Evoked Cavernous Activity: Neuroanatomic Implications

ORIGINAL ARTICLE Evoked cavernous activity: neuroanatomic implications

U Yilmaz1, B Vicars1 and CC Yang1,2

1Department of Urology, University of Washington, Seattle, WA, USA and 2VA Puget Sound Healthcare System, Seattle, WA, USA

We investigated the autonomic innervation of the penis by using evoked cavernous activity (ECA). We recruited seven men with thoracic spinal cord injury (SCI) and sexual dysfunction, and six men who were scheduled to have pelvic surgery (PS), specifically non-nerve-sparing radical cystoprostatectomy. In the PS patients, ECA was performed both pre- and postoperatively. The left median nerve was electrically stimulated and ECA was recorded with two concentric electromyography needles placed into the right and left cavernous bodies. We simultaneously recorded hand and foot sympathetic skin responses (SSRs) as controls. In the SCI group, all but one patient had reproducible hand SSRs. None of these patients had ECA or foot SSRs. All the PS patients had reproducible ECA and SSRs, both preoperatively and postoperatively. There was no difference in the latency and amplitude measurements of ECA and SSRs in the postoperative compared with that of the pre-operative period (P40.05). In conclusion, ECA is absent in men with SCI above the sympathetic outflow to the genitalia. In men, after radical pelvic surgery, ECA is preserved, indicating the preservation of sympathetic fibers. International Journal of Impotence Research (2009) 21, 301–305; doi:10.1038/ijir.2009.34; published online 16 July 2009

Keywords: ECA; autonomic nerve; electrodiagnostic test; erectile dysfunction; penis

Introduction more, there are no direct measures of parasympathetic function for the penis. Intact autonomic neural pathways are prerequisites In earlier studies, we showed the absence of for the vascular changes during erection and ECA in men with dysautonomias2 and in most men detumescence, with signals traveling through the after non-nerve-sparing radical pelvic surgery.4 The spinal cord tracts, the pelvic and hypogastric absence of ECA was interpreted as the disruption plexuses, and the cavernous nerves. Only a few of sympathetic innervation to the corpora, which tests exist to assess the autonomic components of was presumed to affect the erectile function. We genital innervation. One autonomic testing tech- conducted this study to further investigate the nique called evoked cavernous activity (ECA) autonomic innervation of the penis and the associa- measures intrapenile electrical activity after a brief tion to erectile function by using ECA to evaluate noxious stimulus. An applied noxious stimulus two groups of men with erectile dysfunction (ED): results in a generalized, sympathetic nervous system one cohort with ED from central nervous system discharge that manifests throughout the body, disruption due to spinal cord injury (SCI), and one including the corpus cavernosum.1,2 In healthy, group with peripheral (cavernous) nerve disruption potent men of varying ages, ECA is seen with due to pelvic surgery (PS). Our hypothesis was that consistent latency measurements.3 Although the either peripheral or central injury to the autonomic sympathetic discharge mediates corporal vaso- innervation to the penis would result in loss of constriction, and thus is anti-erectile, it is still a ECA. This result would be the basis from reflection of overall autonomic innervation. Further- which ECA could be used as a diagnostic test for neurogenic ED.

Correspondence: Dr CC Yang, Department of Urology, University of Washington, Box 356510, Seattle, WA 98195- Methods 6510, USA. E-mail [email protected] After the study protocol approval by the Human Received 22 April 2009; revised 15 June 2009; accepted 15 Subjects Review Committee of our hospital, we June 2009; published online 16 July 2009 recruited seven men with complete thoracic SCI and ECA: neuroanatomic considerations U Yilmaz et al 302

Figure 1 (a) Diagram of evoked cavernous activity (ECA) reflex. Afferent signals travel through the median nerve to the CNS. The resultant sympathetic discharge is manifested throughout the body, including the erectile tissue. (b) Waveforms of ECA and sympathetic skin responses (SSRs) of a pelvic surgery patient in the pre-operative period. The stimulus is delivered at time 0 and the latency is measured between 0 and L, whereas the amplitudes are measured from the peak of the first deflection to the peak of the after (opposite) deflection. The time-base is 1 s per division, and the entire trace measures 10 s.

