Total anorectal reconstruction with gracilis muscle flap-assisted coloplasty following abdominoperineal resection: a case report

Abstract

Introduction

Total anorectal reconstruction (TAR) can be challenging for surgeons aiming to achieve both anatomic and functional anal reconstruction. Techniques involving coloplasties and graciloplasties have been described but have yet to become routine procedures. In this report, we present a rare case of TAR combining a coloplasty with an innervated gracilis muscle flap. This case aims to illustrate the surgical approach and emphasize the critical role of physiotherapy in postoperative functional recovery.

Methods

We report the case of a young female patient who underwent abdominoperineal resection (APER) following a low rectal carcinoma, resulting in a permanent end colostomy. Five years later, with no evidence of local or distant recurrence, the patient sought to improve her quality of life. TAR was performed by mobilizing the colon into the perineum and creating a neosphincter through an innervated gracilis muscle flap loop around the colon. Postoperatively, the patient underwent 6 months of intensive pelvic floor rehabilitation, including manometry-guided biofeedback therapy. This case report was prepared in accordance with the CARE (CAse REport) guidelines.

Results

Minor revision surgeries were needed to optimize the perineal colostomy scar, and three dilatation procedures were performed over 4 years to manage mild stenosis. The neo-anus remains functional at the 5-year postoperative visit, with fecal continence achieved within 2 months.

Conclusion

TAR is documented in the literature, though still not used systematically for various reasons, including inadequate muscle contraction, fatigability, and lack of neuroplastic adaptations in the central nervous system. Through this report, we demonstrate that an adequate surgical technique, followed by intensive pelvic floor training and manometry-assisted biofeedback to promote neuroplasticity, can provide successful outcomes.

Supplementary information

The online version contains supplementary material available at https://doi.org/10.1007/s00384-026-05185-9.

Introduction

In recent decades, the management of low rectal cancer has undergone a significant paradigm shift, driven by advances in imaging, neoadjuvant therapy, and surgical techniques. High-resolution pelvic magnetic resonance imaging (MRI) has improved local staging accuracy and assessment of circumferential resection margins, enabling better selection of patients for sphincter-preserving strategies [1]. In parallel, the evolution of neoadjuvant approaches, including intensified chemoradiotherapy and total neoadjuvant therapy protocols, has enhanced tumor downstaging and expanded the potential for organ preservation [2]. As a result, procedures such as intersphincteric resection and local excision using transanal platforms, including transanal minimally invasive surgery (TAMIS), as well as watch-and-wait strategies in selected complete responders, have reduced the need for abdominoperineal resection (APER) [3, 4].

Despite these advances, APER remains unavoidable in a subset of patients with low rectal cancer. In these patients, permanent abdominal colostomy imposes a substantial physical and psychological burden, significantly impairing quality of life (QoL) [5–8]. Therefore, restoring anorectal function following APER represents a major reconstructive challenge and an important unmet clinical need. Total anorectal reconstruction (TAR) has emerged as a potential solution for re-establishing continence and improving patient QoL. Among available techniques, neo-anal sphincter reconstruction using graciloplasty is promising, aiming for both anatomical reconstruction and functional restoration.

The concept of TAR is not novel, with early attempts using various muscle transposition techniques, including the gluteus maximus. Among emerging innovations, dynamic graciloplasty as first described by Baeten et al. represented a major step toward functional neosphincter reconstruction [9, 10]. Other approaches involving artificial bowel sphincters have also been investigated, although their long-term use has remained limited by device-related complications and revision requirements [9, 11]. Despite these advances, the functional results of TAR have remained variable, and its current use should be considered highly selective and limited to experienced teams.

A major limitation of TAR lies in the absence of native anorectal sensory function, which plays a central role in continence mechanisms. Previous studies have shown that reconstructed anorectal units often fail to restore normal sensory feedback, thereby limiting the recovery of physiological continence [12]. In addition, long-term outcomes following dynamic graciloplasty have been heterogeneous, with reported success rates around 50–60% and relatively high complication and revision rates [13, 14]. These findings highlight both the complexity of recreating a functional anal sphincter and the importance of careful patient selection.

