Cavernous Nerve Stimulation and Interposition Grafting: A Critical Assessment and Future Perspectives

POST–RADICAL PROSTATECTOMY ERECTILE DYSFUNCTION Cavernous Nerve Stimulation and Interposition Grafting: A Critical Assessment and Future Perspectives John P. Mulhall, MD Department of Urology, Weill Medical College of Cornell University, New York Presbyterian Hospital and Memorial Sloan-Kettering Cancer Center, New York, NY The sexual dysfunction that results from radical prostatectomy for carcinoma of the prostate is well established, with the degree of macroscopic preservation of the cavernous nerves tied to the degree of postoperative recovery of erectile function that is possible. In addition to the use of preoperative neuroprotective drugs and postoperative erectogenic agents, intraoperative nerve stimulation and grafting offer promise. Nerve stimulation may serve as a predictor of postoperative potency, and nerve grafting offers a potential way to correct the damage that occurs during wide resection. This article reviews the current literature on these intraoperative measures and discusses the need for additional studies of their potential benefits in prostatectomy candidates. [Rev Urol. 2005;7(suppl 2):S18-S25] © 2005 MedReviews, LLC Key words: Erectile dysfunction • Radical prostatectomy • Cavernous nerve grafting • Cavernous nerve stimulation n estimated 200,000 men will receive a diagnosis of prostate cancer in 2005, and in more than 70% of these patients, the disease will be clinically localized.1 Of these men, approximately half will undergo radical prostatectomy (RP) for definitive local therapy.2 Although this procedure is technically successful for long-term disease control, the resulting sexual dysfunction A S18 VOL. 7 SUPPL. 2 2005 REVIEWS IN UROLOGY Cavernous Nerve Stimulation and Grafting experienced by many men may be burdensome. The consequences of sexual dysfunction—specifically, reduction in self-esteem, impaired psychological well-being, and relationship discord—are well established.3,4 Before Walsh and Donker’s5 seminal description of the anatomic radical retropubic prostatectomy in 1982, erectile dysfunction (ED) was an invariable complication of RP. Since the introduction of this technique, potency preservation rates ranging from 20% to 90% have been reported.1,6 It is irrefutable that the nervesparing status of an RP is predictive of the degree of recovery of erectile function (EF), bilateral nerve sparing being associated with spontaneous and oral therapy–assisted EF results superior to those seen after unilateral nerve sparing. Unilateral nerve sparing is, in turn, more likely to lead to functional erections than is non–nerve-sparing surgery.7 The term “nerve sparing” generally refers to the macroscopic preservation of the cavernous nerves. Although rarely reported quantitatively (at least to this point), there are clearly degrees of nerve sparing, depending on the amount of nerve handling, nerve stretching, and electrocautery use during the RP, and preliminary reports suggest that such factors may be predictive of EF outcomes.8 Much attention is currently focused on strategies to minimize neural trauma or correct it proactively. Such strategies include preoperative, intraoperative, and postoperative maneuvers. Preoperative strategies potentially include the use of medications with neuroprotective properties to minimize cavernous neurapraxia.9,10 Intraoperative approaches include optimization of surgical technique, cavernous nerve interposition grafting, the use of intraoperative neurostimulation to aid in nerve identification before prostatectomy, and the defining of cavernous nerve function after prostate removal but before closure.11,12 Postoperative efforts are currently focused on the use of neuromodulatory drugs, which may have neuroregenerative properties. This review presents a critical assessment of the status and future of intraoperative nerve stimulation and grafting. Intraoperative Cavernous Nerve Stimulation Rationale Nerve identification is sometimes difficult before and/or after the extirpation of the prostate, because of the patient’s anatomy and/or intraoperative bleeding. The use of intraoperative cavernous nerve stimulation may aid in identifying the cavernous nerves cavernous nerve stimulation.15 Furthermore, assessment of cavernous nerve response to neurostimulation after prostate extirpation may aid in defining nerve preservation status at the end of the operation. Technical Aspects The nerve localization technique was initially described by Lue and coworkers16 and was later formally developed as an intraoperative tool (CaverMapTM, Blue Torch Medical Technologies, Inc., Ashland, MA) for nerve stimulation and tumescence monitoring. The tip of this device, which contains an array of electrodes, is placed over the region where the cavernous nerves are thought to be situated, and a biphasic