Penile Rehabilitation after Radical Prostatectomy: Where Do We Stand and Where Are We Going?

Run Wang, MD, FACS*
*Division of Urology, University of Texas medical School at Houston, Houston, TX, USA ; + MD Anderson Cancer Center, Houston, TX, USA

International Society for Sexual Medicine, April 2007

Post-prostatectomy erectile dysfunction (ED) remains a serious quality -of-life issue. Recent advances in the understanding of the mechanism of post prostatectomy ED have stimulated great attention toward penile rehabilitation. Vacuum Therapy is a 100% drug free option for Rehab after prostate surgery and from short or long term ED.

ABSTRACT
Introduction. Post-prostatectomy erectile dysfunction (ED) remains a serious quality -of-life issue. Recent advances in the understanding of the mechanism of post prostatectomy ED have stimulated great attention toward penile rehabilitation. Aim. This review presents and analyzes a contemporary series of the recent medical literature pertaining to penile rehabilitation therapy after radical prostatectomy (RP). Main Outcome Measures. The laboratory and clinical studies to penile rehabilitation are analyzed. The validity of the methodology and the conclusion of the findings from each study are determined. Methods. The published and presented reports dealing with penile rehabilitation following RP in human and cavernous nerve injury in animal models are reviewed. Results. Exciting scientific discoveries have improved our understanding of post-prostatectomy ED at the molecular level. The rationale for post-prostatectomy penile rehabilitation appears to be logical according to animal studies. However, clinical studies have not consistently replicated the beneficial effects found in the laboratory studies. Currently available clinical studies are flawed die to short- term follow-up, small number of patients in the studies, studies with retrospective nature, or prospective studies without. Rehabilitation programs are also facing a challenge with the compliance, which is critical for success for any rehabilitation program. At the present time, we do not have concrete evidence to recommend what, when, how long and how often a particular penile rehabilitative therapy can be used effectively. Conclusions. Large prospective, multicentered, placebo-controlled trials with adequate follow-up are necessary to determine the cost-effective and therapeutic benefits of particular penile rehabilitative therapy of therapies in patients following the treatment of clinically localized prostate cancer. Until such evidence is available, it is difficult to recommend any particular penile rehabilitation program as a standard of practice.

Introduction
Radical prostatectomy (RP) is a major option for treatment of clinically localized prostate cancers with excellent long-term results. Han et al.(1) reported that the 15-year overall actuarial cancer-specific survival rate was 90% in Gleason 6 or Gleason 7 (3 + 4) prostate cancer treated with RP. Despite being an excellent therapy for prostate cancer, RP is associated with several quality –of-life issues, mainly urinary incontinence and erectile dysfunction (ED) [2,3]. Several studies have demonstrated acceptable continence rates following surgery [4-6]. A recent review reported that at 1 year 93.4% of men who underwent RP use 0-1 pads for urinary incontinence [7]. However, the same cannot be said for erectile function. Currently published clinical studies produce poorly interpretable and inconsistent determinations of erection outcome following the treatment of clinically localized prostate cancer. The ED and erectile function recovery rates are widely disparate. According to a recent analysis conducted be the American Urological Association panel of experts after review of the publications from 1991 through 2004, ED and erectile function were 10-100% and 6-86% for RP [8)]. The experts emphasize that studies must apply scientifically rigorous methodology and standard outcome measures regarding erectile function following treatment of the prostate cancer. Nevertheless, many men will have decreases in penile rigidity and penile length after RP [3, 9]. Recent advances in the understanding of post-prostatectomy ED have led to promising, yet controversial new strategies, namely penile rehabilitation in the management of this serious issue. This article presents and analyzes a contemporary series of the recent medical literature pertaining to penile rehabilitation therapy after RP.

Pathophysiology of Post-prostatectomy ED
Erectile function becomes impaired immediately following RP and it’s thought to be secondary to the damage done to the cavernous nerves, which us known as neuropraxia. Neuropraxia can be caused by mechanically induced nerve stretching that may occur during prostate retraction, thermal damage to nerve tissue caused be electrocautery, ischemia of the nerves secondary to disruption if blood supply while attempting to control surgical bleeding, and local inflammatory effects associated with surgical trauma [10]. Even in the most meticulous nerve-sparing dissection, some degree of neuropraxia is unavoidable because of the close proximity of the nerves to the prostate gland. These nerves tend to recover slowly; it may take as long as 3 years for them to reach a new baseline functional status [6]. Absence or decreased erection and penile size ensue before recovery of the cavernous nerve [11, 12].

