{)022-534? /92/1473-·0883$03,00/0 Vol. 147, 883-887, March 1992
TrtE JOURNAL OF UROLOGY
Printed in U.S.A.
Copyright© 1992 by AMERICAN UROLOGICAL AS30CIATJ:ON, INC.
ical Treatment RADICAL RETROPUBIC PROSTATECTOMY: MORBIDITY AND QUALITY OF LIFE, EXPERIENCE WITH 620 CONSECUTIVE CASES PIERRE LEANDRI, GEORGES ROSSIGNOL, JEAN-ROMAIN GAUTIER AND JACOB RAMON From the Department of Urology, Saint-Jean Languedoc-Cerou, Toulouse, Cedex, France
ABSTRACT
We describe our experience and complications of radical retropubic prostatectomy. From March 1983 through December 1990, 620 consecutive patients have undergone an anatomical radical retropubic prostatectomy for the treatment of prostatic carcinoma. The surgical technique we used is described. In 167 patients the procedure included preservation of the neurovascular bundles. There were no modifications in the surgical technique during this period. There were no operative deaths. Mean operating time was reduced from 3 hours in the first 100 patients to 1.5 hours in the last 220 patients. The average blood loss was reduced remarkably as well. There were only 3 cases of rectal injury, which were closed primarily and healed completely. One patient died of acute myocardial infarction 12 days after an uneventful operation. This patient accounted for the only perioperative death in our experience. Early complications occurred in 43 patients (6.9%), including only 2 cases (0.3%) of anastomotic urinary leakage. The late complication rate, excluding incontinence and impotence, was 1.3%. No patient was totally incontinent. Among the patients who were followed for 1 year or longer 95% achieved complete urinary control and 5% experienced stress urinary incontinence. Preservation of sexual function in patients who underwent a nerve-sparing operation was achieved in 71 %. Our results indicate that radical retropubic prostatectomy can be performed with low morbidity and without affecting the quality of life in the majority of patients. KEY WORDS: prostatic neoplasms, prostatectomy, morbidity
Radical prostatectomy is considered by many to be the treatment of choice for localized carcinoma of the prostate. In France, as in most of Europe, there had been a conservative approach to prostate cancer. Radiation therapy and endocrine therapy had been the predominant alternative in prostate cancer until the last decade. Radical prostatectomy as definitive treatment for prostate cancer was reinstated in France in the early 1980s as a result of significant improvements in the surgical technique, which had been refined to preserve the nerves essential for potency. The technique associated with nerve-sparing radical prostatectomy as proposed by Walsh and Donker has improved the likelihood of continence especially and certainly potency over previous techniques. 1 With a decline in these complications increasing numbers of patients and physicians began to select this surgical method of disease elimination. Reduced patient morbidity and broadened surgical experience also have increased the use of radical prostatectomy in selected patients with locally advanced disease or with limited regional lymph node metastasis. 2- 4 Since 1983 we have performed radical retropubic prostatectomy on all surgical candidates with clinical stages A, B and C prostate cancer. We describe our experience with 620 consecutive patients, and detail our surgical technique, pathological findings, complications and long-term surgical results. MATERIALS AND METHODS
Patient population. A total of 620 consecutive patients underwent radical retropubic prostatectomy for treatment of prostatic carcinoma between March 1983 and December 1990 at our medical institute. Patient age ranged from 46 to 84 years, with a mean age of 68 years. Patients were considered candiAccepted for publication September 27, 1991. 883
dates for radical prostatectomy if they had a biopsy proved prostate tumor without evidence of metastatic disease. Suitable candidates underwent preoperative evaluation, including cystoscopy, serum acid phosphatase determination, radioisotope bone scan, chest films and since January 1988 serum prostate specific antigen (PSA) levels. The upper urinary tracts were evaluated with excretory urography or renal ultrasonography. Transrectal ultrasonography of the prostate was performed as an adjunct to staging in the majority of the patients. Absence of metastasis was established in all patients by negative results from preoperative radionuclide bone scanning, and radiological and biochemical studies. Patients with palpable or ultrasonic evidence of extension of the tumor beyond the capsule or into the seminal vesicles were also considered by us as candidates for radical surgery and, if tumor extension was confirmed pathologically, the patients were further treated with external beam irradiation or androgen ablation. Of the patients 153 (24%) had undergone prior open prostatectomy or transurethral resection of the prostate for benign prostatic hypertrophy, including 38 (6%) who had undergone these procedures more than 2 years before the diagnosis of prostate cancer. In 505 patients (82%) the tumor was confirmed by perinea! prostate biopsy. Following transperineal needle biopsy radical prostatectomy was deferred for 6 to 12 weeks (mean 9 weeks). Following transurethral resection or open prostatectomy surgery was deferred for 3 months. Based upon the preoperative evaluation clinical stage was assigned as follows: Al-tumor discovered by prostatectomy for benign disease and involving less than 5% of the resected tissue, and Gleason score of less than 8, A2-cancer involving more than 5% of the resected tissue or Gleason score of 8 or greater, Bl-palpable tumor confined to the prostate less than 1.5 cm. in size or involving 25% or less of 1 lobe, B2-palpable
884
LEANDRI AND ASSOCIATES
tumor involving more than 25% of 1 lobe but confined to the gland, Cl-minimal palpable extracapsular extension and C2extension of tumor into bladder neck and/or seminal vesicles. Of the patients 29 had clinical stage Al, 86 stage A2, 162 stage B, 210 stage B2, 99 stage Cl and 34 stage C2. Surgical technique. The operative technique, which is a modification of the neuroanatomical approach originated by Walsh and Donker, 1 has been described previously. 