Restorative proctocolectomy: the four loop (W) reservoir

The four loop (W) reservoir was developed with the aim of improving the functional results of the J reservoir and to avoid the need for catheterization often seen with the S reservoir. It is no more difficult to construct and results in better function. Sixty-four patients have undergone the procedure. There were no deaths. Mean hospital stay (including ileostomy closure) was 24 +/- 8 days. The reservoir has been removed in one patient (one week postoperatively) who was found to have a Dukes' B rectal carcinoma. Function has been assessed in 51 patients (mean follow-up 18.6 +/- 8.9 months). Frequency of defaecation per 24 h is 3.3 +/- 1.0 (range 1-8) with night evacuation in 14 per cent. Antidiarrhoeal medication is required by 20 per cent of patients. Continence is normal in 92 per cent, and 8 per cent have minor leakage. All patients defaecate spontaneously.     

Restorative proctocolectomy with ileal reservoir for ulcerative colitis and familial adenomatous polyposis: a comparison of three reservoir designs

One hundred and four patients were treated by restorative proctocolectomy with ileal reservoir for ulcerative colitis and familial polyposis. Three different designs of reservoir were used (triple loop 68, double loop 13, quadruple loop 23). There were no postoperative deaths but six (5.8 per cent) had the reservoir removed. Rates for pelvic sepsis were 25, 15 and 13 per cent, and for intestinal obstruction requiring laparotomy 14.7,0 and 8.6 per cent. Function was assessed in 88 patients (58, 12 and 18) after mean intervals from closure of the ileostomy of 23.7, 12.7 and 4.5 months. Frequency of defaecation per 24 h was 3.7 +/- 1.6, 5.5 +/- 1.6 and 4.1 +/- 1.3, being significantly greater for double loop reservoirs; night evacuation was more prevalent in the same group (26, 58 and 22 per cent). Significantly fewer patients with triple than with double loop reservoirs required antidiarrhoeal medication (19 and 58 per cent). Normal continence occurred in 67, 75 and 89 per cent of patients in the three groups. All patients with double or quadruple loop reservoirs defaecated spontaneously while only 41 per cent with triple loop reservoirs did so. Mean intra-operative reservoir volumes were 177 +/- 64, 172 +/- 58 and 325 +/- 37 ml and volumes after closure of the ileostomy were 416 +/- 176, 197 +/- 69 and 322 +/- 33 ml respectively. Double loop reservoirs were significantly smaller than the other two designs after ileostomy closure. There was an inverse relationship between reservoir volumes and frequency. A quadruple loop reservoir directly connected to the anal sphincter preserved spontaneous evacuation and resulted in function similar to that obtained with the triple loop reservoir.     

Quality of life after restorative proctocolectomy with pelvic ileal reservoir

Fifty-five patients who had undergone restorative proctocolectomy with a pelvic ileal reservoir between 3 and 82 months previously were sent a detailed questionnaire to assess their quality of life. All patients had experienced an ileostomy at some time. Fifty-one (94 per cent) returned completed questionnaires. The average frequency of evacuation from the reservoir was four times per 24 hours. Antidiarrhoeal medication was required by 14 (27.4 per cent). Continence was normal in 32 (62.7 per cent) and only 3 (5.9 per cent) had troublesome faecal leakage. Twenty-two (43.1 per cent) evacuated stool spontaneously, while 24 (47.1 per cent) needed to catheterize the reservoir for each evacuation. Four patients who had had the reservoir removed were added to the 51 responders to give a total number of 55 whose quality of life was considered in this study. The reservoir was preferred to an ileostomy with regard to confidence (87 per cent), cleanliness (89 per cent), sexual self-image (87 per cent), social (85 per cent) and sport activity (87 per cent), and ease of carrying out work (84 per cent). The overall preference was 87 per cent. Thirty-four (66.7 per cent) felt there was no significant disadvantage associated with the reservoir, while 10 (19.6 per cent) saw the long convalescent period and 9 (17.6 per cent) the requirement for catheterization as drawbacks.     