Lack if erections will then lead to poor oxygenation of the corporal bodies, eventually progressing to cavernosal fibrosis and ultimately causing a venous leak seen clinically as venogenic ED [13]. In an experimental model, significant over expression of hypoxia-related substances, like transforming growth factor-B ( TGF-B) and collagen I and III, was found in rats that had undergone bilateral incision of the cavernosal nerves compared with controls [14]. When human penile smooth muscle cell is exposed to prolonged hypoxic environment, TGF-B dependent enfothelin-1 (ET-1) synthesis is increased. ET-1 is a potent constrictor of penile smooth muscle and a profibrotic peptide [15]. Studies also showed that low oxygen tension in human cavernosal tissue inhibits production of prostaglandin-E (PGE). PGE inhibits collagen formation by inhibiting TGF-B that induces collagen synthesis. With the inhibition of PGE, TGF-B is allowed to induce connective tissue synthesis [16]. The trabecular smooth muscle is then replaced with collagen, which leads to the loss of veno-occlusive mechanism [13, 17, 18].

A reduction in arterial inflow has also been reported by several authors. This associated with the ligation of accessory internal pudendal arteries during prostatectomy [19, 20]. Combination of nerve damage with decreased arterial inflow may intensify hypoxia and ultimately lead to apoptosis or programmed cell death, which has recently been linked to the pathophysiology of post prostatectomy ED. Lim et al. [21] demonstrated that the rat with cavernous nerve crush and bilateral internal iliac artery ligation had significant decrease of intracorporeal pressure, loss of cavernous smooth muscle, and neural staining. User et al. [22] performed bilateral neurotomy of rat penis and found that there was significant apoptosis in the subalbugineal smooth muscle cells. With apoptosis in the region of subtunical venular plexus, a defeat in the veno-occlusive mechanism of the corpus cavernous occurs. McVary et al. [23] recently confirmed that both intrinsic and extrinsic apoptotic pathways were activated in rats whose cavernous nerves were disrupted.

Basic Science and Rationale of Postprostatectomy Penile Rehabilitation
With the understanding of mechanism of postprostatectomy ED, scientific studies are focused on ways to increase oxygenation of the cavernosal bodies, decrease tissue fibrosis and apoptosis and essentially improve erectile function. Schwartz et al. [24] assessed the effect of sildenafil on the intracoporeal smooth muscle content of patients after RP. Group 1 received 50 mg of sildenafil and Group 2 received 100 mg of sildenafil every other night for 6 months following catheter removal. A percutaneous biopsy of the intracorporeal smooth muscle was performed prior to incision for RP and under local anesthesia 6 months later. They found a statistically significant increase in smooth muscle content. The limitations of the study were not double-blind and there was no placebo-controlled group.

Chronic use of tadalafil (2mg/kg, daily for 3 months) was studied in rats after bilateral cavernosal neurotomy by Vignozzi et al. [25]. Daily use of tadalafil significantly decreased neurectomy-induced hypoxyprobe labeling and increased the muscle/fiber ration in rat penile tissues. Interestingly, overexpression of the endothelin type B (ETB) receptor was seen in hypoxic tissues, and chronic use of tadalafil reverted ETB overexpression. The physiological role of the ETB in covernosal tissues is not clear, but it may be related to smooth muscle relaxation. Chronic use of tadalafil did not rescue the meurectomy-induced hypoexpression of neuronal nitric oxide synthase (nNOS) and endothelial nitric oxide synthase (eNOS) in this study. Therefore the mechanism of tadalafil preventing bilateral neurectomy-induced hypotic damage to the penile tissues is not clear.

