Treatment Options for Sexual Dysfunction in Patients with

Chronic Kidney Disease: A Systematic Review of

Randomized Controlled Trials
Mariacristina Vecchio,* Sankar D. Navaneethan,†‡ David W. Johnson,§ Giuseppe Lucisano,*
Giusi Graziano,* Marialuisa Querques,储 Valeria Saglimbene,* Marinella Ruospo,*
Carmen Bonifati,* Emmanuele A. Jannini,¶ and Giovanni F.M. Strippoli*‡**††
*Department of Clinical Pharmacology and Epidemiology, Mario Negri Sud Consortium, S Maria Imbaro, Italy;
†
Department of Nephrology and Hypertension, Cleveland Clinic, Cleveland, Ohio; ‡Cochrane Renal Group, Sydney,
Australia; §Department of Nephrology, University of Queensland at Princess Alexandra Hospital, Brisbane, Australia;
储
Nephrology, University of Milan, San Raffaele, Milan, Italy; ¶School of Sexology, Department of Experimental
Medicine, University of L’Aquila, L’Aquila, Italy; **Centre for Kidney Research, NHMRC Centre for Clinical Research
Excellence in Renal Medicine, School of Public Health, University of Sydney, Sydney, Australia; and ††Diaverum
Medical-Scientific Office Lund, Sweden

Background and objectives: Sexual dysfunction is very common in patients with chronic kidney disease (CKD), but
treatment options are limited. The benefits and harms of existing interventions for treatment of sexual dysfunction were
assessed in patients with CKD.
Design, setting, participants, & measurements: MEDLINE (1966 to December 2008), EMBASE (1980 to December 2008), and
the Cochrane Trial Registry (Issue 4 2008) were searched for parallel and crossover randomized and quasi-randomized trials.
Treatment effects were summarized as mean differences (MD) or standardized mean difference (SMD) with 95% confidence
intervals (CI) using a random effects model.
Results: Fourteen trials (328 patients) were included. Phosphodiesterase-5 inhibitors (PDE5i) compared with placebo
significantly increased the overall International Index of Erectile Function-5 (IIEF-5) score (three trials, 101 patients, MD 1.81,
95% CI 1.51 to 2.10), all of its individual domains, and the complete 15-item IIEF-5 (two trials, 80 patients, MD 10.64, 95% CI
5.32 to 15.96). End-of-treatment testosterone levels were not significantly increased by addition of zinc to dialysate (two trials,
22 patients, SMD 0.19 ng/dl, 95% CI ⴚ2.12 to 2.50), but oral zinc improved end-of-treatment testosterone levels. There was no
difference in plasma luteinizing and follicle-stimulating hormone level at the end of the study period with zinc therapy.
Conclusions: PDE5i and zinc are promising interventions for treating sexual dysfunction in CKD. Evidence supporting their
routine use in CKD patients is limited. There is an unmet need for studying interventions for male and female sexual
dysfunction in CKD considering the significant disease burden.
Clin J Am Soc Nephrol 5: 985–995, 2010. doi: 10.2215/CJN.09081209

⬍50 years versus 90% ⱖ50 years) (3). Similar results have been

S
exual dysfunction is a set of disorders characterized by
physical and psychologic changes that result in the in- reported in women with CKD, with 55% of female dialysis
ability to perform satisfactory sexual activities. The con- patients reporting difficulty with sexual arousal (2). Dysmen-
dition has been found to be significantly more common in men orrhea, delayed sexual development, impaired vaginal lubrica-
and women with chronic kidney disease (CKD) than in the tion, dyspareunia, and difficulties in reaching orgasm are also
general population (1). Men with CKD frequently suffer from frequently observed (7,8).
reduced libido, erectile dysfunction, and difficulty reaching Multiple factors contribute to the frequent occurrence of sex-
orgasm (2). Approximately 50% of male predialysis CKD pa- ual dysfunction in CKD patients, including hormonal distur-
tients and 80% of male dialysis patients have erectile dysfunc- bances (such as hyperprolactinemia, hypogonadism in males,
tion (3– 6). Moreover, the prevalence of erectile dysfunction in and changes in hypothalamic-pituitary function in women) (9),
male dialysis patients has been found to increase with age (63% anemia (10), CKD mineral and bone disorder (4), psychosocial
factors (such as depression, anxiety, poor self-esteem, social
Received December 16, 2009. Accepted February 25, 2010. withdrawal, marital discord, body image issues, fear of disabil-
Published online ahead of print. Publication date available at www.cjasn.org.
ity and death, loss of employment, and financial difficulties)
(2,11,12), autonomic neuropathy (13), medications (including
Correspondence: Dr. Giovanni F.M. Strippoli, Diaverum Medical Scientific
Office, Via Solarino 1, Bari, Italy. Phone/Fax: ⫹39 099 4595340; E-mail: antihypertensives, antidepressant, and histamine receptor
research@diaverum.com and strippoli@negrisud.it blockers) (2), and comorbid illness (such as diabetes mellitus,