Although rarely reported quantitatively, there are clearly degrees of nerve sparing, depending on the amount of nerve handling, nerve stretching, and electrocautery use during the radical prostatectomy. before prostate removal so as to minimize trauma to the nerves. Nerve injury can occur at 4 time points during radical retropubic prostatectomy: at the time of urethral dissection, during lateral pedicle division, during dissection along the base of the prostate, and at the time of seminal vesicle dissection. Classically, nerve trauma is thought of as occurring with cautery, ligation, or avulsion; however, overly aggressive traction may lead to neurapraxia, and even this temporary trauma may result in structural sequelae in the corporeal bodies that may limit the recovery of EF.13,14 Indeed, in our laboratory, using a rat model of cavernous nerve injury, we have demonstrated that nerve exposure alone without any direct manipulation results in a significant reduction in EF as measured by intracavernous pressure generation in response to current is applied. Escalating stimulation intensities from 8 to 20 mÅ are then applied.17 The nerves are usually stimulated at multiple sites along their course before and after removal of the prostate. Tumescence is monitored by a penile strain gauge, which records changes in penile girth. The strain gauge system is sensitive enough to detect a 0.5% change (increase or decrease) in circumference. It is worth noting that neurovascular bundle (NVB) stimulation may result in both tumescence and detumescence responses, and sometimes the detumescence response may precede the tumescence response (F. Rabbiani et al, unpublished data, 2000). The physiologic explanation for the latter occurrence is simple. The cavernous nerves are combined parasympathetic (erectogenic) and sympathetic (erectolytic) nerves VOL. 7 SUPPL. 2 2005 REVIEWS IN UROLOGY S19 Cavernous Nerve Stimulation and Grafting continued (J.P.M., unpublished data, 2005). Stimulation of nerve fibers that are predominantly sympathetic may result in a detumescence response, in contrast to parasympathetic nerve fiber stimulation. The original device design incorporated a module that signaled both visually and aurally the level of response. The latest generation presents the level of response on a laptop computer in graphic form. The current protocol involves application of an 8-mÅ stimulus followed by 10 mÅ, 14 mÅ, and 20 mÅ, each for 20 seconds. This approach permits standardization of data analysis. Literature Review In the initial study on CaverMap, reported by Klotz and Herschorn,17 21 men with functional erections before RP had application of the device intraoperatively. Postoperative EF assessment was performed using a questionnaire. A girth change of 0.5% was used as a cutoff for a positive response on CaverMap. Of the 19 patients Two years later, Klotz18 led a Canadian-based multicenter study using 9 different surgeons, which randomized men, in a single-blind fashion, to either use or no use of CaverMap. All patients had normal EF preoperatively, were younger than 70 years, and had clinically organconfined disease. CaverMap was applied both before and following prostate removal. EF was assessed using a questionnaire and by penile tumescence and rigidity monitoring (Rigiscan®; Endocare, Inc., Irvine, CA). Fifty-three patients with a mean age of 60.5 years were originally enrolled (26 patients in the control group, 27 in the CaverMap group), but only 35 were available for the 12-month EF assessment. Of note, the margin positivity rate was 40%, although there was no difference in this rate between the 2 groups. Among the 53 patients, following prostate removal, 31 had bilateral CaverMap response (
/
), 27 had a unilateral response (
/), and 3 had The initial CaverMap study demonstrated the feasibility of using intraoperative neurostimulation and suggested that a CaverMap response may portend a better prognosis for postoperative erectile function. who had EF assessed postoperatively, the 2 men who had no intraoperative CaverMap response did not regain functional erections after surgery. Sixteen (94%) of 17 who had a response had functional erections following surgery. This study demonstrated the feasibility of using intraoperative neurostimulation and suggested that a CaverMap response may portend a better prognosis for postoperative EF. Although this study failed to adequately define the population studied, it clearly demonstrated the feasibility of using this technology and its potential for predicting EF outcomes. S20 VOL. 7 SUPPL. 