Udenafil is a new phosphodiesterase type 5 inhibitor (PDE5-L) that is currently available only in South Korea. A study from South Korea showed that rats treated with unedafil 10 mg/kg twice daily for 2 months following bilateral neurectomy of cavernous nerves had less expression of TGF-B, hypoxia-inducible factor Ia (HIF Ia) and collagen I at the gene and protein levels in corporal tissues, as compared with the rats without undenafil treatment. Udenafil also significantly improved the relaxation response of corpus cavernosum smooth muscle to sodium nitroprusside in this rat model [26].

Ferrini et al. [27] and Kavebacz et al. [28] recently studied vardenafil and sildenafil in prevention of the coprporal fibrosis and loss of smooth muscle in animals after cavernosal nerve resection. Rats were treated with 20 mg/L vardenafil in the drinking water (accurate dosage was not reported by authors) or 20 mg/kg/day sildenafil after cavernous nevbe resection for 45 days. Corporal veno-occlusive dysfunction was demonstrated via dynamic infusion cavernosometry in rats after nerve resection. Compared with the nerve resection only group, rats treated with cardenafil and sildenafil revealed increased smooth muscle content and proliferating cell nuclear antiden expression (smooth replication), normalized dynamic infusion cabernosometry and the smooth muscle/collagen ratio. Vardenafil did not change the apoptotic index induced by cavernosal nerve resection. They believe that long-term treatment with vardenafil or sildenafil prevents corporal veno-ocvclusive dysfunction by preserving smooth muscle content and inhibiting corporal fibrosis. The same group of the researchers also studied tadalafil in prevention of the corporal fibrosis and loss of smooth muscle after bilateral cavernosal nerve resection in the rat [29]. Rats were treated with 5 mg/kg/day of tadalafil (about 0.8 mg/kg/day in men) for 45 days after unilateral or bilateral cavernosal nerve resection. Compared with the control group, rats treated with tadalafil had normalized erectile response (measured by intracorporeal pressure) to intracavernosal injection (ICI) of papaverine. Tadalafil also reduced collagen content by 17.6% as determined by hydroxproline assay.

Mulhall et al. (30) went a step further to look at the timing of the commencement of treatment and the optimal dosing strategy. They used sildenafil prior to injury or a delay in treatment after cavernous nerve crush in the rat to evaluate the impact upon erectile hemodynamics and to further ascertain if high dose sidenafil was of benefit. They found that the rats which received daily sildenaful had higher intracavernosal pressure/mean arterial pressure (ICP/MAP) ratios than the controls (no sildenafil) was given) after the cabernous nerbe injury. Among the sildenafil-daily-treatment group, the ICP/MAP) rations in animals that started sildenafil at a dose of 20 mg/kg for 3 days prior to cavernous nerve crush injury were higher than that in animals started on sildenafil 1 hour prior to or 3 days after cavernous nerve crush. However, the animals receiving 40 mg/kg sildenafil (high dosage) had the similar ICP/MAP ratios regardless of treatment timing. They also evaluated the markers for endothelia (eNOS; CD31) and neural (nerve growth factor, NGF;nNOS) expression in the penile tissues with the use of immunohistochemistry stains. They found that pretreatment with sildendafil before cavernous nerve crush treatment groups. Use of sildenafil before cavernous nerve crush decreased the apoptotic indices when compared with controls. They concluded that using standard sildenafil dosing regimens, pretreatment resulted in greater erectile function preservation. When sildenafil dosing is delayed using higher dosing could reach the similar erectile preservation effect.

In the same form of animal model, Mulhall et al. [31] studied the effect of hyperbaric oxygen therapy on erectile function recovery. The hyperbaric oxygen therapy was given daily for 10 days, starting in the day of cavernous nerve crush. They found that rats treated with hyperbaric oxygen had higher ICP/MAP ratios, penile NGF and eNOS compared with the rats without hyperbaric oxygen therapy mediates preservation f neurotrophic and endothelial factor expression. These findings support the concept of cavernosal oxygenation after RP to improve erectile function.

