Current therapeutic strategies for erectile function recovery after radical prostatectomy – literature review and meta-analysis

Abstract

Radical prostatectomy is the most commonly performed treatment option for localised prostate cancer. In the last decades the surgical technique has been improved and modified in order to satisfy both oncological safety and postoperative functional results. Urinary incontinence and erectile dysfunction (ED) are the main postoperative functional impairments in patients that undergo radical prostatectomy (RP). [1] According to the EAU Guidelines erectile dysfunction is defined as “the persistent inability to attain and maintain an erection sufficient to permit satisfactory sexual performance”. To this moment it is estimated that 25-75% of the patients who underwent RP develop ED with major implications in postoperative quality of life. [2] Various risk factors can be related to post-prostatectomy ED, among which the most important are the patient’s age, comorbidities, preoperative erectile function, but also surgical approach (robot-assisted, laparoscopic, open) and nerve-sparing technique (inter/intrafascial). [2, 3] Among physiopathological modifications that concur to ED, the nerve damage (neuropraxia) appears to be the main domino piece that triggers other alterations like structural changes of the smooth muscle, arterial damage and veno-oclusive dysfunction which lead to cavernosal oxygenation impairments. [4–6] The clinical results of physiopathological alterations are shortening of penile size, impairment of orgasm and ejaculation achievement which have tremendous impact on psychological and mental health. [5] Various treatment options include oral medication with phosphodiesterase 5 inhibitors (PDE5i), intracavernosal injections or intraurethral applications of prostaglandin E1 (PGE1), vacuum or vibratory treatments. Though there are many options for ED treatment, to this date there is no definitive protocol for penile rehabilitation (PR) after prostatectomy. [5, 7] In this regard we have made a literature review and meta-analysis of the most important randomised trials published in the last 5 years in order to find the best strategies of penile rehabilitation after radical prostatectomy. MATERIALS AND METHODS We have systematically reviewed and collected data from randomised trials published in the last 5 years involving patients diagnosed with prostate cancer who experienced ED after radical prostatectomy. In order to find the necessary trials we have searched PubMed and ScienceDirect databases and used several key words like: erectile dysfunction, penile rehabilitation, phosphodiesterase 5 inhibitors, erectile functional recovery, prostate cancer, prostatectomy and nerve sparing – in various combinations. The first aspect for our meta-analysis inclusion criteria was to identify randomized (prospective or retrospective) placebo-controlled trials involving patients with localised prostate cancer that benefited of nerve sparing radical prostatectomy and developed postoperative ED. The key aspect of our research was the penile rehabilitation strategy which included the therapeutic means used for erectile 1 Dimitrie Gerota Emergency Hospital, Bucharest, Romania 2 Dr. Carol Davila University Central Emergency Military Hospital, Bucharest, Romania 3 Carol Davila University of Medicine and Pharmacy, Bucharest, Romania 10 function recovery and the treatment duration. Other aspects followed by us regarded the means of ED assessment, the surgical means and invasiveness and the rate of treatment success reported to ED severity. In this regard we have conducted our research for each year between January 2014 and December 2018 and reviewed a total of 631 articles published on this theme collected from both search engines. For a time-efficient assessment we have read the title and abstract of each article and included only those articles that matched our criteria. We have excluded