5 A modified pelvic lymph node dissection is performed with removal of the lymphatics overlying the obturator fossa and the internal iliac vessels below the bifurcation of the common iliac artery. The lymphatics overlying the external iliac artery are not resected routinely. The lymph nodes are sent for frozen section evaluation and, in the event of microscopic nodal involvement, an immediate adjuvant orchiectomy is performed. The retroperifoneal fat is teased awa.y to expose clearly the endopelvic fascia. With a scalpel a 1 cm. incision is made in the fascia at its reflection over the pelvic side wall. The incision of the fascia is then extended bluntly toward the puboprostatic ligaments. When possible the puboprostatic ligaments are transected. Adherent muscle fibers are clipped, divided and pushed laterally away from the surface of the prostate. At the completion of this procedure bilaterally, the lateral surfaces of the prostate can be easily palpated and the thickness of the fascia covering the apex of the gland is assessed. At this point a series of sutures is placed sequentially to ligate and divide the dorsal vein complex. With a 1-zero polyglactin suture, the first suture ligature is placed as deep as possible towards the prostatic apex but above the urethral wall and anterior to the prostate. Another suture is placed proximally over the prostate to prevent back flow bleeding. The tissue between the 2 sutures is transected tangentially using fine scissors. The superficial dorsal vein and the puboprostatic ligaments are included in the first suture, and they are transected for the first time at this stage of the procedure in many of the cases. A second suture ligature is placed distally and the tissue encompassed by it is transected. The most distal suture is placed behind the posterior surface of the dorsal vein complex at the apex of the prostate and is encompassed by the entire vein. Once the deep venous complex is transected the apex of the prostate and the anterior wall of the urethra are exposed. Between 3 and 5 sequential sutures are needed to expose the prostatic apex. A sponge stick is used to displace the prostate downward, the urethra is transected anteriorly and the apex of the prostate is dissected gently from both sides of the urethral wall. A right angle clamp is passed beneath the posterior musculature of the urethra and an umbilical tape is passed to keep the plane of dissection. In the event of a nerve-sparing procedure, the lateral pelvic fascia containing the neurovascular bundles on either side of the urethra is left undisturbed and the plane created by the right angle clamp behind the urethra is medial to these bundles. Once the plane behind the urethra is developed, 2 stay sutures are placed on the anterior aspect of the distal urethra. These sutures will be used later during the urethral anastomosis. The urethral catheter is then clamped and divided, and that which remains in the bladder is used for upward traction as the procedure progresses. The right angle clamp is passed again in the plane behind the urethral wall and the entire wall of the urethra is transected. Once the urethra has been transected the plane between the posterior face of the prostate and the anterior wall of the rectum (with its investment of°Denonvilliers' fascia) is developed. This plane allows completion of the prostatorectal cleavage without risking the rectum. If the neurovascular bundles are not preserved and once Denonvilliers' fascial plane is developed, release of the prostate is carried cranially without anatomical limits of dissection. The lateral pelvic fascia and the lateral vascular pedicles are ligated and sectioned stepwise, and the seminal vesicles are dissected and entirely released from the attachment to the rectum. If a nerve-sparing procedure
is performed, care is taken to avoid injury to the neurovascular bundles. The anterior face of the bladder neck is incised with preservation of its circular fibers. Once the bladder is opened, the balloon catheter is deflated, its tip is released from the bladder, and both sides of the catheter are clamped together and placed on light traction. The posterior face of the bladder neck is incised with a direct view of the trigone and the ureteral orifices. Slight traction on the catheter in the prostatic urethra provides excellent view of the trigone even in the event of a protruding median lobe. The posterior aspect of the bladder neck is incised circularly as well, thus the bladder neck maintains a smooth circular configuration that enables a direct anastomosis to the urethral stump without the need to close or reconstruct the bladder neck. Once the prostate has been removed-a-nde0rnplete-hem-0stas-is is achieved, the bladder neck is anastomosed to the urethra. A catheter is placed through the urethra with its tip just inside the pelvis and the urethral mucosa is identified by placing gentle traction on the stay sutures that were inserted earlier in the distal urethra. A 3-zero polyglactin suture on a 26 mm. half circle needle is used for the anastomosis. Ten sutures are placed circumferentially in the distal urethra. The anterior sutures are placed from outside the urethral lumen to the inside, and the posterior sutures are placed from inside the urethral lumen to the outside. Caution is exercised to avoid incorporating the neurovascular bundles in these sutures when the nerves are preserved. Once all of the sutures have been placed in the urethra, the sutures are inserted in the corresponding position in the bladder neck. The bladder is advanced to the urethral edge and the sutures are tied without traction. The posterior plane of the anastomosis is tied first with the knots located inside the lumen. A new 20F Foley catheter is placed through the urethra into the bladder and the anterior plane is ligated with the knots resting outside the urethral lumen. The anastomosis is tested for watertightness by distension through the urethral catheter. Two aspirating drains are left on both sides of the anastomosis. The drains and the urethral catheter are removed 7 and 12 days postoperatively, respectively. Pathological examination. The radical prostatectomy specimens were examined macroscopically and then they were sectioned for histological examination. The proximal and distal urethral margins were removed and examined, and the gland was bivalved and sectioned. Each side was sectioned at 5 to 8 mm. intervals perpendicular to the long axis of the gland. An average of 16 slides of the prostate and seminal vesicles per case was examined (range 12 to 24). Followup. All patients were seen by us 2 months postoperatively and then they were followed at 6-month intervals for 2 years and 1-year intervals thereafter. Followup included history, general physical, digital and rectal examinations, transrectal ultrasonography, serial chest x-rays, yearly renal ultrasonography or excretory urography, and serum acid phosphatase and PSA determinations. Bone scans were performed routinely 2 years postoperatively and at 3-year intervals thereafter, or earlier in the event of new onset of bone pain or on the basis of a newly elevated PSA or serum acid phosphatase. Analysis of the surgical results included early and late complications, urinary continence and potency. Operating time was measured from opening until closure of the skin. Blood loss was calculated from the fluids that were collected in the suction pan, with the exclusion of the fluids that were used for irrigation. The status of postoperative urinary continence was divided into 3 categories: normal----,dry during normal and strenuous activities, and no pads or appliances at any time; stress incontinence-loss of urine during strenuous activities and a pad for protection, and total incontinence-no urinary control. The status of postoperative potency as reported by the patients was defined as full potency, partial potency and impotence: full
