Said Fadel Mishriki, Department of Urology, Aberdeen Royal Infirmary Hospital, Aberdeen AB25 2ZN, Scotland, UK. e-mail: firstname.lastname@example.org
Study Type – Symptom prevalence (prospective cohort)
Level of Evidence 1b
What's known on the subject? and What does the study add?
Evidence that transurethral resection of the prostate (TURP) leads to erectile dysfunction (ED) is conflicting. Several studies claimed significant risk of ED after TURP for benign prostatic hyperplasia with some reporting complete loss of erection. Several studies have been retrospective or have not considered levels of pre-operative ED.
ED associated with lower urinary tract symptoms frequently precedes TURP. TURP did not adversely affect erectile function. Pre-operative ED can be improved by TURP and long-term erectile function is maintained following TURP. The improvement was corroborated by the partners in the short, medium and long-term and was statistically significant.
• To evaluate the effect of transurethral resection of the prostate (TURP) on sexual function in the short (6 months), medium (6 years) and long (12 years) term and assess the conformity between patient and partner regarding sexual function.
PATIENTS AND METHODS
• A prospective cohort study set at the Aberdeen Royal Infirmary University Hospital.
• A total of 280 men referred with lower urinary tract symptoms (LUTS) to a university hospital underwent TURP between January 1993 and September 1994; 145 of their partners (partner or spouse) participated.
• Assessment included American Urological Association symptom score, flow rates and validated self-reported sexual questionnaires (SQ).
• Data were collected at baseline, 3 months, 6 months, 6 years and 12 years of follow-up.
• In all, 120 (43%) men were sexually active preoperatively. At 6 months, 73 (61%) of these 120 men completed the SQ and all were sexually active.
• No sexually active patient became impotent after the procedure. Moreover, 27 (15%) with pre-existing erectile dysfunction reported improved sexual activity and erection quality.
• At 6 years 101 men completed the SQ and 31 (30.7%) were sexually active. At 12 years, 36 (31.9%) of 113 who completed the SQ were sexually active.
• Partners agreed with the men's self assessment at all visits.
• Limitations include possible attrition bias and lack of information from non-responders.
• Erectile dysfunction associated with LUTS frequently precedes TURP.
• The TURP did not adversely affect sexual function.
• Pre-operative erectile dysfunction can be improved by TURP and long-term sexual function is maintained after TURP. These findings, corroborated by the partners, were statistically significant.
Erectile dysfunction (ED) is defined as the consistent inability to attain or maintain an erection of sufficient quality to permit satisfactory sexual intercourse . Evidence that transurethral resection of the prostate (TURP) leads to ED is conflicting. Studies claiming significant risk of ED from TURP include 31% experiencing ED after prostatectomy for benign prostatic hyperplasia (BPH) , 27% reporting complete loss of erection and 8.3% becoming impotent on objective testing [3,4]. In one study significant sexual function improvement after TURP was reported . A randomized study showed non-significant improvement in sexual function associated with TURP when compared with watchful waiting . One important clinical question remains unanswered. Is the long-held idea that TURP is detrimental to sexual function correct or could TURP improve sex life?
There are several reasons for the confusing picture. Severity of urinary symptoms was correlated with measures of ED [7,8]. That the improvement in ED after TURP is associated with an improvement in lower urinary tract symptoms (LUTS), is supported by a similar observation of the improvement produced by alpha-blockers [9,10]. Few studies highlighted the fact that significant numbers of patients with LUTS have ED before TURP [7,11]. Several studies have been retrospective or have not considered levels of preoperative ED . Studies of sexual function after TURP have been limited when compared with other outcomes. Long-term randomized trials are awaited.
Sexual function is an important outcome measure to consider before TURP (still regarded by many as the standard for the treatment of BPH). Hence the determination of its true long-term influence on sexual function is essential. The patient characteristics, functional results and complications of TURP for this study have been published previously . Long-term improvement in quality of life and bother scores noted by patients and corroborated by partners in this study was also reported previously .
The objectives of this study were to evaluate the effect of TURP on sexual function in the short (6 months), medium (6 years) and long (12 years) term and to assess the conformity between patient and partner regarding sexual function.