the reviews and meta-analyses and also the trials of ED that was not related to prostate cancer surgery. That left us 194 publications from which to select. We have also excluded the trials withdrawn from database for various reasons and articles that didn’t followed the therapeutic means of ED recovery. Finally we have selected 15 articles matching our criteria for meta-analysis that were thoroughly full text reviewed. Seven out of the 15 were prospective randomised trials, 2 of them being double-blind, placebo-controlled. Another 2 were nonrandomised prospective trials and the rest of 6 articles were retrospective review. Although some studies included by us followed-up both the ED and postoperative incontinence problem, we have concerned strictly on erectile dysfunction issue and collected the necessary data. For statistical analysis we have used ReviewManager 5.3 and Microsoft Excel software with data of treatment strategies and drug dosages collected from the studies. Although there was less heterogeneity between the study designs of the included articles we found important data about postoperative EF recovery. These data allowed us to include the studies in subgroups in order to analyse them. Table 1. Distribution of included studies Trial author Design Patient no. Mean age Cardiovascular comorbidities Diabetes Vickers et al. [8] Randomised prospective 2162 64.2 NoS NoS Montorsi et al. [9] Randomised double-blind placebo-control 423 58 NoS NoS Seo et al. [10] Retrospective review 92 67.9 40 (43.5%) 14 (15.2%) Nakano et al. [11] Retrospective review 103 63.4 NoS NoS Fode et al. [12] Randomised prospective control 68 62 NoS NoS Stolzenburg et al. [13] Randomised double-blind 422 59 NoS NoS Canat et al. [14] Randomised prospective 112 63 42 (37.5%) 18 (16.07%) Yiou et al. [15] Retrospective review 75 59.4 30 (40%) 6 (8%) Haglind et al. [16] Prospective controlled nonrandomised 2625 63 NoS NoS Kim D. et al. [17] Randomised prospective placebo-control 97 54 NoS NoS Capogrosso et al. [18] Retrospective review 2364 61 NoS NoS Sooriakumaran et al. [19] Prospective non-randomised 2545 63.3 556 (35.2%) 99 (6.2%) RESULTS A total of 11,831 patients were enlisted in the 15 studies we have included for analysis. Mean age of all patients was 61.91 years with a range between 54 and 67.9 years. (Table 1) Although not all the studies specifically mentioned the most encountered comorbidity was the cardiovascular disease (37.5% - 100%) followed by diabetes mellitus (6.2% – 16.07%). For most ED patients the presence of cardiovascular comorbidities creates the premises of disease trigger, followed by diabetes and obesity. [] All patients were Vol. CXXII • No. 3/2019 • December • Romanian Journal of Military Medicine 11 diagnosed (by means of PSA, prostate biopsy and pelvic MRI) with localised prostate cancer mainly T1c - T2c, with Gleason score ranging between 6 and 7 and PSA values below 10 ng/ml. (Table 2) These data were correlated with the postoperative pathology findings which in some cases changed the actual diagnostic stage according to d’Amico criteria. As it can be seen in Table 2 in some cases based on the degree of invasion the cancer stage was changed from localised to locally-advanced (T3b-T4) and based on Gleason score from low or intermediate risk to high risk (Gleason 8). All patients were proposed for radical prostatectomy using various approaches: open surgery (n = 4,460), conventional laparoscopy (n = 1,371) or robot assisted laparoscopic prostatectomy (n = 4,929). (Table 2) Most patients benefited of bilateral nerve sparing (BNS) dissection of the nerve bundles (n = 6,852) by al known technical procedures (interfascial/intrafascial), which represented 63.68% of the operated patients. When case did not impose BNS dissection, unilateral nerve sparing technique was attempted (n = 3,209) which was performed in 29.82% of the operated patients. The rest of 6.5% of the patients either had not benefited of NS