- Email: [email protected]
Nerve-Sparing Radical Prostatectomy: Evaluation of Results after 250 Patients
0022-5347 / 90/ 1433 -0538$02.00/ 0 Vol. 143, March
THE JOURNAL OF UROLOGY
Printed in U.S.A.
Copyright © 1990 by AMERICAN UROLOGICAL ASSO CIATION , INC.
NERVE-SPARING RADICAL PROSTATECTOMY: EVALUATION OF RESULTS AFTER 250 PATIENTS WILLIAM J. CATALONA
\
, ..»'-G3\
_ \ )~ ~
0,V
~
'\
/
AND
STEVEN W. BIGG
From the Division of Urologic Surgery, Washington University School of Medicine, St. Louis, Missouri
J._. ~~-:
"
dolo \
ABSTRACT
W'':::J:.. ~ c...,# ~
\)
~
·\
To examine the efficacy of nerve-sparing radical retropubic prostatectomy in preserving sexual potency and urinary continence, and in providing complete tumor excision we analyzed the records of the first 250 consecutive patients with clinical stage A or B prostate cancer treated since this operation was adopted at our institution. Over-all, sexual potency was preserved in 71 of 112 patients (63%) who underwent bilateral nerve-sparing prostatectomy and 13 of 33 (39%) who underwent a unilateral nerve-sparing procedure with a minimum of 6 months of followup. Preservation of potency correlated with patient age (p equals 0.0035, chi-square) and was significantly (p less than 0.001, chi-square) higher in patients with pathologically organ-confined tumors (72%) than in those with pathologically extracapsular tumors (51%). Of 192 patients followed for at least 6 months 188 (98 %) achieved urinary continence postoperatively. Over-all, apparent complete tumor excision as defined by organ-confined tumor with negative surgical margins and undetectable postoperative prostate specific antigen levels was achieved in 14 preoperatively potent patients (42 %) who underwent a unilateral and 67 (59%) who underwent a bilateral nerve-sparing procedure. Completeness of tumor excision correlated with tumor stage. In approximately 45% of the patients incomplete tumor excision was owing to seminal vesicle and/or lymph node involvement or positive bladder neck margins that could not be attributed to the nerve-sparing modification. However, improper application of the nerve-sparing technique may have contributed in the others. We were unable to detect microscopic penetration of the capsule or distinguish between gross extracapsular tumor extension and periprostatic fibrosis at operation. We conclude that with proper application of nerve-sparing radical retropubic prostatectomy, potency can be preserved in the majority of patients without compromising the adequacy of tumor excision. The completeness of tumor excision appears to be determined primarily by the extent of the tumor. Therefore, patient selection is important. Patients with focal, well differentiated tumors are ideal candidates for a nerve-sparing procedure, while those with high volume, poorly differentiated tumors may be at a higher risk for positive surgical margins. The benefits of wide excision of the neurovascular bundles remain to be demonstrated formally. (J. Ural. , 143: 538-544, 1990)
In 1983 Walsh and associates introduced the nerve-sparing modification of radical retropubic prostatectomy in which the prostate is excised without injuring the neurovascular bundles that contain the cavernous nerves and vessels to preserve erectile potency.1 Since then, Walsh has modified and improved upon this operation and has demonstrated that with the nervesparing technique potency can be preserved in the majority of patients with organ-confined tumors.2• 3 Moreover, he has shown that 1 or both neurovascular bundles can be sacrificed deliberately in patients with evidence of extraprostatic tumor extension at operation. In so doing it is possible to obtain wider margins of resection than usually are achieved with either standard radical perineal or radical retropubic prostatectomy.4 Walsh and associates also demonstrated that erectile potency is preserved in the majority of patients in whom only 1 neurovascular bundle has been excised.5 Other than these studies, there have been few reports on the results of nerve-sparing radical prostatectomy.6- 8 We adopted the nerve-sparing radical prostatectomy in May 1983 and reported on our early results in 40 patients in 1985,6 documenting postoperative return of erectile function and correlating the incidence of positive surgical margins with the clinical stage of tumor and the technique of prostatectomy used (standard versus nerve-sparing). In the present study we analyzed and updated our results in achieving preservation of continence, potency and complete tumor excision in 250 consecutive patients treated with radical prostatectomy since the Accepted for publication September 18, 1989.
nerve-sparing modification was adopted at our institution. We correlated our results with relevant clinical parameters, and also evaluated our ability to distinguish accurately between periprostatic fibrosis and extracapsular tumor extension at operation. The results have altered our approach to radical prostatectomy in some patients with clinically localized prostate cancer. PATIENTS AND METHODS
Patients. A total of 250 consecutive private patients treated by 1 of us (W. J. C.) with clinical stage A or B prostate cancer underwent pelvic lymphadenectomy and radical retropubic prostatectomy between May 6, 1983 and November 30, 1988 using a technique similar to that described by Walsh and associates. 1- 3 All patients had negative pelvic lymph nodes on frozen section at operation. Clinical staging. The disease in all patients was staged with enzymatic serum prostatic acid phosphatase determinations using the thymolphthalein monophosphate substrate and radionuclide bone scans. Most patients also were evaluated by prostate specific antigen levels and pelvic computerized tomographic (CT) scans. Many patients also were evaluated with transrectal prostatic ultrasonography or magnetic resonance imaging. With a modification of the Whitmore-Jewett system the patients were classified as having clinical stage Al disease if they had clinically unsuspected well differentiated prostate cancer in fewer than 5% of a prostatectomy specimen removed
538
539
NERVE-SPARING RADICAL PROSTATECTOMY
for presumed benign hyperplasia. They were considered to have stage A2 disease if more than 5% of the specimen was involved, or if there was moderately or poorly differentiated carcinoma. Patients were classified as having clinical stage B 1 disease if they had palpable carcinoma of less than 2 cm. in diameter involving less than 1 lobe of the prostate and judged to be confined within the prostatic capsule. Stage B2 disease denoted palpable tumors involving both lobes of the prostate or induration of greater than 2 cm. in diameter judged to be confined within the prostatic capsule. All patients with clinical stage A or B disease (except 1) had a normal serum acid phosphatase level and a bone scan with confirmatory radiographs that showed no evidence of metastases. Findings on pelvic CT scans, transrectal prostatic ultrasonography or magnetic resonance imaging were not considered sufficient to exclude these patients from radical prostatectomy. Examination of surgical specimens. At operation the bladder neck and urethral margins were removed from the prostate by grasping the retracted urethra, including its mucosa! edges, with small Allis forceps and circumferentially excising the 2 to 3 mm. urethral stump. The bladder neck similarly was excised circumferentially, including the mucosa! edges and bladder neck muscle. These margins were sent as separate specimens. If the histological examination revealed cancer the surgical margins were considered to be positive. The excised prostate was dipped in india ink and immediately fixed in Bouin's solution to prevent the india ink from spreading into the tissue planes. The specimen then was "breadloafed" in 2 to 5 mm. segments, depending on the size of the gland, and fixed overnight in 10% phosphate buffered formalin solution. Then, 3 hematoxylin and eosin-stained sections from the right half and 3 from the left half of the prostate were examined for the peripheral surgical margins. In addition, 2 longitudinal sections through the base of the seminal vesicles were examined on each side. Two longitudinal sections through the apex of the prostate also were examined. The presence of cancer at the prostatic apex but not in the urethral margin was considered to be a positive anterior and/or posterior margin. If the tumor extended to an inked margin it was called a positive lateral, anterior or posterior surgical margin. Pathological staging. Pathological stages A and B were defined as for clinical staging except that the stage was verified histologically. Pathological stage C was divided into substages: stage Cl denoted microscopic extracapsular tumor extension and stage C2 referred to tumor extension to the fascia surrounding the seminal vesicles or the muscular walls of the seminal vesicle(s) proper with or without microscopically positive margins. Pathological stage Dl referred to patients with clinical stage A or B disease who at pelvic lymphadenectomy had histologically documented lymph node metastases. Organ-confined disease referred to patients with pathological stage A or B cancer. Extracapsular disease denoted pathological stage C or D tumor, and included all patients with positive margins of resection. Tumor grading. Tumors were graded according to the system used at Barnes Hospital into well, moderate or poorly differentiated adenocarcinoma. Followup. Patients were followed at 3 to 6-month intervals for 2 years and thereafter at approximately 6-month intervals. The last date of followup for this analysis was January 4, 1989. At followup a history, digital rectal examination, prostate specific antigen level and acid phosphatase determinations were obtained. Bone scans, CT scans or other tests were not performed routinely unless clinically indicated. Preoperative and postoperative erectile function. Preoperative and postoperative erectile function was determined by history. Information also was obtained from sexual partners when feasible. Postoperatively, the patient was classified as potent if the erection was sufficient for vaginal penetration and sexual
intercourse or not potent if the erection was insufficient for vaginal penetration. Urinary continence. The patients were considered to be continent if they did not require a protective pad to keep the clothing dry. Patients who were dry under virtually all normal ) circumstances but who elected to wear a precautionary pad for occasional leakage of a few drops of urine with severe coughing or straining also were considered as being continent. Nerve-sparing procedure. Throughout t his series our primary goal was to achieve complete tumor excision with preservation of potency being of secondary importance. In patients in whom evidence suggested gross extracapsular tumor extension at operation the neurovascular bundle in question was deliberately sacrificed by wide ipsilateral excision, dissecting the bundle off of the anterior surface of the rectum. In patients who were not potent preoperatively no attempt was made to preserve the neurovascular bundles. In these patients the bundles were taken wider (as in our former standard radical prostatectomy) than in potent patients but the entire neurovascular bundles were not deliberately and completely dissected off the anterior rectal wall (as in a superradical prostatectomy) unless there was evidence of extracapsular tumor extension. RESULTS
Patient age. Mean patient age over-all was 64 years (table 1). The mean age was 61 years for patients with clinical stage Al, 65 for those with stage A2, 62 for t hose with stage Bl and 65 for those with stage B2 disease. Clinical stage. Of the patients 13 had clinical stage Al, 40 stage A2, 82 stage Bl and 115 stage B2 disease (table 2). All except 1 patient had normal preoperative serum prostatic acid 1. Comparison of the preservation of potency as a func tion of clinical stage, age and type of nerve-sparing procedure performed in patients who were potent preoperatively and were followed for at least 6 months
TABLE
Clinical Stage
Pt. Age
U nilat. Procedure No. Potent/ Total (%)
40-49 50- 59 60-69 70-79
Al
Totals 40-49 50- 59 60-69 70- 79
A2
40-49 50-59 60- 69 70- 79
Totals B2
(0)
2/3 0/1 2/5
(67) (0) (40)
0/2
(0)
(0) 0/2 1/1 (100) (0) 0/ 1 2/ 3 (67) 2/ 2 (100) 5/7 (71)
Totals Bl
0/1
Bilat. Procedure No. Potent/ Total(%)
40-49 50- 59 60- 69 70-79
Totals
(80) (50)
5/7 0/ 1 2/3 5/ 10 2/ 4 9/18
(71) (0) (67) (50) (50) (50)
(25) (20) (60) (32)
(67) 2/ 3 15/ 17 (88) 13/24 (54) (0) 0/1 30/45 (67) 1/ 1 (100) 10/12 (83) 14/22 (64) (20) 2/7 27/ 42 (64)
13/33 (39)
71/ 112 (63)
1/4 2/10 3/5 6/19
Totals
4/5 1/2
Relationship between clinical and pathological stage in 250 consecutive patients undergoing radical prostatectomy
TABLE 2.