sexual dysfunction, and six men who were scheduled If ECA was present after the first six stimuli, then for PS, specifically non-nerve-sparing radical cysto- no further stimuli were given. If ECA was not readily prostatectomy. All men in the PS group were sexually identifiable after all of the first six stimuli, then active preoperatively. Exclusion criteria for all groups more stimuli were delivered, up to a total of 12. included moderate to severe cardiovascular disease, The latency was measured at the first deflection diabetes mellitus, untreated hypertension, current from baseline after the stimulus delivery, and drug or alcohol abuse, major psychiatric disorder, amplitude measurements were made peak to peak. and unwillingness or inability to complete the Three responses were averaged for final latency questionnaire. Except for spinal cord injury, there and amplitude values. Sympathetic skin responses was no evidence of other neurological disease. (SSRs) were measured simultaneously from the Each participant underwent a history and hand and foot contralateral to the median nerve physical examination, with a detailed urological stimulus. SSRs are slow wave recordings that are and neurological examination of extremities and temporally related to ECA and share similar wave- genitalia. forms (Figure 1). These responses were recorded as a All participants completed the International In- control for the presence of a generalized sympa- dex of Erectile Function (IIEF) questionnaire5 before thetic discharge. Latency and amplitude measure- ECA testing. In the PS group, ECA was performed ments of the SSRs from the hand and foot were preoperatively and at the third post-operative month measured in a standard manner.6 visit.

Statistical analyses ECA testing The groups were compared using the Kruskal– The electrodiagnostic technique for ECA was pre- Wallis and Mann–Whitney U-tests. In the PS group, viously described.2–4 Briefly, a noxious stimulus the parameters were also compared between the was delivered to the median nerve on the left wrist pre-operative and post-operative periods using the from a Viking IV electrophysiology machine (Viasys Wilcoxon test. Healthcare, Madison, WI, USA) with the participant in the supine position. The stimulus amplitude was 5–7 mA, duration 0.5 ms, and delivered at irregular Results intervals of at least 60 s in to avoid habituation. ECA was recorded from both corpora cavernosa through All participants tolerated the procedure well. concentric 28-gauge gold-plated needle electrodes Average age of the SCI group was younger than the ± ± (20-mm long, 0.4-mm diameter, Nicolet Biomedical PS group (34.7 7.6 years vs 59.1 5.6 years, Inc., WI, USA) placed into the corpora at the lateral Po0.001). aspects at the base of the penis, with the tip of the electrodes positioned in the center of each corporal body. The band-pass filters were set to 0.5–100 Hz. A IIEF ground plate was affixed to the skin overlying the The results of the erectile domain of the IIEF were right anterior/superior iliac crest. tabulated (Table 1). The post-operative PS IIEF and

International Journal of Impotence Research ECA: neuroanatomic considerations U Yilmaz et al 303 Table 1 Mean ECA amplitude and latency measurements

PS group (n ¼ 6) SCI group (n ¼ 7)

Preoperative Postoperative Mean (s.d.) Mean (s.d.) Mean (s.d.)

Latency (ms) Right ECA 1841 (405) 2212.2 (1100) No response Left ECA 1842 (437) 1725 (185) No response Hand SSR 1692 (184) 1675 (360) 1716 (170) Foot SSR 2071 (98) 2577 (1383) No response

Amplitude (mV) Right ECA 295 (140) 493 (592) No response Left ECA 288 (238) 451 (577) No response Hand SSR 2670 (2366) 1763 (854) 2793 (2187) Foot SSR 1658 (1136) 1870 (1298) No response IIEF, erectile domain (total possible score ¼ 30) 16.3 (5.6) 5.6 (6.5) 10.7 (10.3)

Abbreviations: ECA, evoked cavernous activity; IIEF, International Index of Erectile Function; PS, pelvic surgery group; SCI, spinal cord injury; SSR, sympathetic skin response.

Figure 2 Evoked cavernous activity (ECA) in patient with spinal cord injury. (a) Diagram of disrupted ECA reflex and (b) ECA waveforms. Note the absence of activity in both corpora cavernosa and foot sympathetic skin response (SSR). Deflections off the baseline before 1 s in the right ECA and foot SSR trace are artefact. Stimulus is delivered at time 0. The time-base is 1 s per division, and the entire trace measures 10 s. Amplitude sensitivities are set higher than those in Figure 1.

SCI IIEF scores were not statistically different from dry skin, and thus ECA was not recordable. None each other (P ¼ 0.268). of the individuals with SCI had foot SSRs or ECA (Figure 2).