2 2005 no response on either side (/). There was no difference in the overall erectile rigidity between the groups, as assessed by Rigiscan (percentage of men with base rigidity  60%), but the mean and median durations spent at this level of base rigidity were significantly greater in the positive response group (15.9 minutes and 3 minutes vs 2 minutes and 1 minute, respectively). This work further supported the use of CaverMap using 0.5% girth change as a cutoff and a Rigiscan parameter of basal rigidity  60% as a positive EF outcome. Weaknesses in the report are that the authors failed to fully define the nature of the patient REVIEWS IN UROLOGY population, had only 35 patients available for 12-month analysis, and did not attempt to correlate degree of CaverMap response to EF outcomes. In 2000, Kim and associates19 from the University of Chicago published their experience with CaverMap. Sixty men with normal preoperative EF (mean age, 59 years) had CaverMap used during RP. Postoperative EF was assessed using items extracted from the UCLA prostate cancer index, which was mailed to the patients at 6 and 12 months postoperatively. EF was also deemed to be intact if more than 50% of attempts resulted in erectile rigidity firm enough for penetration. Thirty-five (58%) of 60 patients had bilateral nerve sparing and 25 (42%) of 60 had unilateral nerve sparing. Eighty percent of the entire group had a positive CaverMap response. In the group that had bilateral nerve sparing, 63% had a bilateral response, 26% had a unilateral response, and 11% had no response. Of note, the authors used a 1% change in penile girth as the cutoff because, in this case, using a cutoff of 0.5% resulted in a significant proportion of men responding to nonspecific (non–cavernous nerve) stimulation. At 12 months after surgery, 11 (18%) of 60 patients had functional erections (5 of 11 in response to sildenafil). In men with a
/
response, 27% had functional erections; in those with a
/ response, 15% did; and among those with a / response, no patient had return of functional erections. This study raises the issue of standardization of CaverMap endpoints and the clinical meaning of nonspecific stimulation, for both of which we still have no definitive answer. Other possible criticisms include the overall low rate of EF recovery (18%) and again the fact that no attempt was made to correlate degree of response with EF outcomes. The Vanderbilt group reported on their experience in 2001.20 Sixty-one Cavernous Nerve Stimulation and Grafting patients with a mean age of 59.8 years underwent CaverMap stimulation during RP. Bilateral nerve-sparing surgery was conducted in 76%, unilateral nerve-sparing surgery in 8%, and non–nerve-sparing surgery in 16%. Following prostate removal, 58% of the stimulations along the right NVB and 58% along the left NVB resulted in penile tumescence. However, 43% of stimulations along the bladder neck (nonspecific stimulation) resulted in some change in penile girth. Based on these data, the authors questioned the value of the device in the form (first-generation) used at the time of this study. This analysis failed to define the preoperative or postoperative EF of the patients and concentrated on the ability of cavernous nerve stimulation to elicit responses in penile girth. This work also raised questions about the concept of “percentage of stimulations resulting in penile girth alterations.” Is this a parameter that has real meaning or should an individual cavernous nerve be considered to have responded if it responds before prostate removal to any stimulation along its path? Where along the cavernous nerve path should stimulation be performed after prostate removal? Should it not be standardized to occur at the base so that the integrity of the distal segment of the nerve can be tested? Distal stimulation gives little assessment of the potential nerve trauma proximally and thus must have minimal clinical significance. In follow-up to this work, Chang and coworkers21 from the same center examined men who underwent prostatectomy with either bilateral or no nerve sparing coupled with intraoperative neurostimulation. Twentytwo men underwent wide resection of the NVBs and all were impotent postoperatively. Intraoperatively, 16 of 22 had no response to neurostimulation, but 6 patients did have a response. Forty-one patients underwent bilateral nerve-sparing procedures; 27 were potent at 1-year followup by patient’s self-report. Thirty of these 41 men had tumescence on intraoperative neurostimulation, and 24 (80%) of these 30 were potent post-operatively. Thus, in this analysis, intraoperative CaverMap response predicted postoperative potency (P  0.017). This work raises the question of whether it is appropriate for the CaverMap response to be compared with surgeon-reported nerve-sparing status. In Chang and associates’ paper,21 6 (27%) of 22 patients undergoing non–nerve-sparing surgery had a positive CaverMap response. Does this mean that this is a nonspecific (non–cavernous nerve) response? What if surgeon reporting of cavernous integrity is inaccurate? For example, approximately 15% of patients in my practice who have been told they have had non–nerve-sparing surgery