Classification of the Distribution of Cavernous Nerve Fibers Around the Prostate by Intraoperative Electrical Stimulation During Laparoscopic Radical Prostatectomy

ORIGINAL ARTICLE Classification of the distribution of cavernous nerve fibers around the prostate by intraoperative electrical stimulation during laparoscopic radical prostatectomy

A Takenaka1, H Soga2, N Hinata1, M Honda1, T Sejima1, M Muramaki2, H Miyake2, K Tanaka2 and M Fujisawa2

1Division of Urology, Department of Surgery, Tottori University Faculty of Medicine, Tottori, Japan and 2Division of Urology, Department of Surgery Related, Kobe University Graduate School of Medicine, Kobe, Japan

We investigated the distribution of cavernous nerve (CN) fibers around the prostate by electrical nerve stimulation during laparoscopic radical prostatectomy to classify the distribution of the CN fibers. Electrical stimulation was performed on 30 consecutive patients with localized prostate cancer; middle of the neurovascular bundle (NVB, point A), base of the NVB (point B), the rectal wall 1 cm posterolateral to the NVB (point C) and the lateral aspect of the prostate (point D). We measured the intraurethral pressure at the midportion to detect the changes in intracavernosal pressure. The mean maximum changes were 10.5±7.9, 11.6±8.8, 9.6±7.4 and 6.7±7.0 cm H2Oat points A, B, C and D, respectively. The patterns of CN fiber distribution were divided into four groups: type 1 (23%), the bundle corresponding to the NVB; type 2 (7%), the bundle from the rectal wall to the prostate; type 3 (27%), the plate including NVB and posterolateral to NVB; and type 4 (43%), the plate between the rectal wall posterolateral to the NVB and the lateral aspect of the prostate. Distribution of the CNs in a bundle-like formation was considered to account for 30%, whereas a plate-like formation accounted for 70%. Understanding these four patterns of CN fiber distribution should facilitate accurate CN-sparing radical prostatectomy. International Journal of Impotence Research (2011) 23, 56–61; doi:10.1038/ijir.2011.4; published online 10 March 2011

Keywords: cavernous nerve; classification; electrical stimulation; nerve sparing; radical prostatectomy

