Complications of laparoscopic radical prostatectomy–A single institute experience

Abstract

Laparoscopic radical prostatectomy (LRP) is a demanding procedure that requires surgical skill, a long learning curve and significant laparoscopic expertise. We report herein our initial experience with 41 laparoscopic radical prostatectomies to assess all perioperative complications in our initial 5‐year experience. We reviewed retrospectively the case records of 41 patients who underwent laparoscopic radical prostatectomy for clinically localized prostate cancer from 2005 to 2010 at our institution. All abnormal symptoms were developed during the 30 days after the operation. Complications were noted as well as their severity according to the Clavien grading system. No conversion to open surgery was necessary in all cases. The median postoperative hospital stay was 10 days (range 9–16). The median duration of bladder catheterization was 8 days (range 7–35). There was no case of multiple‐organ dysfunction or death. Minor complications constituted 82.6% of all complications. The incidence of severe complications was 17.4%. LRP displayed some advantages over open surgery in the perioperative period. Our complication rates were significantly higher than those indicated in the series previously reported. We found that most complications occurred in the first 30 patients who underwent LRP. We suggest that the surgeons should pay more attention to the difficulties of the procedure.

Introduction

In an effort to further decrease the morbidity of radical retropubic prostatectomy (RRP), Schuessler et al. reported the first laparoscopic radical prostatectomy (LRP) in 1992, and in 1997 they reported an initial series of 9 operations [1]. LRP is a demanding procedure that requires surgical skill, a long learning curve and significant laparoscopic expertise. Since November 2005, laparoscopic radical prostatectomy has been performed at our institution. Considering the current interest in this technique, we have chosen to share our difficulties and complications with this procedure, hopefully to reduce morbidity for the patients of other interested urological teams. Thus, herein we report our initial experience with 41 laparoscopic radical prostatectomies, to assess all intraoperative and postoperative complications in our initial 5‐year experience. Complications were classified according to the modified Clavien grading system [2]. In addition, management of the complications was also analyzed.

Patients and methods

We retrospectively reviewed the case records of 41 patients who underwent laparoscopic radical prostatectomy for clinically localized prostate cancer from November 2005 to March 2010 at our institution. A single surgeon performed all of the procedures. Complications were classified in our database according to the modified Clavien grading system.

Only one patient underwent extraperitoneal LRP and the others underwent transperitoneal LRP. Thirty‐three patients underwent a 5‐port LRP and eight patients underwent a 4‐port technique. Bilateral pelvic lymph node dissection (PLND) was performed in 36 patients. Dissection was performed using scissors and harmonic scalpel. The anastomosis was performed using a continuous running suture with two 2‐0 Vicryl sutures tied together. An 18‐Fr Foley catheter was inserted. The specimen was then removed through the primary trocar incision, and a Jackson–Pratt (JP) drain was positioned in the pelvic gutter.

Cystourethrography was performed on the postoperative day 7. The urethral catheter would be removed if there were no contrast extravasation noted in the cystourethrography.

All abnormal symptoms that developed after the operation and required active medical or surgical intervention were documented and considered a complication. Time to onset of the complication was noted as well as its severity according to the Clavien grading system.

Results

No conversion to open surgery was necessary in all cases. Mean serum prostate specific antigen at diagnosis was 9.13 ng/ml (range 5.85–59). The pathological tumor stage, according to the TNM 1997 classification system, revealed 1 pT1 (2.4%), 30 pT2 (73.2%), 10 pT3 (24.3%). The rate of positive margins in stage pT2 was 12.2%, and pT3 was 12.2%. The overall positive margins rate was 24.3%. Low grade Gleason ≤ <6, intermediate grade Gleason = 7 and high grade Gleason ≥ >7 of the surgical specimen was identified in 16 (39%), 11 (26.8%) and 14 (34.1%) cases respectively. Patient characteristics are presented in Table 1.

Table 1.