will elicit some response to phosphodiesterase (PDE)-5 inhibitors, suggesting that some neural tissue remains. In 2001, a multi-institutional study was published reviewing the experience of a group of acclaimed surgeons who perform nerve-sparing prostatectomies.22 Five surgeons enrolled a total of 50 patients for CaverMap stimulation during RP. All patients were younger than 60 years and had clinically organ-confined prostate cancer. For inclusion, patients were required to score 4 or 5 on each question of the EF domain of the International Index of Erectile Function (IIEF) questionnaire (4 refers to “most of the time” and 5 to “always or nearly always” regarding the frequency of various parameters such as development of rigidity adequate for penetration, ability to maintain erection, satisfaction, and confidence). Follow-up of the patients occurred at 3, 6, and 12 months. Ninety percent of patients had bilateral nerve-sparing surgery; the remainder had unilateral nerve-sparing surgery. This report presents data in a similar fashion to the original Vanderbilt report, in that it presents the number of stimulations attempted and the percentage that resulted in a positive CaverMap response, 336 for 100 NVBs (3.3 per nerve). A positive response resulted from 87.8% of all stimulations. Only 6.4% of all stimulations resulted in a tumescence response alone, 32.2% resulted in detumescence, and the remainder resulted in initial tumescence followed by detumescence (or viceversa). Eighty-eight percent of all stimulations (295 of 336) elicited a positive response, but of 54 stimulations overlying nerve bundles that were resected (based on surgeon report), 29 (54%) resulted in no response. The authors concluded based on these figures that CaverMap had a high sensitivity (88%) and a low specificity (54%). This work from high-volume academic RP surgeons, based on specificity and sensitivity data, concluded that there was little role for intraoperative neurostimulation. However, this report can be criticized, as can the original Vanderbilt paper, because sensitivity and specificity were based on the use of “percentage of stimulations resulting in penile girth changes.” What would these parameters have been if the surgeons had counted any positive response as positive? Recent data from Memorial SloanKettering Cancer Center (MSKCC) on the use of a third-generation device demonstrated a 0% potency return (with and without sildenafil use) at 12 months for patients with a / response, a 46% incidence of return of function with
/ response, and an 88% chance of postoperative erections with
/
intraoperative response to neurostimulation (data on VOL. 7 SUPPL. 2 2005 REVIEWS IN UROLOGY S21 Cavernous Nerve Stimulation and Grafting continued file). These figures are in keeping with the Klotz18 and Smith21 data. Critique The greatest challenge for the interpretation of CaverMap data is that the responses are compared with the surgeon’s assessment of the integrity of the cavernous nerve that is being stimulated, and the latter is held as the gold standard. What if surgeons overestimated the integrity of the cavernous nerves or at least overestimated the function of the nerves? Indeed, the whole concept of nerve-sparing surgery has been predicated upon the macroscopic visual assessment by a non-independent evaluator (the surgeon), which clearly has inherent biases associated with it. Furthermore, the concept of bilateral, unilateral and non–nervesparing surgery is clearly a historical one, since we should be moving toward the more meaningful idea of what percentage (proportion) of each indicate that this is a patient for whom oral PDE-5 inhibitor therapy may work. Indeed, this has been our experience at MSKCC. What if all users of CaverMap were to quantify the response to stimulation, for example, 1%, 2% to 4%, or greater than 4%? Might this improve the predictive value of the device? Rabbani and associates23 have demonstrated this, showing that responses of greater than 4% resulted in a greater likelihood of EF recovery compared with responses of less than 1%. Finally, there is a need for some form of consensus regarding the definition of a positive response, specifically, what girth change represents a positive response, as well as on the value as a parameter of reporting percentage of positive responses. The Future There is an interest in developing this tool further. Perhaps endpoints other than penile tumescence/ The concept of bilateral, unilateral, and non-nerve-sparing surgery is clearly a historical one, since we should be moving toward the more meaningful idea of what percentage (proportion) of each cavernous nerve has been preserved. cavernous nerve has been preserved. For example, for a case in which 50% of both cavernous nerves are spared (50% resected bilaterally), is this bilateral or non–nerve-sparing surgery? The CaverMap response tells us important information, that is, that there is some functional neural tissue present at the end of the case, but it fails to quantify the degree of preservation and cannot account for postoperative factors that may be nerve-threatening. Indeed, it is a common event in my practice that a patient has a functional erection within 6 weeks (with or without oral erectogenic agents) but loses this response 2 months later. CaverMap response may, however, S22 VOL. 7 SUPPL. 