Synchronous abdominotrans-sphincteric resection of low rectal cancer: new technique for direct colo-anal anastomosis

Sixty-five patients with low rectal cancer 4-9 cm from the anal margin were treated by rectal resection and direct colo-anal anastomosis. The procedure simultaneously combined abdominal dissection with the patient in the lateral position with a trans-sphincteric approach. In 57 cases a temporary defunctioning colostomy was performed. There were no postoperative deaths. Six patients (9 per cent) developed pelvic sepsis or anastomotic leakage. Faecal continence was normal in 46 of 51 patients (91 per cent) who were operated on at least 1 year previously. The remaining 5 complained of occasional minor soiling. No patients require a permanent colostomy for incontinence. Of 29 patients treated for potential cure greater than 3 years previously, 24 (82 per cent) were alive without recurrence. Local recurrence occurred in four patients (6 per cent). Direct colo-anal anastomosis using a combined abdominotrans-sphincteric approach has produced good functional results without impairing the patient's life expectancy.     

The W ileal reservoir: long-term assessment after proctocolectomy for ulcerative colitis and familial polyposis.

BACKGROUND. This report examines the viability of the W reservoir as a reliable option for the treatment of ulcerative colitis and familial polyposis and studies W reservoir adaptation as reflected by changes in compliance and stool frequency. METHODS. Since 1984, 109 patients have undergone proctocolectomy with W reservoir reconstruction. Ileal reservoir static compliance was measured in 70 and 57 patients at 2 and 12 months after ileostomy takedown and in 25 patients at 3 years. Compliance was calculated as the change in volume over change in pressure. RESULTS. Twenty-four-hour stool frequency decreased from 7.3 +/- 0.2 at 2 months to 4.9 +/- 0.2 at 1 year for patients with ulcerative colitis and from 6.3 +/- 0.4 to 3.4 +/- 0.4 for patients with familial polyposis (p less than or equal to 0.05). Compliance increased from 12.7 +/- 0.6 ml/mm Hg to 14.3 +/- 0.6 ml/mm Hg between 2 months and 1 year. No significant increase in compliance occurred after 1 year. Ninety-six percent of patients were continent during the day at 12 months although 10% experienced occasional minor leakage at night. Average postoperative morbidity (for example, small-bowel obstruction, anastomotic complications) was 35%. No operative deaths, pelvic sepsis, or reservoir loss occurred. CONCLUSIONS. We conclude that W ileal reservoirs (1) are an excellent option for ileal reservoir reconstruction, (2) have optimal functional and compliance properties versus lower capacity designs and straight ileoanal pull-through procedures, and (3) maintain stable compliance characteristics and functional reservoir volume after the initial year of adaptation.     

THE CONTINENT ILEOSTOMY — AN 11 YEAR EXPERIENCE

An 11 year experience of the continent ileostomy is reported. Forty-nine patients have undergone a continent ileostomy during this period. There were 30 females and 19 males. The mean age was 35 years, the youngest 19 and the oldest 58. Forty-four patients were suffering from chronic ulcerative colitis, four from polyposis coli and one from carcinoma of the rectum and multiple colonic polyps. The majority (41) of operations were carried out as secondary procedures but eight were done as primary procedures. The follow-up varies from 11 years to three months. There were no deaths. Seventeen patients (34%) required revisional surgery for complications relating to the valve reservoir or stoma. Eleven of these were related to the valve — slipping four, prolapse five, fistula and perforation one each. Three patients had a para-ileostomy hernia and two a stricture at skin level. Four patients required excision of the reservoir with establishment of an orthodox ileostomy. Eighty-six per cent of patients are continent and never wear an appliance. The average capacity of the reservoir is 400 ml (range 15CL900). Most patients catheterize three times daily (range 2–6). Eighteen per cent catheterize once at night. The average time for catheterization is 6 min. Provided the procedure is technically satisfactory the continent ileostomy is considered the operation of choice for patients requiring a permanent ileostomy.     

The continent ileostomy--an 11 year experience

An 11 year experience of the continent ileostomy is reported. Forty-nine patients have undergone a continent ileostomy during this period. There were 30 females and 19 males. The mean age was 35 years, the youngest 19 and the oldest 58. Forty-four patients were suffering from chronic ulcerative colitis, four from polyposis coli and one from carcinoma of the rectum and multiple colonic polyps. The majority (41) of operations were carried out as secondary procedures but eight were done as primary procedures. The follow-up varies from 11 years to three months. There were no deaths. Seventeen patients (34%) required revisional surgery for complications relating to the valve reservoir or stoma. Eleven of these were related to the valve--slipping four, prolapse five, fistula and perforation one each. Three patients had a para-ileostomy hernia and two a stricture at skin level. Four patients required excision of the reservoir with establishment of an orthodox ileostomy. Eighty-six per cent of patients are continent and never wear an appliance. The average capacity of the reservoir is 400 ml (range 150-900). Most patients catheterize three times daily (range 2-6). Eighteen per cent catheterize once at night. The average time for catheterization is 6 min. Provided the procedure is technically satisfactory the continent ileostomy is considered the operation of choice for patients requiring a permanent ileostomy.     