Gracilis-based TAR remains one of the most extensively studied reconstructive options and has been described in various forms. In its dynamic form, chronic electrical stimulation induces modifications of the gracilis muscle to a fatigue-resistant muscle with a basal resting tone and voluntary contractions, achieving continence [15]. These techniques should be considered to achieve socially acceptable continence and improve the quality of life in selected APER patients.

In this context, we present a case of TAR using graciloplasty following APER, with particular emphasis on postoperative management and the potential role of neuroplasticity in functional adaptation of the reconstructed neosphincter.

Case report

A 36-year-old woman with a history of low rectal cancer was referred to our department for dyspareunia. Five years earlier, she had been diagnosed with a low rectal tumor presenting with functional symptoms, including vaginal flatulence, ileus, and progressive constipation. Initial CT imaging revealed a low rectal mass associated with a rectovaginal fistula. A diagnostic laparotomy confirmed the diagnosis and suggested infiltration of the posterior vaginal wall. A loop sigmoid colostomy was promptly performed to relieve acute symptoms.

Further evaluation, including rectoscopy-guided biopsies, confirmed rectum-specific squamous cell carcinoma (RSSCC). PET-CT imaging demonstrated a locally advanced tumor (T4N0M0) without distant metastasis, with confirmed invasion of the posterior vaginal wall. The patient underwent neoadjuvant chemoradiotherapy, followed 3 months later by an extended APER via open laparotomy, achieving R0 margins. Adjuvant chemotherapy with mitomycin and 5-FU (Nigro-like protocol) was subsequently administered [16]. The postoperative course was marked by a 5-year disease-free interval, after which the patient presented to our department with dyspareunia.

Clinical examination revealed marked cervicovaginal deformity and extensive scarring involving the lower two-thirds of the posterior vaginal wall, while the perineal scar was unremarkable. The patient also had a left-sided abdominal colostomy.

Although the initial complaint was discomfort during vaginal intercourse, further discussion revealed that the patient had adapted to her colostomy despite its significant impact on her sexual life and overall quality of life. We discussed surgical options to address the vaginal deformity and intravaginal scarring, including Z-plasties, with detailed explanation of expected outcomes, limitations, and potential complications. The concept of TAR was also introduced, including its functional goals, risks, and required follow-up. Given the patient’s strong motivation to restore anorectal function, we proceeded with anal sphincter reconstruction using a coloplasty combined with an innervated pedicled gracilis muscle flap to achieve functional TAR.

Surgical procedure

With the patient under general anesthesia in the lithotomy position, the procedure began with the creation of a perineal colostomy. A midline laparotomy was performed to mobilize the descending colon and take down the existing left-sided abdominal colostomy. Intra-abdominal exploration confirmed the absence of disease recurrence, although multiple adhesions were encountered and carefully released. The descending colon was then adequately mobilized to allow tension-free transposition to the perineum. A suitable perineal site was identified and prepared for the future neo-anus, without immediate maturation.

Following colonic mobilization, a pedicled gracilis flap was harvested. An incision was made along the medial aspect of the left thigh, allowing dissection and isolation of the gracilis muscle. The vascular pedicle and motor nerve were identified, preserved, and traced proximally between the adductor longus and adductor magnus muscles. The proximal tendon was maintained, while the distal tendon was divided as distally as possible at the pes anserinus (Fig. 1).

Fig. 1.

Intraoperative view (A) and scheme (B): The coloplasty is dissected and mobilized to the perineum, and a pedicled gracilis muscle flap is harvested while preserving its innervation

The gracilis muscle was then transposed to the perineum through a subcutaneous tunnel. A 360° anticlockwise wrap was designed around the transposed colon to create a dynamic neosphincter (Fig. 2). The perineal colostomy was subsequently secured using interrupted absorbable braided sutures. The procedure was completed with revision of the vaginal cavity, using Z-plasty to release scar contracture and correct the stenosis.

Fig. 2.

Intraoperative view (A) and scheme (B): The gracilis muscle flap is tunneled and looped around the neo-anal canal to create the neosphincter. The loop is fashioned in a anticlockwise configuration

Postoperatively, antibiotic prophylaxis was administered for 7 days, and prophylactic anticoagulation was maintained for 15 days. The patient was kept in a supine position with thigh adduction for 72 h to protect the reconstruction.