Copyright © 2010 by the American Society of Nephrology ISSN: 1555-9041/506 –0985

986 Clinical Journal of the American Society of Nephrology Clin J Am Soc Nephrol 5: 985–995, 2010

cardiovascular disease, and malnutrition) (2,14). Sexual dys- combinations)]. Nonpharmacologic strategies included mechanical de-
function is inversely associated with GFR (7) and is improved vices (vacuum constriction device for inducing erection, penile pros-
after renal transplantation (15,16), suggesting that CKD per se thesis) and psychoeducational interventions such as rational emotive
may contribute to sexual dysfunction in these patients (15). therapy, sex group therapy, modified Masters & Johnson–Kaplan’s
sexual therapy, educational intervention, systematic desensitization,
Studies have also identified significant associations between
and sexual counseling.
sexual dysfunction in CKD patients and depression (8,17), im-
Similarly, for treatment of female sexual dysfunction, pharmacologic
paired quality of life (8,17,18), and adverse cardiovascular out- agents included hormonal therapy [oral or topical (transdermal) estro-
comes (19). Effective treatment of sexual dysfunction in CKD gens, testosterone, progesterone, or tibolone] and drugs (oral PDE5i).
patients may therefore potentially lead to improvement in these Nonpharmacologic strategies included mechanical interventions (estro-
patient-level outcomes, although a causal link has not been gen or nonhormonal lubricating vaginal creams and clitoral therapy
definitively established (18). devices) and psychoeducational interventions (rational emotive ther-
Therapies that have been used to treat sexual dysfunction apy, sex group therapy, modified Masters & Johnson–Kaplan’s sexual
include phosphodiesterase-5 inhibitors (PDE5i), intracavern- therapy, educational intervention, systematic desensitization, and sex-
osal injections, intraurethral suppositories, hormonal therapy, ual counseling).
mechanical devices, and psychotherapy. Although many clini-
cal trials and reviews have explored the role of these interven-
tions for sexual dysfunction in nonuremic patients (20 –24), the Types of Outcome Measures
We planned to obtain the following outcome measures as reported in
effectiveness and safety of these interventions in patients with
the included studies:
CKD have not yet been studied thoroughly. Therefore, we
Male sexual dysfunction outcomes
aimed to evaluate the benefits and harms associated with var-
ious interventions for sexual dysfunction in patients with CKD. • Changes in mean score on any standard validated sexual function
scale. The various scales that were considered for inclusion included
Materials and Methods the 15-item International Index of Erectile Function (IIEF), 5-item
Search Strategy International Index of Erectile Function (IIEF-5), Physic Component
We searched MEDLINE (via OvidSP, 1966 to December 2008), EM- Score, and Mental Component Score.
BASE (via OvidSP, 1980 to December 2008), Cochrane Renal Group’s • Achievement of prolonged penile rigidity satisfactory to enable com-
Specialized Register, and the Cochrane Central Register of Controlled plete sexual intercourse (measurement of genital blood flow and
Trials (CENTRAL) (Cochrane Library Issue 4, 2008) for relevant trials. nocturnal penile tumescence).
CENTRAL and the Renal Group’s specialized register contain the • Number of successful sexual intercourses and number of partici-
handsearched results of conference proceedings from general and spe- pants who showed improved sexual function as measured by pa-
cialty meetings. This is an ongoing activity across the Cochrane Col- tient’s log and reported by study authors.
laboration and is retrospective and prospective. The search strategy • Hormone levels as measured by trialists (including testosterone or
used to obtain titles and abstracts of studies that may be relevant to the other hormone levels).
review is reported in Appendix 1. • Major and minor adverse effects of interventions (coronary ischemia,
headache, flushing for PDE5i, and priapism for intracavernous in-
jections).
Types of Studies
• Treatment compliance (as reported by study authors).
All randomized controlled trials (RCTs) and quasi-RCTs of any treat-
• Number of participants who dropped out
ment (hormone therapy, PDE5i, intracavernous injections, intraurethral
pellets, mechanical devices, and behavioral therapy) for sexual dys-
Female sexual dysfunction outcomes
function in male and female patients with CKD were included. Trials
were considered without language restrictions.
• Changes in score on any standard validated sexual function scales
[Female Sexual Function Index (FSFI), Female Intervention Efficacy
Types of Participants Index (FIEI), and Female Sexual Distress Scale (FSDS)]
Patients aged ⬎18 years and with any stage of CKD, including • Variation of vaginal pressure-volume, genital threshold of percep-
patients who are not receiving renal replacement therapy (predialysis) tion of vibration, vaginal pH, genital blood flow, and prolactin and
and those with end-stage kidney disease who are receiving hemodial- zinc concentrations (units as reported by study authors)
ysis or peritoneal dialysis or have a functioning kidney transplant, were • Hormonal levels as measured by trialists [including luteinizing hor-
considered for inclusion. Studies that enrolled patients without CKD mone (LH), follicle-stimulating hormone (FSH), and prolactin levels
were excluded. or other]
• Levels of other markers (including zinc or other as reported by study
Types of Interventions authors)
All studies of pharmacologic and nonpharmacologic interventions • Number of participants who showed improved sexual function (as
for treating sexual dysfunction in patients with CKD were considered defined by study authors)
for inclusion. For treatment of male sexual dysfunction, we explored • Adverse effects [incidence of coronary ischemia for PDE5i and breast
pharmacologic and nonpharmacologic interventions. Pharmacologic cancer (ductal carcinoma in situ, lobular carcinoma in situ, invasive
agents included hormonal therapy [oral, injected, or topical (transder- ductal carcinoma, invasive lobular carcinoma)] for estrogen replace-
mal) testosterone] and drugs [oral (PDE5i, sildenafil, tadalafil, vard- ment therapy and headache as reported in the study
enafil, and mirodenafil) or topical (intracavernous injections of alpros- • Treatment compliance as defined by study authors
tadil, ␣1-antagonist, and intraurethral alprostadil, prazosin, or their • Number of participants who dropped out
Clin J Am Soc Nephrol 5: 985–995, 2010 Sexual Dysfunction in Chronic Kidney Disease 987

