Anatomic Study of Periprostatic Nerve Distribution: Immunohistochemical Differentiation of Parasympathetic and Sympathetic Nerve Fibres

Transcription

1 EUROPEAN UROLOGY 62 (2012) available at journal homepage: Prostate Cancer Anatomic Study of Periprostatic Nerve Distribution: Immunohistochemical Differentiation of Parasympathetic and Sympathetic Nerve Fibres Roman Ganzer a, *, Jens-Uwe Stolzenburg b, Wolf F. Wieland a, Johannes Bründl a a University of Regensburg, Caritas St. Josef Medical Centre, Regensburg, Germany; b Department of Urology, University of Leipzig, Leipzig, Germany Article info Article history: Accepted March 20, 2012 Published online ahead of print on March 28, 2012 Keywords: High anterior release Prostate nerve distribution Nerve-sparing prostatectomy Parasympathetic Sympathetic Abstract Background: Many authors advocate a high anterior incision during nerve-sparing radical prostatectomy (RP) to improve potency results. Despite a growing number of studies describing autonomic nerves in the ventrolateral position of the prostate, little is known about their quality and their role in erectile function. Objective: The intention of this study was a detailed characterisation of the topographic distribution of periprostatic nerves, including immunohistochemical differentiation of proerectile parasympathetic from sympathetic nerves. Design, setting, and participants: A total of 228 whole-mount sections of 38 prostates (base, middle, apex) from patients following non nerve-sparing laparoscopic RP were analysed. Immunohistochemical analysis was performed using antibodies against tyrosine hydroxylase for sympathetic and vesicular acetylcholine transporter for parasympathetic nerve fibre staining. Outcome measurements and statistical analysis: Quantification of periprostatic parasympathetic and sympathetic nerves was performed after defining prostatic regions via a digital grid. Differences among three independent variables were tested with the nonparametric Kruskal-Wallis test. Results and limitations: The total number of parasympathetic nerves did not decrease from the base to the apex. They were dispersed at the base and mainly located dorsolaterally at the apex, with 14.6% above the horizontal line at the base and only 1.5% at the apex. In contrast, the total number of sympathetic nerves decreased significantly from base to apex, with a constant proportion of ventrolateral nerves between 9% (base) and 6.2% (apex). This anatomic study is limited by the investigation of postprostatectomy specimens and the lack of functional results. Conclusions: Despite the presence of ventrolateral periprostatic nerves, only a minority of these nerves seems to have a parasympathetic proerectile quality. The arguments in favour of a high anterior incision during nerve-sparing prostatectomy might not only include preserved nerves but also other factors, such as reduced traction or improved anatomic support of the neural structures. # 2012 European Association of Urology. Published by Elsevier B.V. All rights reserved. * Corresponding author. Department of Urology, University of Regensburg, Krankenhaus St. Josef, Landshuter Straße 65, D Regensburg, Germany. Tel ; Fax: address: roman.ganzer@gmx.de (R. Ganzer) /$ see back matter # 2012 European Association of Urology. Published by Elsevier B.V. All rights reserved. doi: /j.eururo

2 EUROPEAN UROLOGY 62 (2012) Introduction Since the initial description of a neurovascular bundle (NVB) of the prostate by Walsh and Donker [1], there has been continuous evolution of nerve-sparing techniques in radical prostatectomy (RP) [2 5]. Not only technical modifications in open, laparoscopic, and robotic nervesparing surgery but also an improved understanding of the periprostatic neuroanatomy have led to improved postoperative potency results in current literature. Recent anatomic studies have shown the presence of periprostatic nerves not only in the dorsolateral but also in the ventrolateral position to a variable degree [6 9]. Terms concerning technical modifications aiming to preserve these nerves are curtain dissection [2], high anterior release [10], and most prominent, veil of Aphrodite [11]. All these designations have improved potency results in common. Although there is no doubt of the existence of ventrolateral periprostatic nerves, little is known about their role in erectile physiology. The limitation of most recent anatomic studies is the use of unspecific immunohistochemical staining methods, which do not allow for differentiation between proerectile nerves and those that are involved in the physiology of the prostate and other structures. Despite the fact that pilot functional studies in