dissection either the information about the NS status was not mentioned. [16, 18, 19] The NS status at the end of the surgery was crucial because it ensured the premises for postoperative ED recovery although most of the patients experienced postoperative neuropraxia. Table 2. PCa status and surgical approach for each study Trial author PCa clinical stage Gleason Score Surgical approach Nerve sparing OP CLP RALP BNS UNS Vickers et al. [8] T2a – T3 6 - 8 859 (40%) 546 (25%) 757 (35%) 657 (30%) 1498 (70%) Montorsi et al. [9] T1c – T2c 6 - 7 68 (48.9%) 29 (20.9%) 31 (22.3%) 423 (100%) - Seo et al. [10] T2a – T2c 4 - 8 - - 92 (100%) 57 (62%) 35 (38%) Nakano et al. [11] T2 – T3 6 - 7 NoS NoS NoS 24 (23.3%) 79 (76.7%) Fode et al. [12] T2 – T3 NoS 7 (10.3%) - 61 (89.7%) 37 (54.41%) 31 (45.58%) Stolzenburg et al. [13] T2-T3 ≤ 7 189 (44.78%) 88 (20.85%) 115 (27.25%) 422 (100%) - Canat et al. [14] T2a-T2c 5 - 8 NoS NoS NoS 112 - Yiou et al. [15] T2 - T3 6 - 8 - 75 (100%) - 75 (100%) - Haglind et al. [16] T1 - T3 6 - 8 778 - 1847 1318 750 Kim D. et al. [17] T1c – T3 6 - 8 58 (61.7%) - 36 (38.3%) 96 (98.96%) 1 (1.04%) Capogrosso et al. [18] T2 - T4 NoS 614 (26%) 633 (27%) 1117 (47%) 2253 (95%) 78 (3%) Sooriakumaran et al. [19] T2 – T4 NoS 1792 (70.41%) - 753 (29.58%) 1283 (50.41%) 737 (28.95%) Kim S. et al. [20] T2a – T2c 4 - 8 95 (100%) - - 95 (100%) - Kwon et al. [21] NoS NoS NoS NoS NoS NoS NoS Jo et al. [22] T2a – T2c 5 - 7 - - 120 (100%) NoS NoS NoS – no specification; OS – open surgery; CLP – conventional laparoscopic prostatectomy; RALP – robot assisted laparoscopic prostatectomy; BNS – bilateral nerve-sparing; UNS – unilateral nerve-sparing. 12 Table 3. Preoperative EF assessment and postoperative EF recovery means and results Trial author Type of EF questionnaire Preop. mean EF score Means of penile recovery Subgroups Trial duration Recovery rates Vickers et al. [8] IIEF-6 ³22 - 937 (59%) PDE5i (NoS) PGE1 ICI - 2 years 36% (NoS) Montorsi et al. [9] IIEF-EF ³22 - 423 (100%) PDE5i Tadalafil 5 mg – OaD – 139 (32.86%) Tadalafil 20 mg – PRN – 143 (33.8%) Placebo – 141 (33.3%) 9 months Tadalafil 5 mg – 25.2% Tadalafil 20 mg – 19.7% Placebo – 14.2% Seo et al. [10] IIEF-5 ³22 - 92 (100%) PDE5i Tadalafil 5 mg – OaD Non-Tadalafil 1 year Tadalafil 5 mg – 13.2% Non-Tadalafil – 7.7% Nakano et al. [11] IIEF-5 17.7 PDE5i Vardenafil 10 mg – OD – 30 (29.12%) Vardenafil 20 mg – PRN – 5 (4.85%) No treat – 68 (66.01%) 1 year Vardenafil 10 mg + Vardenafil 20 mg – 21 (60%) No treat – 26 (38.2%) Fode et al. [12] IIEF-5 25 PDE5i PVS PDE5i + PVS - 30 PDE5i - 38 1 year PDE5i + PVS – 16 (53%) PDE5i – 12 (32%) Stolzenburg et al. [13] IIEF-5 ³22 – 422 (100%) PDE5i Tadalafil 5 mg – OaD - 102 Tadalafil 20 mg – PRN - 112 Placebo - 106 9 months Tadalafil 5 mg – OaD – 29 (28.43%) Tadalafil 20 mg – PRN – 24 (21.42%) Placebo – 27 (25.4%) Canat et al. [14] IIEF-6 ³ 22 PDE5i Tadalafil 5 mg – OpW - 38 Tadalafil 20 mg – PRN - 40 Non-Tadalafil - 34 1 year Tadalafil 5 mg – OpW – 19.89 (mean IIEF) Tadalafil 20 mg – PRN – 15.8 (mean IIEF) Non-Tadalafil – 13.47 (mean IIEF) Yiou et al. [15] IIEF-EF EHS ³ 24 ³ 2 PDE5i PGE1 ICI PDE5i + PGE1 ICI – NoS No treatment - NoS 2 years M12 – 19.6 (mean IIEF – treat.) M12 – 18.07 (mean IIEF – no treat.) M24 – 4.63 (mean IIEF – treat.) M24 – 4.92 (mean IIEF – no treat.) Haglind et al. [16] IIEF-5 ³ 21 - RRP RALP 1 year RRP – 124 (15.93%) IIEF-5 ³ 21 RALP – 339 (18.35%) Kim D. et al. [17] IIEF-EF ³ 21 (100%) (Mean 28.1) PDE5i Sildenafil 50 mg - OaD + Sildenafil 100 mg - PRN Placebo + Sildenafil 100 mg - PRN 13 months Sildenafil 50 mg - OaD + Sildenafil 100 mg – PRN – 15 (32.4%) Placebo + Sildenafil 100 mg – PRN – 13 (29%) Capogrosso et al. [18] IIEF-6 ³ 24 (Mean 27) PDE5i PGE1 ICI 12 months - 1779 24 months - 1095 2 years 12 mo – 483 (27%) 24 mo – 377 (34%) Sooriakumaran et al. [19] Penile stiffness Morning erection 1702 (66.9%) PPP. PGE1 ICI PGE1 IUr RARP RRP 2 years RARP – 377 (21%) RRP – 106 (14%) Kim S. et al. [20] IIEF-5 22.4 PDE5i Tadalafil 5 mg – OaD – 2 years Tadalafil 5 mg – OaD – 1 year No Tadalafil 2 years Tadalafil 5 mg OaD – 2 yr. – 16.1 (mean IIEF) Tadalafil 5 mg OaD – 1 yr. – 13.5 (mean IIEF) No Tadalafil – 9.4 (mean IIEF) Vol. CXXII • No. 3/2019 • December • Romanian Journal of Military Medicine 13 Kwon et al. [21] Penile lenght SHIM NoS PDE5i (Sildenafil, Vardenafil, Tadalafil) Complete recovery Incomplete recovery 1 year CR – 318 (60.2%) IR – 210 (39.8%) Jo et al. [22] IIEF - 5 ³ 17 PDE5i (Sildenafil 100 mg) Early treatment - 58 (3 wk. postop.) Delayed treatment - 62 (3 mo. postop.) 