Data Collection Trial Characteristics

The titles and abstracts were screened independently by two review- Two groups of trials were identified: eight parallel trials
ers (M.V., S.D.N.) who discarded studies that were not applicable; (26 –33) and six crossover trials (5,34 –38).
however, studies and reviews that might include relevant data or Of the eight parallel trials, four (50 patients) compared ele-
information on trials were initially retained. The reviewers indepen- mentary zinc or zinc chloride to placebo (26,27,30,32). In three
dently assessed retrieved abstracts and the full text of these studies to of these trials, zinc chloride was added to the dialysis bath
determine which studies satisfied the inclusion criteria. The same re-
(26,27,32); in one trial elementary zinc was administered orally
viewers independently carried out data extraction using standard data
(30). Two trials (99 patients) compared vardenafil with placebo
extraction forms. Studies reported in non-English language journals
were translated before assessment. When more than one publication of
(28,33), and one trial (41 patients) compared sildenafil citrate
one trial existed, only the paper with the most complete data was with placebo (31). One trial (24 patients) compared vitamin E to
included. We only included the data from the first part of the crossover placebo (29). Of the remaining six crossover trials, three trials
trials for this analysis. Further information required from the original (59 patients) compared sildenafil citrate with placebo (5,34,37,).
author was requested by written correspondence and any relevant Two trials (40 patients) compared bromocriptine with placebo
information obtained was included in the review. Disagreements were (36,38), and one trial (15 patients) compared 1,25-dihydroxy-
resolved in consultation with G.F.M.S. cholecalciferol with placebo (35). Five trials included diabetic
patients. Of these, one enrolled patients with diabetic nephrop-
athy. Eleven trials included patients on hemodialysis, one in-
Study Quality
The quality of included studies was assessed independently by M.V. cluded patients on peritoneal dialysis, and the remaining two
and S.D.N. using a checklist that included allocation concealment; studies included renal transplant recipients. Other characteris-
blinding of participants, investigators, outcome assessors, and data tics of the included trials are detailed in Table 1.
analysts; use of intention-to treat-analysis; and completeness to follow-
up. Any discrepancy was resolved by discussion with G.F.M.S. Study Quality
By current methodological standards for reporting, trial qual-
ity was variable. Allocation concealment was adequate in only
Statistical Assessment
For dichotomous outcomes, adverse effects (coronary ischemia due 2 of 14 (14%) trials (30,38) and unclear in 12 of 14 (86%) trials
to PDE5i, priapism, or penile pain due to intraurethal injections, vagi- (27–30,32–38,39). Participants were blinded in 3 of 14 (21%)
nal itching due to vaginal cream, study withdrawal rate due to any trials (33,35,36), investigators were blinded in 1 of 14 (7%) trials
adverse effect) results were expressed as relative risk (RR) with 95% (26), participants and investigators were blinded in 8 of 14
confidence intervals (CI). Data were pooled using the random effects (57%) trials (5,27,29 –32,37–38), but outcome assessors were
model, but the fixed effects model was also analyzed to ensure robust- blinded in none of the trials. Two trials (14%) did not blind all
ness of the model chosen and susceptibility to outliers. Where contin- different groups (28,34). None of the 14 trials (0%) was ana-
uous scales of measurement were used to assess the effects of treatment lyzed on an intention-to-treat basis. The number of patients lost
(15-IIEF, 5-IIEF, FSFI, and FIEI scores; vaginal pH; genital blood flow;
to follow-up ranged from 0% to 36.0%.
variations of vaginal pressure-volume; genital threshold of perception
of vibration; and measurement of testosterone, LH, FSH, prolactin, and
zinc concentration), the mean difference (MD) was used, or the stan- Study Outcomes
dardized mean difference (SMD) if different scale units of measurement PDE5i versus Placebo. There was a consistent improve-
were used. Heterogeneity was analyzed using the ␹2 test on N ⫺ 1 ment in the overall score of the IIEF-5 with PDE5i compared
degrees of freedom (Cochran Q), with an alpha of 0.05 used for statis- with placebo (3 RCTs, 101 patients, MD 1.81, 95% CI 1.51 to
tical significance and the I2 statistic (25). If an adequate number of 2.10) (5,28,31), and a consistent increase of the score of all
studies were identified, we planned for subgroup analysis and metar- individual IIEF-5 tool domains: (1) erection frequency (3 RCTs,
egression to explore the effects of various covariates relating to patient, 149 patients, MD 1.54, 95% CI 1.14 to 1.93) (5,28,31), (2) erection
treatment, and study characteristics such as stage of kidney disease, quality (3 RCTs, 165 patients, MD 1.78, 95% CI 1.04 to 2.53)
treatment duration, and various study quality measures on prespeci-
(5,28,31), (3) penetration ability (3 RCTs, 165 patients, MD 1.70,
fied treatment outcomes. Analyses were performed using Revman 5
95% CI 1.16 to 2.24) (5,28,31), (4) maintenance frequency of
(The Cochrane Collaboration, United Kingdom, 2007).
penetration (4 RCTs, 193 patients, MD 1.60, 95% CI 1.02 to 2.18)
(5,28,31,34), (5) maintenance of erection after penetration (4
Results RCTs, 193 patients, MD 1.83, 95% CI 1.17 to 2.50) (5,28,31,34),
Search Results and (6) erection confidence (3 RCTs, 165 patients, MD 1.39, 95%
We identified 101 articles in MEDLINE, EMBASE, and in CI 0.84 to 1.95) (5,28,31) (Figure 2). When two crossover trials
CENTRAL. Seventy-six studies were excluded at the abstract were excluded from the analysis, no significant changes were
stage because they did not meet the inclusion criteria. Of the shown and the resulting effect remained the same for each
remaining 25 citations (full-text analysis), 9 studies were ex- domain of the overall IIEF-5 score.
cluded because they assessed treatment or outcomes that were There was significant heterogeneity noted in these analyses
not relevant to this review. Finally, 14 trials reported in 16 (Figure 2). These could be attributed to the differences in the
publications and enrolling a total of 328 patients were included patient characteristics such as stage of kidney disease and
in the review (Figure 1) (5,26 –38). Authors of these trials were baseline comorbid disease conditions, treatment characteristics
contacted for additional details, but none responded. such as treatment duration and dose of the agent used, and
988 Clinical Journal of the American Society of Nephrology Clin J Am Soc Nephrol 5: 985–995, 2010

Figure 1. Flow chart showing the number of citations retrieved by individual searches and number of trials included in the
systematic review.