Clinical Stage Al A2 Bl B2 Totals
Pathological Stage T otals Al
A2
7
6 26
7
32
Bl
B2
Cl
C2
01
48
16 39
8
55
0 2 0 30
0 0 0
48
0 12 18 38
68
32
8
13 40 82 115 250
r,.A;,
540
CATALONA AND BIGG
phosphatase levels. The patient with an elevated acid phosphatase titer was believed to have hepatic dysfunction. It was considered that the acid phosphatase titer may have been elevated on this basis. Thus, he was offered radical prostatectomy. A pathologically organ-confined tumor was found and postoperatively this patient maintained an undetectable prostate specific antigen level with persistent mild elevation of acid phosphatase and no evidence of recurrent tumor for 24 months. Of 53 patients classified as having clinical stage A disease 45 (85%) were referred for radical prostatectomy after transurethral prostatectomy had been done elsewhere for clinical benign hyperplasia. Many of these patients had palpable abnormalities in the prostate gland after transurethral resection. However, it was impossible to distinguish between postoperative changes and baseline abnormalities. Therefore, these patients were classified as having clinical stage Al or A2 disease as stated previously. Pathological stage. Pathological stage was Al in 7 patients, vf A2 in 32, Bl in 48, B2 in 55, Cl in 68, C2 in 32 and Dl in 8. (15, All but 1 patient with patholo ical sta e Cl disease had ositive i;;urgica margins. Many of these patients had multiple sites of capsular penetration. We did not observe a case in which ~ e was capsular penetrationexcTusivelyin-tne regionoftlie neurov~ar15un e that was covere -by"tl1e bunaie wiUia negative surgical margin. The distribution of pathological stages as a function of clinical stage is shown in table 2. None of the 13 stage Al cancer patients had extraprostatic tumor spread, although 6 had more extensive or more undifferentiated tumor within the prostate. In contrast, the disease in 14 of 40 patients (30%) with stage A2, 18 of 82 (22%) with stage B 1 and 76 of 115 (68 %) with stage B2 cancer was clinically understaged. Over-all, clinical understaging occurred in 108 of the 250 patients (43%) with clinical stage A or B prostate cancer. Tumor grade in relation to clinical understaging. Occasionally, there was a discrepancy between the tumor grade of the diagnostic biopsy and that of the radical prostatectomy specimen. This topic has been considered in detail previously9 and will not be addressed further. Clinical understaging occurred in 18 of 75 patients (24%) with well differentiated, 65 of 143 (45%) with moderately differentiated and 25 of 32 (78%) with poorly differentiated tumors. The correlation between tumor grade and clinical understaging was significant at a 0.00001 level by chi-square analysis. Nerve-sparing procedure. Deliberate wide excision of the neurovascular bundle was performed unilaterally in 62 patients (25%) and bilaterally in 13 (5%). In the remaining patients who were potent preoperatively an attempt was made to mobilize the prostate medial and anterior to the neurovascular bundles. In most patients the retrograde approach as described by Walsh 3 was used. However, in some patients in whom the plane between the prostate and rectum was obliterated at the prostatic apex an antegrade approach was used. The relationship between clinical stage and the attempt to perform bilateral or unilateral nerve-sparing is shown in table
61
TABLE
3. Type of nerve-sparing procedure performed as a function of clinical stage in 250 consecutive patients Nerve-Sparing Procedure
Clinical Stage Al A2 Bl B2 Totals
Bilat.• No.(%)
Unilat.* No.(%)
8 32 68 67 175
5 5 13 39 62
(61) (80) (83) (58) (70)
(38) (13) (16) (34) (25)
None No.(%) 0 3 1 9 13
(7) (1) (8) (5)
Total No. 13 40 82 115 250
* Not all of these patients were potent preoperatively. In impotent patients neurovascular bundles were not widely excised unless there was evidence of tumor extension.
TABLE 4. Preservation of potency as a function of pathological stage and the type of nerve-sparing procedure performed in 145 patients who were potent preoperatively and followed for 6 months or longer
Pathological Stage A B Cl C2 Dl Totals
U nilat. Procedure No./Total (%) 2/6 5/9 3/11 2/5 1/2 13/33
(33) (56) (27) (40) (50) (39)
Bilat. Procedure No./Total (%) 12/19 36/48 15/29 6/12 2/4 71/112
(63) (75) (52) (50) (50) (63)
3. Bilateral nerve-sparing was attempted in 61 % of the patients with clinical stage Al, 80% with stage A2, 83% with stage Bl but only 58% with stage B2 tumors. Unilateral nerve-sparing was attempted in 38% of the patients with stage Al, 13% with stage A2, 16% with stage Bl and 34% with stage B2 tumors. Followup. Of the patients 58 were followed for less than 6 months postoperatively, 192 for at least 6 months, 161 for 12 months, 102 for 24 months, 61 for 36 months, 28 for 48 months and 6 for 60 months. Complications. The complications of urinary incontinence and sexual impotency are addressed. Other postoperative complications will be the topic of a separate report and will not be discussed further in this study. Postoperative urinary continence. Among 92 patients who were followed for at least 6 months postoperatively 188 (98%) are continent. No patient had total urinary incontinence but 4 have required the use of pads to keep the clothing dry. Preservation of potency. Over-all, 180 of the 250 patients (72%) were potent preoperatively but only 145 fulfilled the criteria of being potent preoperatively, having undergone either a unilateral or bilateral nerve-sparing procedure and being eligible for at least 6 months of followup. Of these patients 71 of 112 (63%) who underwent bilateral nerve-sparing (tables 1, 4 and 5) and 13 of 33 (39%) who underwent unilateral nervesparing (tables 4 and 5) had potency preserved (table 1). Bilateral nerve-sparing procedure: patient age, tumor stage and potency. The interrelationships among patient age, clinic;al stage and preservation of potency in patients who underwent a bilateral nerve-sparing procedure are shown in table 1. Potency was preserved in 71 % of the patients with clinical stage Al, 50% with stage A2, 67% with stage Bl and 64% with stage B2 disease. Potency was preserved in 81 % of the patients 40 to 59 years old, 57% of those 60 to 69 years old and 33% of those 70 to 79 years old. This correlation between patient age and preservation of potency is statistically significant (p = 0.0035, chi-square) . The relationship of pathological stage to preservation of potency in patients who underwent a bilateral nerve-sparing procedure is shown in table 4. Potency was preserved in 63% of the patients with pathological stage A, 75% with stage B, 52% with stage Cl, 50% with stage C2 and 50% with stage Dl disease. The correlation between preservation of potency in patients with organ-confined tumors versus those with extracapsular extension is significant (p <0.0001 by chi-square analysis). Preservation of potency with unilateral nerve-sparing. The interrelationships among clinical stage, patient age and preservation of potency in patients who underwent a unilateral nerve-sparing procedure are shown in table 1. Potency was preserved in 40% of the patients with clinical stage Al, 0% (O of 2) with stage A2, 71 % with stage Bl and 32% with stage B2 tumors. The small numbers of patients in the various groups preclude meaningful correlations. The relationship between pathological stage and preservation of potency in patients who underwent a unilateral nerve-sparing procedure also is shown in table 4. Potency was preserved after unilateral nerve-sparing prostatectomy in 33% of the patients with pathological stage A, 56% with stage B, 27% with
541
NERVE-SPARING RADICAL PROSTATECTOMY TABLE 5. i
~
Relationship among clinical stage, preservation of potency and completeness of tumor excision in 112 patients who were potent preoperatively, underwent a bilateral nerve-sparing procedure and were eligible for 6-month followup
'
I
Ii
Extracapsular Extension + Potent
Pathological Stage Clinical Stage
No.
OrganConfined Disease
Confined, Prostate Specific Antigen :ai0.6
7 12 36 12
7 12 35 12
67
66
Confined, Prostate Specific Antigen :ai0.6, Potent No.(%)
C
D1
0 6 9 26
0 0 0 4
41
4
;i
~
Al A2 Bl B2 Totals
7 18 45 42 112
5 7 25 10 47
(71) (39) (56)* (24) (42)
Prostate Specific Antigen <0.6 No.(%)
Prostate Specific Antigen >0.6 No.(%)
0 1 (6) 4 (9)* 10 (24) 15 (13)
0 1 (6) 0 7 (17) 8 (7)
* The apparent discrepancy between this table listing 29 potent stage Bl cancer patients and table 1 listing 30 potent clinical stage Bl cancer patients is that 1 potent patient with organ-confined tumor had a postoperative prostate specific antigen level of 0.7 ng./ml. and, thus, did not fulfill the criteria for having complete tumor excision plus preservation of potency. TABLE 6.