1 year ET – 24 (41.4%) DT – 11 (17.7%) PPP – preoperative potent patients; IIEF – International Index of Erectile Function; OaD – oance a day; PRN – pro re nata (on demand); ICI – intracavernosal injections; IUr – intrauretral; For pre and postoperative EF assessment the IIEF questionnaire was preferred in most of the studies in different forms: the classical 15 questions IIEF-EF or the shorter IIEF-5/IIEF-6 questionnaires. Most authors preferred the IIEF-5 questionnaire due to its reliability and rapid way of determining the EF status. (Table 3) In most of the studies that used the IIEF questionnaire a cut-of value of 22 points was established for preoperative potent men [8–10, 13, 14] while others established close cut-of values. [16, 20] Only two studies used other means of EF assessment. Sooriakumaran et al. used two questions about penile stiffness and morning erections, both having single choice answer from 5 possible variants. [19] In a particular case Kim et al. compared the penile length both pre and postoperative and additionally used the SHIM questionnaire for ED appraisal. [20] The most frequently used treatment method for penile recovery was the administration of PDE5i which was tested with various ways of administration and dosages and compared with placebo or no treatment control groups. The most commonly used drug in the included trials was Tadalafil [9, 10, 13, 14, 20], followed by Sildenafil [17, 22] and Vardenafil. [11] Some trials studied the single use treatment of PDE5i compared with combinations of PDE5 with PGE1 ICI [8, 15, 18] or penile vibratory stimulation (PVS) devices. [12] There were also particular studies that did not compare the method of PDE5i administration but rather the rate of EF recovery [8, 21], the moment of treatment initiation [22] or the length of treatment administration. [18] Haglid et al. compared the EF recovery based only on the surgical approach type (open vs. RALP) and not interfering with any medication in this regard while Sooriakumaran et al. compared the same surgical approaches using PGE1 treatment setting. [16] Figure 1: Multivariate comparison of PDE5i OaD and PRN treatment compared to control group Although the articles included in our meta-analysis had dichotomous criteria of study design we could find similarities between some of them, which allowed us to organise subgroups of trials. On multivariate analysis we have included trials in which PDE5i was administered as OaD and PRN treatment compared to a pacebo control group. The variables included in analysis were the mean IIEF scores recorded at the end of the trial for each group (OaD, PRN and control). As it can be observed in fig. 1 the higher rates of recovery are in favour of PDE5i OaD group compared to control group (CI 95%, P < 0.0001). On a lower scale, the same result could be observed when PDE5i OaD treatment was compared to PDE5i PRN although in this analysis only 2 studies had the necessary data to be included (CI 95%, P = 14 0.48). In a univariate analysis we also compared the trials were only the percentage of recovered patients after PDE5i treatment was given. (Fig 2) As in previous assessment the PDE5i OaD group had higher recovery rates compared to control group (CI 95% P = 0.40). Also when we have compared PDE5i OaD with PRN group the better results were slightly higher in favour of OaD group thus resulting