Demographic and pathologic features of n = 41 consecutive laparoscopic radial prostatectomy.

Patient characteristics
Median age [y] (range)	64 (52–72)
Median PSA level [ng/mL] (range)	9.13 (5.85–59)
Median prostate weight [g] (range)	38 (22–97)
Pathology stage (%)
PT1	1 (2.4%)
PT2	30 (73.2%)
PT3	10 (24.3%)
Positive margin (%)
PT2	5 (12.2%)
PT3	5 (12.2%)
Overall	10 (24.3%)
Pathology Gleason score (%)
≤6	16 (39%)
7	11 (26.8%)
≥8	14 (34.1%)

Mean blood loss was 200 ml (range 30–1500). With regard to transfusions, 93.4% of patients were not transfused. Median operative time, including lymph node dissection, was 4.9 hours (range 3.2–9.6) overall. The median postoperative hospital stay was 10 days (range 9–16). The median duration of bladder catheterization was 8 days (range 7–35). The rates of one side, both sides and none in neurovascular bundle preservation were 51.2%, 31.7% and 17%, respectively. Only three patients (9.1%) had lymph node metastasis. The peri‐operation parameters are present in Table 2.

Table 2.

Perioperation parameters in laparoscopic radial prostatectomy.

Perioperation parameters
Median blood loss [mL] (range)	200 (30–1500)
Blood transfusion [n] (%)	4 (9.6%)
Median operative time [h] (range)	4.9 (3.2–9.6)
Median catheterization [d] (range)	8 (7–35)
Median hospital stay [d] (range)	10 (9–16)
Neurovascular bundle preservation [n] (%)
One side	21 (51.2%)
Both sides	13 (31.7%)
No	7 (17.0%)
Lymph node positive [n] (%)
Yes	3 (9.1%)
No	33 (90.9%)

Post‐operation parameters were also collected. Median PSA nadir = 0.1 (range < 0.01–1.19). Median time to continence (pad ≤ 1) was 26 days (range 0–196). The biochemical failure rate (PSA ≥ 0.2 ng/ml) after LRP was 21.9%. Only five patients had successful sexual activity.

We observed 13 complications in 41 patients (31.7%). Classifications of the complications are according to the Clavien grading system. There was no case of multiple‐organ dysfunction or death (Grades 4 and 5). Minor complications (Grades 1 and 2) constituted 82.6% of all complications. The incidence of severe complications (Grade 3) was 17.4% (Table 3).

Table 3.

Complications after laparoscopic radical prostatectomy in 41 patients.

Complication	Number	Incidence (%)	Clavien grade
Rectal injury	2	4.9	3
Ureter injury	1	2.4	3
Bladder neck contracture	1	2.4	3
Ileus	3	7.3	2
Deep vein thrombosis	2	4.9	2
Upper gastrointestinal bleeding	2	4.9	2
Wound infection	1	2.4	2
Blood transfusion	3	7.3	2
Anastomosis leakage	7	17.1	1
Incision hernia	1	2.4	1

The most common Grade 1 complications were anastomotic leaks (17.1%), detected on cystourethrography. They had prolonged catheterization due to persistent anastomotic leakage.

Postoperative ileus (7.3%) and blood transfusion (7.3%) were the two most common Grade 2 complications. These patients with ileus were managed with maintenance of intravenous fluid, temporary restriction of oral intake and nasogastric aspiration with resolution of the symptoms. Two patients (4.9%) presented with deep vein thrombosis (DVT) and were managed with intravenous heparin follow by oral warfarin. Two patients (4.9%) presented with upper gastrointestinal bleeding and were managed with esophagogastroduodenoscopy followed by proton pump inhibitor use. One patient (2.4%) presented with wound infection and was managed with intravenous antibiotics and local wound care. The blood transfusion rate was 7.3% (three patients).

Grade 3 complications included two (4.9%) rectal injuries, one ureteral injury (2.4%) and one bladder neck contracture (2.4%). Bladder neck contracture was managed with bougie sounding, with resolution of the symptoms in the case without additional procedures. Rectal injury was identified during operation in one patient. The injury site was sutured immediately in two layers (mucosa and muscle) with 2‐0 catgut, 2‐0 silk and then covered with tissue glue. The integrity of the repair was then checked by filling the rectum with air through an enema tube to distend the lumen and look for air bubbles in the fluid‐filled pelvic cavity. However, urethrorectal fistula was found on postoperative day 5 after many episodes of diarrhea. He received reoperation with transanal repair, temporary intestinal diversion by colostomy and urinary diversion with cystostomy. Although the primary repair failed initially, this fistula had remission spontaneously in the postoperative 18 months. In the other patient, rectal injury had not been found intraoperatively. Diagnosis was made after the development of postoperative urethrorectal fistula. This patient received a second operation of transperineal buccal mucosa repair with colostomy and cystostomy. However, the repair failed after a large quantity of urine was noted in the colostomy bag 7 days later. He received another repair in another hospital 6 months later but this was in vain. To date, the colostomy is still maintained 28 months after laparoscopic radical prostatectomy.

Ureteral complications developed in one patient. Flank pain was noted after discharge. Computed tomography revealed left hydronephrosis with total obstruction at middle third ureter due to the ureter being clamped by an Endoclip. The complication was managed with reconstruction by Boari flap. In this case, there was no further hydronephrosis or other complications after this second intervention.

Discussion

Laparoscopic radical prostatectomy has been developed to reduce complications and patient hospitalization and is emerging as an accepted treatment modality at many centers worldwide. Transperitoneal laparoscopic radical prostatectomy was first performed by Schuessler et al. [3] in 1992. In 1998 Guillonneau et al. [4] reported an initial series of 28 cases with a standardized technique.

Extraperitoneal laparoscopic radical prostatectomy was described by Raboy et al. [5] in 1997, but it was standardized by Bollens et al. [6] in 2000. After the experience of these authors, we began to perform laparoscopic radical prostatectomy with a transperitoneal approach for most patients, and only one patient has received an extraperitoneal approach.

We have not observed any intraoperative or postoperative deaths, which are now a rare event since the mortality of this operation is estimated to be less than 1.5% [[7], [8], [9]]. The various published series report a very low (0–4.4%) conversion rate [[10], [11], [12]]. No conversion to open surgery was necessary in our cases. This provides strong evidence that the laparoscopic approach is feasible.

Decreasing intraoperative bleeding is a constant objective of surgery. The blood loss accounted for the transfusion rate of more than 9% in our series. Guillonneau et al. [13] reported a transfusion rate of less than 5%. The slightly increased transfusion rate during the initial operations was probably related to the surgeon's learning curve. Certain authors [12] have also reported a longer operating time and a higher transfusion rate.

Preservation of sexual function in patients undergoing radical prostatectomy has a significant impact on quality of life after operation [14]. Our data are based on 41 patients who underwent laparoscopic radical prostatectomy with unilateral and bilateral nerve preservation in 21 and 13 of them, respectively, derived after discussion with the patients. In patients with T1 or T2, the bilateral or unilateral nerve preservations were performed. However, all men with stage T3 who were at higher risk for positive margins suggested bilateral neurovascular bundle excision. Of these patients, only five (12%) had intercourse with sildenafil or tadalafil 1–2 times per week. Katz et al. reported the overall rate of sexual function after laparoscopic radical prostatectomy with nerve preservation to be about 53.8% [15]. The low rate of successful sexual activity in our patients could be related to spousal problems and old age. Furthermore, in comparison with men from ‘Western’ regions, the responses from Asian subjects indicate that they are more sexually conservative and less sexually active. It is still controversial whether this procedure compromised surgical margin status (SM). The incidence of positive SMs has been reported to range from 2.5% to 34% in nerve‐sparing radical prostatectomy based on various criteria [16]. In our patients, positive surgical margins occurred at the apex of the prostate and bladder neck.

The most frequent early postoperative complications are anastomotic leakage. Bladder catheter removal is based on surgeon subjective analysis of the anastomosis. When the anastomosis appears with no contrast extravasation on cystography after filling the bladder with 120–180 mL contrast, the catheter was removed on postoperative day 7. Usually, anastomotic leakage is easily managed and the catheter remains in place. The anastomotic leakage and bladder neck contracture are usually related to inexperience, incorrect suture layer, or being without water tie.

The incidence of deep venous thrombosis (DVT) or pulmonary embolism in radical prostatectomy series is between 2% and 3.1% [[8], [17], [18]]. Combining pelvic with laparoscopic surgery increases the risk of thromboembolism. The DVT in our patient may have been contributed to by the lithotomy position, obesity in patients and increasing operation time. Therefore, prevention is based on antithrombotic prophylaxis, compression stockings and early mobilization.

The incidence of rectal injury in laparoscopic prostatectomy is between 0.6% and 1.4% [[7], [12], [13]]. Using the water test can help to identify the rectal wall but cannot ensure risk‐free surgery, because in our study one rectal injury occurred despite no bubble being present during the water test. Some authors confirm the observations that rectal injury can be repaired primarily without colostomy even with the absence of complete mechanical or antibiotic bowel preparation. However, in our experience, one rectal injury was identified and immediately sutured laparoscopically. Two layers (mucosa and musculosa) with tissue glue without colostomy were performed but failed. This patient required reoperation with colostomy and cystostomy. Therefore, we have noted the importance of complete preoperative bowel preparation if rectal closure without colostomy is considered. The other patient was treated with transperineal buccal mucosa repair. However, in the majority of studies, management of the rectal injury itself remains debatable in regard to interposition of healthy tissue in the rectal repair.

In our experience, the risks of complication of these injuries occurred at the end of the procedure during dissection of the apex of the prostate gland. Furthermore, it is noteworthy that our rectal injuries occurred by thermo injury, large volume gland with previous multiple biopsy [[19], [20]] and high stage. Rectal injury during laparoscopic radical prostatectomy can lead to severe postoperative complications. The best treatment is injury prevention with careful dissection of the posterior prostatic planes, particularly the apex [21].

Ureteral injuries are rare. The incidence of ureteral complications in laparoscopic prostatectomy is between 0.3% and 0.7% [[13], [22]]. In our experience, this complication occurred in one patient (2.4%). In this case, the ureter was clamped by an Endoclip after aggressive lymph node dissection. When ureteral injury was not identified during the operation, the diagnosis was suggested by persistent urine leakage, secondary urinary ascites, or hydronephrosis after laparoscopic radical prostatectomy.

From our results, LRP showed some advantages over open surgery in the perioperative period, such as reduced blood loss and a greater chance of safe early catheter removal. However, our complication rates were significantly higher than those indicated in previously reported series. In the study, we demonstrated the overall morbidity associated with laparoscopic radical prostatectomy as 30.7% of patients had complications. With regard to the impact of the number of LRP performed on the rate of complications, we observed a very interesting trend. We found that most complications occurred in the first 30 patients who underwent LRP. Therefore, the numbers of LRP performed by the surgeons were independent predictors of the occurrence of complications. We therefore suggest that surgeons should pay more attention to the procedure, especially in initial cases.

Laparoscopic radical prostatectomy is a technically demanding, but feasible, treatment option for localized prostate cancer. Our series demonstrated that in the hands of an experienced urological surgeon, laparoscopic radical prostatectomy is associated with minimal intraoperative and postoperative morbidity. In our series, a number of complications, which were sometimes serious, may have been avoided by better knowledge of prevention. In this study, we also demonstrated how to manage such complications. Due to growing interest in the laparoscopic approach to radical prostatectomy, it seemed useful to share this experience with other urological teams so that they can apply this technique in their patients with the lowest possible morbidity.