2 2005 detumescence will yield data that are more interpretable. Should we be using intracavernosal pressure measurement or transcutaneous oxygen measurement or Doppler assessment of the cavernosal arteries during cavernous nerve stimulation? It is likely that nerve function assessment using CaverMap will continue to be explored at select centers and that, with further refinements in the technology, improved computer data generation, and a greater understanding of the implications of the data, intraoperative neurostimulation may in the near future assume a role in predicting at least oral therapy response following RP. What is clear is that REVIEWS IN UROLOGY failure to elicit a response on either side will likely translate into the absence of spontaneous EF postoperatively, and this information alone may be of significant value to patients in predicting a return of function as well the need for non–oral therapy–based erectogenic strategies. While it is likely that the absence of CaverMap response augurs poorly for oral PDE-5 inhibitor response, these data are currently not available. Cavernous Nerve Interposition Grafting Rationale Resection of the NVB may be indicated in patients with high-grade disease or advanced unilateral or bilateral disease (multiple positive cores, significant length of positive core, high Gleason grade, suggestion of extracapsular extension on imaging) to decrease the risk of positive surgical margins. Positive surgical margins increase the chance of treatment failure. Extracapsular extension increases the risk of a positive margin, and this occurs most frequently in the area of the NVB. Wide resection, which may include the NVB, decreases the risk of a positive surgical margin but, without intervention to restore continuity of the nerves, increases the chance of postoperative ED. Technical Aspects The initial experience using nerve grafts was in rodents more than a decade ago. As reported in 1991, Quinlan and colleagues24 at Johns Hopkins used the genitofemoral nerve grafted between proximal and distal stumps of the rat cavernous nerve (a single nerve, in contradistinction to the human nerve plexus) to demonstrate that this was technically feasible and that it increased the chances of functional recovery of erections, although formal hemodynamic assessment of erectile response to cavernous nerve injury was Cavernous Nerve Stimulation and Grafting not used then, as is currently standard. During RP, following cavernous nerve resection, an assessment can be made of the ipsilateral or contralateral genitofemoral nerve for caliber matching. If the caliber is acceptable, then the graft can be harvested and sutured with fine nonabsorbable sutures between the 2 stumps of the cavernous nerve. If not, then the sural nerve can be harvested, as described by Kim and associates.25 The harvesting of either nerve adds minimal time to the operation and is done at MSKCC by our plastic surgery colleagues. Literature Review The first report was by Kim and colleagues26 from Baylor College of Medicine, which discusses their experience with 9 men with excellent preoperative EF who underwent non–nerve-sparing RP for locally extensive, high-grade prostate carcinoma. All underwent bilateral sural nerve interposition grafting (SNG). Postoperative EF was assessed using patient interview, a nonvalidated questionnaire, and Rigiscan analysis. Of the 9 patients, 1 had spontaneous functional erections at 14 months postoperatively. This experience demonstrated feasibility and encouraged these investigators to continue to accrue patients for SNG. The following year, they reported on their experience in treating 12 men with SNG and genitofemoral nerve interposition grafting (GNG).27 The data from this analysis indicated that erections unassisted by medication and sufficient for sexual intercourse occurred in one third of patients; partial erections occurred in 42%, and the remainder had no return of erections. The mean IIEF EF domain scores for the 3 groups were 16  10, 9  10, and 7  1, respectively. Erections with sildenafil use were achieved in 50% at 12 months after RP. Twelve potent men who had non–nerve-sparing surgery and had no grafting acted as controls. In this group, none had return of functional erections with or without sildenafil. These data suggest that bilateral interposition nerve grafting improves EF in men undergoing non–nerve-sparing surgery. The major criticism of this