Relationship among clinical stage, preservation of potency and completeness of tumor excision in 33 patients who were potent preoperatively, underwent a unilateral nerve-sparing procedure and were eligible for 6-month followup
Clinical Stage
Al A2 Bl B2 Totals
No.
OrganConfined
Confined, Prostate Specific Antigen :ai0.6
19
5 1 3 6
5 1 2 6
33
15
14
5 2 7
Confined, Prostate Specific Antigen :ai0.6, Potent No.(%) 2 0 2 2 6
(40) (0) (28)* (11) (18)
Pathological Stage
C
Dl
0 1 4 11
0 0 0 2
16
2
Extracapsular Extension Potent
+
Prostate Specific Antigen <0.6 No.(%)
Prostate Specific Antigen >0.6 No.(%)
0 0 2 (29)* 0 2 (6)
0 0 0 4 (21) 4 (12)
* The apparent discrepancy between this table listing 4 potent stage Bl cancer patients and table 1 listing 5 potent stage Bl cancer patients is that 1 patient with organ-confined tumor had a postoperative prostate specific antigen level of 1.5 ng./ml. and, thus, did not fulfill the criteria of having complete tumor excision plus preservation of potency.
stage Cl, 40% with stage C2 and 50% with stage Dl tumors. There was no significant correlation between preservation of potency in patients with organ-confined tumors versus those with extracapsular extension among patients who underwent a unilateral nerve-sparing operation. Relationship among clinical stage, complete tumor excision and preservation of potency. The relationship between complete tumor excision and preserved potency as a function of clinical stage in patients who were potent preoperatively, underwent a bilateral nerve-sparing procedure and were eligible for 6-month followup is shown in table 5. Organ-confined tumor and undetectable postoperative prostate specific antigen levels were noted in all 7 patients with clinical stage Al disease, 12 of 18 potent patients with stage A2 cancer, 35 of 45 potent patients with clinical stage B 1 cancer and 12 of 42 potent patients (29%) with stage B2 disease, and potency was preserved in 5 (71 %), 7 (35%), 25 (56%) and 10 (24%), respectively. In addition, potency was preserved in 2 other stage A2 cancer patients with extracapsular tumor extension (1 with an undetectable and 1 with a detectable postoperative prostate specific antigen level), 4 other stage Bl cancer patients (9%) with microscopic tumor extension (all with undetectable postoperative prostate specific antigen levels) and 17 stage B2 cancer patients with extracapsular tumor extension (10 with undetectable and 7 with detectable prostate specific antigen levels postoperatively). The relationship between complete tumor excision and preserved potency as a function of clinical stage in patients who were potent preoperatively, underwent a unilateral nerve-sparing procedure and were eligible for 6-month followup is shown in table 6. Over-all, of these 33 patients only 6 (18%) had complete tumor excision and preservation of potency. Six patients with extracapsular tumor extension retained potency, including 2 with undetectable and 4 with detectable postoperative prostate specific antigen levels. Over-all, 12 of the 145 eligible patients (8%) had retained potency with proved inadequate tumor excision as defined by
positive margins of resection and detectable postoperative prostate specific antigen levels (tables 5 and 6). Of these patients 6 (50%) had seminal vesicle invasion, 3 (25%) had positive urethral margins and 2 (17%) had positive nodes. One patient (8%) had positive lateral margins only. Intraoperative assessment of tumor extension versus fibrosis. Among 76 patients in whom 1 or both neurovascular bundles were sacrificed 33 (43%) had histologically organ-confined tumors. Conversely, among 175 patients in whom neither neurovascular bundle was sacrificed 66 (38%) had histological evidence of extracapsular tumor extension. Positive tumor margins. Table 7 shows an analysis of the extent of extracapsular tumor extension in the patients who were potent preoperatively and underwent a nerve-sparing procedure for a tumor with positive margins and/or extracapsular tumor extension. Of the 145 preoperatively potent patients 6 (4%) had positive lymph nodes and 21 (14%) had seminal vesicle involvement. Among the patients without seminal vesicle or lymph node involvement 7 (5%) had positive bladder neck margins, 18 (12%) had positive urethral margins and 25 (17%) had positive lateral margins but only 14 (10%) had exclusively positive lateral margins. Only 1 patient had extracapsular tumor extension with negative surgical margins. There was no significant difference between the incidence of positive margins between patients who were and were not potent preoperatively (table 7). All 4 preoperatively potent patients in whom both neurovascular bundles were resected had seminal vesicle or nodal involvement. In contrast, 5 of 9 (55%) preoperatively impotent patients in whom both bundles were excised had organ-confined tumors, while the remaining 4 had seminal vesicle or nodal involvement, or a positive urethral margin. DISCUSSION
A recent report by Walsh of 250 men treated by nervesparing radical prostatectomy who were followed for 1 year or
542
CATALONA AND BIGG
7. Patients with positive margins and/or extracapsular tumor extension who underwent nerve-sparing radical prostatectomy
TABLE
Extracapsular Tumor Extension or Pos. Margins Capsular penetration, neg. margins Pos. margins only:* l lateral Both lateral Posterior Bladder neck Urethra Both lateral + posterior 1 lateral + posterior Posterior + urethra 1 lateral + urethra Both lateral + bladder 1 lateral + bladder Both lateral + urethra Both lateral + bladder + urethra Anterior + bladder + urethra 1 lateral + bladder + urethra 1 lateral + posterior + urethra Totals Pos. seminal vesicles* Pos. lymph nodes* Total extracapsular
Preop. Potent, 145 Pts. No. (%)
Impotent, 63 Pts. No.(%)
1 (1)
0 (0)
11 (8)
8 (13) 1 (2) 1 (2) 0 (0) 4 (6) 1 (2) 0 (0) 0 (0) 1 (2) 1 (2) 0 (0) 1 (2) 0 (0) 1 (2) 1 (2) 1 (2) 21 (33) 6 (10) 0 (0) 27 (43)
3 (2) 1 (1) 3 (2) 13 (9) 3 (2) 2 (1) 2 (1) 2 (1) 2 (1) 1 (1) 0 (0) 1 (1) 0 (0) 0 (0) 0 (0) 44 (30) 21 (14) 6 (4) 72 (50)
* Positive margins in patients with seminal vesicle and/or lymph node involvement are not included.
longer demonstrated that potency was preserved in 72% overall.3 Of these patients 58% recovered potency within 6 months, 95% by 1 year, 99% by 18 months and 100% by 2 years. There was a significant correlation among recovery of potency, patient age and tumor stage. Walsh further reported recovery of potency in 20 of 29 patients (69%) who had 1 neurovascular bundle sacrificed. In a histopathological study of 100 consecutive nerve-sparing radical prostatectomy specimens Eggleston and Walsh reported that 41 % had capsular penetration but only 7% had positive surgical margins.4 In none of t hese patients was the margin positive only at the site of the nervesparing modification. Accordingly, t hese investigators concluded that the adequacy of cancer excision was determined primarily by the extent of the tumor rather than the operative technique. Our results are in general agreement with those of Walsh. 3 However, there are some notable differences between our re1><7 spective series. Our minimal followup intervtl_is-6-months rather than 1 year, which means that our potency rates probably will improve somewhat with further followup. Also, t he age distributions of the 2 series are significantly different, with our series having more older patients (p = 0.0009 by chi-square analysis). Walsh reported that 55% of the patients were less than 60 years old, while only 3% were older than 70 years, compared to only 38 and 11 %, respectively, of our patients. For example, our potency rate for patients with stage A2 disease (50%) is lower than that reported by Walsh (70%). However, in our series only 22% of the stage A2 cancer patients were less than 60 years old compared to 52% reported by Walsh. There is an even more striking disparity in the distribution of clinical stages between the series. In their series 11 % of the patients had stage A, 72% stage Bl and only 17% stage B2 disease, compared to 21, 33 and 46%, respectively, of our patients. This difference in the distribution of clinical stages is highly significant (p <0.00001 by chi-square analysis). These differences may explain, at least partly, our lower over-all rate of preservation of potency (63%) and the fact that Walsh reported 206 of 250 patients (82%) to be potent preoperatively, compared to only 180 of 250 (72%) in our series. Preservation of continence. Urinary continence was achieved by 6 months postoperatively in 188 of 192 patients in our series (98%). No patient had total urinary incontinence but 4 require protective pads. Of the 4 patients who are incontinent 2 had
4J
locally extensive tumors and 1 received early postoperative radiation therapy. Preservation of potency. This discussion on preservation of potency relates to patients who were potent preoperatively, underwent bilateral nerve-sparing radical prostatectomy and were eligible for 6-month followup. The highest incidence of preserved potency was in patients with clinically focal tumors (71 % for t hose with stage Al and 67% for those with stage Bl disease). The incidence of preserved potency was surprisingly low (50%) for patients with stage A2 tumors and surprisingly high (64%) for those with stage B2 tumors. The relatively low rate of preserved potency in patients with clinical stage A disease probably is related to the frequent finding of periprostatic fibrosis or tumor extension that obscured the normal planes of dissection. In several of these patients ant egrade dissection was necessary because of difficulties in establishing the proper plane of dissection at the prostatic apex. Over-all, potency was preserved in 72% of the patients with organconfined tumors compared to only 51 % of those with ext racapsular tumor extension (table 4). There was a more striking correlation between patient age and preservation of potency postoperatively. For example, among patients who underwent a bilateral nerve-sparing operation potency was retained in 81 % of those younger than 50 years, 57% of those 60 to 69 years old and only 33% of those 70 to 79 years old. The highest success was achieved in the 17 patients with clinical stage Bl tumors who were 50 to 59 years old, of whom 15 (88%) retained potency (table 1). Our incidence of preserved potency in patients who had 1 neurovascular bundle deliberately sacrificed was substantially lower (13 of 33 or 39%) than that of Walsh and associates (20 of 29 or 69%). 5 The reason for this discrepancy is not clear. However, among our 7 patients with clinical stage Bl tumors who underwent a unilateral nerve-sparing operation 5 (71%) had preservation of potency, which is roughly comparable to the entire patient group of Walsh (table 1). Clinical understaging. Over-all, t he disease in 43% of the patients was clinically understaged (table 2), which is consistent with previous reports from our and other institutions. 10• 11 Diffuse tumors were understaged more frequently than focal tumors (0% stage Al, 30% stage A2, 22% stage Bl and 68% stage B2, table 2), and high grade tumors were understaged more frequently than low grade tumors (24% of well, 45 % of moderately and 78% of poorly differentiated tumors). Relationship of preservation of potency to completeness of tumor excision. In contrast to the experience of Eggleston and Walsh in which extracapsular tumor extension rarely occurred only in the region of the neurovascular bundles,4 Stamey and associates reported that the majority of patients with ext racapsular tumor extension had extension in t he region of t he ipsilateral neurovascular bundle. 12 Thus, it is of interest to consider the interrelationships among clinical stage, preservation of potency and the apparent completeness of tumor excision, asking what proportion of patients had preservation of potency as well as complete tumor excision. We found that all patients with clinical stage Al prostate cancer who underwent a bilateral nerve-sparing procedure had complete tumor excision as judged by histologically organ-confined tumor and undetectable postoperative prostate specific antigen levels; in 71 % potency also was preserved (table 5). The remaining 2 patients have partial erections that are nearly sufficient for int ercourse and both have been followed for less than 1 year. Among patients with clinical stage A2 tumors 39% had complete tumor excision and preservation of potency compared to 56% of those with stage B 1 tumors and only 24 % of those with clinical stage B2 tumors. Among 33 patients who underwent a unilateral nerve-sparing operation (table 6) only 18% had complete tumor excision and preservation of potency, and potency was preserved in 6% with microscopically positive margins and undetectable prostate specific antigen levels.