All above-mentioned studies were based on cavernosal nerve resection or crush (without removal of the rat prostate) to mimic RP in human. It is not clear at the present time whether the results will be different if the prostate and seminal vesicles are removed in these rats. These studies also raise the speculation of PDE5-Is being active in the erectile tissue after bilateral cavernous nerve resection. PDE5-Is act by inhibiting the metabolism of cyclic guanine monophosphate (cGMP). Production of cGMP in the erectile tissue requires the functional cavernous nerves to produce neuronal nitric oxide (NO). Neuronal NO is critical for initiation of penile erection with the endothelia NO propagating and maintaining penile erection [32]. Without neuronal NO, what is rationale that PDE5-Is work in these animal models? Some authors suggested that PDE5-Is may upregulate the antifibrotic effects of NO produced by inducible nitric oxide synthase (iNOS) that was induced by the neural injury [29]. Others proposed that cavernous neurotomy will only reduce, but not completely abolish, NO signaling. Chronic use of PDE-Is may amplify the depressed NO signaling pathway which inhibits hypoxic fibrosis [25].

With finding that PGE inhibits TGF5-B, Moreland et al. [33] sought to asses the role of cyclic adenosine monophosphate ( cAMP) in the regulation of connective tissue biosynthesis in human corporal smooth muscle cells in culture. They found that cAMP synthesis in human TGF5-B induced collagen synthesis. These data suggest that agents that cause an increased level of cAMP, could lead to a decrease in collagen deposition [13]. This may be the scientific basis that implies that programmed intracavernosal or intraurethral use of PGE is beneficial to cavernous tissues after RP. The key data from representing animal studies are summarized in Table 1.

Penile Rehabilitation Programs and Clinical Evidence
Despite the clear understanding of the mechanism of postprostatectomy ED and well-established rationale for postprostatectomy penile rehabilitation, there is no consensus regarding effective rehabilitation programs. The clinical evidence to support a particular program is also lacking. Telohen et al.[34] studied the attitudes and practice patterns of clinicians who were members of regional and international sexual medicine societies regarding penile rehabilitation. Among 301 physicians from 41 countries 83.7% performed some form of rehabilitation. The rehabilitation strategies included: PDE-Is (95.4%), ICI (75.2%), vacuum erectile device (VED,30.2%) and intrautrethral prostaglandin (9.9%). Most physicians (48.1%) started rehabilitation just after urethral catheter removal, and 36.8% commenced the rehabilitation program within the first 4 months after the RP. The reasons for avoiding rehabilitation included cost (50%), not evidence-based (25%), or not familiar with the rehabilitation programs. Obviously, post-RP penile rehabilitation is widely practiced and commenced relatively early with predominant use of PDE-Is. Penile rehabilitation programs have also extended into some community settings. Caspersin et al. [35] placed patients on sildenafil pr vardenafil daily or tadalafil two times a week 3 weeks after the robotic prostatectomy. Patients recalcitrant to oral therapy were studied with color flow Doppler studies to guide the alternative rehabilitation program, such as VED, intraurethral suppository, and ICI. Their established algorithm for penile rehabilitation is based on no clinical evidence. Use of penile color flow Doppler study to guide the alternative rehabilitation program is also not evidence-based and may be cost-ineffective. Nevertheless current known penile rehabilitation programs include PDE-Is, ICI, intraurethral prostaglandin, VED and combination therapy.

Phosphodiesterase Type 5 Inhibitors
With the advent of PDE5-Is, Montosrsi et al. [36] explored the effectibeness of sildenafil on increasing nocturnal erections in med with ED of various etiologies. Patients were randomized to receive either 100-mg sildenafil or placebo at bedtime. They found that patients treated with sildenafil had a significantly better overall quality of nocturnal erections as recorded by the Rigiscan device when compared with the control group. The next question was whether increased postprostatectomy nocturnal erections secondary to sildenaful could lead to a better recovery of spontaneous erectile function. In a well-known study sponsored by Pfizer Inc., 76 men with normal erectile function preoperatively who had undergone bilateral nerve-sparing RP were randomly assigned to receive sildenafil or placebo nightly. Treatment began 4 weeks after surgery and consisted of sildenafil at 50 or 100 mg or placebo every night for 36 weeks, followed by 2 months without therapy. At the end of the trial, 27% of men in the sildenafil group vs. 4% in the placebo group reported return of spontaneous normal erections. Additionally, there was a progressive increase in the duration and amplitude of nocturnal erections in the sildenafil-treated group [37]. Although the results of this study were widely criticized because of the poor overall quality of erectile function, it is the first study to correlate the return of nocturnal erections with the return of spontaneous sex-stimulated erections that are sufficient for intercourse. The question is whether PDE5-Is are active in the erectile tissues in the early phase following RP. PDE5-Is act through inhibition of metabolism of cGMP, which is the second messenger facilitating erection. It is our current understanding that production of cGMP in the erectile tissue required functional cavernous nerves to produce nitric oxide. It is unlikely that PDE-Is are effective until the cavernous nerve is at least partially recovered from neuropraxia following RP [38].