Introduction nerve-sparing procedure based on anatomic research is widely performed to restore erectile function To clarify the neurogenic factors involved in after radical prostatectomy, postoperative potency impotence in patients following radical prostatect- rates reported by other institutes varied with age, omy, several studies have attempted to demonstrate preoperative sexual function, surgeon and posto- the autonomic innervation of the corpora caverno- perative period. In addition, the potency rate sa.1–4 The concept of the neurovascular bundle after non-nerve-sparing radical prostatectomy was (NVB) was advocated by Walsh and Donker,1 reported to be 0–30%.5,6 whereas Costello et al.2 reported that most of the In our previous studies, we noted that the nerves included in NVB originally descend poster- distribution of cavernous nerve (CN) fibers was ior to the seminal vesicle, converging at the wider than that shown in our image of the NVB.3,7 midprostatic level only to diverge once again as Recently, the validity of our earlier research was they approach the prostatic apex. Although the confirmed by the finding that prostatic fascia- preserving procedure showed excellent outcomes with regard to postoperative erectile function.8,9 Furthermore, anatomical and electrophysiological Correspondence: Professor A Takenaka, Division of Urol- studies showed the distribution of nerve fibers on ogy, Department of Surgery, Tottori University Faculty of 10–14 Medicine, 36-1, Nishi-cho, Yonago, Tottori 683-8504, the ventral side of the prostatic fascia. However, Japan. all of these anatomic and electrophysiological E-mail: [email protected] papers showed interindividual variations. Since Received 22 August 2010; revised 11 January 2011; 2000, we have used a laparoscopic approach for accepted 31 January 2011; published online 10 March 2011 radical prostatectomy. This study investigated the Distribution of cavernous nerve around the prostate A Takenaka et al 57 distribution of CN fibers around the prostate by including 30 mA, 10 Hz and with 0.3 ms duration, electrical nerve stimulation during laparoscopic and performed at four points to identify the CN radical prostatectomy. As clarifying the CN distribu- fibers; middle of the NVB (point A), base of the NVB tion during radical prostatectomy would improve (point B), the rectal wall 1 cm posterolateral to the the accuracy of CN-sparing operations, we under- NVB between points A and B (point C) and took this study to demonstrate periprostatic patterns the lateral aspect (3 O’clock) of the middle of the of CN fibers. prostate (point D; Figure 1). We measured the intraurethral pressure at the midportion of the penile shaft using an inserted balloon catheter (7.5 Fr, Fuji system, Tokyo, Japan) to detect changes Patients and methods in intracavernous pressure (ICP).14,17 This catheter was connected to a nerve-sparing monitor FGK-1 Between November 2006 and August 2007, electri- (Motomura Manufacturer) through a disposable pres- cal stimulation during laparoscopic radical prosta- sure transducer set (Nihon Kohden, Tokyo, Japan). tectomy was performed on 30 consecutive patients Maximum changes in pressure at four points were with clinically localized prostate cancer undergoing compared using Scheffe test. Maximum changes in surgery at Kobe University Hospital, Japan. Patient ICP at points B, C and D were compared with those characteristics are shown in Table 1. None of the at point A in each case, and a positive response was patients had a history of pelvic surgery, pelvic defined as a 70% or greater change compared with irradiation, transurethral surgery or neoadjuvant the maximum change at point A in each case. hormonal therapy. The laparoscopic procedure was This study was approved by the ethics committee performed as reported originally by the Montsouris of Kobe University Graduate School of Medicine, and group.15 General anesthesia was achieved using informed consent was obtained from each patient. propofol, N2O plus sevoflurane and neuromuscular blockade. None of the patients in this study received an epidural catheter for anesthesia. Surgery was Results performed with 12–15 mm Hg of insufflation pressure, but during stimulation, insufflation pressure Every stimulation at each point except point D was decreased to 10 mm Hg. evoked changes in ICP. Stimulation at point D did Electrical stimulation was performed between the not induce any changes in ICP in one patient. bunching procedure and bladder neck transection Median duration of time for the electrical stimula- using a nerve stimulator device FGK-1S (Motomura tions was 13.5 min (range, 12.5–29.0). Several Manufacturer, Tokyo, Japan) and a specially made seconds after the beginning of the stimulation, ICP bipolar electrode for laparoscopic surgery. The increased gradually and then decreased gradually at interval between the two electrodes was 7 mm long. almost all points in all cases (Figure 1). Accompany- As described previously,16 stimulation was admi- ing the rise and decline of ICP, we found that the nistered for 30 s as a monophasic rectangular pulse, penile shaft became minutely tumescent in some cases. The mean maximum changes in ICP were Table 1 Patient characteristics 10.5±7.9, 11.6±8.8, 9.6±7.4 and 6.7±7.0 cm H2Oat points A, B, C and D, respectively (Figure 2). There Age (mean, range; years old) 66.7 (56–75) were no significant differences in ICP among points Preoperative PSA (mean, range; ng mlÀ1) 8.93 (3.46–18.74) A, B and C, although ICP after stimulation of point D Preoperative IIEF-5 score (mean, range) 11.1 (1–25) was significantly lower than that of other points. Clinical T stage Points representing positive responses defined by cT1c 9 our method were mapped in each case. This method cT2a 13 showed that the distribution of CN fibers could be cT2b 8 classified into four patterns (Figure 3): type 1 (points Number of biopsies, Gleason score A and B), the bundle corresponding to the NVB was p613present in 7 cases (23%); type 2 (points A and C), the 716bundle from the rectal wall to the prostate was X81present in 2 cases (7%); type 3 (points A, B and C), Prostate weight (mean, range; g) 45.6 (25–85) the plate including the NVB and posterolateral to Prostate volume (mean, range; ml) 36.3 (20.1–66.5) NVB was present in 8 cases (27%); and type 4 (all points), the plate between the rectal wall poster- Pathological T stage olateral to the NVB and the lateral aspect of the pT2c 23 prostate was present in 13 cases (43%). Conse- pT3a 6 pT3b 1 quently, the distribution of the CNs showing a bundle-like formation (types 1 and 2) was consid- IIEF-5, International Index of Erectile Function-5; PSA, prostate ered to account for 30%, whereas a plate-like specific antigen. formation (types 3 and 4) accounted for 70%.