study characteristics such as use of blinding and intention-to- Zinc versus Placebo
treat analysis. Because of the lack of an adequate number of Gonadotropins. There was no consistent increase in end-
studies, we could not conduct subgroup analysis or metar- of-treatment plasma testosterone concentration by addition of
egression as planned a priori. zinc to the dialysate (2 RCTs, 22 patients, SMD 0.19, 95% CI
We found a significant increase in the overall satisfaction ⫺2.12 to 2.50) (26,27), whereas oral zinc significantly improved
score of the IIEF-15 sexual assessment tool with sildenafil com- end-of-treatment plasma testosterone levels (1 RCT, 20 patients,
pared with placebo (1 RCT, 41 patients, MD 2.64, 95% CI 1.32 to SMD 1.62, 95% CI 0.58 to 2.66) (29). There was no significant
3.96), and a consistent improvement in erectile function (2 reduction in the end-of-treatment plasma FSH concentration (2
RCTs, 80 patients, MD 10.64, 95% CI 5.32 to 15.96), orgasmic RCTs, 28 patients, MD ⫺9.69 mIU/ml, 95% CI ⫺23.72 to 4.34)
function (1 RCT, 41 patients, MD 1.70, 95% CI 0.35 to 3.05), and (26,30) and plasma LH level (2 RCTs, 20 patients, MD 18.80
intercourse satisfaction (1 RCT, 41 patients, MD 1.71, 95% CI mIU/ml, 95% CI ⫺26.16 to 63.76) (26,30) with zinc compared
0.11 to 3.31) (31). with placebo.
Adverse Effects. There was no significant increase in the Sexual Function. One trial showed significant decrease in
risk of headache when PDE5i was compared with placebo (1 the frequencies of intercourse (1 RCT, 20 patients, RR 0.22, 95%
RCT, 41 patients, RR 1.05, 95% CI 0.16 to 6.76) (31). Included CI 0.06 to 0.78) and total or partial impotence (1 RCT, 20
studies did not report incidence of coronary ischemia for patients, RR 0.13, 95% CI 0.02 to 0.82) with zinc compared with
PDE5i, priapism for intracavernous injections, and breast can- placebo (30). There was no consistent variation in libido using
cer for estrogen replacement therapy. zinc compared with placebo (2 RCTs, 34 patients, RR 0.11, 95%
Table 1. Characteristics of studies and participants in the included studies
Population Characteristics Key Outcome Measures

Duration of Type of
Study Dialysis/ Renal Follow-Up
Study ID Age (years) Interventiona N
Design Transplant Percent Replacement (months) IIEF/ Testosterone LH/FSH
Mean NTP Otherb
(months) Diabetics Therapy IIEF-5 Level Level
(SD)/Range
Mean
(SD)/Range
Clin J Am Soc Nephrol 5: 985–995, 2010

Antoniou 1977 Parallel 48.5 (7.9) 49.1 (14.5) NA HD Zinc chloride in 8 3 to 4 x x

(26) dialysate
(400 ␮g/L)
Brook 1980 (27) Parallel 37.6 (8.2) 28.0 (21.2) NA HD Zinc chloride in 14 1.5 x x x x
dialysate
(400 ␮g/L)
Demir 2006 (28) Parallel 48 (7.4) 25 NA Txp Vardenafil 60 1 x x
(10/20 mg)
Yeksan 1992 (29) Parallel 38.4 (10.8) NA NA HD Vitamin E (300 mg) 24 2 x x x
Mahajan 1982 (30) Parallel 38.0 (7.0) 24.6 (18.5) 5.0 HD Elementary oral 20 6 x x x
zinc or zinc
acetate (25/50
mg)
Seibel 2002 (31) Parallel 49.0 (10.0) 42.0 (31.0) NA HD Sildenafil (50 mg) 41 1 x
Wabrek 1982 (32) Parallel 48.7 (8.3) 6.6 25.0c HD Zinc in dialysate 8 NA x x x
(400 mg/L)
Yang 2008 (33) Parallel NA NA 0.0 HD Vardenafil (10/20 39 1 x x
mg)
Bellovich 2000 (34) Crossover 52.4 NA 28.0 HD Sildenafil citrate (25 14 NA x x x
to 50 mg)
Blumberg 1980 Crossover 37.4 (10.9) 29.8 (26.1) 0.0 HD 1,25(OH)2D3 (0.25/ 15 2 to 4 x x x
(35) 0.5/1.5 ␮g/d)
Bommer 1979 (36) Crossover 28.0 to 50.0 26.0 to 86.0 NA HD Bromocriptine 15 8 x x
(5 mg/d)
Mahon 2005 (37) Crossover 55.6 (12.0) 6.0 to 60.0 46.0 PD Sildenafil citrate 13 1 x x
(50/100 mg)
Muir 1983 (38) Crossover 44.5 (10.2) 62.1 (26.2) NA HD Bromocriptine (until 25 3 x x
7.5 mg/day)
Sharma 2006 (5) Crossover 40.0 (8.0) 56.4 (36.0) 21.8 Txp Sildenafil (25/50/ 32 2 x x
100 mg)

NA, not available; Txp, transplant patients; HD, hemodialysis; PD, peritoneal dialysis; 1,25(OH)2D3, 1,25-dihydroxycholecalciferol.
a
All RCT were placebo controlled.
b
Other outcomes: prolactin level, other biochemical parameters, libido, frequency of intercourse, interview, global efficacy question, frequency and quality of
erection, increase of penile shaft.
c
Sexual Dysfunction in Chronic Kidney Disease

Insulin-dependent diabetes.
989
990 Clinical Journal of the American Society of Nephrology Clin J Am Soc Nephrol 5: 985–995, 2010

Figure 2. Effect of PDE5i on IIEF-5 scores in CKD patients.

CI 0.01 to 1.83) (27,30), nor a significant increase in episodes of patients, MD 7.00 pg/ml, 95% CI 4.43 to 9.57) after administra-
nocturnal penile tumescence (1 RCT, 7 patients, RR 0.75, 95% CI tion of vitamin E compared with placebo, but no statistically
0.07 to 7.73) (32). significant decrease in FSH levels (1 RCT, 24 patients, MD
⫺0.65 mU/ml, 95% CI ⫺2.61 to 1.31). No adverse effects were
Vitamin E versus Placebo reported in this study.
One trial (29) showed a consistent decrease in end-of-treat-
ment prolactin (1 RCT, 24 patients, MD ⫺41.23 ng/ml, 95% CI Other Outcomes
⫺50.42 to ⫺32.04), LH (1 RCT, 24 patients, MD ⫺6.77 mU/ml, Other outcomes reported in the included studies are summa-
95% CI ⫺10.15 to ⫺3.39), and testosterone levels (1 RCT, 24 rized in Table 2.
Clin J Am Soc Nephrol 5: 985–995, 2010 Sexual Dysfunction in Chronic Kidney Disease 991

Table 2. Other outcomes related to sexual dysfunction as reported in the included studies
Mean ⫾ SD Mean ⫾ SD
(treatment group or
Study ID N Intervention Outcome (control group or
baseline value) or RR after treatment)
(95% CI)

Antoniou 1977 (26) 8 Oral zinc versus Testosterone concentration 8.00 ⫾ 3.50 3.20 ⫾ 2.00
placebo (ng/ml)b
LH (mIU/ml)b 85.30 ⫾ 81.00 47.30 ⫾ 26.90
FSH (mIU/ml)b 24.00 ⫾ 24.30 33.00 ⫾ 8.70
Bellovich 2000 (34) 14 Sildenafil citrate IIEF
Frequency of penetration 3.85 ⫾ 3.10 4.43 ⫾ 2.92
Maintenance of erection 4.43 ⫾ 2.69 4.93 ⫾ 2.32
penetration
Brook 1980 (27) 14 Zinc chloride versus Improvement of libido 1.00 (0.08, 13.02) NA
placebo to Plasma testosterone 9.00 ⫾ 4.23 12.00 ⫾ 1.32
dialysate (nmol/L)a
Mahajan 1982 (30) 20 Oral zinc acetate Total/partial impotence 0.13 (0.02, 0.82) NA
versus placebo Decreased libido 0.11 (0.01, 1.83) NA
Decreased frequency of 0.22 (0.06, 0.78) NA
intercourse
Increased plasma 5.20 ⫾ 1.58 3.00 ⫾ 0.95
testosteroned
Decreased plasma FSHd 25.00 ⫾ 22.14 35.00 ⫾ 15.81
Decreased plasma LHd 49.00 ⫾ 82.22 38.00 ⫾ 25.3
Mahon 2005 (37) 13 Sildenafil citrate Global efficacy question 2.50 (1.05, 5.96) NA
versus placebo
Muir 1983 (38) 14 Bromocriptine Testosterone (nmol/L)b 16.80 ⫾ 4.49 17.00 ⫾ 4.11
versus placebo
Sharma 2006 (5) 32 Sildenafil Citrate Global efficacy question 4.33 (2.07, 9.08)
versus placebo Blood urea nitrogen (mg/ 18.3 ⫾ 7.6 17.9 ⫾ 51.0
dl)a
Creatinine (mg/dl)a 1.48 ⫾ 0.4 1.4 ⫾ 0.4
Hemoglobin (g/dl)a 12.3 ⫾ 1.5 13.2 ⫾ 1.4
Wabrek 1982 (32) 8 Oral zinc versus Tumescence episodes 0.75 (0.07, 7.73) NA
placebo
Yeksan 1992 (29) 24 Vitamin E versus Prolactin (ng/ml) 15.00 ⫾ 4.28 56.23 ⫾ 15.66
placeboc LH (mU/ml) 4.66 ⫾ 1.80 11.43 ⫾ 5.70
FSH (mU/ml) 4.23 ⫾ 1.83 4.88 ⫾ 2.94
Testosterone (pg/ml) 11.79 ⫾ 4.16 4.79 ⫾ 1.82
a
P value not significant.
b
Significance not reported.
c
Data pre- and post-vitamin E treatment only is reported.
d
P ⬍ 0.05.

Discussion bromocriptine, and dihydroxycholecalciferol in CKD patients

Key Findings and no trials assessed intracavernous injections, transurethral
Our systematic review demonstrated that in small clinical injections, mechanical devices, or behavioral therapy in CKD.
trials, PDE5i improved various aspects of erectile function in The safety and efficacy of interventions for sexual dysfunction
CKD patients. No data about safety of these agents have been in women with CKD were poorly studied.
reported in these studies. Oral zinc supplementation resulted in
a significant increase in plasma testosterone concentration Comparison with Other Studies
along with an increase in the potency and frequency of inter- PDE5i have been extensively studied in the general popula-
course. However, administration of zinc in the dialysate did not tion and have been generally shown to improve erectile re-
improve testosterone or the other biochemical parameters of sponse and to be well tolerated in men with mild to severe
gonadal failure. Only sparse data were available for vitamin E, erectile dysfunction due to varying etiologies (24,39 – 46). In a
992 Clinical Journal of the American Society of Nephrology Clin J Am Soc Nephrol 5: 985–995, 2010