In contrast, Montorsu et al. [39] recently showed that there is no significant difference in erectile function between on-demand PDE5-Is and PDE5-I as rehabilitative treatment in patients after bilateral nerve-sparing RP. The International Index of Erectile Function erectile function domain (IIEF-EF) at the mean 12-month follow-up was 19.5 (plus or minus) 9.4 for patients using PDE5-I on-demand compared with 18.3 (plus or minus) 4.0 in patients using PDE5-I as rehabilitation. Criticism of this study is that compliance to PDE5-I as rehabilitative treatment was not reported. Therefore, the benefits of rehabilitative regimen cannot be evaluated.

Intracavernosal Injection
Programmed ICI was the first modality used by Montorsi et al. [40] as a penile rehabilitation strategy. Thirty patients who underwent bilateral nerve-sparing RP for clinically localized prostate cancer were randomized to either group1, which consisted of ICI of alprostadil three times a week for 12 weeks, or group 2 which consisted of no erectogenic therapy. At 6 months, 67% of the patients in group 1 reported return of spontaneous erections satisfactory for sexual intercourse vs. 20% of the patients in group 2. Two cases (17%) of the failures in group 1 were secondary to cavernous veno-occlusive dysfunction vs. 8 cases (53%) in group 2 based upon Doppler sonography at 6 months. Unfortunately, long term follow-up results with scheduled ICI were lacking in this study.

The only long-term follow-up study with ICI as a rehabilitation strategy was reported by Mulhall et al. [41]. Men with functional preoperative erections who underwent RP were given oral sildenafil early postoperatively. Non-responders were switched to ICI three times a week. Only patients who presented within 6 months post RP, who completed the IIEF on at least three separate occasions after surgery, and who had been followed for at east 18 months were included. Data from men who were committed to rehabilitation (group R) were compared with those of men who did not follow the protocol but continued to be followed serially (group NR) following RP. At 18 months post RP, group R had a significantly higher percentage of patients who were capable of having medication-unassisted intercourse, better mean erectile rigidity, higher mean IIEF-EF domain scores, higher percentage of patients responding to sildenafil, less time to become a sildenafil responder, and higher percentage of patients responding to ICI, when compared with group NR. This study indicated that programmed ICI resulted in higher rates of spontaneous functional erections and erectogenic drug response after RP. The authors did not report the percentages of early sildenafil responders and nonresponders before they entered tie ICI. The study was not randomized and patients were excluded if the did not comply with the follow-up schedule, so selection bias was a limitation of the study. The nature of frequent needle injection to the penis will cause concern regarding patient compliance. A prospective study demonstrated that long-term compliance to penile injection rehabilitation after RP is problematic. Only 52.3%, 25.9%, and 35.3% of patients were performing penile injection for rehabilitation as prescribed at 4,8, and 12-month follow-ups [42].

Intraurethral Prostaglandin
Intraurethral prostaglandin is delivered as suppository of alsrostadil with Medicated Urethral System for Erection ( MUSE©; Vivus Inc., Mountain View, CA USA). At a prospective trial, 91 patients were enrolled, with 56 in the treatment group and 35 in the control group. Patients in the treatment group received 125 mg of aprostadil three times a week. The dose was titrated up to 250 mg three times a week for 4 months. If the patient could not tolerate the higher dose, he was lowered to the 125 mg dose three times a week. At 6 months, 53% of patients in the treatment group vs. 11% of patients in the control group were able to have a natural erection sufficient for vaginal penetration [43]. Long term follow-up results with scheduled intraurethral use of alprostadil were lacking in this study.