Results

Both surgical sites healed uneventfully within 10 days. Bowel function resumed within 72 h, and fecal continence was achieved within 2 months through manometry-assisted biofeedback and intensive pelvic floor rehabilitation, with no need for external or implantable stimulation of the neosphincter (Fig. 3).

Fig. 3.

Long-term result: The anus is successfully reconstructed, and the donor site scar is acceptable

Pelvic floor training primarily aimed to increase the resting tone of the transposed gracilis muscle and residual pelvic musculature. In parallel, manometry-assisted biofeedback sessions [17] were used to enhance neosphincter function by providing real-time feedback on muscle activity, allowing the patient to progressively develop partial voluntary control over the innervated gracilis muscle and improve awareness of intra-abdominal pressure.

In the context of TAR, where normal continence mechanisms and rectal sensation are absent [12], functional continence relies less on generating high voluntary squeeze pressures and more on controlled evacuation. Timed defecation, achieved through voluntary partial relaxation (“de-squeezing”), appears more relevant than forceful contraction, which would only be beneficial in the presence of preserved rectal sensation.

The rehabilitation protocol was initiated at 2 weeks postoperatively and included three weekly biofeedback sessions for the first 12 weeks, followed by two sessions per week until week 24. Each session lasted approximately 45 min and used high-resolution water-perfused manometry catheters to provide real-time visual feedback. In parallel, the patient followed a structured pelvic floor training program of similar duration, consisting of three daily sets of kneeling-squat contractions (10-s contraction, 20-s relaxation), combined with balloon expulsion exercises.

Routine follow-up, including anorectal manometry, was performed annually over a 5-year period to assess neo-anal function (Table 1). In physiological conditions, resting pressures generated by the internal anal sphincter typically range from 60 to 70 mmHg, while voluntary contraction results in an approximate 1.5-fold increase, reaching 120 to 170 mmHg [18]. In this patient, resting intra-anal pressure reached 90 mmHg at 1 month postoperatively, likely reflecting transient intramuscular edema, as confirmed by ultrasound imaging.

Table 1.

Early and long-term anal manometry measurements

Follow-up	1 month	2 months	6 months*	1 year*	3 years*	5 years
Maximal basal pressure (MBP)	90 mmHg	110 mmHg	110 mmHg	130 mmHg	120 mmHg	110 mmHg
Maximal squeeze pressure (MSP)	100 mmHg	130 mmHg	140 mmHg	180 mmHg	170 mmHg	170 mmHg

^*Episodes of neo-anal stricture

Voluntary squeeze pressures increased to approximately 100 mmHg within 4 weeks. These early results indicate the presence of a functional neosphincter capable of voluntary contraction, although still insufficient to ensure continence. Fecal continence was subsequently achieved within 2 months postoperatively, with squeeze pressures reaching normal values (≈130 mmHg). Resting pressures remained elevated, up to 110 mmHg, likely reflecting persistent edema and early fibrotic remodeling.

Three episodes of neo-anal stricture causing obstructive symptoms were observed, associated with progressive scar formation and contributing to persistently elevated resting intra-anal pressures throughout the 5-year follow-up period (Table 1). All episodes were successfully managed with dilation and local corticosteroid injections. The first dilation procedure was performed 6 months postoperatively, following the patient’s complaints of ineffective timed evacuation. Manometric assessment at that time demonstrated persistently elevated resting pressures (110 mmHg) and voluntary squeeze pressures of 140 mmHg. In conjunction with the clinical symptoms, these findings supported the indication for anal dilation and led to discontinuation of the biofeedback protocol. At both 1 and 3 years postoperatively, the patient presented with similar symptoms. Resting pressures measured 130 and 120 mmHg, respectively, while voluntary squeeze pressures reached 180 and 170 mmHg. Anal dilation was performed on both occasions, resulting in restoration of effective timed evacuation. At 5 years postoperatively, resting pressures have been stabilized at 110 mmHg, with voluntary squeeze pressures of 170 mmHg. Given the patient’s functional adaptation and the absence of recurrent evacuation difficulties, further dilation was not indicated. Clinical examination of the neo-anal region revealed a well-healed scar without abnormalities.