Long-term follow-up of the ileoanal anastomosis in children and young adults

The purpose of this study was to carry out a long-term study of the ileoanal anastomosis (IAA) in children and young adults, comparing the straight IAA to the J pouch. One hundred twenty-one young people who had undergone IAA were studied, with 114 available for long-term follow-up. One hundred one were 18 years and under. Forty-nine patients had a straight IAA and 72 had a J-pouch reservoir. There were no deaths. After surgery, three children had intraabdominal sepsis and one had pelvic sepsis, but it did not lead to excision of the IAA. The mean stool frequency in all 114 patients was 5.0 +/- 2.5 per day and 1.2 +/- 1.1 at night. The mean number of stools for the straight IAA was 6 per day and 2.1 at night. The mean number of stools for the straight IAA with balloon dilations was 5.8 per day and 1.2 at night, and for the J pouch it was 4 per day and 1 at night. Patients with both the J pouch and straight IAA had good to excellent sensation, with patients with the J pouch always able to distinguish flatus from stool in 87% of patients and almost always in 13%. Daytime continence was very good in both groups. Moderate nighttime loss of stool occurred in 10 patients, 6 with a straight IAA and 4 with a J pouch. Ninety-five percent of the 114 patients were satisfied or very satisfied, with most children with a J pouch very satisfied. The J pouch remains the procedure of choice in young people.     

The model AS 800 artificial urinary sphincter: Mayo Clinic experience

The model AS 800 artificial urinary sphincter was implanted in 100 male and 9 female patients between 7 and 89 years old. Postoperative followup was 1 to 32 months. The indication for implantation was total urinary incontinence in 86 patients (78.9 per cent), stress incontinence in 22 (21.2 per cent) and urgency incontinence in 1 (0.9 per cent). Of the patients 97 (89 per cent) underwent implantation for the first time, 7 (6.4 per cent) had a previous artificial urinary sphincter model replaced by the AS 800 device and 5 (4.6 per cent) underwent reimplantation of a previous model. The cuff was placed around the bladder neck in all 9 female patients, whereas in the male patients the cuff was implanted around the bladder neck in 20 and around the bulbous urethra in 80. Thirty-one patients (28.4 per cent), 29 of whom were continent at night, were practicing nocturnal deactivation of the device. Complete post-activation continence was achieved in 91 patients (83.5 per cent), some leakage occurred in 10 (9.2 per cent) and 8 (7.3 per cent) remained incontinent. A total of 23 patients required 1 or more revisions, the most common indications for the first revision being loss of cuff compression (9), tubing kink (3), cuff erosion (3) and infection (2). At the time of this report 89 patients (81.7 per cent) were continent, 9 (8.3 per cent) still had some leakage, 3 (2.8 per cent) were incontinent, 5 (4.6 per cent) were awaiting reimplantation and 3 (2.8 per cent) had died of unrelated causes.     

Ileal pouch-anal anastomosis. The Emory University experience.

The ileal pouch-anal anastomosis has become a practical alternative to proctocolectomy for the treatment of ulcerative colitis and polyposis coli. To evaluate its success, the Emory University Affiliated Hospital experience from February 1984 to March 1989 was retrospectively reviewed. There were a total of 50 patients identified; 84 per cent had ulcerative colitis, and 16 per cent had polyposis coli (familial polyposis and Gardner's syndrome). The majority of these patients underwent a two-stage operation, but one-third required a three-stage procedure due to difficulty in mucosal proctectomy or toxic megacolon. J-pouch construction was performed in 72 per cent of patients, S-pouch construction in 14 per cent, straight ileo-anal anastomosis in 8 per cent, and lateral isoperistaltic ileo-anal anastomosis in 6 per cent. Of the 50 patients, 36 (72%) have had closure of the temporary ileostomy. Fourteen patients have not had ileostomy closure due to change in diagnosis to Crohn's disease, operative complications, or ileostomy closure pending. The combined operative morbidity per patient for the ileal pouch-anal anastomosis and the closure of the ileostomy was 32 per cent. This included bowel obstruction, 16 per cent; pelvic abscess, 6 per cent; and ileo-anal separation, 4 per cent. Follow-up on patients with ileostomy closure ranged from 6 months to 4 years (mean, 1.3 years). Stool frequency was 5.9 stools per 24 hours at 6 months and improved with time. During the follow-up period, all patients were eventually completely continent of stool during the day, and most became completely continent of stool at night.(ABSTRACT TRUNCATED AT 250 WORDS)     

Restorative proctocolectomy with a three-loop ileal reservoir for ulcerative colitis and familial adenomatous polyposis. Clinical results in 66 patients followed for up to 6 years.