work, however, is the relatively low EF domain scores achieved, even in the group reporting spontaneous functional erections, which raises concern about the consistency of erectile response in these men given that the IIEF is a frequency questionnaire (eg, “over the past 4 weeks how often were you able to . . . ?”) A follow-up study from the same group included 28 men who underwent bilateral cavernous nerve resection and SNG/GNG, 23 of whom had follow-up of at least 12 months.28 The data demonstrated that 26% of patients had spontaneous medically unassisted erections sufficient for intercourse, 26% had partial erections, and 48% had no erectile activity whatsoever. In these 3 groups, the mean EF domain scores were 20  6, 10  5, and 6  3, respectively. Of note, using sildenafil, the overall potency rate (ability to achieve vaginal penetration) was 43%. This was in contrast to only 1 of 70 patients in their database who underwent bilateral nerve resection and had return of functional erections. Thus, the early work suggested that, using this approach, approximately 50% of patients could have functional erections, at least with the use of sildenafil. Indeed, this latter report showed better EF domain scores in the functional group, indicating that some patients even achieved normalization of their EF domain score ( 26). The MD Anderson Cancer Center experience was published in 2003.29 Thirty men with functional erections preoperatively underwent bilateral cavernous nerve resection and grafting. At a mean follow-up of 23 months, 18 (60%) of 30 had subjective and objective changes in penile tumescence with sexual stimulation. However, only 13 (43%) were capable of having sexual intercourse, 7 (23%) without sildenafil and 6 (20%) with sildenafil. The only report presently published on unilateral nerve grafting comes from Baylor and assesses the impact of unilateral nerve grafting on urinary function.30 In a nonrandomized study, 53 patients who underwent unilateral SNG were compared with 58 men who had no grafting. The endpoints were the time to urinary function recovery above the median for the group and overall urinary control. At 12 months post-RP, complete urinary control was attained by 94.7% of graft patients compared with 58.3% of nongraft patients. The graft group was 10 times more likely to attain a urinary function score above the median. Grafting increased the odds of attaining complete urinary control by 15-fold and 29-fold at 6 and 12 months postoperatively, respectively. There is no report assessing the impact of unilateral cavernous nerve grafting on EF outcomes. There are 2 recent papers discussing the feasibility of performing cavernous nerve interposition grafting using laparoscopic31 and robotic32 approaches, and both demonstrate that it is possible, but no functional data are presented in either report. At MSKCC, in a nonrandomized fashion, we have analyzed patients who underwent RP with either unilateral or bilateral nerve resection in association with nerve grafting, who were potent before surgery and completed the IIEF questionnaire in a serial fashion. In this analysis, 64 patients underwent unilateral nerve resection with nerve grafting, and 21 patients had bilateral nerve resection with VOL. 7 SUPPL. 2 2005 REVIEWS IN UROLOGY S23 Cavernous Nerve Stimulation and Grafting continued nerve grafting. For patients undergoing unilateral nerve grafting, 4-year actuarial recovery of an erection permitting penetrative sexual relations on at least some occasions (EF domain score  17) was 40%, and for patients who underwent bilateral NVB resection with nerve grafting, the 3-year actuarial recovery was also 40%. Cavernous nerve grafting appears to be associated with minimal complications, which are limited to the morbidity associated nerve resection is uncommon, a point emphasized by Dr. Patrick Walsh33 in a recent journal editorial. Another challenge to nerve grafting as a procedure is the “hassle factor” for surgeons. That is, most surgeons do not have the technical experience to do this themselves and do not have an association with an interested plastic surgeon or neurosurgeon who might assist in performing the actual interposition grafting. There is a dire need for the devel- There is a dire need for the development and funding of randomized controlled trials to assess the utility of nerve grafting. with harvesting the sural nerves (genitofemoral nerve harvesting has no significant morbidity), such as incisional pain and hyposthesia over the lateral aspect of the dorsum of the foot (a problem the majority of patients do not complain about) and increased operative time. Critique One of the challenges to the propagation of interposition nerve grafting is the fact that bilateral deliberate opment and funding of randomized controlled trials to assess the utility of nerve grafting. The question cannot be answered definitively in any