recent systematic review of 130 mostly short-term (ⱕ12 weeks) we did not identify studies enrolling predialysis patients. Be-
RCTs of treatments for erectile dysfunction in men, Tsertsvadze cause the prevalence of sexual dysfunction remains high in
et al. (41) reported that PDE5i were significantly more effective predialysis CKD patients whereas the prevalence of cardiovas-
than placebo in improving sexual intercourse success (69% cular disease is lower than in dialysis patients, this group may
versus 35%) and resulted in a higher proportion of men with represent an opportunity to safely conduct clinical trials assess-
improved erections (range 67% to 89% versus 27% to 35%) in ing the safety and efficacy of PDE5i.
mixed study populations and in study populations of men with Our systematic review also identified important opportuni-
specific comorbid conditions, such as diabetes mellitus, stable ties for examining the effects of treatments for sexual dysfunc-
cardiovascular disease, hypertension, depression, multiple scle- tion on patient-level outcomes, such as quality of life and
rosis, rectal excision for bowel cancer, and radical prostatec- cardiovascular events. Previous studies have observed strong
tomy for prostate cancer. The magnitude of improvement in associations between erectile dysfunction, depression, and ad-
erectile function was comparable between sildenafil, vard- verse cardiovascular outcomes (8,17–19,50), although a causal
enafil, and tadalafil. Balanced against these benefits, PDE5i link has not been definitively established (18). Of the trials
were associated with an increased risk of any adverse event (RR included in our meta-analysis, none considered these patient-
1.72, 95% CI 1.53 to 1.93), the most common of which were level outcome parameters.
headaches, flushing, dyspepsia, myalgia, and back pain. Al-
though the reporting of all serious adverse or cardiovascular Strengths and Weaknesses
adverse events was inconsistent and incomplete, the overall Our review had several strengths and weaknesses. The
rate of serious adverse events in men randomly assigned to strengths included systematic searches of medical databases,
PDE5i was ⱕ2% and comparable to those assigned to placebo. data extraction, analysis, and trial quality assessment by two
There was insufficient evidence to determine whether treat- independent reviewers. The key findings were limited by the
ment with PDE5i increased the risks of serious cardiovascular lack of long-term studies analyzing interventions targeting
events or nonarteritic anterior ischemic optic neuropathy. On erectile dysfunction/sexual dysfunction in CKD patients. The
the basis of these findings, the American College of Physicians included studies had relatively small sample sizes and were
issued clinical practice guidelines strongly recommending powered to observe differences in surrogate end points rather
PDE5i for men who seek treatment for erectile dysfunction and than patient-focused outcomes. Five studies had a crossover
who do not have contraindications to PDE5i use (47). design and most did not adequately report study methods to
In keeping with the findings of Tsertsvadze et al. in non-CKD determine trial quality. Although it is plausible that treatment
populations, we found that administration of PDE5i to men effects of various agents might differ depending on the stage of
with CKD caused clinically meaningful and statistically signif- the kidney disease, on the basis of the current available data, it
icant improvements in general sexual satisfaction and erectile is unclear whether such effects truly exist. Further, significant
dysfunction. However, despite the high rate of sexual dysfunc- heterogeneity was observed for many outcomes but further
tion in CKD patients and concerns about the safety of pharma- analysis to explore the reasons for this heterogeneity could not
cologic treatments in the setting of renal impairment, we found be conducted because of the small number of studies. Publica-
only six small clinical trials (including comparative and cross- tion bias might exist; however, given the lack of adequate
over trials) that assessed PDE5i in CKD. The longest study number of studies, formal tests could not be conducted. In the
duration was 8 months. Comparison of the efficacy of different general population, a higher PDE5i dose provides a better
PDE5i was not possible because only limited data were avail- response, but whether such effects existed in CKD was unclear.
able for sildenafil and vardenafil and no data were available for In short, these issues highlight the fact that treatment of sexual
tadalafil or mirodenafil. Unfortunately, there was also a com- dysfunction in CKD has received inadequate attention by re-
plete lack of safety data for PDE5i in CKD patients, a popula- searchers to date.
tion that is at high risk for silent cardiovascular disease.
Similar to the general population, our review did not identify Conclusions
any clinical trial analyzing the safety and efficacy of PDE5i in Implications for Practice
female CKD patients, despite the ubiquitous occurrence of sex- Modest evidence exists for the efficacy of PDE5i in CKD
ual dysfunction in this group. Biologic plausibility exists to patients. The safety profile of these agents has not been exten-
support the use of PDE5i in female patients with sexual dys- sively analyzed in CKD. Clinicians may use PDE5i in CKD
function (48), but efficacy and safety are unclear and there is no patients who do not have any contraindications for PDE5i use.
consensus on the best treatment options for sexual dysfunction Oral zinc seems to increase testosterone levels and improve
in female patients (49). Further studies in this important clinical sexual dysfunction, but this evidence needs to be confirmed in
area are warranted. future larger trials.
Our review found some earlier studies supporting the use of
oral zinc therapy in CKD to improve sexual dysfunction. These Implications for Research
were short-term investigations that assessed the effect of zinc Given the high prevalence of sexual dysfunction in CKD and
on surrogate end points, such as gonadal hormone levels. With lack of clinical trials, further and larger trials exploring various
the declining use of zinc in clinical practice, further studies treatment options in male and female CKD patients are needed.
were not conducted. Most trials enrolled dialysis patients, and These studies should focus on biochemical and patient-cen-
Clin J Am Soc Nephrol 5: 985–995, 2010 Sexual Dysfunction in Chronic Kidney Disease 993