International Journal of Impotence Research Distribution of cavernous nerve around the prostate A Takenaka et al 58

Figure 1 Intracavernosal pressure monitoring after electrical stimulation at each point; (a) at the middle of the NVB, (b) at the base of the NVB, (c) the rectal wall 1 cm posterolateral to the NVB between points A and B, and (d) the lateral aspect (3 O’clock) of the middle of the prostate. BL, bladder; ICP, intracavernous pressure (cm H2O); LA, levator ani muscle; NVB, neurovascular bundle; PR, prostate; RF, perirectal fat; U, urethra.

International Journal of Impotence Research Distribution of cavernous nerve around the prostate A Takenaka et al 59

Figure 2 Mean maximum changes in intracavernous pressure.

Discussion

The objective of nerve-sparing radical prostatectomy is to remove the prostate without injuring either the prostatic capsule or the CN. Using fresh and fixed adult cadavers, we studied the neuroanatomy of the branches from the pelvic plexus to reconsider the NVB. We demonstrated anatomically and electro- physiologically that the NVB was likely to differ from the actual course of the CN fibers.3,7,18 Recently, histological investigation of the prostatic fascia suggested that nerve fibers were distributed at the ventral side of the prostatic fascia beyond the NVB.11–13 However, Costello et al.2 demonstrated the concept that some nerve fibers running in the NVB penetrated the prostatic capsule, whereas other fibers innervated other structures including the rectum and levator ani muscle, as well as the corpora cavernosa. Consequently, the distribution and interindividual variations of CN responsible for erectile function have not been elucidated. By electrical stimulation at the ventral side of the prostate, Kaiho et al.14 confirmed that the CN fibers were widely spread beyond the NVB. Thus, these anatomical and functional investigations suggested that the CN fibers were also distributed on the Figure 3 Distribution of cavernous nerve fibers and ratio of ventral side of the prostate. This might explain why intracavernous pressure in each pattern. Type 1 (points A and B; prostatic fascia-preserving procedure has shown the bundle corresponding to the neurovascular bundle (NVB)), excellent outcomes with regard to postoperative type 2 (points A and C; the bundle from the rectal wall to the 8 prostate), type 3 (points A, B and C; the plate including NVB and erectile function. In the present study, we con- posterolateral to NVB) and type 4 (all points; the plate between firmed interindividual variations of CN distribution the rectal wall posterolateral to the NVB and the lateral aspect of and classified them into four groups to improve the the prostate). accuracy of CN-sparing while maintaining oncolo- gical control. definite NVB structure on magnetic resonance Walsh et al.19 described excellent outcomes with imaging compared with that in those with an regard to postoperative erectile function after a identifiable NVB structure on magnetic resonance bilateral nerve-sparing procedure based on the imaging, suggesting that individual variations in the classical concept of NVB. However, Lee et al.20 distribution of the CN fibers affected the recovery of demonstrated that the rates of erectile function erectile function. Lunacek et al.4 demonstrated that recovery after classical nerve-sparing surgery were CN fibers became displaced more anteriorly along worse in patients who did not demonstrate a the convex surface of the prostate with growth of the

International Journal of Impotence Research Distribution of cavernous nerve around the prostate A Takenaka et al 60 prostatic parenchyma. Our previous study also Conclusion showed significant individual variations in the We demonstrated that the distribution of CN course of the CN near the apex of the prostate.18 fibers could be classified into four patterns Based on the present analysis, the distribution using intraoperative electrical stimulation. The of CN fibers was classified into four patterns. To distribution of the CN in a bundle-like formation maintain erectile function, the prostatic fascia- was considered to account for 30%, whereas a preserving procedure might be especially useful plate-like formation accounted for 70%. Under- in cases showing a type 4-distribution pattern. standing these four patterns of CN fiber distribution On the contrary, potency after non-nerve-sparing should facilitate accurate CN-sparing radical pros- radical prostatectomy was reported to be 0–30%.5,6 tatectomy. Histological study showed that branches from the pelvic splanchnic nerves passed through the para- rectal space to reach the inferior level slightly below the prostatic apex, which was likely to result in Conflict of interest unexpected nerve sparing after non-nerve-sparing radical prostatectomy.18 Patients with postoperative The authors declare no conflict of interest. erections after non-nerve-sparing procedure might have belonged mainly to the group showing a type 2 distribution pattern. Stimulation was performed under 10 mm Hg of References CO2 pressure to keep the stimulated area dry. 1 Walsh PC, Donker PJ. Impotence following radical prostatect- Abdominal CO2 insufflation in the pneumoperito- omy: insight into etiology and prevention. J Urol 1982; 128: neum and subsequent increase in abdominal 492–497. pressure cause changes in the intrapelvic hemody- 2 Costello AJ, Brooks M, Cole OJ. Anatomical studies of the namics, and venous compression due to positive neurovascular bundle and cavernosal nerves. BJU Int 2004; 94: pressure from CO insufflation during laparoscopic 1071–1076. 2 3 Takenaka A, Murakami G, Soga H, Han SH, Arai Y, radical prostatectomy might decrease the blood Fujisawa M. Anatomical analysis of the neurovascular supply to the corpora cavernosa. Initially, we bundle supplying penile cavernous tissue to ensure a attempted to perform the stimulation without reliable nerve graft after radical prostatectomy. J Urol 2004; 172: 1032–1035. CO2 pressure; however, the stimulated area became 4 Lunacek A, Schwentner C, Fritsch H, Bartsch G, Strasser H. wet with venous oozing in some cases and the Anatomical radical retropubic prostatectomy: ‘curtain intraurethral pressure was significantly lower than dissection’ of the neurovascular bundle. BJU Int 2005; 95: that in patients with dry field. Consequently, 1226–1231. keeping the stimulated area dry was more important 5 Penson DF, McLerran D, Feng Z, Li L, Albertsen PC, Gilliland for the reproducibility of these proportions than the FD et al. 5-year urinary and sexual outcomes after radical prostatectomy: results from the prostate cancer outcomes influence of insufflation pressure. study. J Urol 2005; 173: 1701–1705. As this was the first study investigating patterns 6 Dubbelman YD, Dohle GR, Schroder FH. Sexual function of CN fiber distribution, the classification of CN before and after radical retropubic prostatectomy: a systematic distribution contains some limitations. First, we review of prognostic indicators for a successful outcome. Eur Urol 2006; 50: 711–718; discussion 718–720. arbitrarily defined positive response to electrical 7 Takenaka A, Tewari A, Hara R, Leung RA, Kurokawa K, stimulation when maximum changes in ICP at each Murakami G et al. Pelvic autonomic nerve mapping point were 70% or greater compared with the around the prostate by intraoperative electrical stimulation maximum changes at point A in each case. There with simultaneous measurement of intracavernous and is no evidence to support this definition. Second, intraurethral pressure. J Urol 2007; 177: 225–229; discussion 229. we did not assess the tumescence of the penile shaft 8 Menon M, Kaul S, Bhandari A, Shrivastava A, Tewari A, by any objective method. In the present study, we Hemal A. Potency following robotic radical prostatectomy: a attempted to measure changes in ICP without questionnaire based analysis of outcomes after conventional considering tumescence of the penile shaft. When nerve sparing and prostatic fascia sparing techniques. J Urol 2005; 174: 2291–2296; discussion 2296. the penile shaft becomes tumescent because of 9 Stolzenburg JU, Rabenalt R, Do M, Schwalenberg T, Winkler stimulation, it takes 10–20 min before the next M, Dietel A et al. Intrafascial nerve-sparing endoscopic stimulation can be performed, as the tumescence extraperitoneal radical prostatectomy. Eur Urol 2008; 53: recedes with the decline of ICP. The changes in ICP 931–940. may not reflect the status of erectile function, but 10 Savera AT, Kaul S, Badani K, Stark AT, Shah NL, Menon M. Robotic radical prostatectomy with the ‘Veil of Aphrodite’ demonstrate the distribution of CN fibers. Third, the technique: histologic evidence of enhanced nerve sparing. Eur present study consisted of a small number of Urol 2006; 49: 1065–1073; discussion 1073–1064. patients. However, understanding these four pat- 11 Eichelberg C, Erbersdobler A, Michl U, Schlomm T, terns of CN fiber distribution should facilitate Salomon G, Graefen M et al. Nerve distribution along the prostatic capsule. Eur Urol 2007; 51: 105–110; discussion accurate CN-sparing prostatectomy and may lead 110–101. to real-time monitoring of the CN during radical 12 Ganzer R, Blana A, Gaumann A, Stolzenburg JU, Rabenalt R, prostatectomy. Bach T et al. Topographical anatomy of periprostatic and

International Journal of Impotence Research Distribution of cavernous nerve around the prostate A Takenaka et al 61 capsular nerves: quantification and computerised planimetry. 17 Tsujimura A, Miyagawa Y, Takao T, Matsumiya K, Nakayama Eur Urol 2008; 54: 353–361. M, Tsujimoto Y et al. Significance of electrostimulation in 13 Lee SB, Hong SK, Choe G, Lee SE. Periprostatic distribution of detecting neurovascular bundle during radical prostatectomy. nerves in specimens from non-nerve-sparing radical retro- Int J Urol 2006; 13: 926–931. pubic prostatectomy. Urology 2008; 72: 878–881. 18 Takenaka A, Murakami G, Matsubara A, Han SH, Fujisawa M. 14 Kaiho Y, Nakagawa H, Saito H, Ito A, Ishidoya S, Saito S et al. Variation in course of cavernous nerve with special reference to Nerves at the ventral prostatic capsule contribute to erectile details of topographic relationships near prostatic apex: histolo- function: initial electrophysiological assessment in humans. gic study using male cadavers. Urology 2005; 65: 136–142. Eur Urol 2009; 55: 148–154. 19 Walsh PC, Marschke P, Ricker D, Burnett AL. Patient-reported 15 Guillonneau B, Vallancien G. Laparoscopic radical pro- urinary continence and sexual function after anatomic radical statectomy: the Montsouris experience. J Urol 2000; 163: prostatectomy. Urology 2000; 55: 58–61. 418–422. 20 Lee SE, Hong SK, Han JH, Han BK, Yu JH, Jeong SJ et al. 16 Kurokawa K, Suzuki T, Suzuk IK, Terada N, Ito K, Yoshikawa Significance of neurovascular bundle formation observed on D et al. Preliminary results of a monitoring system to preoperative magnetic resonance imaging regarding post- confirm the preservation of cavernous nerves. Int J Urol operative erectile function after nerve-sparing radical retro- 2003; 10: 136–140. pubic prostatectomy. Urology 2007; 69: 510–514.

International Journal of Impotence Research