885
RADICAL RETROPUBIC ?ROSTATECTOMY'. lV!ORBIDITY AND QUALITY OF UFE
potency-able to achieve an erection sufficient for penetration and orgasm with the same quality as preoperatively, partial potency-erections and able to have intercourse but the quality of these erections is worse compared to erections before the surgery, and impotence-partial or no erections and unable to have intercourse. All data were analyzed using the chi-square or Student t test. RESULTS
Pathological findings. Among the 487 patients who presented with tumor confined within the prostate (clinical stages A and B) 208 (43 % ) had pathological evidence of disease extension beyond the prostatic capsule or had evidence of regional node extension (table 1), including 29 of 115 patients (25%) who presented with clinical stage A and 187 of 372 patients (50%) who presented with stage B disease. Overall in 371 patients (60%) the disease was upstaged, including 11 of 29 (38%) with clinical stage Al, 66 of 86 (77%) with stage A2, 69 of 162 (43%) with stage Bl, 150 of 210 (73%) with stage B2, 30 of 99 (30%) with stage Cl and 13 of 34 (38%) with stage C2. A total of 93 patients had microscopic involvement of the pelvic lymph nodes (pathological stage Dl disease). Of these 93 patients 1 (3%) presented with clinical stage Al, 5 (6%) with stage A2, 8 (5%) with stage Bl, 32 (15%) with stage B2, 34 (34%) with stage Cl and 13 (38%) with stage C2 disease. Downstaging occurred in 12 patients (2%), including 4 with clinical stage B2 and 8 with clinical stage C disease. Morbidity and mortality. There were no operative deaths. Data pertaining to operative morbidity are summarized in table 2. The average operating time was 10 minutes longer when nerve preservation was performed but there was no influence on the perioperative blood loss. As demonstrated in table 2, the operative time and blood loss were reduced significantly as we gained experience. Complete data on the postoperative hematocrit were not available for evaluation. Table 3 lists the early and late complications. There were 3 cases of rectal injury during the operation, which were among the first 100 cases of our experience and included 2 patients with pathological stage C2 disease and 1 with stage B2 disease. The rectal injuries were closed primarily and healed completely. One patient died of acute myocardial infarction 12 days after an uneventful operation and during a normal convalescence. This patient accounted for the only perioperative death in our experience. Early complications occurred in 43 patients (6.9%) and were attributed in part to pelvic lymphadenectorny. Prolonged urinary leakage from the urethral anastomosis occurred in 2 patients during our early In these cases the TABLE
Clinical Stage
1. Comparison between clinical and pathological stages Pathological Stage
No. Pts.
Al A2 Bl B2
Cl C2
Totals
29 86 162 210 99 34 620
Al
A2
Bl
B2
Cl
18
5
0 0
1 20 0
0
0
0 0
0 0
Is
21
1 14 32 56 3 1 107
2 9 19 57 30 0 117
27 93
4 4 0 133
279 (45%)
TABLE
Dl
11
1 5
10
8
61 28 20
32 34 13
ill
93
341 (55%)
2. Operative morbidity
First 100 Pts. Operative time (hrs, mins.) Estimated blood loss
C2
3 (2.20-3.50)
Next 300 Pts. 2
700 (400-2,000) 600
(1.30-2.50)
Last 220 Pts. 1.30 (0.50-2.20)
(200-2,000) 300
(100-1,500)
(cc)
Blood transfusion (units)
3 (0-6)
1.6 (0-4)
Numbers in parentheses represent range.
0.2 (0-2)
TABLE
3. Complications after radical prostatectomy in 620 patients No. Pts. (%)
Operative complications: Intraop. mortality Rectal injury, closed primarily Early complications: Myocardial infarction, day 12 postop. (fatal) Pulmonary emboli (nonfatal) Lower extremity deep vein thrombophlebitis Lymphocele, drained surgically Anastomotic urinary leakage, closed spontaneously Wound infection following a subcutaneous hematoma Severe urosepsis (nonfatal) Late complications: Bladder neck contracture treated with 1 dilation Postop. scar hernia, repaired surgically Hydrocele, repaired surgically Ureteral transection or obstruction Total incontinence
0 3 (0.48)
1 (0.16) 5 (0.8)
14 (2.26) 14 (2.26) 2 (0.32) 6 (0.96)
1 (0.16) 3 (0.48) 2 (0.32) 3 (0.48) 0
0
urethral catheter was left in place for 4 weeks and the leakage resolved without sequelae. Pulmonary emboli occurred in 5 patients and 14 patients had lower extremity deep vein thrombophlebitis. The vascular complications resolved without significant permanent sequelae. The late complication rate, excluding incontinence and impotence, was 1.3%. Anastomotic contracture occurred in only 3 patients (0.5% ). All 3 cases responded to a single dilation that has not had to be repeated. Continence. Followup information on urinary continence 6 months and 1 year or longer after surgery was available on 484 and 398 patients, respectively. No patient was totally incontinent. A total of 434 patients (90%) achieved complete urinary control within 6 months postoperatively, the majority of whom were dry at 3 months after surgery. One year postoperatively 377 patients (95%) achieved normal urinary continence, and this rate remained consistent with longer followup. A total of 50 patients (10%) experienced stress incontinence 6 months after surgery but 1 year or longer postoperatively only 21 patients (5%) were experiencing stress urinary incontinence. Among the 50 patients with stress incontinence 6 months after surgery 29 (58%) were 70 years old and older, compared to 10 of 21 (48%) 1 year postoperatively. Overall, complete return of urinary control was achieved in 97% of the patients less than 70 years old and in 94 % of those older than 70 years. Potency. Nerve preservation radical prostatectomy was performed on a selected group of patients who had tumor confined to the prostate, with or without capsular extension but without seminal vesicle involvement (stages A to who were potent and who were interested in preservation of sexwas sent to the patients who surgery for completion by them and their partners, the status of postoperative potency was evaluated according to their reports. Nerve-sparing radical prostatectomy was performed in 167 patients. Followup information regarding potency 1 year or longer after surgery was available on 106 men. Among the 106 patients 33 were younger than 60 years, 68 were 60 to 70 years old and 5 were older than 70 years. Of the patients 32 (30%) achieved full potency and 40 (38%) were partially potent at 6 months after surgery, A year after surgery 59 patients (56%) were completely potent a.nd 16 (15%) were partially potent. The total number of patients who have erections sufficient for vaginal penetration (full and partial potency) was 72 (68%) and 75 (71%) at 6 months and 1 year after surgery, respectively. The impact of age on the return of sexual function is presented in table 4. DISCUSSION
An additional advantage of radical prostatectomy is the benefit of pathological study of the tumor. In this series 60% of the prostate tumors were upstaged, demonstrating the inac-
886
LEANDRI AND ASSOCIATES TABLE
4. Influence of age on the return of sexual function Status of Potency*
Age (yrs.)