NERVE -SPARING RADICAL PROSTAT ECTOMY
1
543
There were 12 patients in the series in whom potency was In a recent editorial comment Walsh indicated that he delibpreserved who had proved inadequate tumor excision as defined erately sacrificed 1 neurovascular bundle in 16% of the patients by positive margins of resection and detectable postoperative with stage BIN, 35% with stage Bl and 80% with stage B2 prostate specific antigen levels. However, only 1 patient (8%) prostate cancer, while he sacrificed both neurovascular bundles had positive lateral margins only. The remaining 11 patients in 5, 4 and 12%, respectively. 13 In our series the neurovascular had either seminal vesicle invasion, positive nodes or positive bundles were sacrificed in a significantly smaller proportion of urethral margins that may not have been influenced by wider patients (table 3). Based upon these results we adopted a more , aggressive approach towards sacrificing the neurovascular bun- ,,surgical excision. It may be argued that the nerve-sparing modification may dles. However, this policy may lead not only to wider tumor compromise the urethral and posterior margins as well as the excision in patients with extracapsular tumor extension but lateral margins. In dissecting out the membranous urethra also to more frequent unnecessary sacrifice of the bundles in without disturbing the paraurethral bands that contain fibro- patients with intracapsular tumors associated with periprosmuscular attachments to the prostatic apex anteriorly and the tatic fibrosis. neurovascular bundles laterally, the surgeon may inadvertently CONCLUSIONS dissect the urethra proximally into the substance of the prosOur results generally confirm previous reports that with the tatic parenchyma. If this occurs the urethra may be transected too proximally, producing a positive urethral margin. A subcap- nerve-sparing technique of radical retropubic prostatectomy sular plane then may be developed along the posterior surface potency can be preserved in the majority of patients without of the prostate while attempting to mobilize the prostate off compromising the adequacy of tumor excision. However, proper the rectum. This procedure denudes the posterior surface of patient selection is important. We believe that the adequacy of the prostate and, thus, also may produce an iatrogenically tumor excision is determined primarily by the extent of the positive posterior margin. We have made this error on several tumor but we also believe that patients who undergo a nerveoccasions, perhaps accounting for some of our positive margins. sparing procedure may be at a higher risk for positive surgical Once the error is recognized our policy is to switch to an margins. The extent to which the incidence of positive margins antegrade dissection, beginning at the bladder neck to establish can be decreased by wide excision of the neurovascular bundles the proper posterior plane of dissection from above and to remains unquantified. Our data provide little evidence that ensure complete excision of the posterior prostatic capsule. To excision of the neurovascular bundles was of much benefit, avoid this problem with apical dissection we routinely expose since most patients in whom the bundles were sacrificed had 1 to 2 cm. of membranous urethra distal to the insertion of the seminal vesicle or nodal involvement, or positive urethral or f r .\ · paraurethral bands to the apex of the prostate. The membra- bladder neck margins that may not have been altered by wider . · nous urethra is transected distal to the apex of the prostate at excision. Patients with focal, well differentiated tumors are the same location as in the performance of a standard radical ideal candidates for a nerve-sparing radical prostatectomy, retropubic prostatectomy. The rectourethralis muscle is iso- while those with bulky, diffuse and poorly differentiated tumors lated with a right-angle clamp and divided distal to the posterior are not good candidates. These latter patients seldom can be lip of the prostatic apex. If these precautions are taken the cured with any form of radical prostatectomy. urethral and posterior margins should not be compromised by REFERENCES the nerve-sparing modification. 1. Walsh, P. C., Lepor, H. and Eggleston, J.C.: Radical prostatectomy Of 145 patients who were potent preoperatively and had a with preservation of sexual function: anatomical and pathologinerve-sparing procedure 72 (50%) had extracapsular tumor or cal considerations. Prostate, 4: 4 73, 1983. tumor with positive margins, 6 (4%) had positive nodes, 21 2. Walsh, P. C.: Radical retropubic prostatectomy. In: Campbell's (14%) had seminal vesicle involvement, 18 (12%) had positive Urology, 5th ed. Edited by P. C. Walsh, R. F . Gittes, A. D. urethral margins and 7 (5 % ) had positive bladder neck margins. Perlmutter and T. A. Stamey. Philadelphia: W. B. Saunders Co., Only 14 patients ( 10%) had exclusively positive lateral margins vol. 3, sect. XV, chapt. 76, p. 2754, 1986. and not all of these were exactly in the region of the neurovas3. Walsh, P . C.: Radical prostatectomy, preservation of sexual function, cancer control: the controversy. Urol. Clin. N. Amer., 14 : cular bundles (table 7) . Thus, it appears that although some of 663, 1987. the potent patients who underwent a nerve-sparing procedure could possibly have had negative surgical margins with a wider ~ 4. Eggleston, J. C. and Walsh, P. C.: Radical prostatectomy with preservation of sexual function: pathological findings in the first lateral dissection and more extensive apical dissection, it is_ ./1 100 cases. J. Urol., 134: 1146, 1985. uncertain how many, if any, would have been rendered free of[t 5 Walsh, P . C., Epstein, J . I. and Lowe, F. C.: Potency following tumor. The anticipated benefits of resecting the neurovascular · radical prostatectomy with wide unilateral excision of the neubundles have not been demonstrated formally. The high incirovascular bundle. J. Urol., 1 38: 823, 1987. dence of positive margins remains a concern. 6. Catalona, W . J. and Dresner, S. M.: Nerve-sparing radical prostaOur results demonstrate that we were unable to distinguish tectomy: extraprostatic tumor extension and preservation of erectile function. J . Urol., 134: 1149, 1985. between capsular penetration of tumor and periprostatic fibro7. Fowler, J . E ., Jr., Clayton, M., Sharifi, R., Mouli, K., Ojeda, L. and sis by the gross appearance at operation. Among 13 patients Ray, P . S.: Early experience with Walsh technique of radical who had bilateral neurovascular bundle excision for presumed retropubic prostatectomy. Urology, 29: 242, 1987. extracapsular tumor extension 5 (38%) had histologically or8. Weldon, V. E. and Tave!, F. R.: Potency-sparing radical perinea! gan-confined cancer and 28 of 62 (45%) who had unilateral prostatectomy: anatomy, surgical technique and initial results. neurovascular bundle excision had intracapsular disease. ConJ. Urol., 140: 559, 1988. versely, 66 of 175 patients (38%) in whom both neurovascular 9. Catalona, W. J., Stein, A. J . and Fair, W. R.: Grading errors in bundles were preserved had histological evidence of extracapprostatic needle biopsies: relation to the accuracy of tumor grade sular tumor extension. Thus, errors were commonly made in in predicting pelvic node metastases. J. Urol., 127: 919, 1982. 10. Catalona, W. J. and Stein, A. J.: Staging errors in clinically both directions. localized prostate cancer. J. Urol., 127: 452, 1982. All of our patients who had capsular penetration in the region Lange, P. H. and Narayan, P.: Understaging and undergrading of of the neurovascular bundles had positive surgical margins. We ll. prostate cancer: argument for postoperative radiation as adjudid not observe capsular penetration of tumor only in the area ~ vant therapy. Urology, 21: 113, 1983. of the neurovascular bundle that was completely covered by an . Stamey, T. A., McNeal, J. E., Freiha, F. S. and Redwine, E.: excised neurovascular bundle. We cannot exclude the possibilMorphometric and clinical studies on 68 consecutive radical ity that had we been more aggressive in resecting the neuroprostatectomies. J. Urol., 139: 1235, 1988. 13. Walsh, P. C.: Editorial comment. J. Urol., 139: 1240, 1988. vascular bundles we might have observed this phenomenon.