The donor site (inner thigh) scar remained aesthetically acceptable. The patient reported high satisfaction, noting an overall QoL comparable to her pre-disease state (Table 2). At the last follow-up, the patient demonstrated a Wexner score of 3/20, consistent with mild incontinence, with only occasional gas leakage and minimal lifestyle impact. FI-related quality of life remained well preserved across all domains, reflecting good functional adaptation.

Table 2.

Functional and quality of life outcomes at last follow-up

Outcome measure	Domain/parameter	Score/result	Interpretation
Wexner Incontinence Score (0–20)	Solid stool	0	Never
	Liquid stool	0	Never
	Gas	2	Sometimes
	Pad use	0	Never
	Lifestyle alteration	1	Rarely
	Total score	3/20	Mild incontinence
FIQL (quality of life assessment)	Lifestyle	Preserved	Minimal impact
	Coping/behavior	Adequate	Good adaptation
	Depression/self-perception	Normal	No psychological impairment
	Embarrassment	Minimal	Occasional concern

The Wexner Incontinence Score ranges from 0 (perfect continence) to 20 (complete incontinence). Scores of 1–5 indicate mild, 6–10 moderate, and 11–20 severe incontinence. FIQL domains assess the impact of fecal incontinence on quality of life, including lifestyle, coping behavior, emotional well-being, and social embarrassment

Discussion

Patients undergoing total anorectal resection experience a significant decline in quality of life following this curative but amputational procedure. Over the past decades, multiple approaches to TAR have been proposed; however, none has consistently emerged as a reliable alternative to permanent end colostomy [9]. While this technique may not restore normal continence due to the characteristics of the gracilis muscle replacing the anal sphincter and the disruptions of rectal anatomy, including local innervation [12, 19], TAR can significantly improve quality of life, allowing patients to regain a lifestyle closer to their pre-APER condition. This case report illustrates this potential benefit.

We reviewed current TAR techniques with a focus on functional outcomes, patient quality of life, and procedure-related morbidity. Complications can be broadly categorized into digestive, perineal, and neosphincter-related events. Digestive complications include colonic ischemia or necrosis, anastomotic obstruction, and fistula formation—particularly rectovaginal fistulas—often related to compromised vascular supply or excessive tension during colonic mobilization [9, 19]. Perineal complications are also common and include abscess formation, wound dehiscence, perineal or groin infections, and cutaneous stenosis, reflecting both the complexity of the surgical field and local tissue conditions [9].

Neosphincter-related complications remain a major concern and include gracilis muscle dehiscence, fibrosis, stenosis, and dysfunction of the reconstructed sphincter mechanism. In addition, device-related complications, such as infection, electrode failure, or exposure in dynamic graciloplasty with electrostimulation, have been widely reported and contribute significantly to overall morbidity [9, 15]. Reported complication rates remain high across techniques, ranging from 30% to over 70%, and exceeding 80% in some series when long-term functional failure and revision procedures are considered [9, 15, 20]. In our experience, both the incidence and severity of these complications appear closely related to patient selection, prior surgical history, and postoperative management. Patients with multiple previous abdominal surgeries or prior pelvic irradiation often present increased technical difficulty and a higher risk of complications. These observations are consistent with the literature and underscore the importance of careful patient selection, multidisciplinary assessment, and meticulous perioperative planning when considering TAR.

Finally, recent refinements in surgical technique have aimed to reduce donor-site morbidity associated with gracilis harvesting. Minimally invasive approaches have been described, with lower rates of wound complications—including infection, dehiscence, and postoperative pain—while preserving adequate muscle viability for reconstruction [21]. Although not used in the present case, these techniques may further improve the safety of graciloplasty and should be considered in selected patients.

A simple perineal colostomy is generally associated with suboptimal continence, which has led to the development of adjunctive procedures such as Malone appendicostomy to facilitate antegrade enema irrigation. Despite reported morbidity rates ranging from 33 to 56%, many patients achieve satisfactory quality of life through structured bowel management [9, 19, 22]. Functional outcomes remain variable, with some series reporting effective pseudocontinence, although overall patient satisfaction is inconsistent [9].