that the benefit of PDE5i OaD compared to PRN treatment was minor (CI 95%, P = 0.95). Figure 2: Recovery rates in trials using PDE5i OaD and PRN treatment compared to control group DISCUSSION Erectile dysfunction is a major postoperative complication for men who undergo radical prostatectomy, alongside urinary incontinence. At this moment prostate cancer is the second most diagnosed neoplasia in men due to the systematic determination of PSA. [23] Radical prostatectomy remains the most important therapeutic procedure for localised PCa having a postoperative life-expectancy of 10 years. [2] Although the technical enhancements of prostatectomy reduced the rates of postoperative complication, ED still remains one of the most complex problem for PCa patient’s quality of life. Even if the nerve sparing dissection technique of the prostate bundles partially improved the ED rates, there are still problems regarding the postoperative penile rehabilitation. Neuropraxia is the main pathological alteration that leads to major structural changes of the penile tissue. This can also lead to veno-oclusive dysfunctions of blood drainage. Due to the constant state of hypo-oxygenation the penile smooth muscles suffers a process of apoptosis and fibrosis which in time are producing irreversible penile malformations such as penile shortening of 2-3 cm and Peyronie disease. [5, 24] In this regard it is advisable that urologists should discuss with their patients about the possibility of postoperative ED in various rates – from temporary to permanent. [25] Two trials included in our meta-analysis studied the rate of ED occurrence and penile recovery degree after RALP compared to RRP in high volume centres from Sweden. In a 2015 prospective study, Haglid et al. found a slightly higher rate of penile recovery for RALP, but the differences were not significant for these two surgical approaches. [16] In a recently conducted prospective trial, Sooriakumaran et al. (2018) found earlier recovery rates of EF for RALP (21%) compared to RRP (14%) due to a more precise identification and preservation of nerve bundles during RALP. Also the rates of post-surgical positive margins were similar to open surgery which is an important aspect for oncological radicality. [19] Both studies were prospective non-randomised. One key aspect of ED diagnosis is the use of pre and postoperative EF assessment questionnaires. As we have shown above the most frequently used questionnaire was IIEF-5 which is the short version of the IIEF-EF questionnaire. Other questionnaires like Sexual Health Inventory for Men (SHIM) or Expanded Prostate Cancer Index Composite (EHGS) have similar predictive results to IIEF and should be routinely used for postoperative ED diagnosis. [25, 26] It is a known fact that older age, cardiovascular comorbidities and diabetes mellitus are the most important predictive factors for ED onset. [27, 28] On the other hand the younger age and the lower body mass index are Vol. CXXII • No. 3/2019 • December • Romanian Journal of Military Medicine 15 demonstrated protective factors for ED recovery. [2] Even though not all the studies included in our meta-analysis reported the rate of cardiovascular disease and diabetes mellitus we could emphasize that in the post prostatectomy setting these two comorbidities could have significant impact on ED onset and the subsequent penile recovery. [10, 14, 15, 19, 20, 21] Penile rehabilitation is the cumulus of therapeutically means which includes medication, medical devices or actions, simple or in various combination recommended for erectile function recovery. [5] PDE5i are the most frequently used drugs in clinical practice for penile rehabilitation and to this date is the most studied therapeutic method administered to ED patients. The main advantage of PDE5i therapy is that there are quick and easy to administer but it requires the integrity of at least 1 cavernosal nerve bundle. Also the treatment cost