McCullough [44] explored the possible mechanism of penile rehabilitation with intraurethral alprostadil. Twenty-two men enrolled in a randomized comparative penile rehabilitation trial of nightly alprostadil vs sildenafil, underwent preoperative corporal oximetery as well as corporal oximetry before and after their first doses of alprosdadil 125 or 250 mg. The first does of alprosdadil 125 mg was given the day the catheter was removed. One month later after using alprisdadil 125 mg nightly, the alptosdadil significantly improved corporal and glandular oxygen saturation despite lack of penile rigidity. This study provided a scientific rationale for the use of low-dose intraurethral alprostadil in penile rehabilitation. The author did not report any data regarding the patients receiving sildenafil in this report. Currently, a multicenter study with the use of intraurethral alprostadiul as a rehabilitation strategy after bilateral nerve-sparing RP is in progress in the United States. We are eagerly awaiting the results.

Vacuum Erectile Device
The VED is recently gaining popularity as a rehabilitative tool. In an early prospective study, patients were started on daily use of VED, biweekly injection of combined PGE, papaverine, and phentolamine (Trimix), and patient-directed use of PDE5-Is following unilateral nerve-sparing RP with or without sural grafting. Patients were evaluated preoperatively, 6 weeks postoperatively, and at 4 month intervals thereafter for up to 2 years for rehabilitation regimen compliance and penile length. At 4 months, only the patients with good VED compliance had a 0.4-cm increase n penile length vs 0.3-cm decrease in length in patients with poor VED compliance relative to their 6 week measurement. No beneficial effect was found for patients using ICI regarding the recovery of penile length [45]. The follow-up study in this cohort demonstrated impressive recovery of erectile function (over 71%) that was attributed to the penile rehabilitation [46]. This finding clearly needs to be studied in a larger trial. The compliance with the rehabilitation regimen was also studied in these patients. In total, 73.3%, 66.7%, and 47.1% of patients were using VED for rehabilitation as prescribed at 4-, 8-, and 12-month follow-ups, which was significantly higher than the compliance to ICI as mentioned previously. Interestingly, the compliance was better in patients older than 57 years than that in those who were younger. This factor highlights the need to counsel all patients about treatment compliance, but especially younger men who may be more likely to neglect therapy [42].

Monga et al. [47] contacted a multicentered randomized study to compare early (1 month post-nerve sparing RP) with traditional (6months after surgery) use of VED. They only have the 6 month follow-up data available at the present time. The preliminary results showed that early use of VED for rehabilitation (10 minutes a day without the construction ring) significantly improves the IIEF-EF scores, (12-+ 8.8 vs. 2.8+-2.0, P<0.005) and preserves penile length (mean change -0.4vs.-2.0 cm) compared with control group. Long term results of this study are not available at this time. Thee authors believe that use of VED is more cost-effective compared with frequent use of expensive PDE-Is or frequent penile injection. This method required active patient participation in the rehabilitation process compared with taking a pill.

Most criticism for VED as a rehabilitation method is the unknown mechanism in improving spontaneous erection. Bosshardt et al. [48] used blood gas analysis to evaluate the origins of blood for erection with the use of VED. Blood gas analyses were obtained from the corpora cavernosa with VED-induced erection immediately after application of constriction ring. The measurements were repeated 15 and 30 minutes later with the constriction ring in place. The blood gas results of the corpora cavernosa were compared with the arterial blood from arterial radialis and venous blood from vena cubiti. The result showed that mean O2 saturation of corporeal blood immediately after VED induced erection was 79.2%, compared with 94.5% from arteries and 54.7% from veins. They calculated that 58% of blood with VED-induced erection was arterial and 42% of blood was venous in origin. The O2 saturation decreased significantly after 30 minutes with the ring in place. This finding established the rationale that we do not recommend using the ring when VED is used for penile rehabilitation purpose. Arterial blood may not only provide oxygen to the corporal tissues, it may also carry other nutrients such as certain growth factors to the tissues (Tom Lue, personal communication).

Another advantage of the VED is to ensure multiple erections on a daily basis. The average man obtains three to six erections per night during the rapid-eye-movement sleep. We do not know how many erections at a given time are required to maintain the health of erectile tissue. Therefore, the number of erections required per week for penile rehabilitation is not known. Use of VED as a rehabilitation modality can maximize the number of erections that other current rehabilitation methods cannot reach.