The results of restorative proctocolectomy with a three-loop ileal reservoir were reviewed. Of 66 patients treated between 1976 and 1982, 52 had ulcerative colitis and 14 had familial adenomatous polyposis. The temporary ileostomy was closed between 2 and 78 months previously in 63 cases. Of these, three patients had had the reservoir removed and two were lost to follow-up. One other patient was subsequently found to have Crohn's disease. Function was assessed in 55 patients who had undergone closure of the ileostomy more than 8 weeks previously. Mean frequency of defecation was 3.7 per 24 hours (range 1-9.5) and 11 patients (20%) were taking antidiarrheal medication. Spontaneous defecation occurred in 22 patients (40%) while 29 (52.7%) had to use a catheter passed per anum. Four patients defecated spontaneously but sometimes used a catheter. Continence was normal in 36 (65.4%) and minor leakage once every 2 to 3 days occurred at night in 16 (29.1%). Three patients (5.4%) had some soiling during day and night. Troublesome perianal soreness (five patients, 9.1%) necessitated a defunctioning ileostomy in one. Fifty-four of the 55 patients assessed preferred their quality of life to that with an ileostomy.     

Single layer anastomosis in the upper gastrointestinal tract

A total of 466 single layer upper gastrointestinal anastomoses were made in 349 patients during an 11-year period. Six (1.3 per cent) anastomoses leaked. Radiological leakage was seen in one of 24 (4.2 per cent) oesophagogastric/jejunal anastomoses. There were no leaks after 66 gastrojejunal anastomoses; one of 84 (1.2 per cent) gastroduodenal anastomoses leaked and was converted to a gastrojejunal anastomosis. Two of 121 (1.7 per cent) biliary-enteric anastomoses leaked and both were successfully managed without reoperation. Two of 171 (1.2 per cent) enteroenteric anastomoses leaked, both in patients with established intraperitoneal sepsis which proved fatal. Of the 349 patients, 13 (3.7 per cent) died in hospital or within 30 days of operation but in only two was anastomotic leakage implicated. Single layer appositional upper gastrointestinal anastomoses are simple, safe and economic.     

去带盲结肠可控性膀胱术18例报告

目的 介绍去带盲结肠可控性膀胱术的临床经验。方法 对18例膀胱癌患根治性全膀胱切除后,施行了去带可控性盲结肠膀胱术。结果 获访17例,平均随访16（6－28）个月。16例能完全控制排尿,1例夜间轻微溢尿。插管均容易。术后12月膀胱容量平均415ml。充盈压力平均22cmH2O。无输尿管狭窄及反应。肾功能、 血生化、尿液化验正常。结论 去带盲结肠可控性膀胱术具有手术简便、容量大、内压低、无反流、可控排尿、并发症少的优点。     

Comparison of clinical and compliance characteristics between S and W ileal reservoirs

Ileal reservoir reconstruction has become the preferred technique for restoration of bowel continuity in most patients after colectomy for ulcerative colitis or familial adenomatous polyposis. We analyzed and compared compliance characteristics of triple-limb S and quadruple-limb W reservoir designs and correlated changes in capacity with overall function. Fifty patients underwent colectomy and reservoir construction for ulcerative colitis or familial adenomatous polyposis; 12 received S reservoirs and 38 received W reservoirs. Reservoir compliance was assessed by means of a specially designed condom catheter that continuously recorded intrareservoir pressure and changes in perfused volume. During reservoir infusion, volumes and pressures at initial fullness, normal sensation of evacuation, and maximum tolerated volume were noted. Studies were performed at 2 and 12 months after ileostomy takedown. An increase in normal evacuation volume from 218 +/- 9 mL to 310 +/- 12 mL between 2 and 12 months (p less than or equal to 0.05) was observed in patients with W reservoirs. Similar changes were recorded in S reservoir reconstructions (201 +/- 14 mL to 291 +/- 22 mL, p less than or equal to 0.05). No significant differences were observed in the mean pressure at normal evacuation volume between the S and W groups at 2 and 12 months. The 24-hour stool frequency decreased an average of 3 per day for both reservoir designs between the 2- and 12-month study period (p less than or equal to 0.05). This frequency was most directly predicted by normal evacuation volume (r = 0.90 for W and 0.88 for S). The decrease in stool frequency correlated with increased reservoir compliance, as shown by larger tolerated volumes at similar pressures. Restorative proctocolectomies with S or modified W reservoirs are both acceptable alternatives and demonstrate similar compliance characteristics and functional results.     