ECA All the PS patients had reproducible ECA and SSRs, Discussion both preoperatively and postoperatively (Figure 1). There was no difference in the latency and ampli- Evoked cavernous activity is a measure of sympa- tude measurements of ECA and SSRs in the post- thetic (autonomic) discharge in the erectile tissue of operative compared with that of the pre-operative the penis. In this study, we examined the measure- period (Table 1, P40.05). Both pre-operative and ments in men with ED due to central nervous system post-operative values were within the normal disruption from spinal cord injury, and men with ED limits.3 Despite the reported loss of erectile function due to peripheral (cavernous) nerve disruption after in the post-surgery group, there was not a corres- radical pelvic surgery. The rationale of selecting ponding change in ECA parameters, indicating no these groups was to explore the pathways involved consistent association between ECA and erectile in the transmission of autonomic signals to the function. corpora cavernosa, and the possible association with In the SCI group, all but one participant had erectile function. In men with spinal injury above reproducible hand SSRs. This participant had very the level of the sympathetic outflow to the penis

International Journal of Impotence Research ECA: neuroanatomic considerations U Yilmaz et al 304 (T10-L2), there was no ECA response, and in men participants in our study also suggests a lateral with peripheral nerve injury after surgery, there was location of the sympathetic nerves to the penis, a persistence of ECA. These findings can be used to away from the main surgical site, thereby preserving further define autonomic penile innervation. sympathetically mediated neural activity (Figure 3). The absence of ECA and foot SSR in men with The autonomic fibers in the neurovascular bundle SCI between T1-T9 is what would be expected: immediately adjacent to the prostate are likely the sympathetic outflow to the penis and lower primarily parasympathetic fibers. This finding of extremities is derived from the T10-L2 spinal differentially located autonomic fibers may also segments, and any injury to the spinal cord proximal explain ED with shortening of penile length after to these levels precludes normal autonomic outflow non-nerve-sparing radical prostatectomy proce- to these structures. Although the noxious stimulus dures.10 Injury of parasympathetic fibers, but pre- is able to enter the CNS through intact somatic servation of more laterally located sympathetic pathways through the median nerve (C5-C7), the fibers, can cause an autonomic imbalance within sympathetic outflow of the reflex is disrupted the penis, resulting in the dominance of sympa- proximal to the nerve roots to the penis and lower thetically mediated smooth muscle contraction of extremities (Figure 2). Thus, central disruption of the corpus cavernosum, and subsequent ED and the autonomic pathways results in no recordable penile shortening. All these findings further ECA or SSR. support the concept that the cavernous nerve is Peripheral injury to the autonomic fibers presents not a singular structure within a neurovascular a different scenario. In this study, each individual in bundle, but a ‘net’ or ‘layer’ of fibers extending from the surgical cohort served as his own control, and the anterior surface of the prostate, and extending we found that ECA is reproducibly present with no posterolaterally to the lateral rectal wall.11 Another changes in the event latency, after non-nerve- explanation for the presence of ECA can be based sparing pelvic surgery that should have ablated the on an alternate route of penile sympathetic inner- path of the autonomic fibers to the penile erectile vation, such as through the dorsal nerve of the tissue. In our previous study, 8 of 11 participants penis.12 after non-nerve-sparing prostatectomy did not have Owing to the persistence of ECA in this cohort of measurable ECA,4 which we interpreted as denerva- men after radical pelvic surgery, it is not likely that tion after surgery. However, the average age of these ECA is going to be able to diagnose men with ED due men was significantly older than the nerve-sparing to iatrogenic peripheral nerve injury, and cannot be cohort, and it may have been that senescent changes used to prognosticate on the return of erectile to the erectile tissue was the primary determinant function, as we previously believed. Our original for the loss of ECA, rather than the fact that the men assumption was incorrect that ECA, which is a had undergone non-nerve-sparing radical prosta- measure of sympathetic activity, was a surrogate tectomy. measure for the parasympathetic function of the The persistence of ECA after the radical pelvic autonomic fibers, because we believed that the surgery can possibly be explained by the highly parasympathetic and sympathetic fibers ran together variable course of the parasympathetic and sympa- along the cavernous nerve neurovascular bundle thetic fibers through the pelvic plexus to the erectile alongside the prostate gland. However, ECA still tissue. One study showed the distribution of may be useful in diagnosing denervation due to autonomic fibers to be wider than expected, extend- neurological disease.2 ing posterolaterally on the rectal wall, separate from the neurovascular bundle immediately adjacent to the prostate.7 It was also suggested that the cavernous nerve had interindividual variations in its course near the apex of the prostate and the surgically defined neurovascular bundle is often likely to differ from the actual axial course of the cavernous nerve fibers passing through the para- rectal space and the rectourethral muscle. Therefore, an axial course of the cavernous nerve fibers lateral to the surgical field seemed to provide ‘unexpected nerve-sparing’ after non-nerve-sparing radical prostatectomy, because the fibers are likely to be located 8 9 outside the surgical margin. Terada et al. reported Figure 3 The surgical field of a radical cystoprostatectomy is that of the 16 instances in which the neurovascular schematically presented. Autonomic fibers innervating the penis bundle was macroanatomically resected, a positive are known to travel on the posterolateral aspects of the prostate gland. On the basis of the evoked cavernous activity responses, intracavernous pressure increase after intraoperative we posit that the sympathetic fibers lie lateral to the area of electrical stimulation was detected in five cases resection, and thus are spared, as compared with that of the more (31%). We think the persistence of ECA in the medial parasympathetic fibers, which are resected.