humans support the assumption that ventrolateral nerves do contribute to erectile function [12], other authors do not find any benefit in sparing these nerves, an aspect which is reflected by terms such as veil of mystery [13]. A few recent studies focussing on periprostatic nerve type differentiation indicate that only a minority of proerectile parasympathetic nerves is to be found ventrolaterally, thereby questioning the purpose of a high anterior incision of the periprostatic fascia during nerve sparing. Despite the significance of these findings, the low number of specimens investigated must be taken into consideration [14]. The intention of this study was to investigate the topographic distribution of periprostatic nerves, including immunohistochemical differentiation of proerectile parasympathetic from sympathetic nerves. 2. Material and methods 2.1. Specimens Whole-mount sections of the prostate were created from patients undergoing non nerve-sparing endoscopic extraperitoneal RP, including wide excision of the prostatic pedicle and the NVB [15]. All prostatectomies were performed between October 2009 and June 2010 by two surgeons. The technique of whole-mount processing of the specimens has been described previously [7]. For this study, five consecutive 10-mm whole-mount sections were cut from the base, the middle, and the apex of the prostate, respectively. One section from each region of the prostate was stained with haematoxylin and eosin. Of the remaining adjacent sections, two were used for immunohistochemical sympathetic (tyrosine hydroxylase [TH]) and two for parasympathetic (vesicular acetylcholine transporter [VAChT]) nerve fibre staining. Only specimens with complete integrity of both the prostatic capsule and the periprostatic tissue were considered for analysis Parasympathetic nerve staining To stain proerectile parasympathetic nerves, we used a rabbit antibody (Sigma, code no. V 5387, dilution 1/2000) against VAChT. ACh is synthesised in the cytoplasm by acetylcholine transferase and is transported by VAChT to the synaptic vesicles, where it is stored and released following activation. Immunohistochemical staining of VAChT is considered specific for the presence of parasympathetic cholinergic neurons and was used by other groups [16] Sympathetic nerve staining A polyclonal rabbit antibody (Abcam, code no. ab112, dilution 1/750, Sapphire Bioscience Pty Ltd, Waterloo NSW, Australia) against TH was used to stain sympathetic nerve fibres. This enzyme is involved in catecholamine synthesis and responsible for conversion of phenylalanine to dopamine [16] Nerve quantification All sections were digitised using a high-resolution photo scanner (Perfection V750 Pro, Epson, Meerbusch, Germany) and processed with Adobe Photoshop v.6.0 software (San Jose, CA, USA). For nerve quantification, digital copies of all immunohistochemically stainedwholemount sections were centrally covered with a grid dividing each section into six sectors numbered clockwise. The periprostatic tissue was thus divided into the following regions: anterior (A), ventrolateral (VL), dorsolateral (DL), and dorsal (D; Fig. 1A). Thereafter, two investigators performed periprostatic nerve quantification within the different regions, distinguishing between parasympathetic and sympathetic nerves. All calculations were performed with mean values from the results of regions of two adjacent whole-mount sections Statistical analysis Statistical analysis was performed with SPSS v.15 (IBM Corp., Armonk, NY, USA). Differences among three independent variables were tested with the nonparametric Kruskal-Wallis test. P values <0.05 were considered statistically significant. The nerve count correlation between both investigators was assessed using Kendall s correlation coefficient. 3. Results Whole-mount sections of 49 prostates were created. After excluding all prostates with whole-mount sections showing artefacts of the prostatic capsule and periprostatic tissue, 228 whole-mount sections of 38 prostates were available for investigation. Baseline characteristics are shown in Table 1. The nerve count correlated between both investigators showed high statistical significance for all regions ( p < 0.001). Therefore, all calculations were done with mean values of both investigators. A total of 999 and 1531 parasympathetic and sympathetic nerve fibres were counted, respectively. All details of nerve quantification are shown in Table 2. The majority of both nerve fibre types was found in the dorsolateral region of the presumed NVB Parasympathetic nerve fibre distribution The total number of parasympathetic nerves did not decrease from the base on the way to the apex (Table 2).