Comparison of the function of triplicated and duplicated pelvic ileal reservoirs after mucosal proctectomy and ileo-anal anastomosis for ulcerative colitis and adenomatous polyposis

Among 39 consecutive patients who underwent colectomy, mucosal proctectomy and ileo-anal anastomosis, a triplicated pelvic ileal pouch was constructed in 17, and a duplicated pouch in 22 patients. There was no mortality, but complications such as anastomotic dehiscence and pelvic sepsis led to removal of the pouch in seven patients (18 per cent). The functioning of the pouch and anal sphincter was assessed in 31 patients 6 months, and in 22 patients 12 months after closure of the diverting ileostomy. By 6 months, all patients were either completely continent or experienced only minor leakage and defaecation could be deferred for more than 15 min by 81 per cent of patients and flatus distinguished from faeces by 90 per cent of patients. No significant differences between triplicated and duplicated pouches were discernible at 6 months. At 12 months defaecation was significantly less frequent (P less than 0.05) in patients with triplicated pouches (median, 5 times in 24 h) than in patients with duplicated pouches (7 times in 24 h). All patients with triplicated pouches and all except one with duplicated pouches were able to defaecate spontaneously, without needing to intubate the reservoir. Thus, provided the early postoperative problems can be overcome, most patients achieve good anal function after mucosal proctectomy combined with a pelvic ileal reservoir. No evidence was found in this study that the functional results of duplicated pouches were superior to those of triplicated pouches; in fact, the triplicated pouches proved to be slightly superior.     

Prognostic value of urodynamic testing in myelodysplastic patients

We herein describe the clinical progress of 42 myelodysplastic patients studied urodynamically and followed for a mean of 7.1 years. Urodynamic evaluation included urethral pressure profilometry, simultaneous determination of urethral pressure, intravesical pressure and external anal or external urethral sphincter electromyography with fluoroscopic voiding cystourethrography. Assessment of urethral function showed 36 patients (86 per cent) with an open vesical outlet and nonfunctional proximal urethra. Cystometrography revealed that 7 of 42 patients (17 per cent) had reflex detrusor activity: 4 with coordinated micturition and 3 with detrusorsphincter dyssynergia. Thirty-five patients (83 per cent) had areflexic detrusor dysfunction: 5 with atonic detrusor response and 30 with a progressive increase in pressure with increasing volume. The intravesical pressure at the time of urethral leakage was 40 cm. water or less in 20 patients and at pressures greater than this value in 22 patients. No patient in the low pressure group had vesicoureteral reflux and only 2 showed ureteral dilatation on excretory urography. In contrast, of the patients in the higher pressure group 15 (68 per cent) showed vesicoureteral reflux and 18 (81 per cent) showed ureteral dilatation on excretory urography. Thus, a striking relationship between the urethral closure pressure and intravesical pressure at the time of urethral leakage and the clinical course in this group of myelodysplastic patients is demonstrated. Every patient with a normally closed vesical outlet was continent on intermittent catheterization and an anticholinergic agent, while only 60 per cent of patients with open bladder outlets similarly treated achieved good urinary control and none was dry. An artificial sphincter device would seem to be a reasonable method to achieve urinary control in the latter patients but the detrusor response to filling also must be considered. Detrusor hypertonia should be controlled or controllable before a sphincter augmenting device can be used safely. Treatment options for patients with high urethral closure pressures include intermittent catheterization and anticholinergic medications or a sphincter ablative procedure to decrease the outlet resistance combined with anticholinergic therapy and implantation of an artificial sphincter. However, only longer followup will determine if these therapeutic regimens will prevent upper urinary tract deterioration.     