PATIENTS AND METHODS
In all, 280 men with LUTS referred to Aberdeen Royal Infirmary University Hospital between January 1993 and September 1994 underwent TURP as a definitive treatment in this prospective-protocol-based study. Treatment decision was based on combined assessment of symptoms, AUA symptom score, flow rates and residual urine. Of their partners (partners and spouses), 145 agreed to participate.
Personal details, DRE, PSA, flow studies, post-micturition residuals and TRUS for prostate size were recorded. Patients and partners completed the Sex Questionnaire (Appendices 1 and 2). Possible minimum and maximum individual scores were 0 and 8.
Men diagnosed with carcinoma of the prostate after TURP and patients presenting with urinary retention or requiring open prostatectomy were excluded from the study.
Patients and partners have been followed up for 12 years after TURP. Partners were asked to complete the questionnaire independently. This took place when accompanying the patient to the clinic, at home in the presence of a research nurse or questionnaires were returned by post. Patients and partners completed these questionnaires preoperatively and at 3 months, 6 months, 6 years and 12 years. During the 12 years, some of the men who were single or widowed at the start of the study have married and some married men have divorced or been widowed.
Ethical approval was granted for the initial study and all follow-up stages and all participants provided written informed consent.
A questionnaire suitable for self-administration designed to assess sexual desire, sexual activity and erection quality was based on early non-validated questionnaires used in 1992. The international index of erectile function was not available till February 1997 . Reliability of the questionnaires and stability of responses over time were internally validated by a test–retest analysis, achieved by conducting the questionnaires twice in the same group with a 3-month interval between tests during which no change in therapy was instituted. This method has been used in at least one similar study .
The test–retest analysis compared with Spearmann's rank analysis, to assess consistency in scoring, showed strong correlation (patients: r = 0.753, P = 0.001; partners: r = 1.0, P = 0.01).
Inter-rater reliability is a measure used to examine the agreement between two ‘raters/observers’ on the assignment of categories of a categorical variable. Cohen's Kappa coefficient was used to determine the consistency of responses between patients and their partners who had responded at the various time-points (pre-TURP, 6 months, 6 years and 12 years). Patients without partners or whose partners had not responded were excluded from the analysis because the patients' and partners' responses were considered in pairs. There were four possible paired outcomes: ‘Yes–Yes’, ‘No–No’, ‘Yes–No’, ‘No–Yes’, with the ‘Yes–Yes’ and ‘No–No’ groups reflecting agreement between patients and partners.
The number of patients and partners at each stage of follow-up is shown in Fig. 1. Mean age of patients at the time of TURP was 68 years (range 49–88 years).
Table 1 shows patient and partner numbers and percentage of partners agreeing/differing with the patients' sexual status. At baseline, 217 men were married, 10 were single, 31 were widowed and 22 did not declare a marital status.
Table 1. Partner agreeing or differing regarding sexual status
Preoperatively 120 (43%) men (mean age 65 years) were sexually active with 81.8% of 66 partners in agreement. A total of 160 (57%) men (mean age 70 years) had ED with 96.2% of 79 partners in agreement.
At the 6 months follow-up, 168 men were followed up. One hundred men (mean age 69 years) were sexually active with 86% of 36 partners in agreement and 68 men (mean age 70 years) had ED with 96.8% of 31 partners in agreement. Of the 180 men who were sexually inactive preoperatively, 27 (15%) men with pre-existing ED reported improved sexual activity. Of the 120 men who were sexually active preoperatively, 73 (61%) completed the 6 months follow-up. All (100%) were still sexually active after TURP. No patient who was fully sexually active preoperatively reported developing ED.
At the 6 years follow-up, 215 of the original 280 patients were still alive and 101 men (mean age 72 years) completed the SQ. Thirty-one (mean age 69 years) were sexually active with 77.3% of 22 partners in agreement and 70 men (mean age 73 years) had ED with 94.4% of 36 partners in agreement. Of the 73 sexually active patients who completed the 6 months follow-up, 47 (mean age 64 years) completed the 6 years follow-up. Of these, 30 (64%) were still sexually active.
At the 12 years follow-up 164 of the original 280 patients were still alive. In all, 113 (mean age 77 years) completed the SQ and 60 partners completed the questionnaires. This response, considering the significant number of single or widowed men, reflected the zeal in tracing patients and partners for the final follow-up. Of the 113, 36 (mean age 73 years) were still sexually active with 91.7% of 24 partners in agreement and 77 (mean age 79 years) had ED with 97.2% of 36 partners in agreement.
Inter-rater reliability scores (kappa coefficient) for patients and their partners are shown in Table 2, which shows a high level of agreement between patients and partners regarding the patients' sexual activity status.
Table 2. Inter-rater reliability analysis using Cohen's kappa coefficient
95% confidence interval
Interpretation of Kappa coefficient values: 0.40–0.59, moderate; 0.60–0.79, good; ≥0.80, excellent. TURP, transurethral resection of the prostate.
To our knowledge, no such data involving partners have been published, although one-shot sexual activity assessment by both man and partner showed good correlation . In this study TURP was not associated with ED postoperatively. The long-term follow-up and corroboration by the partners are the main strong points of this study. These findings have several important implications for clinical practice.
First, TURP was traditionally thought to be associated with ED so counselling was advocated during consent for the procedure. Erectile dysfunction has been reported in 8.3–32.5% [2–4] with one study noting postoperative ED as high as 81% . The large variation is the result of differing definitions of ED/impotence, use of varied outcome measures (subjective and objective), varying follow-up duration and inclusion of procedures other than TURP (bladder neck incision/open prostatectomy) . In the light of this study, counselling may need to be revisited.
Second, this study has highlighted a crucial point; ED is common in patients with LUTS before treatment with 57% reporting ED preoperatively. A study of patients with BPH found a significant difference in sexual function between those with severe LUTS (44.2% unsuccessful coitus) and mild LUTS (13.1%) . Reports of ED post-TURP should be contextualized with the preoperative condition; reporting 81% postoperative ED is qualified by a preoperative figure of 86% . A 12% ED after TURP when compared with 3% for ‘equivalent’ general surgical procedures was noted  and ED was reported in 11.8% of 158 patients undergoing TURP or suprapubic/transvesical prostatectomy .
The importance of age in relation to ED is apparent in this study. Sexually active men were younger preoperatively and at all follow-up stages although this was not significant. The Multinational Survey of the Aging Male involving over 14 000 men linked ED to age and LUTS severity. ED correlated with age and the BPH impact index and with individual components of the International Prostatic Symptom Score (IPSS) notably nocturia and urgency .
Third, this study contradicts the concept that TURP damages sexual function. In one previous study, 56% of 168 men with BPH had poor sexual function and 46% were functionally impotent . Objective assessment showed that patients who were fully potent preoperatively had 34.7% ED at 4 days but this had decreased to 8.3% at 3 months . Rigiscan of nocturnal tumescence in addition to International Index of Erectile Function-15 reported 14% ED in 63 preoperatively potent men .
Fourth, this study has confirmed that TURP is associated with a 15% improvement in pre-existing ED contradicting several previous studies. Some studies are in agreement, noting postoperative improvement . One previous study using the International Prostatic Symptom Score, BPH impact index and a sexual function questionnaire for four surgical options including TURP reported that none of the procedures caused ED although 48.6% in the TURP subcohort experienced reduction in ejaculation . Another reported no significant change in postoperative (24%) vs preoperative (22%) impotence in 388 men undergoing TURP . No change in sexual satisfaction and libido was found between preoperative and 12 months values after TURP in 155 men  and no difference was found between men undergoing TURP and inguinal hernia repair except for retrograde ejaculation . Another study reported a significant increase in morning erections after TURP .
In the present study, patients were adequately counselled regarding postoperative retrograde ejaculation. This may account for the fact that none confused retrograde ejaculation with ED. There are no guidelines regarding retrograde ejaculation. It should be noted that even those studies showing no ED recorded absence or reduction of ejaculate [17,23]. Post-TURP retrograde ejaculation may explain the continued perception by patients that TURP is detrimental to sexual function. Studies comparing TURP and alternative surgical modalities show similar outcomes in all areas except retrograde ejaculation [17,23]. Details regarding ejaculation were not collected in this study and it would have been impossible for the partners to comment on or verify ejaculatory findings.
Could TURP be actually beneficial in terms of sexual function outcome? The pathophysiology of ED after TURP remains unknown. Psychological factors such as loss of sexual self-confidence after TURP play a role . Various treatments can both increase and decrease post-treatment sexual activity . Few studies identified factors that may increase the risk of ED. These include age , diabetes mellitus , cardiovascular disease , relatively poor International Index of Erectile Function-15 and nocturnal tumescence scores , small adenoma size [3,4], intraoperative capsular perforation , and damage to penile small nerve fibres . Some have shown mixed outcomes in the same study . The US Food and Drug Administration approved sildenafil citrate, the first phosphodiesterase 5 inhibitor in March 1998 . Active Viagra prescribing started in the UK in 1999, which is when the 6 years follow-up took place so it could not have affected the results. The 12 years follow-up may have been affected.
The improvement in ED in this study is related to the improvement in LUTS after TURP, which was also corroborated by the partners and the results were published previously [12,13]. This is akin to the improvement in ED produced by alpha-blockers [9,10]. Surgical alternatives to TURP show a mixed picture. Transurethral microwave thermotherapy, interstitial laser coagulation, transurethral needle ablation and Holmium laser enucleation gave similar results to TURP regarding sexual function [24,30]. Laser therapy had a beneficial effect on pre-existing ED .
There was consistent agreement between patient and partner scores before and after TURP. The same SQ was completed by the patient and partner preoperatively, at 6 months, 6 years and 12 years. In this respect this study is unique. It is interesting to note that the partner had no difficulty commenting on specific topics in the SQ. This may provide an independent sexual function assessment source worthy of note during counselling. Statistically significant conformity was maintained throughout.
In less than 3% of cases, at various follow-up stages, a man's test score was 8 whereas his partner's score was 0. Data analysis failed to find a reason for the variation and there is no plausible explanation. Readers can draw their own conclusions.
Potential limitations include the design, conduct and reporting by a single group on a single site and no age-matched men not undergoing TURP. Attrition bias and lack of information from non-responders are also factors. It is possible to consider the non-responders as sexually inactive and therefore not interested in the questionnaire or as sexually active and simply not attending. The percentage of partners responding throughout the study was between 36% and 71%. Other limitations include the lack of an internationally accepted questionnaire at the time of the study initiation. The 12 years follow-up may have been biased by the presence of phosphodiesterase 5 inhibitors.
In conclusion, this long-term prospective study in which partners actively participated confirms that ED is associated with LUTS and frequently precedes TURP. Improvement in ED was associated with improvement in LUTS. TURP probably does not adversely affect sexual function and in some cases preoperative ED can be improved by TURP. Furthermore, long-term sexual function continued to be maintained after TURP. These findings, corroborated by the partners, were significant.
The authors wish to thank Mrs Barbara Gibbons for encouraging partner involvement during the initial period, Mrs Edna Ledingham for patient and partner long-term follow-up and Mrs Penny Vale for proofreading the manuscript.
CONFLICT OF INTEREST
APPENDIX 1. SEX QUESTIONNAIRE FOR PATIENT
Patient of Study number
1Over the past month or so have you had any interest in sex?
2Over the past month or so, have you had any kind of sexual activity (intercourse, masturbation etc.)?
3Over the past month or so, how often have you been able to have sexual erections when you were sexually stimulated?
Not at all
Less than 1 time in 5
Less than half the time
About half the time
More than half the time
4Over the past month or so, how would you rate your erections?
Not firm enough to have intercourse
Firm enough to have intercourse
APPENDIX 2. SEX QUESTIONNAIRE FOR PARTNER
Sex questionnaire partner
Partner of Patient Study Number
1Over the past month or so has your partner had any interest in sex?
0 (if no, ignore questions 3 and 4)
2Over the past month or so, has your partner had any kind of sexual activity (intercourse, masturbation etc.)?
0 (if no, ignore questions 3 and 4)
3Over the past month or so, how often has your partner been able to have sexual erections when sexually stimulated?
Not at all
Less than 1 time in 5
Less than half the time
About half the time
More than half the time
4Over the past month or so, how would you rate the erections?