorme operations when advanced age and/or poor overall health mitigated against an abdominal approach. Mean age was 82 (range, 30–100) years. Eighty-eight percent were females. Surgery was performed in the prone jackknife position utilizing intravenous sedation and local anesthesia. RESULTS: Follow-up ranges from 1 year to 2 years (mean, 47 months). There have been 9 recurrences in 8 patients (22 percent). Mean time to recurrence was 13 months (range, 1 month to 6 years). One death occurred in an 81-year-old patient within 24 hours of surgery from cardiopulmonary arrest. Minor complications occurred in 25 percent of patients. CONCLUSION: Satisfactory prolapse repair was safely performed in 78 percent of this high-risk group. Pitfalls in performing this procedure relate primarily to associated perineal and colonic conditions. Most prominent among these conditions are weak or absent anal sphincter tone, perineal descent, and previous sphincter injury. Extensive diverticular disease may prohibit effective and complete proximal mucosectomy. An inadequate mucosectomy sets the stage for early recurrence of prolapse.     

Delayed repair of obstetric injuries of the anorectum and vagina

PURPOSE: We categorized the various types of postobstetric injuries of the anorectum and vagina encountered in a five-year period. The operative procedures used to repair these injuries and the functional outcome after surgery were assessed. METHODS: Between 1986 and 1991, 52 patients were surgically treated for obstetric injuries of the anorectum and vagina; 48 patients were available for follow-up study. Four clinical injury types were identified: Type I, incontinent anal sphincter (11 patients); Type II, rectovaginal fistula (16 patients); Type III, rectovaginal fistula and incontinent anal sphincter (11 patients); and Type IV, cloaca-like defect (10 patients). The mean age of the patients was 30 years, the mean duration of symptoms before surgery was 13 months, and the mean follow-up period was 16 months. The major component of surgical repair for each injury type was: Type I, overlap repair of external anal sphincter; Type II, rectal mucosal advancement flap; Type III, overlap repair of external anal sphincter and rectal mucosal advancement flap; and Type IV, overlap repair of external anal sphincter, anterior levatorplasty, and anal and vaginal mucosal reconstruction. Fecal diversion was not performed in any patient. Specific questions were asked at the most recent follow-up assessment to determine results. RESULTS: Continence status postoperatively was classified as perfect, impaired, or poor; poor was defined as no improvement or worse. Postoperative continence (perfect, impaired, or poor) was, respectively: Type I (11 patients), 64 percent, 36 percent, and 0 percent; Type II (16 patients), 56 percent, 0 percent, and 44 percent; Type III (11 patients), 64 percent, 36 percent, and 0 percent; and Type IV (10 patients), 90 percent, 10 percent, and 0 percent. Vaginal discharge of stool was eliminated in all patients with a rectovaginal fistula. Subjectively, 92 percent of the patients had excellent or good results. Complications included wound hematoma (n=2), fecal impaction (n=2), urinary retention (n=1), and urinary tract infection (n=1). CONCLUSION: Patients with Type II injuries had the worst results (P < 0.001). These patients should be evaluated for anal incontinence before surgery to assess the need for a concomitant sphincteroplasty.     

Fistula-in-ano in Crohn's disease

The outcome of aggressive surgical treatment of 64 symptomatic anal fistulas in 55 patients with Crohn's disease has been studied. Forty-one fistulas, in 33 patients, were treated by conventional fistulotomy (17 subcutaneous, 19 intersphincteric, 5 low transsphincteric fistulas). Thirty wounds (73 percent) healed within 3 months and eight more wounds (93 percent) healed within 6 months. Three wounds did not heal within 12–18 months. Two of these patients subsequently required proctocolectomy. Wound healing was not influenced by the presence of rectal Crohn's disease or granulomatous inflammation in the tract. No change in continence was experienced by 26 of the 33 patients who underwent fistulotomy. Three patients required proctocolectomy and the remaining four patients experienced minor degrees of incontinence postoperatively. Sixteen high transsphincteric, five suprasphincteric, and one extrasphincteric fistula in 22 patients were treated by laying open external tracts and placing a noncutting seton through the sphincter, which was left in place for prolonged periods to maintain drainage. During follow-up (6 months to 10 years, median 2.5 years), three fistulas healed and seven remained quiescent. Nine patients required further treatment by a new seton and three patients required proctocolectomy. Eight of the 22 patients who had a seton inserted had no change in continence, and six patients in this group developed minor changes in continence, mostly related to diarrhea associated with intestinal disease. Anal fistulas in Crohn's disease, which involve minimal sphincter muscle, can be successfully treated by fistulotomy. High fistulas should be treated with seton drainage to limit recurrent suppuration and preserve sphincter function.Read at the 89^th meeting of the American Society of Colon and Rectal Surgeons, St Louis, Missouri April 29–May 4 1990.     

Anorectal pressure gradient in patients with anal incontinence

Anorectal pressures in patients with fecal incontinence have been investigated. With anal manometry, 34 percent of patients with fecal incontinence had maximal resting pressure and 39 percent had maximal squeeze pressure within the normal range. When a pressure gradient was calculated as the pressure difference between maximal resting pressure and rectal pressuring during filling of a rectal balloon, patients with fecal incontinence could be better distinguished from controls: 20 percent of patients with fecal incontinence had values within the normal range when the rectal pressure at the earliest defecation urge was used (P <0.05), and 12 percent had values within the normal range when the rectal pressure at maximal tolerable volume was used (P <0.01). Anorectal pressure gradient measurements seem to distinguish patients with fecal incontinence from controls better than maximal resting pressure or maximal squeeze pressure alone.     

New grading and scoring for anal incontinence

A grading system of anal incontinence (AI) is described that takes into account both degree and frequency of symptoms. A, B, and C indicate AI for flatus/mucus, liquid stool, and solid stool, respectively; 1, 2, and 3 indicate occasional, weekly, and daily AI. A scoring system, ranging from 0 (continence) to 6 (severe AI,i.e.,daily AI for solid stool or C3) also is reported. Three hundred thirty-five patients have been evaluated by this method in our institution: 30 percent had severe AI, graded as C3; only 9 percent had mild symptoms graded as A. Both males and females could not control diarrhea (Grade B) in 44 percent of cases. Nearly half of the 110 patients who underwent surgery had a C3 incontinence before treatment. Positive results were achieved in 75 percent of cases after surgery:e.g.,AI score significantly improved from 4.2±1.6 to 1.5±1.9 (P <0.001) in those with AI and rectal prolapse. Most of the failures were the patients with idiopathic C3 incontinence. In conclusion, this grading and scoring system allowed a satisfactory assessment of patients' AI before and after treatment. It may also be used to achieve an objective comparison between different series.