and side effects (headache, flushing and palpitations) are not negletable. [24] In their common trial Montorsi et al. (2014) and Stolzenburg et al. (2015) studied the effect of Tadalafil administered once a day (OaD) and on demand (PRN) compared to a placebo-controlled group and found that the OaD administration of Tadalafil 5 mg has the best results for penile recovery after prostatectomy. [9, 13] Similar results for Tadalafil OaD administration were also found by Seo et al. (2014), Canat et al. (2015) and Kim S. et al. (2018) based on the mean IIEF score at the end of the trial. [14, 20] Nakano et al. (2014) and Kim D. et al. (2016) found similar results at the end of their trials although they have studied the effect of Vardenafil respectively Sildenafil. [11, 17] In our multivariate analysis the PDE5i OaD treatment has evident advantages over placebo and on demand PDE5i administration for postoperative penile recovery rate. Regarding the choice of PDE5i treatment there are no specific trials that compare the efficiency of currently available substances in postoperative ED. In a 2005 open-label randomised trial Eardley et al. emphasized that Tadalafil is the preferred PDE5i used for ED treatment because of long-lasting effect in time compared to Sildenafil that allowed patients to have less concerns about spontaneity of erection. [29] Further more Hyndman et al. demonstrated in their randomised trial that Sildenafil could have major impact over urinary continence, thus impairing the overall recovery of PCa operated patients. [30] Prostaglandin E1 intracavernosal injection (PGE1 ICI) was the first type of treatment for postoperative EF recovery, introduced and studied by Prof. Montorsi et al. [31] The most important advantage of PGE1 ICI is that is very effective in any type of surgical approach even if nerve sparing was not indicated, though the cost, the need to refrigerate the product, the invasiveness of the treatment and the side effects are a majos counterbalance for treatment initiation. [24] In our meta-analysis, Vickers et al., Yiou et al. and Capogrosso et al. have used PGE1 ICI in combination with PDE5i treatment for penile recovery. The results achieved by Yiou at 12 respectively 24 months of combined treatment were very similar in both drug-administered and control group. [15] On the other hand Capogrosso et al. found that at 24 months of combined PGE1 ICI and PDE5i treatment the rates of recovery were higher than at 12 months (34% vs. 27%). [18] Vacuum and vibratory devices are non-invasive and very effective in EF recovery especially when combined with PDE5i treatment. Fode et al. demonstrated in their randomised prospective trial that EF recovery could be very effective when penile vibratory stimulation devices are used in conjunction with PDE5i (53% recovery rate) compared to PDE5i only group (32%). [12] Regarding treatment initiation Jo et al. demonstrated that early penile rehabilitation commencement (at 3 weeks after surgery – immediately after urinary catheter removal) has better recovery results than delayed treatment initiation (at 3 months postoperative). [22] CONCLUSIONS The postoperative EF recovery of PCa patients can be achieved using a multitude of penile recovery strategies. For this purpose a preoperative assessment is very important, by using erectile function questionnaires. Patients should be informed about the postoperative ED risk and should be counselled in this area. Nerve-sparing robot assisted laparoscopic prostatectomy is the most protective surgical approach for EF preservation. PDE5i is the main therapeutic approach for postoperative penile recovery. Tadalafil 5 mg once a day was the most effective long term therapy for ED treatment both single use or in combination with PGE1 and other devices (vacuum or vibratory stimulation). Penile prosthesis implants should be regarded as last resort treatment option for cases when no penile recovery strategy has been successful.