other way, and the detractors will continue to criticize the procedure until such data are available. In the world of prostate cancer, however, this strategy ranks low on the list of priorities. The major criticisms of the data available thus far, besides the noncontrolled nature of the studies, are the short duration of follow-up and the small numbers of patients. It is possible that the short follow-up periods undermine the value of the approach. Indeed, of the post-RP patients in my practice, the nerve graft patients are the group to most commonly experience continued recovery of EF past 2 years. The patient number issue is a function of the relatively uncommon need for nerve resection in the current era of patient screening. Despite these limitations, the data indicate that men undergoing non–nerve-sparing surgery who have bilateral nerve grafting retain a significant chance of a functional response to oral erectogenic therapy, in the range of 50%, compared with zero in (nonrandomized) control groups. The Future Given the association between bilateral cavernous nerve resection and the long-term failure to regain spontaneous functional erections or failure to have a significant response to oral therapy, cavernous nerve interposition grafting seems to be a reasonable strategy to restore EF in men for whom bilateral nerve resection is either planned or is required intraoperatively. Thus, in the Main Points • The use of intraoperative cavernous nerve stimulation may aid in identifying the cavernous nerves before prostate removal so as to minimize trauma. The CaverMap device is used to stimulate the nerves before and after removal of the prostate; simultaneous penile tumescence measurements have been shown to be predictive of potency post-surgery. Such information may be useful to patient and physician in planning a postoperative treatment strategy. • While intraoperative cavernous nerve stimulation has been called unnecessary by some, and questions about its optimal application remain, one could argue that the gold standard, visual assessment of cavernous nerve integrity by the surgeon, is less foolproof. • Researchers at Johns Hopkins originally showed that the genitofemoral nerve grafted between proximal and distal stumps of the rat cavernous nerve increased the chances of functional recovery of erections. In humans in whom the neurovascular bundle is resected, the ipsilateral or contralateral genitofemoral nerve or sural nerve may be similarly grafted to improve the chances of postoperative erectile function. • Reports from Baylor College showed that erections unassisted by medication and sufficient for sexual intercourse occurred in up to one third of patients who had undergone bilateral nerve grafting—potentially half with the use of sildenafil—but there is concern about the consistency of the erectile response in these men, the noncontrolled nature of the studies, and the short duration of follow-up. More studies are needed. • Whether unilateral nerve grafting has the potential to restore erectile function after prostatectomy remains to be seen. S24 VOL. 7 SUPPL. 2 2005 REVIEWS IN UROLOGY Cavernous Nerve Stimulation and Grafting small group of men in whom bilateral cavernous nerve resection is necessary, we believe it is appropriate to offer cavernous nerve interposition grafting. Whether unilateral nerve grafting has the potential to have a similar impact remains to be seen; however, for the patient who is acutely interested in postoperative EF, a discussion of the potential pros and minimal cons of unilateral nerve grafting should be considered by the urologic surgeon during preoperative counseling. 4. Summary 9. Strategies to optimize nerve preservation and nerve function are likely to translate into better long-term EF outcomes, at least in response to oral erectogenic pharmaceuticals. Although there is supportive data for both intraoperative neurostimulation and interposition cavernous nerve grafting, widespread utilization of the 2 strategies will not happen without data that definitively prove a link to improved sexual function outcomes in the RP population, and this will be achieved only though large, randomized, multi-institutional studies. 5. 6. 7. 8. 10. 11. 12. 13. 14. 15. References 1. 2. 3. Stanford JL, Feng Z, Hamilton AS, et al. Urinary and sexual function after radical prostatectomy for clinically localized prostate cancer: the Prostate Cancer Outcomes Study. JAMA. 2000; 283:354-360. Harlan LC, Potosky A, Gilliland FD, et al. Factors associated with initial therapy for clinically localized prostate cancer: Prostate Cancer Outcomes Study. J Natl Cancer Inst. 2001;93: 1864-1871. Bokhourm BG, Clark JA, Inui TS, et al. Sexuality after treatment for early prostate cancer: exploring the meanings of “erectile dysfunction.” J Gen Int Med. 2001;16:649-655. 16. 17. 