Combination of ICI and PDE5-Is
Combination rehanilitation prodteams aim to stimulate early and maximum recovery of erectile function. Montorsi et al.{49] randomized patients to receive three ICI of alprostadil per week for 3 months, followed ny oral sildenafil as needed for 3 months vs. sildenafil as needed starting 3 months after nerve-sparing RP as monotherapy. At 6 months, it was found thayt the patients in the combination arm had an 82% response rate to sildenafil vs. only 52% in the sildenafil only arm. Controversially, in this study fewer patients in the combination arm had spontaneous penile tumescence compared with the patients in the sildenafil only group at 3-a nd6-month follow-up. Yet, the patients in the combination arm had higher IIEF-EF and the Erectile Dysfunction Inventory of Treatment Satisfaction scores than those in the sildenafil only group. Nandipati et al. [50] conducted a prospective study with 22 patients who underwent bilateral nerve-sparing RP. The study set out to evaluate how the combination of ICI and PDE5-I would affect spontaneous return of erectile function. The patients were started on a sildenafil dose of 50mg on the day of discharge from the hospital. Eighteen patients were then started on 4 mg of PGE1 2-3 times a week, and the remaining four patients were started on low-dose Trimis (20U) 2-3 times a week. The patients were allowed to have doses titrated secondary to efficacy and side effects. At mean follow-up of 6 months, 11/22 (50%) patients had return of spontaneous partial erections. In this study, it was noted that there were no patients who dropped out. This possibly could be due to the allowance of dose titration. Unfortunately, there was not a mono-therapy group as a control in this study.

Combination of PDE5Is and VED
Yassin et al. [51] reported their retrospective study with combination of PDE5-Is and VED for early penile rehabilitation following nerve="sparing" RP. Patients started 25-mg sildenafil 3 times a week or 5-mg tadalafil twice weekly with VED use at least twice a day, 11 days after the RP, for 3 months. They found that 56% of patients on PDE5-I and VED rehabilitation program obtained erection sufficient for sexual intercourse. Patients on sildenafil rehabilitation reported higher successes than the patients on tadalail (78% vs. 64%). This study was not well designed and the follow-up was very short. There was no control group to compare the success rate of on-demand PDE5-Is vs. scheduled PDE5-Is.

New Horizons
To overcome the detrimental effect of neuropraxia following RP, investigators are presently looking at different nerve regeneration regimens for applicability in stimulating early return of erectile function. Neuroimmunophilin ligands such as FK506 have neuroportective and neuroregenerative activities in vitro and in vivo by protecting neurons from chemotoxin-induced cell death and stimulating axonal outgrowth [52,53]. Sezen et al. [54] used cavernous nerve crush model in rats to mimic the nerve damage resulting from bilateral nerve-sparing RP. They demonstrated that FK506 given at the time as the crush and on successive days prevented the loss of the erectile function compared with the saline-treated rats. Mulhall et al. [55] recently found that the ICP/MAP ration was significantly higher in the 3.2 mg/kg FK506 treatment group than the control and lower dose groups after cavernous nerve crush injury. FK506 treatment also preserves the smooth muscle content as evaluated with Masson’s trichome staining. Lagoda et al. [56] studied the mechanism of FK506 and sildenafil on the erectile function recovery after cavernous nerve injury in rats. After unilateral cavernous nerve crush, rats were treated with sildenafil ( 20 mg/kg) once daily for 5 days. At day 14, erectile function was measured by electrical stimulation of the cavernous nerve. The penis was processed for Western blot analysis of glutathione perixidase (GPX; antioxidant), nitro-tyrosine (NT;marker of tissue damage), and total/phosphor-Akt (antiapoptotic factor). They found that sildenafil and FK506 significant;ly improved the erectile function as measured by the maximum ICP and normalized area under the curve compared with the saline-treated group. No significant difference were found for NT levels and total/phosphor-Akt levels. This study suggested that FK506 plays a protective role of regulating oxidative damage and sidenafil may work through other mechanisms yet to be discovered. Disappointedly, the multicentered, placebo-controlled clinmical trial of GPI 1485 ( a novel neuroimmunophilin ligand) in patients who underwent RP did not show obvious benefits for early and better return of erectile function compared with the placebo group ( Arthur Burnett, personal communication).