tered end points along with establishing their safety profile in 14. dyspareunia.tw.
dialysis and predialysis patients. Comparative studies of the 15. orgasm$.tw.
efficacy and safety of different PDE5i in CKD patients are also 16. or/1 to 15
warranted. 17. exp Renal Replacement Therapy/
18. (hemodialysis or hemodialysis).tw.
Appendix 1: Search Strategy 19. (hemofiltration or hemofiltration).tw.
We searched MEDLINE using the following search strategy: 20. (hemodiafiltration or hemodiafiltration).tw.
1. exp Sexual Dysfunction, Physiologic/ 21. dialysis.tw.
2. exp Sexual Dysfunctions, Psychologic/ 22. (PD or CAPD or CCPD or APD).tw.
3. Orgasm/ 23. Kidney Disease/
4. sexual dysfunction$.tw. 24. Chronic Kidney Disease/
5. sex$ disorder$.tw. 25. Kidney Failure/
6. frigid$.tw. 26. Chronic Kidney Failure/
7. (erectile$ adj (disorder$ or dysfunction$)).tw. 27. Uremia/
8. (sexual adj (arousal or aversion$)).tw. 28. (chronic kidney or chronic renal).tw.
9. fsfi.tw. 29. (CKF or CKD or CRF or CRD).tw.
10. Female Sexual Function Index.tw. 30. (end-stage renal or end-stage kidney or endstage renal or
11. International Index of Erectile Function.tw. endstage kidney).tw.
12. iief.tw. 31. (ESRF or ESKF or ESRD or ESKD).tw.
13. Female Sexual Distress Score.tw. 32. ur?emi$.tw.
14. FSDS.tw. 33. exp Kidney Transplantation/
15. (impotent or impotence).tw. 34. or/17 to 33
16. dyspareunia.tw. 35. and/16,34
17. orgasm$.tw.
18. or/1 to 17 Acknowledgments
19. exp Renal Replacement Therapy/ The authors thank Narelle Willis (Review Group Coordinator of the
20. (hemodialysis or hemodialysis).tw. Cochrane Renal Group), Gail Higgins and Ruth Mitchell (Trials Search
21. dialysis.tw. Coordinators of the Cochrane Renal Group), and the Cochrane Renal
22. (PD or CAPD or CCPD or APD).tw. Group for assistance with preparation of this study.
23. Renal Insufficiency/
24. Kidney Failure/ Disclosures
25. exp Renal Insufficiency, Chronic/ None.
26. Kidney Diseases/
27. Uremia/
28. (end stage renal or end stage kidney or endstage renal or References
endstage kidney).tw. 1. Laumann EO, Paik A, Rosen RC. Sexual dysfunction in the
29. (ESRF or ESKF or ESRD or ESKD).tw. United States. Prevalence and predictors. JAMA 281: 6,
30. (chronic kidney or chronic renal).tw. 1999
31. (CKF or CKD or CRF or CRD).tw. 2. Finkelstein F, Shirani S, Wuerth D, Finkelstein SH: Therapy
32. ur?emi$.tw. insight: Sexual dysfunction in patients with chronic kidney
disease. Nat Clin Pract Nephrol 3: 200 –207, 2007
33. or/19 to 32
3. Rosas S, Joffe M, Franklin E, Strom BL, Kotzker W, Brens-
34. and/18,33
inger C, Grossman E, Glasser D, Feldman H: Prevalence
We searched EMBASE using the following search strategy: and determinants of erectile dysfunction in hemodialysis
patients. Kidney Int 59: 2259 –2266, 2001
1. exp Sexual Dysfunction/ 4. Anantharaman P, Schmidt RJ: Sexual function in chronic
2. sexual dysfunction$.tw. kidney disease. Adv Chronic Kid Dis 14: 119 –125, 2007
3. sex$ disorder$.tw. 5. Sharma RK, Prasad N, Gupta A, Kapoor R: Treatment of
4. frigid$.tw. erectile dysfunction with sildenafil citrate in renal allograft
5. (erectile$ adj (disorder$ or dysfunction$)).tw. recipients: A randomized double-blind, placebo-con-
6. (sexual adj (arousal or aversion$)).tw. trolled, cross-over trial. Am J Kidney Dis 48: 128 –133, 2006
6. Procci WR, Goldstein DA, Adelstein J, Massry SG: Sexual
7. fsfi.tw.
dysfunction in the male patients with uremia: A reap-
8. Female Sexual Distress Score.tw.
praisal. Kidney Int 19: 317–323, 1981
9. International Index of Erectile Function.tw. 7. Bellinghieri G, Santoro D, Mallamace A, Savica V: Sexual
10. iief.tw. dysfunction in chronic renal failure. J Nephrol 21: s113–
11. Female Sexual Function Index.tw. s117, 2008
12. fsds.tw. 8. Peng YS, Chiang CK, Kao TW, Hung KY, Lu CS, Chiang
13. (impotent or impotence).tw. SS, Yang CS, Huang YC, Wu KD, Wu MS, Lien YR, Yang
994 Clinical Journal of the American Society of Nephrology Clin J Am Soc Nephrol 5: 985–995, 2010