18. Guest JF, Das Gupta R. Health-related quality of life in a UK-based population of men with erectile dysfunction. Pharmacoeconomics. 2002; 20:109-117. Walsh PC, Donker PJ. Impotence following radical prostatectomy: insight into etiology and prevention. J Urol. 1982;128:492-497. Catalona WJ, Carvalhal GF, Mager DE, Smith GS. Potency, continence and complication rates in 1,870 consecutive radical retropubic prostatectomies. J Urol. 1999;162:433-438. Quinlan DM, Epstein JI, Carter BS, Walsh PC. Sexual function following radical prostatectomy: influence of preservation of neurovascular bundles. J Urol. 1991;145:998-1002. Touijer AK, Trabulsi E, Hassen W, et al. Evaluation of erectile function at 3 months in a contemporary series of laparoscopic radical prostatectomy. J Sex Med. 2004;1(suppl):108. Burnett AL, Becker RE. Immunophilin ligands promote penile neurogenesis and erection recovery after cavernous nerve injury. J Urol. 2004; 171:495-500. Sezen SF, Hoke A, Burnett AL, Snyder SH. Immunophilin ligand FK506 is neuroprotective for penile innervation. Nat Med. 2001;7: 1073-1074. Burnett AL. Neuroprotection and nerve grafts in the treatment of neurogenic erectile dysfunction. J Urol. 2003;170:S31-S34. Scardino PT, Kim ED. Rationale for and results of nerve grafting during radical prostatectomy. Urology. 2001;57:1016-1019. User HM, Hairston JH, Zelner DJ, et al. Penile weight and cell subtype specific changes in a post-radical prostatectomy model of erectile dysfunction. J Urol. 2003;169:1175-1179. Carrier S, Zvara P, Nunes L, et al. Regeneration of nitric oxide synthase-containing nerves after cavernous nerve neurotomy in the rat. J Urol. 1995;153:1722-1727. Mullerad M, Donohue JF, Li PS, et al. Functional sequelae of cavernous nerve injury in the rat: model dependency. J Sex Med. 2004;1(suppl 1); 39. Abstract O35. Lue TF, Gleason CA, Brock GB, et al. Intraoperative electrostimulation of the cavernous nerve: technique, results and limitations. J Urol. 1995; 154:1426-1428. Klotz L, Herschorn S. Early experience with intraoperative cavernous nerve stimulation with penile tumescence monitoring to improve nerve sparing during radical prostatectomy. Urology. 1998;52:537-542. Klotz L, Heaton J, Jewett M, et al. A randomized phase 3 study of intraoperative cavernous nerve stimulation with penile tumescence monitoring to improve nerve sparing during radical prosta- 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. tectomy. J Urol. 2000;164:1573-1578. Kim HL, Stoffel DS, Mhoon DA, Brendler CB. A positive cavermap response poorly predicts recovery of potency after radical prostatectomy. Urology. 2000;56:561-564. Holzbeierlein J, Peterson M, Smith JJ. Variability of results of cavernous nerve stimulation during radical prostatectomy. J Urol. 2001;165:108-110. Chang SS, Peterson M, Smith JA. Intraoperative nerve stimulation predicts postoperative potency. Urology. 2001;58:594-597. Walsh PC, Marschke P, Catalona WJ, et al. Efficacy of first-generation Cavermap to verify location and function of cavernous nerves during radical prostatectomy: a multi-institutional evaluation by experienced surgeons. Urology. 2001;57:491-494. Rabbani F, Cozzi P, Scardino PT. Quantitative assessment of the response to CaverMap nerve stimulation at radical prostatectomy [abstract]. J Urol. 2002;167:356. Quinlan DM, Nelson RJ, Walsh PC. Cavernous nerve grafts restore erectile function in denervated rats. J Urol. 1991;145:380-383. Kim ED, Scardino PT, Kadmon D, et. al. Interposition sural nerve grafting during radical retropubic prostatectomy. Urology. 2001;57:211-216. Kim ED, Scardino PT, Hampel O, et al. Interposition of sural nerve restores function of cavernous nerves resected during radical prostatectomy. J Urol. 1999;161:188-192. Kim ED, Nath R, Kadmon K, et al. Bilateral nerve graft during radical retropubic prostatectomy: 1-year follow-up. J Urol. 2001;165:1950-1956. Kim ED, Nath R, Slawin KM., et al. Bilateral nerve grafting during radical retropubic prostatectomy: extended follow-up. Urology. 2001;58: 983-987. Chang DW, Wood CG, Kroll SS, et al. Cavernous nerve reconstruction to preserve erectile function following non-nerve-sparing radical retropubic prostatectomy: a prospective study. Plast Reconstr Surg. 2003;111:1174-1181. Singh H, Karakiewicz P, Shariat SF, et al. Impact of unilateral interposition sural nerve grafting on recovery of urinary function after radical prostatectomy. Urology. 2004;63:1122-1127. Turk IA, Deger S, Morgan WR, et al. Sural nerve graft during laparoscopic radical prostatectomy: initial experience. Urol Oncol. 2002;7:191-194. Kaouk JH, Desai MM, Abreu SC, et al. Robotic assisted laparoscopic sural nerve grafting during radical prostatectomy: initial experience. J Urol. 2003;170:909-912. Walsh PC. Nerve grafts are rarely necessary and are unlikely to improve sexual function in men undergoing anatomic radical prostatectomy. Urology. 2001;57:1020-1024. VOL. 7 SUPPL. 2 2005 REVIEWS IN UROLOGY S25