Abstract
Objective:
This study aimed to demonstrate our management of rectal injury (RI) and rectourinary fistula (RUF) from radical prostatectomy (RP) and identify a possible factor that increased the chance of developing RUF.
Materials and Methods:
Between January 2011 and December 2019, a total of 14 cases of RI were retrospectively reviewed and analyzed, including preoperative, perioperative, and postoperative information.
Results:
In all 14 cases of RI, the average age at RP was 66.3 years (54–77). During the study period, 8 of 14 cases of RI occurred in our hospital, and the incidence of RI was 0.42%. RI was intraoperative recognition in 8 cases and delayed diagnosis in 6 cases. For immediate recognition, 4 of 8 cases were primarily repaired without developing RUF and did not require diverting colostomy and suprapubic cystostomy. RUF occurred in 10 cases including 4 cases of intraoperative recognition and all cases of delayed diagnosis. In a subgroup analysis of RI that occurred in our hospital, the timing for diagnosis was clinically and statistically significant difference (P = 0.029). Instantly detected RI during RP and intraoperative rectal repair resulted in no postoperative complication. Among all 10 cases of RUF, 5 cases were successfully repaired by modified York–Mason procedure with dartos tissue flap interposition. No major complications were reported.
Conclusions:
Incidence of RI was 0.42% and intraoperative recognition of RI was a key to prevent the development of RUF. Modified York–Mason procedure with dartos tissue flap interposition was an effective treatment for RUF.
Keywords: Prostate cancer, radical prostatectomy, rectal injury, rectourinary fistula
INTRODUCTION
Radical prostatectomy (RP) is another standard treatment for prostate cancer which can be performed in open, laparoscopic, or robotic techniques. One of the serious complications is rectal injury (RI), and population-based incidence was 0.5%.[1] Some studies could not identify risk factors associated with RI.[2] Some studies demonstrated a risk reduction of RI during RP, including obesity, high-volume center, and robotic technique.[1] Consequently, only 12% of RI, which was performed primary repair, developed rectourinary fistula (RUF), and the overall incidence was <0.1%.[3] Moreover, RUF may occur from delayed recognition of RI.[2] Factors associated with the development of RUF were the extent of rectal laceration, prior radiation, intraoperative signs of rectal infiltration,[3] or gross fecal spillage.[4] On the other hand, intraoperative recognition and primary repair with vascularized tissue interposition prevented the development of RUF.[5] Treatments consisted of conservative management with or without diverting colostomy (DC) and surgical intervention.[3] Majority of series reported a low rate of spontaneous closure.[6] Suitable RUF case for spontaneous closure with or without DC was new occurred fistula, diameter <2 cm, no prior pelvic radiation, and normal nutritional status.[6,7] Due to the low incidence and varieties of management, this study aims to demonstrate our management of RI and RUF and identify a possible factor that is associated with the development of RUF.
MATERIALS AND METHODS
After approval of our institutional review board, a total of 14 cases of RI and RUF from RP between January 2011 and December 2019 were reviewed. Data including age at RP, pathological tumor (pT) stage, RP techniques, timing of diagnosed RI (immediate or delayed recognition), intraoperative findings during RP including documented severe adhesion and unsecured vesicourethral anastomosis, perioperative management of RI, DC, suprapubic cystostomy (SPC), adverse features, and adjuvant treatment for prostate cancer as well as management of RUF were collected and analyzed. Immediate recognition was defined as intraoperatively diagnosed RI during RP as well as delayed recognition was defined as no intraoperatively identified or mentioned RI. Unsecured vesicourethral anastomosis was defined by demonstrating leakage during intraoperative bladder irrigation or urine leakage from drain during the postoperative period. In addition, DC and SPC were performed either during RP or postoperative period. Adverse features included detectable prostate-specific antigen (PSA) or positive surgical margin or seminal vesicle invasion or extraprostatic extension. To identify a factor associated with the development of RUF, the patients were divided into Group 1 – RI with RUF and Group 2 – RI without RUF.
Statistical analysis
Mean and standard deviation were used to describe quantitative data, while number and percentage were used to describe qualitative data. To compare data between RI patients with and without RUF (Group 1 and 2), an independent sample t-test was used for quantitative data, while Pearson Chi-square or Fisher’s exact test was used for qualitative data. Data were collected and analyzed using SPSS Inc. Released 2009. PASW Statistics for Windows, Version 18.0. Chicago, IL, USA: SPSS Inc. All tests were two sided and P < 0.05 was considered to be statistical significance.
RESULTS
Between January 2011 and December 2019, there were 14 cases of RI and RUF, 8 cases were performed RP in our hospital and 6 cases were performed RP from other hospitals. For 9 years, 1905 cases were performed RP in our hospital so that the incidence of RI was 0.42%. The mean age at RP was 66.3 years (54–77). pT staging was missed in 3 cases and 8 of 11 cases (57.1%) were locally advanced diseases (pT3 or pT4). Majority of cases (71.4%) were performed laparoscopic techniques. RI was immediately diagnosed in 8 cases (57.1%) as well as 6 cases (42.9%) were delayed [Table 1].
Table 1.
Demographic data
| Parameters | n (%) |
|---|---|
| Mean age (range) (years) | 66.3 (54-77) |
| Hospital performed RP | |
| Our hospital | 8 (57.1) |
| Other hospitals | 6 (42.9) |
| pT staging | |
| pT2 | 3 (21.4) |
| pT3 | 5 (35.7) |
| pT4 | 3 (21.4) |
| Missing | 3 (21.4) |
| RP technique | |
| Open | 3 (21.4) |
| Laparoscopic | 10 (71.4) |
| Robotic | 1 (7.1) |
| Timing of diagnosed RI | |
| Immediate | 8 (57.1) |
| Delayed | 6 (42.9) |
| Intraoperative findings during RP | |
| Unsecured vesicourethral anastomosis* | 1 (7.1) |
| Documented severe adhesion | 7 (50.0) |
| Developed RUF | 10 (71.4) |
| DC | 10 (71.4) |
| SPC | 9 (64.3) |
| Having both DC and SPC | 7 (50.0) |
*Incomplete information due to retrospective chart review. pT: Pathological tumor, RP: Radical prostatectomy, RI: Rectal injury, RUF: Rectourinary fistula, DC: Diverting colostomy, SPC: Suprapubic cystostomy
In immediate recognition, 3 of 8 cases (37.5%) were documented severe adhesion around the prostate, especially posterior, and no case had unsecured vesicourethral anastomosis. All cases were primarily repaired. DC and SPC were performed in 5 cases (62.5%) and 4 cases (50.0%), respectively. Three cases were performed both DC and SPC. A total of 2 of 8 (25.0%) cases were successfully repaired without developing RUF and did not require DC and SPC.
In delayed recognition, 4 of 6 cases (66.7%) were documented severe adhesion and only 1 case was reported unsecured vesicourethral anastomosis. Four cases were performed both DC and SPC and 1 case was performed only DC. The remaining case was performed only SPC and finally spontaneous closure.
Overall, 10 cases (71.4%) developed RUF including 4 cases (50.0%) of immediate recognition and 6 cases (100.0%) of delayed recognition. Six cases (60%) were successfully repaired by modified York–Mason procedure with dartos tissue interposition [Figure 1] and 1 case was spontaneous closure. Three cases did not repair due to prostate cancer progression and were lost to follow-up. No major complications including rectal stenosis, fecal incontinence, or urinary incontinence required surgical correction were reported after repair. Reversal colostomy was performed within 3–6 months after successful closure RUF.
Figure 1.
Surgical step of Modified York–Mason procedure with dartos tissue flap interposition. (a) After the midline, incision was made from tip of coccyx to anus, intergluteal cleft was opened layer by layer. Posterior rectal was clearly dissected from the levator plate, (b) Fistula opening was identified (in the circle), (c) Foley catheter No. 12 Fr. was inserted via the fistula, and balloon was inflated to pull the fistula during dissection of the anterior rectal wall, (d) After the bladder wall defect was repaired with interrupted absorbable suture No. 2-0, dartos tissue flap was dissected and passed through the created tunnel between anterior rectal wall and urethra, (e) After the bladder defect was covered by the dartos tissue flap, proximal anterior rectal wall flap was pulled and covered the defect, (f) Posterior rectal wall was approximately closed by starting from the mucocutaneous junction and repairing of anal sphincter
To identify a factor associated with the development of RUF, the patients were divided into Group 1 – RI with RUF and Group 2 – RI without RUF. On analysis, RP technique and DC demonstrated a statistically significant difference [Table 2].
Table 2.
Patient’s characteristics of all rectal injuries with or without rectourinary fistula
| Characteristics | Group 1 RI with RUF (n=10), n (%) | Group 2 RI without RUF (n=4), n (%) | P |
|---|---|---|---|
| Age at RP (years), mean±SD | 66.8±6.0 | 65.0±4.5 | 0.600 |
| Hospital performed RP | |||
| Our hospital | 4 (40.0) | 4 (100.0) | 0.085 |
| Other hospitals | 6 (60.0) | 0 | |
| pT staging* | |||
| T2 | 2 (28.6) | 1 (25.0) | 1.000 |
| T3 and T4 or node positive | 5 (71.4) | 3 (75.0) | |
| RP technique | |||
| Open | 0 | 2 (50.0) | 0.011 |
| Laparoscopic | 10 (100.0) | 1 (25.0) | |
| Robotic | 0 | 1 (50.0) | |
| Timing of diagnosed RI | |||
| Immediate | 4 (40.0) | 4 (100.0) | 0.085 |
| Delayed | 6 (60.0) | 0 | |
| DC | 9 (90.0) | 1 (50.0) | 0.041 |
| SPC | 7 (70.0) | 2 (50.0) | 0.580 |
| Post-RP treatments | |||
| Adjuvant radiation therapy | 1 (10.0) | 2 (50.0) | 0.176 |
| Adjuvant hormonal therapy | 4 (40.0) | 3 (75.0) | 0.559 |
| Adjuvant chemotherapy | 1 (10.0) | 1 (25.0) | 0.505 |
| Adverse features§ | 4 (40.0) | 3 (75.0) | 0.559 |
*Missing data 3 cases in Group 1 (n=7), §Adverse features included detectable PSA or positive surgical margin or seminal vesicle invasion or extraprostatic extension. pT: Pathological tumor, RP: Radical prostatectomy, RI: Rectal injury, RUF: Rectourinary fistula, DC: Diverting colostomy, SPC: Suprapubic cystostomy, SD: Standard deviation, PSA: Prostate-specific antigen
All RI cases from other institutes were referred to our hospital because of RUF so that only 8 RI cases that occurred in our institute were selected and analyzed. In subgroup analysis, 4 of 8 cases (50%) were delayed diagnosed intraoperative RI and all of these cases developed RUF, while in the remaining 4 cases, the surgeon can detect RI during RP and immediate repair rectum, resulting no postoperative complication. Timing for diagnosis RI was a clinically different and statistically significant difference (P = 0.029). Between these two groups, regardless of age, pT staging (T), technical of RP, DC, and SPC, there was no statistically significant difference [Table 3].
Table 3.
Subgroup analysis of rectal injury with or without rectourinary fistula in our hospital
| Characteristics | Group 1 RI with RUF (n=4), n (%) | Group 2 RI without RUF (n=4), n (%) | P |
|---|---|---|---|
| Age at RP (years), mean±SD | 71.3±4.3 | 65±4.5 | 0.094 |
| pT staging* | |||
| T2 | 1 (25.0) | 1 (25.0) | 1.00 |
| T3 and T4 or node positive | 3 (75.0) | 3 (75.0) | |
| RP technique | |||
| Open | 0 | 2 (50.0) | 0.091 |
| Laparoscopic | 4 (100.0) | 1 (25.0) | |
| Robotic | 0 | 1 (25.0) | |
| Timing of diagnosed RI | |||
| Immediate | 0 | 4 (100.0) | 0.029 |
| Delayed | 4 (100.0) | 0 | |
| DC | 3 (75.0) | 1 (25.0) | 0.486 |
| SPC | 3 (75.0) | 2 (50.0) | 1.00 |
pT: Pathological tumor, RP: Radical prostatectomy, RI: Rectal injury, RUF: Rectourinary fistula, DC: Diverting colostomy, SPC: Suprapubic cystostomy, SD: Standard deviation
DISCUSSION
RI is a rare complication and usually occurs in RP for locally advanced prostate cancer or salvage RP after radiation therapy.[3,8] Many studies reported that the incidence was <1%.[1-3] Incidence of our study was 0.42%. Not all RI developed RUF,[3] and it probably depended on some factors and RI management.
According to a very low incidence of RUF, it was less likely to identify a robust factor associated with the development of RUF. In the past studies, many factors had been reported including extent of rectal laceration, prior radiation, intraoperative signs of rectal infiltration,[3] gross fecal spillage,[4] and delayed recognition of RI.[2] On the other hand, intraoperative recognition and primary repair with vascularized tissue interposition prevented the development of RUF.[5] In this study, we noticed that all delayed recognition developed RUF as well as half of immediate recognition developed RUF. However, the timing of diagnosed RI was more likely to be a factor and reach a statistically significant correlation, although a small number of the population participated in this study due to the very low incidence of RI.
From a population-based study, RI incidence was higher in open technique than laparoscopic or robotic technique. Moreover, robotic techniques reduced the risk of RI.[1] In this study, majority of RI occurred during laparoscopic technique which was gained popularity in our country, as well as robotic technique was available in few hospitals.
DC might be required in some cases and it depends on the patient’s condition and surgeon preference. One study recommended in cases with infiltration of the rectum, prior prostate surgery, large rectal defect, or salvage RP.[3] In this study, DC was performed in a patient who had intraoperative fecal contamination, postoperative urinary tract infection, waiting for spontaneous closure after development of RUF or surgeon preference.
Treatment of RUF consisted of conservative treatment and surgical intervention. From the systematic review, conservative treatment has a lower success rate than surgical invention. There were four surgical approaches for repairing RUF including trans-perineal, transsphincteric, transabdominal, and transanal approaches, and the success rate was 90%, 88%, 87%, and 59%, respectively.[6] In this study, only one RUF was spontaneous closure after conservative treatment. Six cases were repaired by transsphincteric approach, described as modified York–Mason procedure,[9] with dartos tissue interposition, and all cases were success. No major complications including rectal stenosis, fecal incontinence, or urinary incontinence that required surgical correction were reported.
The limitation of this study was a retrospective design in which some important information was incomplete. The incidence of RI and RUF was very low so that it was difficult to make a robust conclusion that which factor related to the development of RUF.
CONCLUSIONS
In summary, the incidence of RI from RP was 0.42%, and majority of cases occurred from the laparoscopic technique. Development of RUF after RI depended on many factors, one of which may include timing for diagnosis RI. To prevent fecal contamination and infection, DC may be required. Modified York–Mason procedure with dartos flap interposition was an effective treatment for RUF.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
REFERENCES
- 1.Barashi NS, Pearce SM, Cohen AJ, Pariser JJ, Packiam VT, Eggener SE. Incidence, risk factors, and outcomes for rectal injury during radical prostatectomy: A population-based study. Eur Urol Oncol. 2018;1:501–6. doi: 10.1016/j.euo.2018.06.001. [DOI] [PubMed] [Google Scholar]
- 2.Wedmid A, Mendoza P, Sharma S, Hastings RL, Monahan KP, Walicki M, et al. Rectal injury during robot-assisted radical prostatectomy:Incidence and management. J Urol. 2011;186:1928–33. doi: 10.1016/j.juro.2011.07.004. [DOI] [PubMed] [Google Scholar]
- 3.Mandel P, Linnemannstons A, Chun F, Schlomm T, Pompe R, Budaus L, et al. Incidence, risk factors, management, and complications of rectal injuries during radical prostatectomy. Eur Urol Focus. 2018;4:554–7. doi: 10.1016/j.euf.2017.01.008. [DOI] [PubMed] [Google Scholar]
- 4.Kheterpal E, Bhandari A, Siddiqui S, Pokala N, Peabody J, Menon M. Management of rectal injury during robotic radical prostatectomy. Urology. 2011;77:976–9. doi: 10.1016/j.urology.2010.11.045. [DOI] [PubMed] [Google Scholar]
- 5.Roberts WB, Tseng K, Walsh PC, Han M. Critical appraisal of management of rectal injury during radical prostatectomy. Urology. 2010;76:1088–91. doi: 10.1016/j.urology.2010.03.054. [DOI] [PubMed] [Google Scholar]
- 6.Hechenbleikner EM, Buckley JC, Wick EC. Acquired rectourethral fistulas in adults: A systematic review of surgical repair techniques and outcomes. Dis Colon Rectum. 2013;56:374–83. doi: 10.1097/DCR.0b013e318274dc87. [DOI] [PubMed] [Google Scholar]
- 7.Thomas C, Jones J, Jäger W, Hampel C, Thüroff JW, Gillitzer R. Incidence, clinical symptoms and management of rectourethral fistulas after radical prostatectomy. J Urol. 2010;183:608–12. doi: 10.1016/j.juro.2009.10.020. [DOI] [PubMed] [Google Scholar]
- 8.Calleris G, Marra G, Dalmasso E, Falcone M, Karnes RJ, Morlacco A, et al. Is it worth to perform salvage radical prostatectomy for radio-recurrent prostate cancer?A literature review. World J Urol. 2019;37:1469–83. doi: 10.1007/s00345-019-02749-z. [DOI] [PubMed] [Google Scholar]
- 9.Stephenson RA, Middleton RG. Repair of rectourinary fistulas using a posterior sagittal transanal transrectal (modified York-Mason) approach: An update. J Urol. 1996;155:1989–91. [PubMed] [Google Scholar]