The circular smooth muscle cuff (CSMC) technique uses autologous colonic muscle to create a neosphincter. Despite reported morbidity rates ranging from 22 to 76% [9, 23], it remains a viable option, as many patients achieve functional continence postoperatively. Long-term follow-up studies indicate that up to 80% of patients reach satisfactory continence, often supported by neo-anal irrigation. QoL metrics following CSMC are similar to those achieved with dynamic graciloplasty, although morbidity and the need for additional interventions remain considerable [9].

Gluteoplasty, involving the transposition of the gluteal muscle to be reassigned as a sphincter, has garnered renewed interest, particularly with techniques that include pudendal nerve anastomosis to improve functionality [24]. This approach demonstrates promising manometric outcomes but is associated with substantial morbidity [9]. While manometry often indicates adequate pressures, a significant proportion of patients require irrigation or other continence aids. Limited studies suggest that some patients may achieve spontaneous defecation, though further research is needed to substantiate these findings.

Dynamic graciloplasty involves relatively straightforward dissection of the gracilis vascular pedicle; however, transposition of the descending colon to the perineum may be technically demanding, particularly in patients with extensive intra-abdominal adhesions. The use of external electrical stimulation or implantable pulse generators is widely described, although associated with significant morbidity [9, 15]. While this technique can achieve functional continence, postoperative complication rates—including infection and device exposure—range from 14 to 74%, with long-term morbidity reported in up to 90% of cases.

Manometric evaluation of the neosphincter varies according to the type of stimulation. External stimulation combined with biofeedback typically results in lower resting intra-anal pressures compared to physiological values but allows effective voluntary contraction. In contrast, internal stimulation is associated with higher resting tone but less efficient voluntary contraction and carries a substantial complication burden. Despite these limitations, long-term quality of life may remain acceptable in selected patients, with some achieving near-complete functional recovery [25]. Recent studies have also demonstrated the long-term durability of this approach, with acceptable continence maintained over decades, although often requiring adjunctive strategies such as irrigation and ongoing functional adaptation [26].

Challenges related to flap innervation and its functional performance as a neosphincter are well recognized. Several studies have explored the use of internal or external electrical stimulation to enhance muscle function; however, consistent restoration of full continence remains difficult to achieve. Postoperative pelvic floor training is commonly recommended, whereas the use of manometry-assisted biofeedback is less frequently reported. In our approach, the primary objective was to promote voluntary activation of the neosphincter and avoid reliance on artificial stimulation, both to encourage neuroplastic adaptation and to limit device-related morbidity, which the patient declined. Fecal continence was achieved within 2 months, supported by manometry-assisted biofeedback, which may have facilitated neuroplasticity and contributed to the development of partial voluntary control over the neosphincter.

Careful patient selection is a key determinant of success in TAR. Based on both the literature and our experience, TAR should be reserved for highly motivated patients with a strong preference to avoid permanent colostomy, stable oncologic status, and sufficient physical and cognitive capacity to participate in postoperative rehabilitation. Favorable local tissue conditions and the absence of extensive pelvic fibrosis are also important for technical feasibility. In contrast, TAR is generally contraindicated in patients with active or recurrent disease, poor general condition, severe comorbidities, or limited compliance, as these factors adversely affect both surgical outcomes and long-term functional adaptation. Overall, TAR should not be considered a standard alternative to colostomy, but rather a highly individualized reconstructive option aimed at improving quality of life in carefully selected patients [9, 19].

In conclusion, although several techniques for total anorectal reconstruction have been described, none consistently achieves optimal functional and quality-of-life outcomes. Each approach is associated with specific limitations, including significant morbidity and variable continence, often necessitating adjunctive therapies. Advances in tissue engineering offer promising perspectives for the development of more physiologically integrated continence mechanisms that may reduce the complications observed with current techniques. At present, TAR remains a complex and highly individualized decision, guided by patient preference, risk tolerance, surgical expertise, and the ability to engage in structured postoperative management.

Supplementary information

Below is the link to the electronic supplementary material.

Author contribution

N.G wrote the main manuscript text, prepared Figs. (1A, 2 A, 3) and tables (1 and 2). J.C. prepared Figs. 1B-2B. All authors reviewed the manuscript.

Funding

None.

Data availability

All data available have been included in the manuscript and the table.

Declarations

Ethical approval

Not required for this type of study.

Competing interests

The authors declare no competing interests.