3 1152 [(Fig._1)TD$FIG] EUROPEAN UROLOGY 62 (2012) Fig. 1 Immunohistochemical staining of sympathetic nerves (tyrosine hydroxylase). (A) Whole-mount section, middle of the prostate, division of the periprostatic tissue in defined regions. (B) Ten-fold magnification of a sympathetic nerve in the ventrolateral region. (C) Forty-fold magnification of the nerve in (B) showing the neurovesicles. V = ventral; VL = ventrolateral; DL = dorsolateral; D = dorsal. At the base, parasympathetic nerve fibres were more dispersed than at the apex. At the base, 78% of parasympathetic nerves where found in the dorsolateral position but in 96.5% at the apex ( p < ; Fig. 2C). In contrast, 14.6% of parasympathetic nerves were found above the horizontal line at the base, but at the apex (Fig. 2B), the proportion of ventrolateral parasympathetic nerves was negligible, with only 1.5% ( p < ; Fig. 3B) Sympathetic nerve fibre distribution In contrast to parasympathetic nerves, we found a statistically significant decrease in the total number of quantified sympathetic nerves from the base on their course to the apex (Table 2). The majority of sympathetic nerves were Table 1 Baseline data Data point Value Age, yr, PSA, ng/ml, Prostate volume, cm 3, Pathologic T stage, no. (%): pt2a 2 (5.3) pt2b 2 (5.3) pt2c 14 (36.7) pt3a 12 (31.6) pt3b 8 (21.1) Gleason score, median (range) 7 (5 9) SD = standard deviation; PSA = prostate-specific antigen. found dorsolaterally. The proportion of ventrolateral nerves was constant, with 9%, 5.4%, and 6.2% at the base, the middle, and the apex, respectively (Fig. 1B and 1C; Fig. 3A). 4. Discussion Before the pioneer publication of the NVB anatomy by Walsh and Donker in 1982, it was widely assumed that erectile nerves run through the prostate on their way to the corpora cavernosa and are inevitably damaged during RP [1]. Despite an on-going evolution of nerve-sparing techniques, erectile dysfunction (ED) remains the healthrelated quality of life (QoL) domain most commonly impaired after prostatectomy [17,18]. This fact illustrates the necessity of optimising potency outcomes following RP. Improved understanding of the prostatic neuroanatomy is of paramount importance to developing technical modifications in nerve sparing. During the past decade, numerous publications have described a more variable distribution of periprostatic nerves than had been previously thought to exist, thereby questioning the classic concept of a purely dorsolateral NVB [2,6,7,19,20]. In their study of 79 prostatic specimens, Kyoshima et al. were able to find a distinct NVB in only 52% of cases. In the remaining specimens, the nerves were more dispersed around the prostate [19]. Eichelberg et al. found up to 28.5% of nerves above the horizontal line [6], an aspect that we were able to confirm in a previous study containing whole-mount sections of 30 prostates [7]. Lunacek et al. performed gross and histologic preparation of 29 male foetuses and eight adult specimens. They

4 EUROPEAN UROLOGY 62 (2012) Table 2 Quantification and distribution of periprostatic nerves VAChT (parasympathetic) TH (sympathetic) p value Apex, no. (%); Middle, no. (%); p value Base, no. (%); Apex, no. (%); Middle, no. (%); Base, no. (%); Ventral 7 (2.2); (2.1); (0); (0); (0); (0.2); ns Ventrolateral 40 (12.4); (8.0); (1.5); < (9); (5.4); (6.2); ns Dorsolateral 251 (78.0); (86.4); (96.5); < (86.2); (91.4); (92.6); Dorsal 24 (7.4); (3.5); (2.0); (4.8); (3.2); (1.0); Total ns < VAChT = vesicular acetylcholine transporter; TH = tyrosine hydroxylase; SD = standard deviation; ns = not significant. observed that cavernous nerves run as a distinct structure during the embryonic phase but disperse over the lateral prostatic surface with gestation. The group transferred these findings to a modified surgical technique, calling it curtain dissection [2]. Kaiho et al. supported these anatomic findings with a functional study performed in 12 patients, showing that intraoperative electrostimulation of the ventral prostatic circumference leads to an increase in urethral pressure [12]. Techniques with a high anterior incision of the periprostatic fascia have been described in open, laparoscopic, and robot-assisted RP with excellent functional results [3,4,10,21]. The group from the Hamburg Martini-Clinic reported on 1150 men following retropubic nerve-sparing prostatectomy with erections