Table 16.2 shows the complications and functional results of perineal colostomy when constructed from a colonic smooth-muscle wrap associated with colonic irrigations. Early specific complications are mainly related to coloperineal partial wound dehiscence, peristomal infection, coloperineal anastomotic necrosis, colon ischemia, and colovaginal fistula. In some patients, these complications require conversion to a definitive abdominal colostomy. Mucosal prolapse and ­stricture of the neoanus are the commonest late ­complications [26, 27].

Gamagami et al. [27] demonstrated manometrically that the cuff maintains a high pressure zone, with a mean pressure of 39 cm of water, and Lasser et al. [26] found a tonic neosphincter in 60 % of patients; however, all patients in both series utilized colonic irrigation so that pseudocontinent status was presumably achieved in large measure by irrigation of the perineal colostomy daily or every other day rather than by the presence of the neo-internal sphincter itself. Lasser et al. also pointed out that the absence of tonicity was not always correlated with a bad functional result and that the contribution to continence of the smooth-muscle neosphincter seemed to be related more to the cerclage effect for the avoidance of gross leakage. Portier et al. [28] also performed a perineal colostomy associated with a Malone procedure in 18 patients after abdominoperineal resection for cancer, and at 6 months of follow-up the mean continence score was 6.4 (of a total of 20) with a mean fecal incontinence quality of life total score of 12.5 (of a total of 16). None of the patients requested an abdominal colostomy because of a poor functional outcome. In this series, it is somewhat surprising that there were no reported perineal complications, which might be explained by the fact that nonirradiated tissue is brought into the pelvis by the pull-through colon. Pocard et al. [74] compared the functional results and quality of life of coloanal anastomosis and perineal colostomy with an autotransplant of a free flap of colonic muscle around the perineal stoma, as described by Lasser and colleagues [26], and found that the two techniques resulted in comparable results for both function and quality of life. None of those studies show the comparative functional status of patients with and without enemas. In a recent article, Hirche et al. [75] reported the long-term outcome of a series of 27 patients who had undergone neosphincter reconstruction [23] associated with perineal and neosphincteral training, external electrostimulation of the perineal cuff for a minimum of 6 months, home biofeedback (in most cases), and colonic irrigation. Objective assessment of normal continence was achieved in 80 % of patients, with statistically significant differences noted between resting and compression pressures in all patients.

Table 16.3 shows the complications and functional results of an implanted artificial sphincter for creation of a continent perineal colostomy. Romano et al. [52] published the largest series, including eight patients, and that is the only series in which no complications related to the implant of the sphincter occurred (follow-up range, 6–28 months). The rest of the reports [13, 53, 76] are anecdotal with regard to patient outcomes, and many patients subsequently had the device explanted because of skin or colon erosion.

Construction of a pouch in TAR has been reported only as an additional procedure to sphincter re-creation without associated morbidity in a few series and in a small number of patients. Unfortunately, studies showing an objective comparison among patients with and without the pouch are still lacking. In these studies the types of pouch differ, making analysis difficult. Geerdes et al. [11] reported that the addition of a pouch associated with a double dynamic graciloplasty in 4 patients when compared with 11 patients without a pouch did not result in improved continence. In another series of 26 patients, Vorobiev et al. [10] described a technique of smooth-muscle plasty of the internal anal sphincter in combination with a colonic C-shaped pouch, reporting perfect continence in 85 % of patients assessed 12 months after ileostomy closure. The mean resting anal pressure was 39 mmHg at 3 months and increased to 54 mmHg after 1 year. Vorobiev et al. also observed similar increases in the threshold and maximal tolerated volumes of neorectal distension. The role of a coloplasty in combination with re-creation of an internal sphincter and the implant of an artificial sphincter, as in our case [13], cannot be objectively elucidated.

In an experimental study, Hughes et al. [7] studied the effects of intramuscular electrical stimulation of a colonic J pouch in combination with an electrically stimulated gracilis after abdominoperineal excision of the rectum of seven dogs, demonstrating that electrical stimulation of the pouch generated contraction and sufficient intraluminal pressure to effect pouch evacuation. This option, however, has not yet been applied to humans.

Table 16.4 shows the complications and functional results of ACE alone or combined with other procedures, excluding smooth-muscle neosphincter recreation, already shown in Table 16.2. The most frequent early complications related to the Malone procedure include abscesses and appendiceal necrosis, both of which require re-intervention and conversion to a cecostomy. The ileal/cecal/colonic conduit procedures are technically more complex and carry a higher morbidity rate consequent upon the presence of an intra-abdominal anastomosis. Late complications usually are related to stomal stenosis, which can be easily managed by a temporary catheter at night or by a surgical V-Y plasty. Other late complications include stomal leakage and reflux. To prevent orifice strictures and mucosal prolapse, Ardelean et al. [77] performed the VQZ plasty, and to prevent leakage and reflux, Levitt et al. [78] plicated the cecum around the appendix or the neoappendix—somewhat like a Nissen fundoplication. Williams et al. [79] described a colonic conduit with an intussuscepted Kock-type valve to prevent reflux of the effluent. Overall, the complication rate for continent small-diameter stomas is high but most complications are relatively easy to treat [80].

Saunders et al. [67] proposed the combination of an electrically stimulated gracilis and a continent colonic conduit for TAR. Seven of 14 patients who had completed treatment with sufficient follow-up were able to control solid and liquid stool. In their series of perineal colostomy and appendicostomy, Farroni et al. [64] reported that fecal loss did not occur 1 h or more after antegrade continence enema in 11 out of 13 patients; the other patients reported only limited leakage. During follow-up, no single patient requested conversion to an abdominal stoma. In another group of patients with anorectal malformations undergoing reoperation, Alderlean et al. [62] reported that all of their nine patients in whom a Malone’s procedure was performed were able to stay dry for at least 24 h. The ACE procedure contributes to the avoidance of constipation after TAR when external sphincter reconstruction or substitution has been performed. It seems that in all the procedures in which ACE was associated, the good functional results were due to colonic irrigation rather than the other aspects of the technique, although authors do not report the status of a given patient with and without the irrigation.

The analysis of all these retrospective series shows that patients who underwent TAR after abdominoperineal resection prefer a perineal colostomy rather than an abdominal colostomy, even if the functional result has not been optimal. However, it must be taken into account that all patients constitute a selected group, and no randomized studies compare both alternatives. It is well known that after coloanal anastomosis, a significant number of patients experience variable degrees of disabling functional results, especially when radiochemotherapy was administered [81]. With respect to this point, Grumann and colleagues [82] reported a better quality of life in patients undergoing APR when compared with those undergoing a low anterior resection and coloanal anastomosis. Although this study has methodologic biases, it underlines the importance of quality of life after sphincter-sparing operations, and rectal cancer studies should focus more on criteria concerning these parameters as opposed to broad sphincter preservation rates [83–85].