,(5
·--·-" -- 1
THE JOURNAL OF UROLOGY
Printed in U.S.A.
Copyright © 1990 by AMERICAN UROLOGICAL ASSO CIATION , INC.
NERVE-SPARING RADICAL PROSTATECTOMY: EVALUATION OF RESULTS AFTER 250 PATIENTS WILLIAM J. CATALONA
\
, ..»'-G3\
_ \ )~ ~
0,V
~
'\
/
AND
STEVEN W. BIGG
From the Division of Urologic Surgery, Washington University School of Medicine, St. Louis, Missouri
J._. ~~-:
"
dolo \
ABSTRACT
W'':::J:.. ~ c...,# ~
\)
~
·\
To examine the efficacy of nerve-sparing radical retropubic prostatectomy in preserving sexual potency and urinary continence, and in providing complete tumor excision we analyzed the records of the first 250 consecutive patients with clinical stage A or B prostate cancer treated since this operation was adopted at our institution. Over-all, sexual potency was preserved in 71 of 112 patients (63%) who underwent bilateral nerve-sparing prostatectomy and 13 of 33 (39%) who underwent a unilateral nerve-sparing procedure with a minimum of 6 months of followup. Preservation of potency correlated with patient age (p equals 0.0035, chi-square) and was significantly (p less than 0.001, chi-square) higher in patients with pathologically organ-confined tumors (72%) than in those with pathologically extracapsular tumors (51%). Of 192 patients followed for at least 6 months 188 (98 %) achieved urinary continence postoperatively. Over-all, apparent complete tumor excision as defined by organ-confined tumor with negative surgical margins and undetectable postoperative prostate specific antigen levels was achieved in 14 preoperatively potent patients (42 %) who underwent a unilateral and 67 (59%) who underwent a bilateral nerve-sparing procedure. Completeness of tumor excision correlated with tumor stage. In approximately 45% of the patients incomplete tumor excision was owing to seminal vesicle and/or lymph node involvement or positive bladder neck margins that could not be attributed to the nerve-sparing modification. However, improper application of the nerve-sparing technique may have contributed in the others. We were unable to detect microscopic penetration of the capsule or distinguish between gross extracapsular tumor extension and periprostatic fibrosis at operation. We conclude that with proper application of nerve-sparing radical retropubic prostatectomy, potency can be preserved in the majority of patients without compromising the adequacy of tumor excision. The completeness of tumor excision appears to be determined primarily by the extent of the tumor. Therefore, patient selection is important. Patients with focal, well differentiated tumors are ideal candidates for a nerve-sparing procedure, while those with high volume, poorly differentiated tumors may be at a higher risk for positive surgical margins. The benefits of wide excision of the neurovascular bundles remain to be demonstrated formally. (J. Ural. , 143: 538-544, 1990)
In 1983 Walsh and associates introduced the nerve-sparing modification of radical retropubic prostatectomy in which the prostate is excised without injuring the neurovascular bundles that contain the cavernous nerves and vessels to preserve erectile potency.1 Since then, Walsh has modified and improved upon this operation and has demonstrated that with the nervesparing technique potency can be preserved in the majority of patients with organ-confined tumors.2• 3 Moreover, he has shown that 1 or both neurovascular bundles can be sacrificed deliberately in patients with evidence of extraprostatic tumor extension at operation. In so doing it is possible to obtain wider margins of resection than usually are achieved with either standard radical perineal or radical retropubic prostatectomy.4 Walsh and associates also demonstrated that erectile potency is preserved in the majority of patients in whom only 1 neurovascular bundle has been excised.5 Other than these studies, there have been few reports on the results of nerve-sparing radical prostatectomy.6- 8 We adopted the nerve-sparing radical prostatectomy in May 1983 and reported on our early results in 40 patients in 1985,6 documenting postoperative return of erectile function and correlating the incidence of positive surgical margins with the clinical stage of tumor and the technique of prostatectomy used (standard versus nerve-sparing). In the present study we analyzed and updated our results in achieving preservation of continence, potency and complete tumor excision in 250 consecutive patients treated with radical prostatectomy since the Accepted for publication September 18, 1989.
nerve-sparing modification was adopted at our institution. We correlated our results with relevant clinical parameters, and also evaluated our ability to distinguish accurately between periprostatic fibrosis and extracapsular tumor extension at operation. The results have altered our approach to radical prostatectomy in some patients with clinically localized prostate cancer. PATIENTS AND METHODS
Patients. A total of 250 consecutive private patients treated by 1 of us (W. J. C.) with clinical stage A or B prostate cancer underwent pelvic lymphadenectomy and radical retropubic prostatectomy between May 6, 1983 and November 30, 1988 using a technique similar to that described by Walsh and associates. 1- 3 All patients had negative pelvic lymph nodes on frozen section at operation. Clinical staging. The disease in all patients was staged with enzymatic serum prostatic acid phosphatase determinations using the thymolphthalein monophosphate substrate and radionuclide bone scans. Most patients also were evaluated by prostate specific antigen levels and pelvic computerized tomographic (CT) scans. Many patients also were evaluated with transrectal prostatic ultrasonography or magnetic resonance imaging. With a modification of the Whitmore-Jewett system the patients were classified as having clinical stage Al disease if they had clinically unsuspected well differentiated prostate cancer in fewer than 5% of a prostatectomy specimen removed
538
539
NERVE-SPARING RADICAL PROSTATECTOMY
for presumed benign hyperplasia. They were considered to have stage A2 disease if more than 5% of the specimen was involved, or if there was moderately or poorly differentiated carcinoma. Patients were classified as having clinical stage B 1 disease if they had palpable carcinoma of less than 2 cm. in diameter involving less than 1 lobe of the prostate and judged to be confined within the prostatic capsule. Stage B2 disease denoted palpable tumors involving both lobes of the prostate or induration of greater than 2 cm. in diameter judged to be confined within the prostatic capsule. All patients with clinical stage A or B disease (except 1) had a normal serum acid phosphatase level and a bone scan with confirmatory radiographs that showed no evidence of metastases. Findings on pelvic CT scans, transrectal prostatic ultrasonography or magnetic resonance imaging were not considered sufficient to exclude these patients from radical prostatectomy. Examination of surgical specimens. At operation the bladder neck and urethral margins were removed from the prostate by grasping the retracted urethra, including its mucosa! edges, with small Allis forceps and circumferentially excising the 2 to 3 mm. urethral stump. The bladder neck similarly was excised circumferentially, including the mucosa! edges and bladder neck muscle. These margins were sent as separate specimens. If the histological examination revealed cancer the surgical margins were considered to be positive. The excised prostate was dipped in india ink and immediately fixed in Bouin's solution to prevent the india ink from spreading into the tissue planes. The specimen then was "breadloafed" in 2 to 5 mm. segments, depending on the size of the gland, and fixed overnight in 10% phosphate buffered formalin solution. Then, 3 hematoxylin and eosin-stained sections from the right half and 3 from the left half of the prostate were examined for the peripheral surgical margins. In addition, 2 longitudinal sections through the base of the seminal vesicles were examined on each side. Two longitudinal sections through the apex of the prostate also were examined. The presence of cancer at the prostatic apex but not in the urethral margin was considered to be a positive anterior and/or posterior margin. If the tumor extended to an inked margin it was called a positive lateral, anterior or posterior surgical margin. Pathological staging. Pathological stages A and B were defined as for clinical staging except that the stage was verified histologically. Pathological stage C was divided into substages: stage Cl denoted microscopic extracapsular tumor extension and stage C2 referred to tumor extension to the fascia surrounding the seminal vesicles or the muscular walls of the seminal vesicle(s) proper with or without microscopically positive margins. Pathological stage Dl referred to patients with clinical stage A or B disease who at pelvic lymphadenectomy had histologically documented lymph node metastases. Organ-confined disease referred to patients with pathological stage A or B cancer. Extracapsular disease denoted pathological stage C or D tumor, and included all patients with positive margins of resection. Tumor grading. Tumors were graded according to the system used at Barnes Hospital into well, moderate or poorly differentiated adenocarcinoma. Followup. Patients were followed at 3 to 6-month intervals for 2 years and thereafter at approximately 6-month intervals. The last date of followup for this analysis was January 4, 1989. At followup a history, digital rectal examination, prostate specific antigen level and acid phosphatase determinations were obtained. Bone scans, CT scans or other tests were not performed routinely unless clinically indicated. Preoperative and postoperative erectile function. Preoperative and postoperative erectile function was determined by history. Information also was obtained from sexual partners when feasible. Postoperatively, the patient was classified as potent if the erection was sufficient for vaginal penetration and sexual
intercourse or not potent if the erection was insufficient for vaginal penetration. Urinary continence. The patients were considered to be continent if they did not require a protective pad to keep the clothing dry. Patients who were dry under virtually all normal ) circumstances but who elected to wear a precautionary pad for occasional leakage of a few drops of urine with severe coughing or straining also were considered as being continent. Nerve-sparing procedure. Throughout t his series our primary goal was to achieve complete tumor excision with preservation of potency being of secondary importance. In patients in whom evidence suggested gross extracapsular tumor extension at operation the neurovascular bundle in question was deliberately sacrificed by wide ipsilateral excision, dissecting the bundle off of the anterior surface of the rectum. In patients who were not potent preoperatively no attempt was made to preserve the neurovascular bundles. In these patients the bundles were taken wider (as in our former standard radical prostatectomy) than in potent patients but the entire neurovascular bundles were not deliberately and completely dissected off the anterior rectal wall (as in a superradical prostatectomy) unless there was evidence of extracapsular tumor extension. RESULTS
Patient age. Mean patient age over-all was 64 years (table 1). The mean age was 61 years for patients with clinical stage Al, 65 for those with stage A2, 62 for t hose with stage Bl and 65 for those with stage B2 disease. Clinical stage. Of the patients 13 had clinical stage Al, 40 stage A2, 82 stage Bl and 115 stage B2 disease (table 2). All except 1 patient had normal preoperative serum prostatic acid 1. Comparison of the preservation of potency as a func tion of clinical stage, age and type of nerve-sparing procedure performed in patients who were potent preoperatively and were followed for at least 6 months
TABLE
Clinical Stage
Pt. Age
U nilat. Procedure No. Potent/ Total (%)
40-49 50- 59 60-69 70-79
Al
Totals 40-49 50- 59 60-69 70- 79
A2
40-49 50-59 60- 69 70- 79
Totals B2
(0)
2/3 0/1 2/5
(67) (0) (40)
0/2
(0)
(0) 0/2 1/1 (100) (0) 0/ 1 2/ 3 (67) 2/ 2 (100) 5/7 (71)
Totals Bl
0/1
Bilat. Procedure No. Potent/ Total(%)
40-49 50- 59 60- 69 70-79
Totals
(80) (50)
5/7 0/ 1 2/3 5/ 10 2/ 4 9/18
(71) (0) (67) (50) (50) (50)
(25) (20) (60) (32)
(67) 2/ 3 15/ 17 (88) 13/24 (54) (0) 0/1 30/45 (67) 1/ 1 (100) 10/12 (83) 14/22 (64) (20) 2/7 27/ 42 (64)
13/33 (39)
71/ 112 (63)
1/4 2/10 3/5 6/19
Totals
4/5 1/2
Relationship between clinical and pathological stage in 250 consecutive patients undergoing radical prostatectomy
TABLE 2.
Clinical Stage Al A2 Bl B2 Totals
Pathological Stage T otals Al
A2
7
6 26
7
32
Bl
B2
Cl
C2
01
48
16 39
8
55
0 2 0 30
0 0 0
48
0 12 18 38
68
32
8
13 40 82 115 250
r,.A;,
540
CATALONA AND BIGG
phosphatase levels. The patient with an elevated acid phosphatase titer was believed to have hepatic dysfunction. It was considered that the acid phosphatase titer may have been elevated on this basis. Thus, he was offered radical prostatectomy. A pathologically organ-confined tumor was found and postoperatively this patient maintained an undetectable prostate specific antigen level with persistent mild elevation of acid phosphatase and no evidence of recurrent tumor for 24 months. Of 53 patients classified as having clinical stage A disease 45 (85%) were referred for radical prostatectomy after transurethral prostatectomy had been done elsewhere for clinical benign hyperplasia. Many of these patients had palpable abnormalities in the prostate gland after transurethral resection. However, it was impossible to distinguish between postoperative changes and baseline abnormalities. Therefore, these patients were classified as having clinical stage Al or A2 disease as stated previously. Pathological stage. Pathological stage was Al in 7 patients, vf A2 in 32, Bl in 48, B2 in 55, Cl in 68, C2 in 32 and Dl in 8. (15, All but 1 patient with patholo ical sta e Cl disease had ositive i;;urgica margins. Many of these patients had multiple sites of capsular penetration. We did not observe a case in which ~ e was capsular penetrationexcTusivelyin-tne regionoftlie neurov~ar15un e that was covere -by"tl1e bunaie wiUia negative surgical margin. The distribution of pathological stages as a function of clinical stage is shown in table 2. None of the 13 stage Al cancer patients had extraprostatic tumor spread, although 6 had more extensive or more undifferentiated tumor within the prostate. In contrast, the disease in 14 of 40 patients (30%) with stage A2, 18 of 82 (22%) with stage B 1 and 76 of 115 (68 %) with stage B2 cancer was clinically understaged. Over-all, clinical understaging occurred in 108 of the 250 patients (43%) with clinical stage A or B prostate cancer. Tumor grade in relation to clinical understaging. Occasionally, there was a discrepancy between the tumor grade of the diagnostic biopsy and that of the radical prostatectomy specimen. This topic has been considered in detail previously9 and will not be addressed further. Clinical understaging occurred in 18 of 75 patients (24%) with well differentiated, 65 of 143 (45%) with moderately differentiated and 25 of 32 (78%) with poorly differentiated tumors. The correlation between tumor grade and clinical understaging was significant at a 0.00001 level by chi-square analysis. Nerve-sparing procedure. Deliberate wide excision of the neurovascular bundle was performed unilaterally in 62 patients (25%) and bilaterally in 13 (5%). In the remaining patients who were potent preoperatively an attempt was made to mobilize the prostate medial and anterior to the neurovascular bundles. In most patients the retrograde approach as described by Walsh 3 was used. However, in some patients in whom the plane between the prostate and rectum was obliterated at the prostatic apex an antegrade approach was used. The relationship between clinical stage and the attempt to perform bilateral or unilateral nerve-sparing is shown in table
61
TABLE
3. Type of nerve-sparing procedure performed as a function of clinical stage in 250 consecutive patients Nerve-Sparing Procedure
Clinical Stage Al A2 Bl B2 Totals
Bilat.• No.(%)
Unilat.* No.(%)
8 32 68 67 175
5 5 13 39 62
(61) (80) (83) (58) (70)
(38) (13) (16) (34) (25)
None No.(%) 0 3 1 9 13
(7) (1) (8) (5)
Total No. 13 40 82 115 250
* Not all of these patients were potent preoperatively. In impotent patients neurovascular bundles were not widely excised unless there was evidence of tumor extension.
TABLE 4. Preservation of potency as a function of pathological stage and the type of nerve-sparing procedure performed in 145 patients who were potent preoperatively and followed for 6 months or longer
Pathological Stage A B Cl C2 Dl Totals
U nilat. Procedure No./Total (%) 2/6 5/9 3/11 2/5 1/2 13/33
(33) (56) (27) (40) (50) (39)
Bilat. Procedure No./Total (%) 12/19 36/48 15/29 6/12 2/4 71/112
(63) (75) (52) (50) (50) (63)
3. Bilateral nerve-sparing was attempted in 61 % of the patients with clinical stage Al, 80% with stage A2, 83% with stage Bl but only 58% with stage B2 tumors. Unilateral nerve-sparing was attempted in 38% of the patients with stage Al, 13% with stage A2, 16% with stage Bl and 34% with stage B2 tumors. Followup. Of the patients 58 were followed for less than 6 months postoperatively, 192 for at least 6 months, 161 for 12 months, 102 for 24 months, 61 for 36 months, 28 for 48 months and 6 for 60 months. Complications. The complications of urinary incontinence and sexual impotency are addressed. Other postoperative complications will be the topic of a separate report and will not be discussed further in this study. Postoperative urinary continence. Among 92 patients who were followed for at least 6 months postoperatively 188 (98%) are continent. No patient had total urinary incontinence but 4 have required the use of pads to keep the clothing dry. Preservation of potency. Over-all, 180 of the 250 patients (72%) were potent preoperatively but only 145 fulfilled the criteria of being potent preoperatively, having undergone either a unilateral or bilateral nerve-sparing procedure and being eligible for at least 6 months of followup. Of these patients 71 of 112 (63%) who underwent bilateral nerve-sparing (tables 1, 4 and 5) and 13 of 33 (39%) who underwent unilateral nervesparing (tables 4 and 5) had potency preserved (table 1). Bilateral nerve-sparing procedure: patient age, tumor stage and potency. The interrelationships among patient age, clinic;al stage and preservation of potency in patients who underwent a bilateral nerve-sparing procedure are shown in table 1. Potency was preserved in 71 % of the patients with clinical stage Al, 50% with stage A2, 67% with stage Bl and 64% with stage B2 disease. Potency was preserved in 81 % of the patients 40 to 59 years old, 57% of those 60 to 69 years old and 33% of those 70 to 79 years old. This correlation between patient age and preservation of potency is statistically significant (p = 0.0035, chi-square) . The relationship of pathological stage to preservation of potency in patients who underwent a bilateral nerve-sparing procedure is shown in table 4. Potency was preserved in 63% of the patients with pathological stage A, 75% with stage B, 52% with stage Cl, 50% with stage C2 and 50% with stage Dl disease. The correlation between preservation of potency in patients with organ-confined tumors versus those with extracapsular extension is significant (p <0.0001 by chi-square analysis). Preservation of potency with unilateral nerve-sparing. The interrelationships among clinical stage, patient age and preservation of potency in patients who underwent a unilateral nerve-sparing procedure are shown in table 1. Potency was preserved in 40% of the patients with clinical stage Al, 0% (O of 2) with stage A2, 71 % with stage Bl and 32% with stage B2 tumors. The small numbers of patients in the various groups preclude meaningful correlations. The relationship between pathological stage and preservation of potency in patients who underwent a unilateral nerve-sparing procedure also is shown in table 4. Potency was preserved after unilateral nerve-sparing prostatectomy in 33% of the patients with pathological stage A, 56% with stage B, 27% with
541
NERVE-SPARING RADICAL PROSTATECTOMY TABLE 5. i
~
Relationship among clinical stage, preservation of potency and completeness of tumor excision in 112 patients who were potent preoperatively, underwent a bilateral nerve-sparing procedure and were eligible for 6-month followup
'
I
Ii
Extracapsular Extension + Potent
Pathological Stage Clinical Stage
No.
OrganConfined Disease
Confined, Prostate Specific Antigen :ai0.6
7 12 36 12
7 12 35 12
67
66
Confined, Prostate Specific Antigen :ai0.6, Potent No.(%)
C
D1
0 6 9 26
0 0 0 4
41
4
;i
~
Al A2 Bl B2 Totals
7 18 45 42 112
5 7 25 10 47
(71) (39) (56)* (24) (42)
Prostate Specific Antigen <0.6 No.(%)
Prostate Specific Antigen >0.6 No.(%)
0 1 (6) 4 (9)* 10 (24) 15 (13)
0 1 (6) 0 7 (17) 8 (7)
* The apparent discrepancy between this table listing 29 potent stage Bl cancer patients and table 1 listing 30 potent clinical stage Bl cancer patients is that 1 potent patient with organ-confined tumor had a postoperative prostate specific antigen level of 0.7 ng./ml. and, thus, did not fulfill the criteria for having complete tumor excision plus preservation of potency. TABLE 6.