Bella et al. [57] applied neurturin, a member of the glial cell line-derived neurotrophic factor family of ligands, directly at the crushed cavernous nerve sited in the rat model. They found that neurturin facilitated recover of erectile function with 50% increase if ICP compared with the controls. This study supports further investigation of neurturin as a neuroprotective and/or neroregenerative agent. Kendirci et al. [58] injected nonhematopoietic bone marrow stem cells isolated by the p75 NGF receptor into the rat penis after the cavernous nerve crush. They found that the ICP/MAP rations and the total ICP values are significantly higher in the rats that received the stem cells injection compared with those without injection. Their stuffy supports further investigation of the potential use of cell-based approaches to improve the erectile function after cavernous nerve injury. Burnett et al. [59] recently explored the feasibility of using an implantable electrode array for cavernous nerve stimulation for patients undergoing nerve-sparing RP. The implantable cavernous nerve electrode array was placed over the neurovascular bundles in nine patients undergoing retropubic RP. Four of the nine (44%) patients demonstrated a significant increase in penile circumference after cavernous nerve stimulation. Lack of response in some subjects may relate to differences in responses to anesthesia, as anesthetic agents are known to suppress tumescence. The findings suggest that an implantable system for programmed cavernous nerve stimulation after nerve-sparing RP is a feasible therapeutic option to improve post- RP penile erection.

Conclusion
Recent exciting scientific discoveries have improved our understanding of postprostatectomy ED at the molecular level. Therationale for postprostatectomy penile rehabilitation appears very logical according to animal studies. However, clinical studies have not consistently replicated the beneficial effects found in the laboratory studies. Currently available clinical studies are flawed due to short-term follow-up, small number of patients in the studies with retrospective nature, or prospective studies without control. Current rehabilitation programs are also facing a challenge with the compliance, which may be the reason for ineffectiveness of rehabilitation in some reports. The impact of patient partners’ attitude and participation in the rehabilitation programs is not known. At the present time we are unable to recommend what, when, how long, and how often a particular penile rehabilitative therapy can be used effectively. Bannowsky et al. [60] recommended that nocturnal penile tumescence and rigidity testing could serve as a guide for early penile rehabilitation in the form of either a PDE5-L (if the patient had erectile rigidity). This concept has yet to be tested. Mulhall et al. [61] analyzed the predictors of out-comes with pharmacological penile rehabilitation after RP. The patients ate from the database which participated in a post-RP rehabilitation program with oral erectogenic agents or ICI therapy. In total, 56.5% patients had partner-corroborated functional erections without PDE5-Is at 18 months post RP. They found that patients without nerve-sparing surgery, commencement of rehabilitation>6 months post RP, use of Trimix dose>50 units, age>60 years, and presence of>1 vascular comorbidity had high risk for failure of return of nonmedication-assisted erections.

Who else may need penile rehabilitation? There is no publication regarding penile rehabilitation after RP without sparing of the cavernosal nerves; therefore, the benefits of penile rehabilitation in this group of patients are not known. Rehabilitation may not necessarily improve erectile function in these patients; it may, however, prevent the decrease in penile length that is very common after RP [12, 46, 62]. In addition, about 64% of patients with three-dimensional conformal external-beam radiation for prostate cancer will have ED [63]. We do not have answers at this time whether early penile rehabilitation will preserve erectile function in these patients.

Clearly, large prospective, multi-centered, placebo-controlled trials with adequate follow-up ate necessary to determine the cost-effective and therapeutic benefits of particular penile rehabilitative therapy or therapies in patients following the treatment of clinically localized prostate cancer. Until such evidence is available, it is difficult to recommend any particular penile rehabilitation program as a standard of practice. Exploration of novel rehabilitation approaches is necessary to overcome the problem of poor compliance associated to most currently used rehabilitation programs.

Acknowledgment
The author would like to thank Mrs. Dorothy Stradinger for her editorial assistance.

Corresponding Author: Run Wang, MD, UT-Houston Medical School-Urology, 6431 Fannin Street Suite 6018, Houston TX 77030, USA Tel:713-500-7337; Fax 713-500-7319 e-mail: run.wang@uth.tmc.edu