International Journal of Impotence Research ECA: neuroanatomic considerations U Yilmaz et al 305 Conclusion 3 Yang CC, Yilmaz U, Vicars BG. Evoked cavernous activity: normal values. J Urol 2008; 179: 2312–2316. 4 Yilmaz U, Ellis W, Lange P, Yang C. Evoked cavernous activity: Evoked cavernous activity is not present in men measuring penile autonomic innervation following pelvic with SCI above the sympathetic outflow to the surgery. Int J Impot Res 2006; 18: 296–301. genitalia, indicating that it is a centrally generated 5 Rosen RC, Riley A, Wagner G, Osterloh IH, Kirkpatrick J, signal. In men, after radical pelvic surgery, ECA is Mishra A. The international index of erectile function (IIEF): a preserved, indicating that at least sympathetic fibers multidimensional scale for assessment of erectile dysfunction. Urology 1997; 49: 822–830. are preserved, which warrants further studies to 6 Vetrugno R, Liguori R, Cortelli P, Montagna P. Sympathetic delineate periprostatic autonomic fibers relative to skin response: basic mechanisms and clinical applications. the sympathetic/parasympathetic components. ECA Clin Auton Res 2003; 13: 256–270. does not seem to be clinically useful for diagnosing 7 Takenaka A, Tewari A, Hara R, Leung RA, Kurokawa K, Murakami G et al. Pelvic autonomic nerve mapping around peripheral penile autonomic disruption. the prostate by intraoperative electrical stimulation with simultaneous measurement of intracavernous and intra- urethral pressure. J Urol 2007; 177: 225–229. Conflict of interest 8 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 The authors declare no conflict of interest. apex: histologic study using cadavers. Urology 2005; 65: 136–142. 9 Terada N, Arai Y, Kurokawa K, Ohara H, Ichioka K, Matui Y Acknowledgments et al. Intraoperative electrical stimulation of cavernous nerves with monitoring of intracorporeal pressure to confirm nerve- This work was supported by NIH NIDDK 1 R21 sparing during radical prostatectomy: early clinical results. Int J Urol 2003; 10: 251–256. DK069315. 10 Savoie M, Kim SS, Soloway MS. A prospective study measuring penile length in men treated with radical prostatectomy for prostate cancer. J Urol 2003; 169: 1462–1464. References 11 Sievert KD, Hennenlotter J, Laible I, Amend B, Schilling D, Anastasiadis A et al. The periprostatic autonomic nerves— 1 Yarnitsky D, Sprecher E, Barilan Y, Vardi Y. Corpus caverno- bundle or layer? Eur Urol 2008; 54: 1109–1116. sum electromyogram: spontaneous and evoked electrical 12 Giuliano F, Rampin O, Jardin A, Rousseau JP. Electrophysio- activities. J Urol 1995; 153(3 Pt 1): 653–654. logical study of relations between the dorsal nerve of the penis 2 Yilmaz U, Soylu A, Ozcan C, Kutlu R, Gunes A. Evoked and the lumbar sympathetic chain in the rat. J Urol 1993; 150: cavernous activity. J Urol 2002; 167: 188–191. 1960–1964.

International Journal of Impotence Research