Relationship among clinical stage, preservation of potency and completeness of tumor excision in 33 patients who were potent preoperatively, underwent a unilateral nerve-sparing procedure and were eligible for 6-month followup
Clinical Stage
Al A2 Bl B2 Totals
No.
OrganConfined
Confined, Prostate Specific Antigen :ai0.6
19
5 1 3 6
5 1 2 6
33
15
14
5 2 7
Confined, Prostate Specific Antigen :ai0.6, Potent No.(%) 2 0 2 2 6
(40) (0) (28)* (11) (18)
Pathological Stage
C
Dl
0 1 4 11
0 0 0 2
16
2
Extracapsular Extension Potent
+
Prostate Specific Antigen <0.6 No.(%)
Prostate Specific Antigen >0.6 No.(%)
0 0 2 (29)* 0 2 (6)
0 0 0 4 (21) 4 (12)
* The apparent discrepancy between this table listing 4 potent stage Bl cancer patients and table 1 listing 5 potent stage Bl cancer patients is that 1 patient with organ-confined tumor had a postoperative prostate specific antigen level of 1.5 ng./ml. and, thus, did not fulfill the criteria of having complete tumor excision plus preservation of potency.
stage Cl, 40% with stage C2 and 50% with stage Dl tumors. There was no significant correlation between preservation of potency in patients with organ-confined tumors versus those with extracapsular extension among patients who underwent a unilateral nerve-sparing operation. Relationship among clinical stage, complete tumor excision and preservation of potency. The relationship between complete tumor excision and preserved potency as a function of clinical stage in patients who were potent preoperatively, underwent a bilateral nerve-sparing procedure and were eligible for 6-month followup is shown in table 5. Organ-confined tumor and undetectable postoperative prostate specific antigen levels were noted in all 7 patients with clinical stage Al disease, 12 of 18 potent patients with stage A2 cancer, 35 of 45 potent patients with clinical stage B 1 cancer and 12 of 42 potent patients (29%) with stage B2 disease, and potency was preserved in 5 (71 %), 7 (35%), 25 (56%) and 10 (24%), respectively. In addition, potency was preserved in 2 other stage A2 cancer patients with extracapsular tumor extension (1 with an undetectable and 1 with a detectable postoperative prostate specific antigen level), 4 other stage Bl cancer patients (9%) with microscopic tumor extension (all with undetectable postoperative prostate specific antigen levels) and 17 stage B2 cancer patients with extracapsular tumor extension (10 with undetectable and 7 with detectable prostate specific antigen levels postoperatively). The relationship between complete tumor excision and preserved potency as a function of clinical stage in patients who were potent preoperatively, underwent a unilateral nerve-sparing procedure and were eligible for 6-month followup is shown in table 6. Over-all, of these 33 patients only 6 (18%) had complete tumor excision and preservation of potency. Six patients with extracapsular tumor extension retained potency, including 2 with undetectable and 4 with detectable postoperative prostate specific antigen levels. Over-all, 12 of the 145 eligible patients (8%) had retained potency with proved inadequate tumor excision as defined by
positive margins of resection and detectable postoperative prostate specific antigen levels (tables 5 and 6). Of these patients 6 (50%) had seminal vesicle invasion, 3 (25%) had positive urethral margins and 2 (17%) had positive nodes. One patient (8%) had positive lateral margins only. Intraoperative assessment of tumor extension versus fibrosis. Among 76 patients in whom 1 or both neurovascular bundles were sacrificed 33 (43%) had histologically organ-confined tumors. Conversely, among 175 patients in whom neither neurovascular bundle was sacrificed 66 (38%) had histological evidence of extracapsular tumor extension. Positive tumor margins. Table 7 shows an analysis of the extent of extracapsular tumor extension in the patients who were potent preoperatively and underwent a nerve-sparing procedure for a tumor with positive margins and/or extracapsular tumor extension. Of the 145 preoperatively potent patients 6 (4%) had positive lymph nodes and 21 (14%) had seminal vesicle involvement. Among the patients without seminal vesicle or lymph node involvement 7 (5%) had positive bladder neck margins, 18 (12%) had positive urethral margins and 25 (17%) had positive lateral margins but only 14 (10%) had exclusively positive lateral margins. Only 1 patient had extracapsular tumor extension with negative surgical margins. There was no significant difference between the incidence of positive margins between patients who were and were not potent preoperatively (table 7). All 4 preoperatively potent patients in whom both neurovascular bundles were resected had seminal vesicle or nodal involvement. In contrast, 5 of 9 (55%) preoperatively impotent patients in whom both bundles were excised had organ-confined tumors, while the remaining 4 had seminal vesicle or nodal involvement, or a positive urethral margin. DISCUSSION
A recent report by Walsh of 250 men treated by nervesparing radical prostatectomy who were followed for 1 year or
542
CATALONA AND BIGG
7. Patients with positive margins and/or extracapsular tumor extension who underwent nerve-sparing radical prostatectomy
TABLE
Extracapsular Tumor Extension or Pos. Margins Capsular penetration, neg. margins Pos. margins only:* l lateral Both lateral Posterior Bladder neck Urethra Both lateral + posterior 1 lateral + posterior Posterior + urethra 1 lateral + urethra Both lateral + bladder 1 lateral + bladder Both lateral + urethra Both lateral + bladder + urethra Anterior + bladder + urethra 1 lateral + bladder + urethra 1 lateral + posterior + urethra Totals Pos. seminal vesicles* Pos. lymph nodes* Total extracapsular
Preop. Potent, 145 Pts. No. (%)
Impotent, 63 Pts. No.(%)
1 (1)
0 (0)
11 (8)
8 (13) 1 (2) 1 (2) 0 (0) 4 (6) 1 (2) 0 (0) 0 (0) 1 (2) 1 (2) 0 (0) 1 (2) 0 (0) 1 (2) 1 (2) 1 (2) 21 (33) 6 (10) 0 (0) 27 (43)
3 (2) 1 (1) 3 (2) 13 (9) 3 (2) 2 (1) 2 (1) 2 (1) 2 (1) 1 (1) 0 (0) 1 (1) 0 (0) 0 (0) 0 (0) 44 (30) 21 (14) 6 (4) 72 (50)
* Positive margins in patients with seminal vesicle and/or lymph node involvement are not included.
longer demonstrated that potency was preserved in 72% overall.3 Of these patients 58% recovered potency within 6 months, 95% by 1 year, 99% by 18 months and 100% by 2 years. There was a significant correlation among recovery of potency, patient age and tumor stage. Walsh further reported recovery of potency in 20 of 29 patients (69%) who had 1 neurovascular bundle sacrificed. In a histopathological study of 100 consecutive nerve-sparing radical prostatectomy specimens Eggleston and Walsh reported that 41 % had capsular penetration but only 7% had positive surgical margins.4 In none of t hese patients was the margin positive only at the site of the nervesparing modification. Accordingly, t hese investigators concluded that the adequacy of cancer excision was determined primarily by the extent of the tumor rather than the operative technique. Our results are in general agreement with those of Walsh. 3 However, there are some notable differences between our re1><7 spective series. Our minimal followup intervtl_is-6-months rather than 1 year, which means that our potency rates probably will improve somewhat with further followup. Also, t he age distributions of the 2 series are significantly different, with our series having more older patients (p = 0.0009 by chi-square analysis). Walsh reported that 55% of the patients were less than 60 years old, while only 3% were older than 70 years, compared to only 38 and 11 %, respectively, of our patients. For example, our potency rate for patients with stage A2 disease (50%) is lower than that reported by Walsh (70%). However, in our series only 22% of the stage A2 cancer patients were less than 60 years old compared to 52% reported by Walsh. There is an even more striking disparity in the distribution of clinical stages between the series. In their series 11 % of the patients had stage A, 72% stage Bl and only 17% stage B2 disease, compared to 21, 33 and 46%, respectively, of our patients. This difference in the distribution of clinical stages is highly significant (p <0.00001 by chi-square analysis). These differences may explain, at least partly, our lower over-all rate of preservation of potency (63%) and the fact that Walsh reported 206 of 250 patients (82%) to be potent preoperatively, compared to only 180 of 250 (72%) in our series. Preservation of continence. Urinary continence was achieved by 6 months postoperatively in 188 of 192 patients in our series (98%). No patient had total urinary incontinence but 4 require protective pads. Of the 4 patients who are incontinent 2 had
4J
locally extensive tumors and 1 received early postoperative radiation therapy. Preservation of potency. This discussion on preservation of potency relates to patients who were potent preoperatively, underwent bilateral nerve-sparing radical prostatectomy and were eligible for 6-month followup. The highest incidence of preserved potency was in patients with clinically focal tumors (71 % for t hose with stage Al and 67% for those with stage Bl disease). The incidence of preserved potency was surprisingly low (50%) for patients with stage A2 tumors and surprisingly high (64%) for those with stage B2 tumors. The relatively low rate of preserved potency in patients with clinical stage A disease probably is related to the frequent finding of periprostatic fibrosis or tumor extension that obscured the normal planes of dissection. In several of these patients ant egrade dissection was necessary because of difficulties in establishing the proper plane of dissection at the prostatic apex. Over-all, potency was preserved in 72% of the patients with organconfined tumors compared to only 51 % of those with ext racapsular tumor extension (table 4). There was a more striking correlation between patient age and preservation of potency postoperatively. For example, among patients who underwent a bilateral nerve-sparing operation potency was retained in 81 % of those younger than 50 years, 57% of those 60 to 69 years old and only 33% of those 70 to 79 years old. The highest success was achieved in the 17 patients with clinical stage Bl tumors who were 50 to 59 years old, of whom 15 (88%) retained potency (table 1). Our incidence of preserved potency in patients who had 1 neurovascular bundle deliberately sacrificed was substantially lower (13 of 33 or 39%) than that of Walsh and associates (20 of 29 or 69%). 5 The reason for this discrepancy is not clear. However, among our 7 patients with clinical stage Bl tumors who underwent a unilateral nerve-sparing operation 5 (71%) had preservation of potency, which is roughly comparable to the entire patient group of Walsh (table 1). Clinical understaging. Over-all, t he disease in 43% of the patients was clinically understaged (table 2), which is consistent with previous reports from our and other institutions. 10• 11 Diffuse tumors were understaged more frequently than focal tumors (0% stage Al, 30% stage A2, 22% stage Bl and 68% stage B2, table 2), and high grade tumors were understaged more frequently than low grade tumors (24% of well, 45 % of moderately and 78% of poorly differentiated tumors). Relationship of preservation of potency to completeness of tumor excision. In contrast to the experience of Eggleston and Walsh in which extracapsular tumor extension rarely occurred only in the region of the neurovascular bundles,4 Stamey and associates reported that the majority of patients with ext racapsular tumor extension had extension in t he region of t he ipsilateral neurovascular bundle. 12 Thus, it is of interest to consider the interrelationships among clinical stage, preservation of potency and the apparent completeness of tumor excision, asking what proportion of patients had preservation of potency as well as complete tumor excision. We found that all patients with clinical stage Al prostate cancer who underwent a bilateral nerve-sparing procedure had complete tumor excision as judged by histologically organ-confined tumor and undetectable postoperative prostate specific antigen levels; in 71 % potency also was preserved (table 5). The remaining 2 patients have partial erections that are nearly sufficient for int ercourse and both have been followed for less than 1 year. Among patients with clinical stage A2 tumors 39% had complete tumor excision and preservation of potency compared to 56% of those with stage B 1 tumors and only 24 % of those with clinical stage B2 tumors. Among 33 patients who underwent a unilateral nerve-sparing operation (table 6) only 18% had complete tumor excision and preservation of potency, and potency was preserved in 6% with microscopically positive margins and undetectable prostate specific antigen levels.