The efficacy of free tissue transfer in the treatment of osteomyelitis

We evaluated the efficacy of free tissue transfer in the treatment of osteomyelitis in thirty-three patients who were followed for an average of 41.4 months (range, twenty-four to fifty-nine months). The bone lesions were classified in two ways: first, according to whether there had been drainage for less or more than six months and second, according to whether the lesion was Type I, II, or III. The thirty-three patients underwent a total of thirty-seven free tissue transfers, and thirty (79 per cent) of the transfers survived. Major complications were encountered after 41 per cent of the surgical procedures, consisting of failure of the free tissue transfer in seven patients (21 per cent) and recurrent sepsis in six (20 per cent) at follow-up. Four of the seven patients in whom the free tissue transfer failed underwent an amputation. Six of the patients with a failed transfer and nine of the ten with recurrent sepsis had a Type-III lesion. Previously published reports have suggested that free tissue transfer is an extremely successful and reliable procedure for the treatment of osteomyelitis. In our series limited success was achieved in patients with a Type-I or II lesion, but six of the patients with a Type-III lesion had recurrence of infection despite a successful transfer.     

Management of perianal sepsis in immunosuppressed patients

Despite improvements in the supportive care of immunosuppressed patients controversy still surrounds the surgical management and outcome of anorectal sepsis in these patients. We reviewed 83 immunocompromised patients with diagnosis of perianal sepsis from 1995 to 1997. Sixty-six patients (80%) were followed for a mean of 15 months. Mean age was 44 years and 76 per cent were males. Twenty-eight per cent were HIV+, 34 per cent had inflammatory bowel disease on steroids, 20 per cent had malignancies, and 18 per cent had diabetes. Twenty-eight per cent had anal fistula, 2 per cent had perianal abscess, and 40 per cent had both. Primary sites of fistula were: transsphincteric (38%), intersphincteric (33%), superficial (20%), and suprasphincteric (3%), and multiple tracks (6%). Horseshoeing was present in 14 per cent of cases. The most commonly practiced surgical procedures were primary fistulotomy (n = 23) and fistulotomy plus drainage (n = 28). Seven patients underwent fistulotomy and ostomy and eight patients were treated with fistulectomy plus drainage. Most wounds (91%) healed within 8 weeks. Incontinence (6%) and recurrence (7%) were the most commonly observed complications. These results are similar to those seen in the general population. Perianal sepsis can be safely managed in immunocompromised patients, with high rates of healing and low complication rates. An aggressive sphincter-preserving approach in the management of these patients may be undertaken.     

Conservative proctocolectomy: a dubious option in ulcerative colitis

Conservative proctocolectomy was performed for ulcerative colitis in 19 patients, Crohn's disease in three and familial adenomatous polyposis in one. Healing was uncomplicated in only three patients (13 per cent). Eleven developed an anal discharge and nine an infected pelvic haematoma despite peranal drainage. Fourteen patients developed pelvic sepsis and, despite surgical curettage in 11, none healed. Six of these patients have had the anal sphincter divided, with healing in only one, and the anal canal has been excised in two. Eleven patients have ultimately healed at a median time of 28 months and eight have persistent sepsis after a median period of 45 months. Two patients with sepsis have had a successful ileoanal anastomosis. Conservative proctocolectomy cannot be recommended as a definitive operation for ulcerative colitis even though it may permit a subsequent restorative procedure.     

Proctocolectomy without ileostomy: ileo-anal anastomosis with an ileal reservoir

Six patients with polyposis coli and five with chronic ulcerative colitis underwent total colectomy and mucosal proctectomy with preservation of the anal sphincter and levator ani muscle. An ileal reservoir, constructed from the terminal ileum, is brought out through the anal sphincter for anastomosis to to the anus at the dentate line. Two patients in the series had a three-limb reservoir with anastomosis of a short efferent ileal limb to the anus (Parks technique). Nine patients had a two-limb J-shaped reservoir with the apex of the reservoir anastomosed to the anus (Utsunomiya technique). All patients had a temporary defunctioning ileostomy. There were no deaths. Two patients suffered from severe infection within the rectal muscle cuff. One resolved completely with spontaneous drainage into the reservoir; the other had continuing sepsis and eventually required excision of the anus and conversion to a continent ileostomy. All patients remain in good general condition with no disturbance of urinary or sexual function. Continence is satisfactory in all patients but two wear a pad at night. All evacuate their reservoirs spontaneously: none requires the use of a catheter. For most patients stool frequency varies from four to eight times daily but two patients have more frequent bowel motions. Colitis patients have more frequent bowel activity than those with polyposis. Three patients take Imodium tablets to lessen bowel frequency. The operation should be reserved for specialized centres and is still under trial; however, it appears likely to become the operation of choice for all patients with polyposis coli and for many patients with chronic ulcerative colitis.