sufficient for intercourse after 1 yr in 84 92% of patients [21]. In a retrospective analysis by Nielsen et al, patients undergoing high anterior release were more likely to achieve a Sexual Health Inventory for Men score of 16 or greater and/or a satisfaction score of 4 compared to those undergoing standard nerve sparing (93% vs 77%) [10]. In a series of 2652 patients, Menon et al. reported that 70% of patients had intercourse 1 yr following veil of Aphrodite nerve-sparing surgery compared to 40% of patients with the conventional technique [3]. It should be emphasised that all these centres report on their results at an advanced level of experience, where possibly numerous other technical steps have been optimised during the learning curve that ultimately contribute to these outstanding results. Furthermore, all these series are retrospective in character. Despite enthusiasm about these results, we cannot ignore that there are hardly any well-designed, controlled studies comparing a high anterior release with the classic dorsolateral incision. There is only one retrospective comparative study with 137 patients in each group comparing curtain dissection and standard nerve-sparing techniques in laparoscopic RP (LRP). Interestingly, the authors could not find any difference in potency outcome, suggesting the term veil of mystery [13]. If a functional benefit will be proved in the future, this might not only be explained by preserved nerves but also by other factors such as reduced traction and improved anatomic support of the neural structures. All previously cited anatomic studies come to a similar conclusion: Periprostatic nerves are found not only in the classic location of the NVB but also to a variable degree on the ventrolateral aspect of the prostate. However, all publications have a considerable limitation: Nerve quantification was either performed without [2,6,19] or with unspecific nerve fibre staining methods. In most studies, immunohistochemical staining against the neuroprotein S100 was used [7,8,20]. S100 is a general immunolabelling marker for all nerves that specifically identifies Schwann cells in formalinfixed paraffin-embedded tissue [22]. The dilemma is represented by growing evidence on ventrolateral nerves of the prostate without any clear understanding of their role in the physiology of erectile function. To improve the understanding of the physiology of these nerves, it is important to differentiate proerectile parasympathetic nerves from other nerves. Such a differentiation was

5 1154 [(Fig._2)TD$FIG] EUROPEAN UROLOGY 62 (2012) Fig. 2 Immunohistochemical staining of parasympathetic nerves (vesicular acetylcholine transporter). (A) Whole-mount section, apex of the prostate, division of the periprostatic tissue in defined regions. (B) Magnification of ventrolateral periprostatic tissue showing no nerves. (C) Magnification of parasympathetic nerves in the dorsolateral region. V = ventral; VL = ventrolateral; DL = dorsolateral; D = dorsal. performed by Costello et al. by immunohistochemical staining in four male cadavers. Although they found 27.8% of all nerves on the ventral half of the prostate, only 4 6.8% were of parasympathetic origin [14]. Limitations were the small number of specimens, among which two were embalmed cadavers, which might have compromised the quality of immunohistochemical staining. Alsaid et al. performed a nice three-dimensional reconstruction of the location and course of pelvic adrenergic, cholinergic, and sensory nerve fibres. However, the study was limited by the [(Fig._3)TD$FIG] low number of specimens as well as the fact that it was performed in cadaveric material of male foetuses and not adults [23]. As an amendment to previous studies using unspecific nervefibrestaining,weintendedtoundertakeadetailed mapping study of periprostatic nerves with immunohistochemical differentiation of parasympathetic from sympathetic nerves. To our knowledge, this is the study with the largest number of specimens in current literature. Although the true role of parasympathetic and sympathetic Fig. 3 Comparison of the distribution of sympathetic and parasympathetic nerves in the dorsolateral and ventrolateral regions at the base, middle, and apex. (A) Sympathetic nerves; (B) parasympathetic nerves. TH = tyrosine hydroxylase; VAChT = vesicular acetylcholine transporter; ns = not significant.