NERVE -SPARING RADICAL PROSTAT ECTOMY
1
543
There were 12 patients in the series in whom potency was In a recent editorial comment Walsh indicated that he delibpreserved who had proved inadequate tumor excision as defined erately sacrificed 1 neurovascular bundle in 16% of the patients by positive margins of resection and detectable postoperative with stage BIN, 35% with stage Bl and 80% with stage B2 prostate specific antigen levels. However, only 1 patient (8%) prostate cancer, while he sacrificed both neurovascular bundles had positive lateral margins only. The remaining 11 patients in 5, 4 and 12%, respectively. 13 In our series the neurovascular had either seminal vesicle invasion, positive nodes or positive bundles were sacrificed in a significantly smaller proportion of urethral margins that may not have been influenced by wider patients (table 3). Based upon these results we adopted a more , aggressive approach towards sacrificing the neurovascular bun- ,,surgical excision. It may be argued that the nerve-sparing modification may dles. However, this policy may lead not only to wider tumor compromise the urethral and posterior margins as well as the excision in patients with extracapsular tumor extension but lateral margins. In dissecting out the membranous urethra also to more frequent unnecessary sacrifice of the bundles in without disturbing the paraurethral bands that contain fibro- patients with intracapsular tumors associated with periprosmuscular attachments to the prostatic apex anteriorly and the tatic fibrosis. neurovascular bundles laterally, the surgeon may inadvertently CONCLUSIONS dissect the urethra proximally into the substance of the prosOur results generally confirm previous reports that with the tatic parenchyma. If this occurs the urethra may be transected too proximally, producing a positive urethral margin. A subcap- nerve-sparing technique of radical retropubic prostatectomy sular plane then may be developed along the posterior surface potency can be preserved in the majority of patients without of the prostate while attempting to mobilize the prostate off compromising the adequacy of tumor excision. However, proper the rectum. This procedure denudes the posterior surface of patient selection is important. We believe that the adequacy of the prostate and, thus, also may produce an iatrogenically tumor excision is determined primarily by the extent of the positive posterior margin. We have made this error on several tumor but we also believe that patients who undergo a nerveoccasions, perhaps accounting for some of our positive margins. sparing procedure may be at a higher risk for positive surgical Once the error is recognized our policy is to switch to an margins. The extent to which the incidence of positive margins antegrade dissection, beginning at the bladder neck to establish can be decreased by wide excision of the neurovascular bundles the proper posterior plane of dissection from above and to remains unquantified. Our data provide little evidence that ensure complete excision of the posterior prostatic capsule. To excision of the neurovascular bundles was of much benefit, avoid this problem with apical dissection we routinely expose since most patients in whom the bundles were sacrificed had 1 to 2 cm. of membranous urethra distal to the insertion of the seminal vesicle or nodal involvement, or positive urethral or f r .\ · paraurethral bands to the apex of the prostate. The membra- bladder neck margins that may not have been altered by wider . · nous urethra is transected distal to the apex of the prostate at excision. Patients with focal, well differentiated tumors are the same location as in the performance of a standard radical ideal candidates for a nerve-sparing radical prostatectomy, retropubic prostatectomy. The rectourethralis muscle is iso- while those with bulky, diffuse and poorly differentiated tumors lated with a right-angle clamp and divided distal to the posterior are not good candidates. These latter patients seldom can be lip of the prostatic apex. If these precautions are taken the cured with any form of radical prostatectomy. urethral and posterior margins should not be compromised by REFERENCES the nerve-sparing modification. 1. Walsh, P. C., Lepor, H. and Eggleston, J.C.: Radical prostatectomy Of 145 patients who were potent preoperatively and had a with preservation of sexual function: anatomical and pathologinerve-sparing procedure 72 (50%) had extracapsular tumor or cal considerations. Prostate, 4: 4 73, 1983. tumor with positive margins, 6 (4%) had positive nodes, 21 2. Walsh, P. C.: Radical retropubic prostatectomy. In: Campbell's (14%) had seminal vesicle involvement, 18 (12%) had positive Urology, 5th ed. Edited by P. C. Walsh, R. F . Gittes, A. D. urethral margins and 7 (5 % ) had positive bladder neck margins. Perlmutter and T. A. Stamey. Philadelphia: W. B. Saunders Co., Only 14 patients ( 10%) had exclusively positive lateral margins vol. 3, sect. XV, chapt. 76, p. 2754, 1986. and not all of these were exactly in the region of the neurovas3. Walsh, P . C.: Radical prostatectomy, preservation of sexual function, cancer control: the controversy. Urol. Clin. N. Amer., 14 : cular bundles (table 7) . Thus, it appears that although some of 663, 1987. the potent patients who underwent a nerve-sparing procedure could possibly have had negative surgical margins with a wider ~ 4. Eggleston, J. C. and Walsh, P. C.: Radical prostatectomy with preservation of sexual function: pathological findings in the first lateral dissection and more extensive apical dissection, it is_ ./1 100 cases. J. Urol., 134: 1146, 1985. uncertain how many, if any, would have been rendered free of[t 5 Walsh, P . C., Epstein, J . I. and Lowe, F. C.: Potency following tumor. The anticipated benefits of resecting the neurovascular · radical prostatectomy with wide unilateral excision of the neubundles have not been demonstrated formally. The high incirovascular bundle. J. Urol., 1 38: 823, 1987. dence of positive margins remains a concern. 6. Catalona, W . J. and Dresner, S. M.: Nerve-sparing radical prostaOur results demonstrate that we were unable to distinguish tectomy: extraprostatic tumor extension and preservation of erectile function. J . Urol., 134: 1149, 1985. between capsular penetration of tumor and periprostatic fibro7. Fowler, J . E ., Jr., Clayton, M., Sharifi, R., Mouli, K., Ojeda, L. and sis by the gross appearance at operation. Among 13 patients Ray, P . S.: Early experience with Walsh technique of radical who had bilateral neurovascular bundle excision for presumed retropubic prostatectomy. Urology, 29: 242, 1987. extracapsular tumor extension 5 (38%) had histologically or8. Weldon, V. E. and Tave!, F. R.: Potency-sparing radical perinea! gan-confined cancer and 28 of 62 (45%) who had unilateral prostatectomy: anatomy, surgical technique and initial results. neurovascular bundle excision had intracapsular disease. ConJ. Urol., 140: 559, 1988. versely, 66 of 175 patients (38%) in whom both neurovascular 9. Catalona, W. J., Stein, A. J . and Fair, W. R.: Grading errors in bundles were preserved had histological evidence of extracapprostatic needle biopsies: relation to the accuracy of tumor grade sular tumor extension. Thus, errors were commonly made in in predicting pelvic node metastases. J. Urol., 127: 919, 1982. 10. Catalona, W. J. and Stein, A. J.: Staging errors in clinically both directions. localized prostate cancer. J. Urol., 127: 452, 1982. All of our patients who had capsular penetration in the region Lange, P. H. and Narayan, P.: Understaging and undergrading of of the neurovascular bundles had positive surgical margins. We ll. prostate cancer: argument for postoperative radiation as adjudid not observe capsular penetration of tumor only in the area ~ vant therapy. Urology, 21: 113, 1983. of the neurovascular bundle that was completely covered by an . Stamey, T. A., McNeal, J. E., Freiha, F. S. and Redwine, E.: excised neurovascular bundle. We cannot exclude the possibilMorphometric and clinical studies on 68 consecutive radical ity that had we been more aggressive in resecting the neuroprostatectomies. J. Urol., 139: 1235, 1988. 13. Walsh, P. C.: Editorial comment. J. Urol., 139: 1240, 1988. vascular bundles we might have observed this phenomenon.
,(5
·--·-" -- 1