N
o.
Pt
s.
6 Mos. Postop., No. Pts. (%)
Full Less than 60 60-70
Older than 70 Totals
33 68 5 106
Partial Impotent
9 (27) 17 23 (34) 22 0 1 32 (30) 40
(52) (32) (20) (38)
7 23 4 34
(21) (34) (80) (32)
1 Yr. Postop., No. Pts. (%)
Full
Partial Impotent
21 (64) 4 (12) 38 (56) 11 (16) 0 1 (20) 59 (56) 16 (15)
8 (24) 19 (28) 4 (80) 31 (29)
* Full potency-erection with the same quality as preoperatively. Partial potency-the patient has erections and is able to have intercourse but with a lesser quality than preoperatively.
curacy of the available clinical staging techniques, particularly with regard to identifying local extension of disease beyond the capsule or region.al riodeextension. Our findings are notaiffer~ ent from those reported by others. s-a Staging errors in other series varied but as many as 70% of the patients with clinical localized lesions have pathological stage C disease and/or microscopic pelvic lymph node metastasis. These findings need to be viewed in studies that attempt to compare different modalities of treatment based upon clinical tumor staging. As concepts regarding the treatment of advanced local stages (C and Dl) are changing from single modality treatment to multiple modality treatment, with radical prostatectomy and hormonal therapy or radiotherapy as an immediate adjuvant treatment, 2- 4, 9 the importance of the exact pathological stage is further emphasized. Of the 29 cases of stage Al disease 38% were upstaged to A2 to Dl disease and of the 86 cases of stage A2 disease 77% were upstaged as well. These data suggest that if these patients had been left untreated they were at high risk for disease progression and support the concept of recommending definitive treatment for stage A disease. Our surgical results compare favorably to those of previous reports. 8 •10- 12 Of particular note is our low rate of intraoperative morbidity. We had no operative mortality and only 3 cases of rectal injury occurred during our early experience. Rectal perforation during radical prostatectomy varies between 1 and 6% in reported series with 1% requiring colostomy.8 •10• 12 In our last 500 cases we had no rectal injury, which results from the technique we use to expose and dissect the prostatic apex and from the way we keep the plane of Denonvilliers' fascia during dissection of the posterior aspect of the prostate. The calculated blood loss in our series was acceptable and reduced markedly as we continued to gain more experience. The sequential sutures we use to encompass the deep dorsal venous plexus with the corresponding sutures on the prostatic side to prevent back flow bleeding reduce to a minimum the blood loss during exposure and transection of the urethra. A similar technique was recently described by Myers.13 Meticulous separation of the seminal vesicles and the ejaculatory ducts from the base of the bladder furthermore reduces the total blood loss. In some cases some bleeding occurs when the urethra is transected, with the blood coming from the urethral walls. Attempts at stopping this bleeding by suture ligations risk continence or damage the neurovascular bundles during nerve preservation procedures. This bleeding stops once the urethra is anastomosed to the bladder neck. By reducing the operative time the anastomosis can be performed in an acceptable delay after urethral transection, which also contributes to the decline in the total blood loss.The decrease in operating time also contributed to the low rate of thromboembolic complications that occurred in our series by shortening the time of venous stasis in the lower extremities. Early mobilization after surgery also helps to prevent these complications. Anastomotic complications occurred in only 5 of the 620 patients (0.8%). In 2 patients we noted anastomotic leakage that disappeared within 4 weeks without surgical intervention. Using simple drainage with an indwelling catheter the leakage resolved without permanent sequelae. Only 3 patients (0.48%)
had anastomotic contracture that required only 1 dilation. Anastomotic strictures as a late complication of radical prostatectomy varied in previous reports from 1.3 to 27%. 10• 12 The low incidence of anastomotic stricture formation in our series demonstrates the reliability of the anastomosis when it is performed by the technique we described. Of importance in this surgical technique are the preservation of the bladder neck with its circular musculature, and the creation of a wide anastomosis between the urethra and the bladder neck using as many as 10 circumferential sutures. The wide anastomosis provides a watertight closure that prevents urinary leakage and, thus, decreases the possibility of periurethral scarring that may contribute to the development of a late stricture. The posterior suture line of the anastomosis is composed of 5 intraluminal ties. Using this technique the 2 edges can be approximated without-te-ns-i6n-a-nd, as--demtmst-l'-ated-, th.efa
TrtE JOURNAL OF UROLOGY
Printed in U.S.A.