6 EUROPEAN UROLOGY 62 (2012) nerves around the prostate is not yet understood, we considered parasympathetic nerves as proerectile in this study. We confirmed that the majority of nerves is located in a dorsolateral position. Interestingly, >50% of nerves in this location were of sympathetic origin. In 2004, Costello et al. traced both parasympathetic and sympathetic nerve fibres within the NVB of cadaver specimens not only innervating the corpora cavernosa but also the levator ani, the rectum, and the prostate [24]. This finding correlates with our findings. We found 7.8% of all counted nerves in the ventral half of the prostate. Here, too, around half of them were sympathetic. At the apex, only a minority of ventrolateral nerves were of proerectile parasympathetic quality. Based on our findings, a high incision might help preserve more parasympathetic nerves at the base and the middle but is of little help at the apex. Although there are no data showing increased positive margin rates with this approach, we have to consider that the apex lacks a clear anatomic capsule, with a potentially increased risk of a positive margin in cases of apical anterior tumours, when performing a high incision at the apex. Our finding of a significant reduction in the number of sympathetic nerves from the base to the apex might be explained by a branching of nerves to innervate the prostate and adjacent structures. In contrast, there was no reduction in the proerectile parasympathetic nerves on the way to the corpora cavernosa. Therefore, it seems inappropriate to use the terms neurovascular bundle and cavernous nerves synonymously. Our study contains some limitations: First, it is based on the investigation of prostatectomy specimens. Even when wide excision is performed, it is essential to consider that the surgical procedure might influence anatomic findings. A wide excision of the NVB will differ among surgeons. Therefore, it cannot be guaranteed that all periprostatic nerves will be with the specimen. Furthermore, the selection of patients not qualifying for nerve sparing because of impaired baseline potency status or risk of non organconfined disease might affect our results. We did not investigate the correlation between prostate size and distribution of different nerve fibre types. In a previous study, we were able to show that the distribution of S100-stained periprostatic nerves is not dependent on prostate size [8]. However, this result could be different when considering parasympathetic and sympathetic nerves separately. When comparing our results of quantified nerves with those in our previous investigations with S100 staining, it becomes evident that the total number of quantified nerves in this study is much lower. This might be explained by the different approaches: Staining against the neuroprotein S100 guarantees high effectiveness because of the staining of nerve sheaths. In contrast, immunohistochemical staining of VAChT and TH is designed for the content of vesicles, which might explain a lower quantification output. However, we consider the ratio of both nerve fibre types with respect to their location to be meaningful. 5. Conclusions The main goal in modifying nerve-sparing techniques in RP is to improve functional results and postoperative QoL. Despite growing insights into the anatomic foundations of potency preservation, anatomic and physiologic details remain unclear. Despite proven evidence of ventrolateral prostatic nerves, our study reveals that only a minority of them seems to have proerectile parasympathetic quality. The advantage of a high anterior release over conventional nerve sparing is proved neither by comparative clinical studies nor by anatomic studies. Future prospective randomised trials should elucidate whether there is a benefit in terms of return to continence and erectile function. Author contributions: Roman Ganzer had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Ganzer, Stolzenburg. Acquisition of data: Ganzer, Bründl. Analysis and interpretation of data: Ganzer, Bründl, Stolzenburg. Drafting of the manuscript: Ganzer, Bründl. Critical revision of the manuscript for important intellectual content: Wieland, Stolzenburg. Statistical analysis: Ganzer, Bründl. Obtaining funding: Ganzer. Administrative, technical, or material support: None. Supervision: Wieland. Other (specify): None. Financial disclosures: Roman Ganzer certifies that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None. Funding/Support and role of the sponsor: This research project was supported by a grant from the Deutsche Krebshilfe e.v. Acknowledgment statement: The authors would like to thank Nina Niessl for her excellent technical support. References [1] Walsh PC, Donker PJ. Impotence following radical prostatectomy: insight into etiology and prevention. J Urol 1982;128: [2] Lunacek A, Schwentner C, Fritsch H, Bartsch G, Strasser H. Anatomical radical retropubic prostatectomy: curtain dissection of the neurovascular bundle. BJU Int 2005;95: [3] Menon M, Shrivastava A, Kaul S, et al. Vattikuti Institute prostatectomy: contemporary technique and analysis of results. Eur Urol 2007;51: [4] Stolzenburg JU, Rabenalt R, Do M, et al. Intrafascial nerve-sparing endoscopicextraperitonealradical prostatectomy. Eur Urol 2008; 53: [5] Graefen M, Walz J, Huland H. Open retropubic nerve-sparing radical prostatectomy. Eur Urol 2006;49: [6] Eichelberg C, Erbersdobler A, Michl U, et al. Nerve distribution along the prostatic capsule. Eur Urol 2007;51: [7] Ganzer R, Blana A, Gaumann A, et al. Topographical anatomy of periprostatic and capsular nerves: quantification and computerised planimetry. Eur Urol 2008;54:

7 1156 EUROPEAN UROLOGY 62 (2012) [8] Ganzer R, Blana A, Stolzenburg JU, et al. Nerve quantification and computerized planimetry to evaluate periprostatic nerve distribution-does size matter? Urology 2009;74: [9] Takenaka A, Murakami G, Matsubara A, Han SH, Fujisawa M. Variation in course of cavernous nerve with special reference to details of topographic relationships near prostatic apex: histologic study using male cadavers. Urology 2005;65: [10] Nielsen ME, Schaeffer EM, Marschke P, Walsh PC. High anterior release of the levator fascia improves sexual function following open radical retropubic prostatectomy. J Urol 2008;180: , discussion [11] Menon M, Tewari A, Peabody J. Vattikuti Institute prostatectomy: technique. J Urol 2003;169: [12] Kaiho Y, Nakagawa H, Saito H, et al. Nerves at the ventral prostatic capsule contribute to erectile function: initial electrophysiological assessment in humans. Eur Urol 2009;55: [13] Chabert CC, Merrilees DA, Neill MG, Eden CG. Curtain dissection of the lateral prostatic fascia and potency after laparoscopic radical prostatectomy: a veil of mystery. BJU Int 2008;101: [14] Costello AJ, Dowdle BW, Namdarian B, Pedersen J, Murphy DG. Immunohistochemical study of the cavernous nerves in the periprostatic region. BJU Int 2011;107: [15] Stolzenburg JU, Rabenalt R, Do M, et al. Endoscopic extraperitoneal radical prostatectomy: the University of Leipzig experience of 1,300 cases. World J Urol 2007;25: [16] Alsaid B, Bessede T, Karam I, et al. Coexistence of adrenergic and cholinergic nerves in the inferior hypogastric plexus: anatomical and immunohistochemical study with 3D reconstruction in human male fetus. J Anat 2009;214: [17] Shikanov SA, Zorn KC, Zagaja GP, Shalhav AL. Trifecta outcomes after robotic-assisted laparoscopic prostatectomy. Urology 2009; 74: [18] Alemozaffar M, Regan MM, Cooperberg MR, et al. Prediction of erectile function following treatment for prostate cancer. JAMA 2011;306: [19] Kiyoshima K, Yokomizo A, Yoshida T, et al. Anatomical features of periprostatic tissue and its surroundings: a histological analysis of 79 radical retropubic prostatectomy specimens. Jpn J Clin Oncol 2004;34: [20] Sievert KD, Hennenlotter J, Laible IA, Amend B, Nagele U, Stenzl A. The commonly performed nerve-sparing total prostatectomy does not acknowledge the actual nerve courses. J Urol 2009;181: [21] Budäus L, Isbarn H, Schlomm T, et al. Current technique of open intrafascial nerve-sparing retropubic prostatectomy. Eur Urol 2009;56: [22] Hollabaugh Jr RS, Dmochowski RR, Steiner MS. Neuroanatomy of the male rhabdosphincter. Urology 1997;49: [23] Alsaid B, Karam I, Bessede T, et al. Tridimensional computerassisted anatomic dissection of posterolateral prostatic neurovascular bundles. Eur Urol 2010;58: [24] Costello AJ, Brooks M, Cole OJ. Anatomical studies of the neurovascular bundle and cavernosal nerves. BJU Int 2004;94: