Edward D. Kim, M.D.¹

Peter T. Scardino, M.D.1

Ori Hampel, M.D.1

Noel Mills, M.D.2

Thomas M. Wheeler, M.D.3

Rahul K. Nath, M.D.4

From the Matsunaga-Conte Prostate Cancer Research Center and

the Scott Department of Urology, ¹

the Department of Neurosurgery and Division of Plastic Surgery, Department of Surgery⁴

and the Department of Pathology,³ Baylor College of Medicine,

The Methodist Hospital, Houston, TX

and

the Department of Surgery,² Tulane University,

New Orleans, LA

Running head: Nerve grafting during radical prostatectomy

Keywords: Nerve grafts; radical prostatectomy; potency; sural nerve

Address correspondence to:

Edward D. Kim, M.D.

Scott Department of Urology

6560 Fannin, Suite 2100

Houston, TX 77030

Tel: 713-798-7737

Fax: 713-798-6007

email: ekim@bcm.tmc.edu

Purpose: The permanent loss of erectile function when both neurovascular bundles are widely resected at the time of radical prostatectomy, and the successful use of autologous nerve grafts in reconstructive surgery, led us to perform bilateral nerve grafts in an effort to restore erectile function in potent patients treated for prostate cancer with radical retropubic prostatectomy (RRP) and resection of both neurovascular bundles (NVB).

Materials and Methods: RRP with deliberate resection of both NVBs was recommended for treatment of high grade, locally extensive prostate cancers in selected sexually active patients who reported normal erectile function. After the prostate was removed, but before the vesicourethral anastomosis, an autologous sural nerve graft was interposed between the divided ends of the cavernous nerves bilaterally. Erectile function was monitored by patient interview, questionnaire, and nocturnal penile tumescence testing after the operation.

Results: At 4-5 months postoperatively, patients observed slowly improving spontaneous erections, manifest by mild tumescence every several hours with regularity. Nocturnal penile tumescence testing with the RigiScan at 4-6 months in 2 patients demonstrated erections that approach minimal criteria for normalcy. Approximately 14 months after surgery, the first patient developed a rigid erection sufficient for penetration and intercourse. The patient described this erection as "an erection of substance--hard, not just fluffy."

Conclusions: We have developed a technique for using grafts of the sural nerve to restore continuity of the cavernous nerves resected during radical prostatectomy. The early return of spontaneous partial erections in our patients suggests that interposition nerve grafts may enhance recovery of erectile function when the neurovascular bundles are resected.

Technical advances over the last 15 years in radical retropubic prostatectomy have enabled surgeons to perform this procedure with diminished risks of troublesome morbidity, such as urinary incontinence and erectile dysfunction.1,2 Yet erectile dysfunction in varying degrees remains a common concern of patients after the operation. When both neurovascular bundles are preserved during radical retropubic prostatectomy, potency rates of up to 71%,2 but generally in the range of 30-60% have been observed.3-9 Despite the ability to achieve intromission occasionally, many such men report seriously impaired quality of erections. Risk factors for failure to recover adequate erections include increasing age, diminished prior erectile function, increasing clinical and pathologic stage, resection of the neurovascular bundle tissue, and even nuances in the surgical technique.2, 9-11

The pathophysiology of erectile dysfunction after radical prostatectomy is primarily neurogenic. The cavernous nerves contain the parasympathetic fibers, originating in spinal cord segments S2-S4, as well as sympathetic fibers from T11-L2. Detailed human cadaver studies have precisely identified these autonomic nerves from the pelvic plexus to the corpora cavernosa, thus enabling the performance of successful nerve-sparing procedures.12 After radical prostatectomy, the probability of recovery of spontaneous erections adequate for intercourse is "quantitatively related to preservation of autonomic innervation." 7 With resection of both neurovascular bundles, no patients have been reported to recover spontaneous erections adequate for intercourse.2,4,7,13

Despite the success of non-invasive forms of treatment for erectile dysfunction, men desire the return of spontaneous erections after radical prostatectomy. Studies in rat models have demonstrated the feasibility and enhancement of erectile function after denervation and placement of cavernous nerve grafts.14,15 While there has been extensive clinical experience with nerve grafts in other anatomical sites, there have been no published reports in humans of nerve grafts to replace resected cavernous nerves.

Based on experience with nerve grafts during the Vietnam war, observations of the expendable sural nerve during lesser saphenous vein harvest for coronary artery bypass grafting, and personal experience with radical prostatectomy, one of the authors (NLM) suggested the use of sural nerve interposition for restoration of erectile function. Consequently, we were stimulated to develop a technique for autologous nerve grafts when both cavernous nerves are resected during radical prostatectomy.

We performed the first bilateral interposition grafts from the sural nerve to both cavernous nerves in March 1997. A 57-year-old married, sexually active man with a history of chronic prostatitis had a palpable nodule in the left lobe of the prostate. The serum PSA level was 2.3 ng/ml. Transrectal ultrasound showed a 1.7 cm. hypoechoic lesion in the left peripheral zone. Needle biopsies of the prostate showed adenocarcinoma in 5 of 6 cores (2 of 3 from the right, Gleason grade 2+3=5; 3 of 3 from the left, Gleason grade 3+4=7) with perineural invasion and extraprostatic extension in the cores from the left. A bone scan and CT scan of the abdomen and pelvis showed no evidence of metastases. Clinical stage was assigned as T3aN0M0. Treatment was initiated with Lupron (leuprolide acetate; TAP Pharmaceuticals, Deerfield, IL) and Eulexin (flutamide; Schering Corp, Kenilworth, NJ) with the patient reporting a gradual decline in libido and in the rigidity of his erections over the next 3 months.

He was referred for definitive treatment 4 months after the biopsy. After careful consideration of the alternatives, he elected treatment with radical prostatectomy. Given the size, extent and grade of his initial lesion, resection of one or both neurovascular bundles was likely to be necessary to minimize the risks of a positive surgical margin. The use of nerve grafts derived from the sural nerve to replace any resected cavernous nerves was discussed in detail with the patient and his family, and he was fully informed of the unknown benefits and the potential risks, including: 1) at the site of the sural nerve procurement: hematoma, infection, pain, temporary sensory deficit in the lateral aspect of the foot, neuroma formation, and delayed ambulation: and 2) at the site of the interposition grafting: infection and blood loss related to increased duration of the procedure.

Five months after the biopsy, we performed a bilateral pelvic lymphadenectomy and radical retropubic prostatectomy. The prostate felt diffusely indurated and there was considerable periprostatic fibrosis, as is common in patients treated with neoadjuvant hormonal therapy. The neurovascular bundles were densely adherent to the prostate posteriolaterally. Both neurovascular bundles were resected widely. All tissue alongside the rectum was excised to assure adequate margins to asssure complete nerve resection. The sural nerve was harvested and used to place interposition grafts between the divided ends of both the left and right cavernous nerves. The total operative time was 4 hours and 20 minutes. The anesthesiologist estimated blood loss at 750 cc. and no transfusions were given. The postoperative recovery in the hospital was uneventful and the patient was discharged on the morning of the second postoperative day.

The prostate was fixed in 10% neutral buffered formalin and processed into whole mount sections in a plane perpendicular to the rectal surface as described previously.16 The final pathology report described a pathological stage T3aN0M0, Gleason grade 3+4=7 adenocarcinoma in the peripheral zone of the prostate measuring 2.5 cm. in diameter. There was established extraprostatic extension near the base on the left, but the surgical margins were not involved. There was extensive periprostatic tissue attached to the prostate posterolaterally on both sides. Microscopically, both neurovascular bundles were seen to be attached to the surgical specimen (fig. 1).

The patient has been followed regularly since the operation. He reported only an area of sensory deficit in the skin over the lateral aspect of the ankle. The patient reported complete urinary control 1 month after the operation and his PSA remains undetectable 1 year later. RigiScan testing (Osbon; Augusta, GA) with the RigiScan Plus software was used for nocturnal penile tumescence monitoring.

With the patient in the supine position for a radical retropubic prostatectomy17 a bilateral pelvic lymphadenectomy is performed. The prostate is mobilized, the dorsal vein complex is divided over the urethra and secured, and the prostate carefully palpated. The course of the neurovascular bundles in relation to the prostate is observed and the degree of adherence of the neurovascular bundles to the prostate assessed. If one or both neurovascular bundles must be widely resected to provide adequate margins around the cancer, a small surgical clip is gently applied for marking purposes across the distal ends of the neurovascular bundle just beyond the point of transection distal to the apex of the prostate. The distal anastomotic sutures are placed in the urethra immediately after it is divided. While prostate removal continues, the microsurgical team is called to harvest approximately 20 cm. of the right sural nerve.

Near the base of the prostate at the lateral vascular pedicles, additional surgical clips are gently placed across the proximal ends of the neurovascular bundle. Once the prostatectomy specimen is removed, hemostasis is obtained, the surgical field is irrigated, and the bladder neck is reconstructed. Before the anastomotic sutures are placed in the bladder neck, a 6.5 cm. length of sural nerve is interposed as a graft between the transected ends of the cavernous nerves on each side (fig. 2).

Gentle handling of the nerve with microsurgical instruments is recommended. A relatively bloodless field is required for precise identification of the transsected ends of the neurovascular bundles, which are not a discrete nerve, but a plexus of fine nerve fibers and vessels. Three interrupted 8-0 nylon sutures are placed in simple interrupted fashion without tension, proximally and distally, facilitated by the surgeon’s use of 4.0 X loupe magnification. Gentle application of the surgical clips minimizes the risk of crush injury to the nerve. However, in cases where crush injury appeared certain, the damaged nerve fibers were trimmed in the area of injury using sharp microsurgical scissors. The bleeding from the vessels of the neurovascular bundles stops after clip application. After clip removal, bleeding from the edges has not been a problem.

Special care must be taken while tying the sutures of the vesicourethral anastomosis to prevent disrupting the nerve grafts. A single Jackson-Pratt suction drain is placed and kept away from the area of the grafts. The additional operative time for the nerve grafting in the first 8 cases averaged 80 minutes. Patients are discharged from the hospital 2-3 days later.

To date we have performed bilateral interposition sural nerve grafts in 9 patients, but only 3 have been followed for 5 months or longer. Herein we report the results in the first 2 patients who have been followed sufficiently to document recovery of erectile function in the early postoperative period (4-12 months). The patient described in detail above observed slowly improving spontaneous erections beginning about 5 months after the operation. He started to experience unstimulated erections resulting in fullness of the penis, but not rigidity, 6-9 times daily. At 7 months post-operatively, sexual stimulation did not increase the quality of erections, and the erections were not sufficient for intercourse. At 11 months post-operatively, the patient described a "30%" spontaneous erection, at times almost sufficient for penetration. During this time, he had been using a vacuum constriction device or MUSE (Vivus; Menlo Park, CA) approximately 1-4 times weekly to obtain an erection satisfactory for intercourse.

Approximately 14 months after surgery, this patient developed a rigid erection sufficient for penetration and intercourse. The patient described this erection as "an erection of substance--hard, not just fluffy." The event lasted several minutes. This patient represents the first documentation of the return of erections sufficient for intercourse after nerve grafting.

Nocturnal penile tumescence testing with the RigiScan was performed 6 months post-operatively. The best night of testing demonstrated two unequivocal erections (fig. 3). The best erection duration was 7.25 minutes, with a tip and base rigidity of 60-79% sustained for 5 minutes. Tip baseline circumference increased from 6.1 to 7.3 cm., while the base circumference increased from 6.5 to 8.1 cm. The base rigidity activity units (RAU) for both erections combined was 10, while the tumescence activity units (TAU) was 4. The combined tip RAU was 8, while the TAU was 3.

The second patient was evaluated 4 months after surgery. Because of a profound clinical depression, he had not attempted any type of sexual activity post-operatively. He did not have spontaneous or stimulated erections and was under the care of a psychiatrist for depression. However, RigiScan testing 4 months post-operatively of this patient demonstrated one prominent partial erection on the second night of use. The base demonstrated 80-100% rigidity for 14 minutes and 60-79% rigidity for 11 minutes. The tip rigidity was 60-79% for 1 minute and 40-59% for 4 minutes. The base RAU was 20, while the TAU was 33. The tip RAU was 9 and the TAU was 20.

To the best of our knowledge this is the first report in humans of documented recovery of complete erectile function after interposition of nerve grafts to replace both resected cavernous nerves. We have performed this procedure in a total of 9 patients. The longest follow-up is 12 months and only 3 patients have been followed for 5 months or longer. Two patients report recovery of partial spontaneous erections at 4-5 months and RigiScan testing documented partial nocturnal erections in a third patient.

The most important finding of this report is the return of spontaneous, unassisted erectile activity satisfactory for penetration and intercourse. His return of function is the first documentation of the ability of nerve grafts to restore erections after deliberate, wide bilateral nerve resections associated with radical retropubic prostatectomy. The functional endpoint of a rigid erection is clearly clinically more important than RigiScan tracings. But with further study, RigiScan findings may have practical benefit in predicting and documenting which patients will have return of erectile function.

Although a "partial" erection can encompass a wide range of function, the RigiScan tracings in figure 3, A obtained 2-6 months after the operation show substantial erectile activity if a minimum cutoff value of 60% or greater erection for 5 minutes is employed.18 In the second patient (fig. 3, B) there is evidence of erectile function, with even more sustained rigidity and tumescence than the first patient. In contrast, from 1985-1997, only one of 70 patients in our database having both nerves deliberately resected in similar fashion states that he has recovered partial erections after radical prostatectomy. This patient does not speak English and has not been available for detailed questioning or RigiScan evaluation, so we view his response with skepticism. The data for our historical control group had been collected prospectively, even before the nerve graft project had started, minimizing any type of potential bias. We are not aware of any reports of the recovery of documented erectile function in patients after intentional bilateral neurovascular bundle resection. 2,4,7,13 Nerve grafts could promote earlier, fuller and more frequent recovery of erectile function in patients with unilateral resection of a neurovascular bundle, whose chances of spontaneous recovery are substantially reduced8,9 since recovery of end organ function is directly related to the volume of functioning nerve fibers.4,7,13,19 However, it would be difficult to document a benefit of nerve grafting in this population without a prospective controlled trial. Bilateral nerve grafts offer an opportunity to demonstrate proof of principal that nerve grafts are technically feasible and can successfully restore erectile function in men who otherwise would have little or no chance of recovery.

Laboratory studies clearly demonstrate the ability of interposition nerve grafts to mediate recovery of erectile function in denervated rats. In Quinlan, Nelson and Walsh’s study, the genitofemoral nerve was used as the replacement for the cavernous nerves after ablation of a 5 mm. segment.14 Potency was assessed post-operatively using mating studies and pelvic nerve electric stimulation. While no improvement was evident at two months, by four months post-operatively, the mating studies were similar between the sham group and the graft group. Similarly, electrical stimulation showed improvement in erectile function at 4 months, but not at 2 months. In a separate investigation using a similar model, 65% of rats had recovery of potency using a genitofemoral nerve graft.20

Ball et al replaced the resected nerves in rats with a silastic tube nerve growth conduit filled with nerve growth enhancing media.15 Using electrical stimulation with resultant tumescence as criteria, these investigators found return of potency in 50-58% of rats, in comparison to 5-11% for the nerve ablated group at 2 and 4 months, respectively.

The principles of nerve grafting are well established.19 The nerve graft provides a conduit that regenerating nerve fibers may use to eventually connect with the transected distal end. With nerve transection the capacity of the proximal stump axons to regenerate is excellent. However, when wide gaps are present between nerve ends, functional recuperation is negligible. Thus, a graft converts disorderly growth into an orderly process by providing a structured framework for regenerating axons to bridge the gap. The graft may further contribute neurotropic agents through its Schwann cells, products of axon-myelin breakdown, or activity of specialized cells of mesodermal origin.21

The ideal donor graft should be autologous to avoid rejection. The deficit as a result of harvesting the graft should be minimal. The diameter should be as small as possible to enhance nerve revascularization, which is accomplished with vessels entering not only the proximal and distal junctions, but also from the tissue of the graft bed. The donor graft should contain many large nerve fibers and have few branching points. The sural nerve meets these criteria and its harvesting results in only a minimal sensory deficit on the lateral ankle.21

The sural nerve is considered standard for grafting purposes. It has been used clinically and studied extensively with brachial plexus, facial nerve, and peripheral nerve injuries.21 Although a myelinated nerve, the sural nerve functionally becomes demyelinated during grafting.21 Nerve regeneration then relies on the Schwann cell basement membrane and its basal lamina.

Important technical details for nerve grafts include: 1) the creation of a tension-free repair, assisted by a graft 10-20% longer than the defect to compensate for shrinkage, 2) the use of a graft slightly larger than the transected nerve ends to capture and channel regenerating axons, 3) use of meticulous tissue handling and loupe magnification to avoid damage to the graft, 4) use of nylon microsuture, sparingly.21 These principles were used during this case by a reconstructive microsurgeon (RKN) with extensive fellowship training in nerve grafting procedures. We have not routinely resected the portion of the nerve to which the surgical clip had been applied. Although a crush injury is likely, the degree of injury is felt to be minor and of a temporary nature. ²¹ However, gentle application of the clip is highly recommended.

The rate of nerve regeneration is about one millimeter per day in adult humans and may vary depending upon age and hormonal factors.22 Time to functional recovery is proportional to the distance the nerve is required to regenerate. A 6 cm. graft will generally need 2-3 months to traverse initially, although earlier or later crossing times are sometimes seen. Full reinnervation of the end organ then takes place over a further 6 to 15 months.

Nerve grafting is only one component in the preservation and recovery of erectile function. Although the pathophysiology of erectile dysfunction after radical prostatectomy is traditionally considered neurogenic in etiology, arterial and corporal smooth muscle alterations have also been implicated. 23-26 These causes may contribute to the failure of some patients to recover erectile function even after meticulous bilateral nerve-sparing procedures and may cause nerve grafts to be less than completely successful in restoring erectile function in some patients.

To confirm the findings of this report, we are continuing to monitor the erectile function with interviews, validated questionnaires, and nocturnal penile tumescence testing in patients with bilateral nerve resection and grafts. In addition, we are assessing in similar fashion patients who had bilateral resection without grafts. The implication from our early experience is that nerve grafts may help to restore spontaneous erectile function in some men being treated surgically for prostate cancer. Although full spontaneous erections may not always be restored with bilateral nerve grafts, an enhancer of erectile quality such as sildenafil citrate may produce sufficient rigidity for patient satisfaction. In theory, sildenafil should not work in patients with bilateral complete nerve resections because of the absence of erections, although clinical experience with sildenafil in this setting remains limited.

We describe for the first time in humans a technique for placement of bilateral interposition nerve grafts at the time of radical prostatectomy with wide resection of both neurovascular bundles. The patients recovered partial spontaneous erections within 4-5 months, and continued to improve. Objective testing at 4-12 months showed spontaneous night time erections exhibiting criteria for normalcy. At 14-15 months post-operatively, the first patient developed rigid erections sufficient for intercourse.

1. Eastham, J. A., Goad, J. R., Rogers, E., Ohori, M., Kattan, M. W., Boone, T. B., and Scardino, P. T.: Risk factors for urinary incontinence after radical prostatectomy. J. Urol., 156: 1707, 1996.

2. Walsh, P. C.: Radical retropubic prostatectomy. In: Campbell’s Urology, 6 ^th ed. Edited by P. C. Walsh, A. B. Retik, T. A. Stamey, and E. D. Vaughan, Jr. Philadelphia: W. B. Saunders Co., vol. 3, chapt. 78, pp 2865-2886, 1992.

3. Davidson, P. J., van den Ouden, D., and Schroeder, F. H.: Radical prostatectomy: prospective assessment of mortality and morbidity. Eur. Urol., 29: 168, 1996.

4. Geary, E. S., Dendinger, T. E., Freiha, F. S., and Stamey, T. A.: Nerve sparing radical prostatectomy: a different view. J. Urol., 154: 145, 1995.

5. Murphy, G. P., Mettlin, C., Menck, H., Winchester, D. P., and Davidson, A. M.: National patterns of prostate cancer treatment by radical prostatectomy: results of a survey by the American College of Surgeons Commission on Cancer. J. Urol., 152: 1817, 1994.

6. Schover, L. R.: Sexual rehabilitation after treatment for prostate cancer. Cancer, suppl 4, 71: 1024, 1993.

7. Quinlan, D. M., Epstein, J. I., Carter, B. S., and Walsh, P. C.: Sexual function following radical prostatectomy: influence of preservation of neurovascular bundles. J. Urol., 145: 998, 1991.

8. Bigg, S. W., Kavoussi, L. R., and Catalona, W. J.: Role of nerve-sparing radical prostatectomy for clinical stage B2 prostate cancer. J. Urol., 144: 1420, 1990.

9. Walsh, P. C., Epstein, J. I., and Lowe, F. C.: Potency following radical prostatectomy with wide unilateral excision of the neurovascular bundle. J. Urol., 138: 823, 1987.

10. Walsh, P. C.: Technique of vesicourethral anastomosis may influence recovery of sexual function following radical prostatectomy. Atlas Urol. Clin. North Am., 2: 59, 1994.

11. Palapattu, G. S., Stapleton, A. M. F., Seale-Hawkins, C. K., Kattan, M. W., and Scardino, P. T.: A change in the technique of radical retropubic prostatectomy markedly improves postoperative potency. J. Urol., 155: 647A, abstract 1346, 1996.

12. Lepor, H., Gregerman, M., Crosby, R., Mostofi, F. K., and Walsh, P. C.: Precise localization of the autonomic nerves from the pelvic plexus to the corpora cavernosa: a detailed anatomical study of the adult male pelvis. J. Urol., 133: 207, 1985.

13. Catalona, W. J. and Bigg, S. W.: Nerve-sparing radical prostatectomy: evaluation of results after 250 patients. J. Urol., 143: 538, 1990.

14. Quinlan, D. M., Nelson, R. J., Walsh, P.C.: Cavernous nerve grafts restore erectile function in denervated rats. J. Urol., 145: 380, 1991.

15. Ball, R. A., Lipton, S. A., Dreyer, E. B., Richie, J. P., and Vickers, M. A.: Entubulization repair of severed cavernous nerves in the rat resulting in return of erectile function. J. Urol., 148: 211, 1992.

16. Wheeler TM. Anatomical considerations in carcinoma of the prostate. Urol. Clin. North Am., 16: 623, 1989.

17. Goad, J. R., and Scardino, P. T.: Modifications in the technique of radical retropubic prostatectomy to minimize blood loss. Atlas Urol. Clin. North Am., 2: 65, 1994.

18. Benet, A. E., Rehman, J., Holcomb, R. G., and Melman, A.: The correlation between the new RigiScan Plus software and the final diagnosis in the evaluation of erectile dysfunction. J. Urol., 156: 1947, 1996.

19. Sunderland, S.: Nerve grafting and related methods of nerve repair. In: Nerve Injuries and Their Repair: A Critical Appraisal. Edited by S. Sunderland. Edinburgh: Churchill Livingstone, pp 467-497, 1991.

20. Ball, R. A., Richie, J., and Vickers, M. A.: Correction of surgically induced erectile dysfunction in the rat by microsurgical interposition nerve graft repair of the cavernosal nerves. Am. Coll. Surg. Surg. Forum, 16: 678, 1990.

21. Mackinnon, S. E., and Dellon, A. L.: Nerve injury and regeneration. In: Surgery of the Peripheral Nerve. S. E. Mackinnon and A. L. Dellon. New York: Thieme Medical Publishers, 1988.

22. Millesi, H.: Healing of nerves. Clin. Plast. Surg., 4: 459, 1977.

23. Polascik, T. J., Walsh, P. C.: Radical retropubic prostatectomy: the influence of accessory pudendal arteries on the recovery of sexual function. J. Urol., 154: 150, 1995.

24. Nehra, A., Hakim, L. S., Mulhall, J., Daller, M., Zeitman, A., Moreland, R., and Goldstein, I.: Radiation therapy for prostate cancer: penile hemodynamic and structural alterations. J. Urol., 157: 358, abstract 1399, 1997.

25. Aboseif, S., Shinohara, K., Breza, J., Benard, F., and Narayan, P.: Role of penile vascular injury in erectile dysfunction after radical prostatectomy. Br. J. Urol., 73: 75, 1994.

26. Klein, L. T., Miller, M. I., Buttyan, R., Raffo, A. J., Burchard, M., Devris, G., Cao, Y. C., Olsson, C., Shabsigh, R.: Apoptosis in the rat penis after penile denervation. J. Urol., 158: 626, 1997.

LEGENDS

Figure 1. Whole mount section of prostate near the base of the gland, from the patient described in the case report. Orientation markers: A – anterior; P – posterior; R – right; L – left; U – urethra. Open spaces within prostate represent areas where tissue was removed for tissue bank. Open and solid arrows mark boundary of prostatic capsule and neurovascular bundle, respectively. Solid line marks area of prostate cancer. Some of the periprostatic tissue on the left lateral prostate is missing in this section due to processing artifact. (Hematoxylin and eosin stain, X2).

Figure 2. A. Intraoperative photograph taken after sural nerve grafts were interposed between the severed ends of both cavernous nerves. Pr=proximal, Di=distal, closed arrow=left sural nerve graft, open arrow=right sural nerve graft.

B. Illustration of left interposition nerve graft after radical retropubic prostatectomy. The nerve has been reversed and the distal branches coapted to the proximal ends of the cavernous nerves at the lateral vascular pedicle. Note the surgical clip across the distal end of the severed right neurovascular bundle.

Figure 3. Best night of RigiScan tracing 6 months after nerve grafting (see text for details). This study demonstrates 2 distinct erections (arrows). Tick marks on the horizontal axis represent 1 hour time intervals.

Figure 1.