Copyright© 1992 by AMERICAN UROLOGICAL AS30CIATJ:ON, INC.
ical Treatment RADICAL RETROPUBIC PROSTATECTOMY: MORBIDITY AND QUALITY OF LIFE, EXPERIENCE WITH 620 CONSECUTIVE CASES PIERRE LEANDRI, GEORGES ROSSIGNOL, JEAN-ROMAIN GAUTIER AND JACOB RAMON From the Department of Urology, Saint-Jean Languedoc-Cerou, Toulouse, Cedex, France
ABSTRACT
We describe our experience and complications of radical retropubic prostatectomy. From March 1983 through December 1990, 620 consecutive patients have undergone an anatomical radical retropubic prostatectomy for the treatment of prostatic carcinoma. The surgical technique we used is described. In 167 patients the procedure included preservation of the neurovascular bundles. There were no modifications in the surgical technique during this period. There were no operative deaths. Mean operating time was reduced from 3 hours in the first 100 patients to 1.5 hours in the last 220 patients. The average blood loss was reduced remarkably as well. There were only 3 cases of rectal injury, which were closed primarily and healed completely. One patient died of acute myocardial infarction 12 days after an uneventful operation. This patient accounted for the only perioperative death in our experience. Early complications occurred in 43 patients (6.9%), including only 2 cases (0.3%) of anastomotic urinary leakage. The late complication rate, excluding incontinence and impotence, was 1.3%. No patient was totally incontinent. Among the patients who were followed for 1 year or longer 95% achieved complete urinary control and 5% experienced stress urinary incontinence. Preservation of sexual function in patients who underwent a nerve-sparing operation was achieved in 71 %. Our results indicate that radical retropubic prostatectomy can be performed with low morbidity and without affecting the quality of life in the majority of patients. KEY WORDS: prostatic neoplasms, prostatectomy, morbidity
Radical prostatectomy is considered by many to be the treatment of choice for localized carcinoma of the prostate. In France, as in most of Europe, there had been a conservative approach to prostate cancer. Radiation therapy and endocrine therapy had been the predominant alternative in prostate cancer until the last decade. Radical prostatectomy as definitive treatment for prostate cancer was reinstated in France in the early 1980s as a result of significant improvements in the surgical technique, which had been refined to preserve the nerves essential for potency. The technique associated with nerve-sparing radical prostatectomy as proposed by Walsh and Donker has improved the likelihood of continence especially and certainly potency over previous techniques. 1 With a decline in these complications increasing numbers of patients and physicians began to select this surgical method of disease elimination. Reduced patient morbidity and broadened surgical experience also have increased the use of radical prostatectomy in selected patients with locally advanced disease or with limited regional lymph node metastasis. 2- 4 Since 1983 we have performed radical retropubic prostatectomy on all surgical candidates with clinical stages A, B and C prostate cancer. We describe our experience with 620 consecutive patients, and detail our surgical technique, pathological findings, complications and long-term surgical results. MATERIALS AND METHODS
Patient population. A total of 620 consecutive patients underwent radical retropubic prostatectomy for treatment of prostatic carcinoma between March 1983 and December 1990 at our medical institute. Patient age ranged from 46 to 84 years, with a mean age of 68 years. Patients were considered candiAccepted for publication September 27, 1991. 883
dates for radical prostatectomy if they had a biopsy proved prostate tumor without evidence of metastatic disease. Suitable candidates underwent preoperative evaluation, including cystoscopy, serum acid phosphatase determination, radioisotope bone scan, chest films and since January 1988 serum prostate specific antigen (PSA) levels. The upper urinary tracts were evaluated with excretory urography or renal ultrasonography. Transrectal ultrasonography of the prostate was performed as an adjunct to staging in the majority of the patients. Absence of metastasis was established in all patients by negative results from preoperative radionuclide bone scanning, and radiological and biochemical studies. Patients with palpable or ultrasonic evidence of extension of the tumor beyond the capsule or into the seminal vesicles were also considered by us as candidates for radical surgery and, if tumor extension was confirmed pathologically, the patients were further treated with external beam irradiation or androgen ablation. Of the patients 153 (24%) had undergone prior open prostatectomy or transurethral resection of the prostate for benign prostatic hypertrophy, including 38 (6%) who had undergone these procedures more than 2 years before the diagnosis of prostate cancer. In 505 patients (82%) the tumor was confirmed by perinea! prostate biopsy. Following transperineal needle biopsy radical prostatectomy was deferred for 6 to 12 weeks (mean 9 weeks). Following transurethral resection or open prostatectomy surgery was deferred for 3 months. Based upon the preoperative evaluation clinical stage was assigned as follows: Al-tumor discovered by prostatectomy for benign disease and involving less than 5% of the resected tissue, and Gleason score of less than 8, A2-cancer involving more than 5% of the resected tissue or Gleason score of 8 or greater, Bl-palpable tumor confined to the prostate less than 1.5 cm. in size or involving 25% or less of 1 lobe, B2-palpable
884
LEANDRI AND ASSOCIATES
tumor involving more than 25% of 1 lobe but confined to the gland, Cl-minimal palpable extracapsular extension and C2extension of tumor into bladder neck and/or seminal vesicles. Of the patients 29 had clinical stage Al, 86 stage A2, 162 stage B, 210 stage B2, 99 stage Cl and 34 stage C2. Surgical technique. The operative technique, which is a modification of the neuroanatomical approach originated by Walsh and Donker, 1 has been described previously. 5 A modified pelvic lymph node dissection is performed with removal of the lymphatics overlying the obturator fossa and the internal iliac vessels below the bifurcation of the common iliac artery. The lymphatics overlying the external iliac artery are not resected routinely. The lymph nodes are sent for frozen section evaluation and, in the event of microscopic nodal involvement, an immediate adjuvant orchiectomy is performed. The retroperifoneal fat is teased awa.y to expose clearly the endopelvic fascia. With a scalpel a 1 cm. incision is made in the fascia at its reflection over the pelvic side wall. The incision of the fascia is then extended bluntly toward the puboprostatic ligaments. When possible the puboprostatic ligaments are transected. Adherent muscle fibers are clipped, divided and pushed laterally away from the surface of the prostate. At the completion of this procedure bilaterally, the lateral surfaces of the prostate can be easily palpated and the thickness of the fascia covering the apex of the gland is assessed. At this point a series of sutures is placed sequentially to ligate and divide the dorsal vein complex. With a 1-zero polyglactin suture, the first suture ligature is placed as deep as possible towards the prostatic apex but above the urethral wall and anterior to the prostate. Another suture is placed proximally over the prostate to prevent back flow bleeding. The tissue between the 2 sutures is transected tangentially using fine scissors. The superficial dorsal vein and the puboprostatic ligaments are included in the first suture, and they are transected for the first time at this stage of the procedure in many of the cases. A second suture ligature is placed distally and the tissue encompassed by it is transected. The most distal suture is placed behind the posterior surface of the dorsal vein complex at the apex of the prostate and is encompassed by the entire vein. Once the deep venous complex is transected the apex of the prostate and the anterior wall of the urethra are exposed. Between 3 and 5 sequential sutures are needed to expose the prostatic apex. A sponge stick is used to displace the prostate downward, the urethra is transected anteriorly and the apex of the prostate is dissected gently from both sides of the urethral wall. A right angle clamp is passed beneath the posterior musculature of the urethra and an umbilical tape is passed to keep the plane of dissection. In the event of a nerve-sparing procedure, the lateral pelvic fascia containing the neurovascular bundles on either side of the urethra is left undisturbed and the plane created by the right angle clamp behind the urethra is medial to these bundles. Once the plane behind the urethra is developed, 2 stay sutures are placed on the anterior aspect of the distal urethra. These sutures will be used later during the urethral anastomosis. The urethral catheter is then clamped and divided, and that which remains in the bladder is used for upward traction as the procedure progresses. The right angle clamp is passed again in the plane behind the urethral wall and the entire wall of the urethra is transected. Once the urethra has been transected the plane between the posterior face of the prostate and the anterior wall of the rectum (with its investment of°Denonvilliers' fascia) is developed. This plane allows completion of the prostatorectal cleavage without risking the rectum. If the neurovascular bundles are not preserved and once Denonvilliers' fascial plane is developed, release of the prostate is carried cranially without anatomical limits of dissection. The lateral pelvic fascia and the lateral vascular pedicles are ligated and sectioned stepwise, and the seminal vesicles are dissected and entirely released from the attachment to the rectum. If a nerve-sparing procedure
is performed, care is taken to avoid injury to the neurovascular bundles. The anterior face of the bladder neck is incised with preservation of its circular fibers. Once the bladder is opened, the balloon catheter is deflated, its tip is released from the bladder, and both sides of the catheter are clamped together and placed on light traction. The posterior face of the bladder neck is incised with a direct view of the trigone and the ureteral orifices. Slight traction on the catheter in the prostatic urethra provides excellent view of the trigone even in the event of a protruding median lobe. The posterior aspect of the bladder neck is incised circularly as well, thus the bladder neck maintains a smooth circular configuration that enables a direct anastomosis to the urethral stump without the need to close or reconstruct the bladder neck. Once the prostate has been removed-a-nde0rnplete-hem-0stas-is is achieved, the bladder neck is anastomosed to the urethra. A catheter is placed through the urethra with its tip just inside the pelvis and the urethral mucosa is identified by placing gentle traction on the stay sutures that were inserted earlier in the distal urethra. A 3-zero polyglactin suture on a 26 mm. half circle needle is used for the anastomosis. Ten sutures are placed circumferentially in the distal urethra. The anterior sutures are placed from outside the urethral lumen to the inside, and the posterior sutures are placed from inside the urethral lumen to the outside. Caution is exercised to avoid incorporating the neurovascular bundles in these sutures when the nerves are preserved. Once all of the sutures have been placed in the urethra, the sutures are inserted in the corresponding position in the bladder neck. The bladder is advanced to the urethral edge and the sutures are tied without traction. The posterior plane of the anastomosis is tied first with the knots located inside the lumen. A new 20F Foley catheter is placed through the urethra into the bladder and the anterior plane is ligated with the knots resting outside the urethral lumen. The anastomosis is tested for watertightness by distension through the urethral catheter. Two aspirating drains are left on both sides of the anastomosis. The drains and the urethral catheter are removed 7 and 12 days postoperatively, respectively. Pathological examination. The radical prostatectomy specimens were examined macroscopically and then they were sectioned for histological examination. The proximal and distal urethral margins were removed and examined, and the gland was bivalved and sectioned. Each side was sectioned at 5 to 8 mm. intervals perpendicular to the long axis of the gland. An average of 16 slides of the prostate and seminal vesicles per case was examined (range 12 to 24). Followup. All patients were seen by us 2 months postoperatively and then they were followed at 6-month intervals for 2 years and 1-year intervals thereafter. Followup included history, general physical, digital and rectal examinations, transrectal ultrasonography, serial chest x-rays, yearly renal ultrasonography or excretory urography, and serum acid phosphatase and PSA determinations. Bone scans were performed routinely 2 years postoperatively and at 3-year intervals thereafter, or earlier in the event of new onset of bone pain or on the basis of a newly elevated PSA or serum acid phosphatase. Analysis of the surgical results included early and late complications, urinary continence and potency. Operating time was measured from opening until closure of the skin. Blood loss was calculated from the fluids that were collected in the suction pan, with the exclusion of the fluids that were used for irrigation. The status of postoperative urinary continence was divided into 3 categories: normal----,dry during normal and strenuous activities, and no pads or appliances at any time; stress incontinence-loss of urine during strenuous activities and a pad for protection, and total incontinence-no urinary control. The status of postoperative potency as reported by the patients was defined as full potency, partial potency and impotence: full
885
RADICAL RETROPUBIC ?ROSTATECTOMY'. lV!ORBIDITY AND QUALITY OF UFE
potency-able to achieve an erection sufficient for penetration and orgasm with the same quality as preoperatively, partial potency-erections and able to have intercourse but the quality of these erections is worse compared to erections before the surgery, and impotence-partial or no erections and unable to have intercourse. All data were analyzed using the chi-square or Student t test. RESULTS
Pathological findings. Among the 487 patients who presented with tumor confined within the prostate (clinical stages A and B) 208 (43 % ) had pathological evidence of disease extension beyond the prostatic capsule or had evidence of regional node extension (table 1), including 29 of 115 patients (25%) who presented with clinical stage A and 187 of 372 patients (50%) who presented with stage B disease. Overall in 371 patients (60%) the disease was upstaged, including 11 of 29 (38%) with clinical stage Al, 66 of 86 (77%) with stage A2, 69 of 162 (43%) with stage Bl, 150 of 210 (73%) with stage B2, 30 of 99 (30%) with stage Cl and 13 of 34 (38%) with stage C2. A total of 93 patients had microscopic involvement of the pelvic lymph nodes (pathological stage Dl disease). Of these 93 patients 1 (3%) presented with clinical stage Al, 5 (6%) with stage A2, 8 (5%) with stage Bl, 32 (15%) with stage B2, 34 (34%) with stage Cl and 13 (38%) with stage C2 disease. Downstaging occurred in 12 patients (2%), including 4 with clinical stage B2 and 8 with clinical stage C disease. Morbidity and mortality. There were no operative deaths. Data pertaining to operative morbidity are summarized in table 2. The average operating time was 10 minutes longer when nerve preservation was performed but there was no influence on the perioperative blood loss. As demonstrated in table 2, the operative time and blood loss were reduced significantly as we gained experience. Complete data on the postoperative hematocrit were not available for evaluation. Table 3 lists the early and late complications. There were 3 cases of rectal injury during the operation, which were among the first 100 cases of our experience and included 2 patients with pathological stage C2 disease and 1 with stage B2 disease. The rectal injuries were closed primarily and healed completely. One patient died of acute myocardial infarction 12 days after an uneventful operation and during a normal convalescence. This patient accounted for the only perioperative death in our experience. Early complications occurred in 43 patients (6.9%) and were attributed in part to pelvic lymphadenectorny. Prolonged urinary leakage from the urethral anastomosis occurred in 2 patients during our early In these cases the TABLE
Clinical Stage
1. Comparison between clinical and pathological stages Pathological Stage
No. Pts.
Al A2 Bl B2
Cl C2
Totals
29 86 162 210 99 34 620
Al
A2
Bl
B2
Cl
18
5
0 0
1 20 0
0
0
0 0
0 0
Is
21
1 14 32 56 3 1 107
2 9 19 57 30 0 117
27 93
4 4 0 133
279 (45%)
TABLE
Dl
11
1 5
10
8
61 28 20
32 34 13
ill
93
341 (55%)
2. Operative morbidity
First 100 Pts. Operative time (hrs, mins.) Estimated blood loss
C2
3 (2.20-3.50)
Next 300 Pts. 2
700 (400-2,000) 600
(1.30-2.50)
Last 220 Pts. 1.30 (0.50-2.20)
(200-2,000) 300
(100-1,500)
(cc)
Blood transfusion (units)
3 (0-6)
1.6 (0-4)
Numbers in parentheses represent range.
0.2 (0-2)
TABLE
3. Complications after radical prostatectomy in 620 patients No. Pts. (%)
Operative complications: Intraop. mortality Rectal injury, closed primarily Early complications: Myocardial infarction, day 12 postop. (fatal) Pulmonary emboli (nonfatal) Lower extremity deep vein thrombophlebitis Lymphocele, drained surgically Anastomotic urinary leakage, closed spontaneously Wound infection following a subcutaneous hematoma Severe urosepsis (nonfatal) Late complications: Bladder neck contracture treated with 1 dilation Postop. scar hernia, repaired surgically Hydrocele, repaired surgically Ureteral transection or obstruction Total incontinence
0 3 (0.48)
1 (0.16) 5 (0.8)
14 (2.26) 14 (2.26) 2 (0.32) 6 (0.96)
1 (0.16) 3 (0.48) 2 (0.32) 3 (0.48) 0
0
urethral catheter was left in place for 4 weeks and the leakage resolved without sequelae. Pulmonary emboli occurred in 5 patients and 14 patients had lower extremity deep vein thrombophlebitis. The vascular complications resolved without significant permanent sequelae. The late complication rate, excluding incontinence and impotence, was 1.3%. Anastomotic contracture occurred in only 3 patients (0.5% ). All 3 cases responded to a single dilation that has not had to be repeated. Continence. Followup information on urinary continence 6 months and 1 year or longer after surgery was available on 484 and 398 patients, respectively. No patient was totally incontinent. A total of 434 patients (90%) achieved complete urinary control within 6 months postoperatively, the majority of whom were dry at 3 months after surgery. One year postoperatively 377 patients (95%) achieved normal urinary continence, and this rate remained consistent with longer followup. A total of 50 patients (10%) experienced stress incontinence 6 months after surgery but 1 year or longer postoperatively only 21 patients (5%) were experiencing stress urinary incontinence. Among the 50 patients with stress incontinence 6 months after surgery 29 (58%) were 70 years old and older, compared to 10 of 21 (48%) 1 year postoperatively. Overall, complete return of urinary control was achieved in 97% of the patients less than 70 years old and in 94 % of those older than 70 years. Potency. Nerve preservation radical prostatectomy was performed on a selected group of patients who had tumor confined to the prostate, with or without capsular extension but without seminal vesicle involvement (stages A to who were potent and who were interested in preservation of sexwas sent to the patients who surgery for completion by them and their partners, the status of postoperative potency was evaluated according to their reports. Nerve-sparing radical prostatectomy was performed in 167 patients. Followup information regarding potency 1 year or longer after surgery was available on 106 men. Among the 106 patients 33 were younger than 60 years, 68 were 60 to 70 years old and 5 were older than 70 years. Of the patients 32 (30%) achieved full potency and 40 (38%) were partially potent at 6 months after surgery, A year after surgery 59 patients (56%) were completely potent a.nd 16 (15%) were partially potent. The total number of patients who have erections sufficient for vaginal penetration (full and partial potency) was 72 (68%) and 75 (71%) at 6 months and 1 year after surgery, respectively. The impact of age on the return of sexual function is presented in table 4. DISCUSSION
An additional advantage of radical prostatectomy is the benefit of pathological study of the tumor. In this series 60% of the prostate tumors were upstaged, demonstrating the inac-
886
LEANDRI AND ASSOCIATES TABLE
4. Influence of age on the return of sexual function Status of Potency*
Age (yrs.)
N
o.
Pt
s.
6 Mos. Postop., No. Pts. (%)
Full Less than 60 60-70
Older than 70 Totals
33 68 5 106
Partial Impotent
9 (27) 17 23 (34) 22 0 1 32 (30) 40
(52) (32) (20) (38)
7 23 4 34
(21) (34) (80) (32)
1 Yr. Postop., No. Pts. (%)
Full
Partial Impotent
21 (64) 4 (12) 38 (56) 11 (16) 0 1 (20) 59 (56) 16 (15)
8 (24) 19 (28) 4 (80) 31 (29)
* Full potency-erection with the same quality as preoperatively. Partial potency-the patient has erections and is able to have intercourse but with a lesser quality than preoperatively.
curacy of the available clinical staging techniques, particularly with regard to identifying local extension of disease beyond the capsule or region.al riodeextension. Our findings are notaiffer~ ent from those reported by others. s-a Staging errors in other series varied but as many as 70% of the patients with clinical localized lesions have pathological stage C disease and/or microscopic pelvic lymph node metastasis. These findings need to be viewed in studies that attempt to compare different modalities of treatment based upon clinical tumor staging. As concepts regarding the treatment of advanced local stages (C and Dl) are changing from single modality treatment to multiple modality treatment, with radical prostatectomy and hormonal therapy or radiotherapy as an immediate adjuvant treatment, 2- 4, 9 the importance of the exact pathological stage is further emphasized. Of the 29 cases of stage Al disease 38% were upstaged to A2 to Dl disease and of the 86 cases of stage A2 disease 77% were upstaged as well. These data suggest that if these patients had been left untreated they were at high risk for disease progression and support the concept of recommending definitive treatment for stage A disease. Our surgical results compare favorably to those of previous reports. 8 •10- 12 Of particular note is our low rate of intraoperative morbidity. We had no operative mortality and only 3 cases of rectal injury occurred during our early experience. Rectal perforation during radical prostatectomy varies between 1 and 6% in reported series with 1% requiring colostomy.8 •10• 12 In our last 500 cases we had no rectal injury, which results from the technique we use to expose and dissect the prostatic apex and from the way we keep the plane of Denonvilliers' fascia during dissection of the posterior aspect of the prostate. The calculated blood loss in our series was acceptable and reduced markedly as we continued to gain more experience. The sequential sutures we use to encompass the deep dorsal venous plexus with the corresponding sutures on the prostatic side to prevent back flow bleeding reduce to a minimum the blood loss during exposure and transection of the urethra. A similar technique was recently described by Myers.13 Meticulous separation of the seminal vesicles and the ejaculatory ducts from the base of the bladder furthermore reduces the total blood loss. In some cases some bleeding occurs when the urethra is transected, with the blood coming from the urethral walls. Attempts at stopping this bleeding by suture ligations risk continence or damage the neurovascular bundles during nerve preservation procedures. This bleeding stops once the urethra is anastomosed to the bladder neck. By reducing the operative time the anastomosis can be performed in an acceptable delay after urethral transection, which also contributes to the decline in the total blood loss.The decrease in operating time also contributed to the low rate of thromboembolic complications that occurred in our series by shortening the time of venous stasis in the lower extremities. Early mobilization after surgery also helps to prevent these complications. Anastomotic complications occurred in only 5 of the 620 patients (0.8%). In 2 patients we noted anastomotic leakage that disappeared within 4 weeks without surgical intervention. Using simple drainage with an indwelling catheter the leakage resolved without permanent sequelae. Only 3 patients (0.48%)
had anastomotic contracture that required only 1 dilation. Anastomotic strictures as a late complication of radical prostatectomy varied in previous reports from 1.3 to 27%. 10• 12 The low incidence of anastomotic stricture formation in our series demonstrates the reliability of the anastomosis when it is performed by the technique we described. Of importance in this surgical technique are the preservation of the bladder neck with its circular musculature, and the creation of a wide anastomosis between the urethra and the bladder neck using as many as 10 circumferential sutures. The wide anastomosis provides a watertight closure that prevents urinary leakage and, thus, decreases the possibility of periurethral scarring that may contribute to the development of a late stricture. The posterior suture line of the anastomosis is composed of 5 intraluminal ties. Using this technique the 2 edges can be approximated without-te-ns-i6n-a-nd, as--demtmst-l'-ated-, th.efa