CC, Tsai DM, Chen PY, Liao CS, Tsai TJ, Chen WY: Sexual 27. Brook AC, Johnston DG, Ward MK, Watson MJ, Cook DB,
dysfunction in female haemodialysis patients: A multi- Kerr DN: Absence of a therapeutic effect of zinc in the
center study. Kidney Int 68: 760 –765, 2005 sexual dysfunction of haemodialysed patients. Lancet 2:
9. Palmer BF: Sexual dysfunction in uremia. J Am Soc Nephrol 618 – 620, 1980
10: 1381–1388, 1999 28. Demir E, Balal M, Paydas S, Sertdemir Y, Erken U: Efficacy
10. Lawrence IG, Price DE, Howlett TA, Harris KP, Feehally J, and safety of vardenafil in renal transplant recipients with
Walls J: Erythropoietin and sexual dysfunction. Nephrol erectile dysfunction. Transplant Proc 38: 1379 –1381, 2006
Dial Transplant 12: 741–747, 1997 29. Yeksan M, Polat M, Türk S, Kazanci H, Akhan G, Erdogan
11. Kutner NG: Quality of life and daily haemodialysis. Sem Y, Erkul I: Effect of vitamin E therapy on sexual functions
Dial 17: 92–98, 2004 of uremic patients in hemodialysis. Int J Artif Organs 15:
12. Kimmel P, Peterson RA, Weihs KL, Immense SJ, Boyle WO, 648 – 652, 1992
Kovac JA, Alleyne S, Cruz I, Veis JH. Psychologic function- 30. Mahajan SK, Abbasi AA, Prasad AS, Rabbani P, Briggs
ing, quality of life, and behavioural compliance in patients WA, McDonald FD: Effect of oral zinc therapy on gonadal
beginning haemodialysis. J Am Soc Nephrol 7: 2152–2159, function in hemodialysis patients. A double-blind study.
1996 Ann Intern Med 97: 357–361, 1982
13. Campese VM. Autonomic nervous system dysfunction in 31. Seibel I, Poli de Figueiredo CE, Telöken C, Moraes JF:
uraemia. Nephrol Dial Transplant 5[Suppl 1]: s98 –s101, 1990 Efficacy of oral sildenafil in hemodialysis patients with
14. Naya Y, Soh J, Ochiai A, Ushijima S, Kamoi K, Ukimura O, erectile dysfunction. J Am Soc Nephrol 13: 2770 –2775, 2002
Kawauchi A, Fujito A, Ono T, Iwamoto N, Aoki T, Imada 32. Wabrek AJ: Zinc: A possible role in the reversal of uremic
N, Marumo M, Miki T: Significant decrease of the interna- impotence. Sexuality e Disability 5: 213–221, 1982
tional index of erectile function in male renal failure pa- 33. Yang J, Ju W, Zeng FQ, Xiao YJ, Zhang XP, Xiao CG:
tients treated with haemodialysis. Int J Impot Res 14: 172– Efficacy and safety of vardenafil for kidney transplant
177, 2002 recipients with erectile dysfunction. Zhonghua Nan Ke Xue
15. Mehrsai A, Mousavi S, Nikoobakht M, Khanlarpoor T, 14: 911–913, 2008
Shekarpour L, Pourmand G: Improvement of erectile dys- 34. Bellovich K, Provenzano R, Wankhede S. Effectiveness of
function after kidney transplantation: The role of the asso- sildenafil citrate in male hemodialysis patients. J Am Soc
ciated factors. Urol J 3: 240 –244, 2006 Nephrol 11: 139A, 2000
16. Al Khallaf HH. Analysis of sexual functions in male non- 35. Blumberg A, Wildbolz A, Descoeudres C, Hennes U, Dam-
diabetic hemodialysis patients and renal transplant recip- bacher MA, Fischer JA, Weidmann P: Influence of 1,25
ients. Transpl Int. 23: 176 –181, 2009 dihydroxycholecalciferol on sexual dysfunction and re-
17. Seidman SN, Roose SP: The sexual effects of testosterone lated endocrine parameters in patients on maintenance
replacement in depressed men: Randomized, placebo-con- hemodialysis. Clin Nephrol 13: 208 –214, 1980
trolled clinical trial. J Sex Marital Ther 32: 267–273, 2006 36. Bommer J, Ritz E, del Pozo E, Bommer G: Improved sexual
18. Turk S, Guney I, Altintepe L, Tonbul Z, Yildiz A, Yeksan function in male haemodialysis patients on bromocriptine.
M: Quality of life in male haemodialysis patients. Nephron Lancet 2: 496 – 497, 1979
Clin Pract 96: c21– c27, 2004 37. Mahon A, Sidhu PS, Muir G, Macdougall IC: The efficacy
19. Goldestein I. The mutually reinforcing triad of depressive of sildenafil for the treatment of erectile dysfunction in
symptoms, cardiovascular disease, and erectile dysfunc- male peritoneal dialysis patients. Am J Kidney Dis 45: 381–
tion. Am J Cardiol 86[Suppl]: 41F– 45F, 2000 387, 2005
20. Padma-Nathan H, Stecher VJ, Sweeney M, Orazem J, 38. Muir JW, Besser GM, Edwards CR, Rees LH, Cattell WR,
Tseng LJ, Deriesthal H: Minimal time to successful inter- Ackrill P, Baker LR: Bromocriptine improves reduced li-
course after sildenafil citrate: Results of a randomized, bido and potency in men receiving maintenance hemodi-
double-blind, placebo-controlled trial. Urology 62: 400 – 403, alysis. Clin Nephrol 20: 308 –314, 1983
2003 39. Keating GM, Scott LJ: Vardenafil: A review of its use in
21. Linet O, Ogrinc F. Efficacy and safety of intracavernosa erectile dysfunction. Drugs 63: 2673–2703, 2003
alprostadil in men with erectile dysfunction. N Engl J Med 40. Markou S, Perimenis P, Gyftopoulos K, Athanasopoulos A,
334: 873– 877, 1996 Barbalias G: Vardenafil (Levitra) for erectile dysfunction: A
22. Esposito K, Giugliano F, Di Palo C, Giugliano G, Marfella systematic review and meta-analysis of clinical trial re-
R, D’Andrea F, D’Armiento M, Giugliano D: Effect of ports. Int J Impot Res 16: 470 – 478, 2004
lifestyle changes on erectile dysfunctions in obese men: A 41. Tsertsvadze A, Fink HA, Yazdi F, Macdonald R, Bella AJ,
randomized controlled trial. JAMA 291: 2978 –2984, 2004 Ansari MT, Garritty C, Soares-Weiser K, Daniel R, Samp-
23. Melnik T, Garcia Oliveira Soares B, Nasello AG. Psychos- son M, Fox S, Moher D, Wilt TJ. Oral phosphodiesterase-5
ocial interventions for erectile dysfunction. In: Cochrane inhibitors and hormonal treatments for erectile dysfunc-
Database of Systematic Review 2007, Issue 3, CD004825 tion: A systematic review and meta-analysis. Ann Intern
24. Vardi M, Nono A. Phosphodiesterase inhibitors for erectile Med 151: 650 – 661, 2009
dysfunction in patients with diabetes mellitus. In: Cochrane 42. Miles CL, Candy B, Jones L, Williams R, Tookman A, King
Database of Systematic Review 2007, Issue 1, CD002187 M. Interventions for sexual dysfunction following treat-
25. Higgins JP, Thompson SG: Quantifying heterogeneity in a ments for cancer. In: Cochrane Database of Systematic Review
meta-analysis. Stats Med 21: 1539 –1558, 2002 2007, Issue 4, CD005540
26. Antoniou LD, Shalhoub RJ, Sudhakar T, Smith JC Jr: Re- 43. Porst H, Rosen R, Padma-Nathan H, Goldstein I, Giuliano
versal of uraemic impotence by zinc. Lancet 2: 895– 898, F, Ulbrich E, Bandel T: The efficacy and tolerability of
1977 vardenafil, a new phosphodiesterase type 5 inhibitor, in
Clin J Am Soc Nephrol 5: 985–995, 2010 Sexual Dysfunction in Chronic Kidney Disease 995

patients with erectile dysfunction: The first at home clinical Assessment Subcommittee of the American College of
trial. Int J Impot Res 13: 192–199, 2001 Physicians. Hormonal testing and pharmacologic treat-
44. Carson CC, Rajfer J, Eardley I, Carrier S, Denne JS, Walker ment of erectile dysfunction: A clinical practice guide-
DJ, Shen W, Cordell WH: The efficacy and safety of line from the American College of Physicians. Ann Intern
tadalafil: An update. BJU Int 93: 1276 –1281, 2004 Med 151: 639 – 649, 2009
45. Hellstrom WJ, Gittelman M, Karlin G, Segerson T, Thibon- 48. D’Amati G, di Gioia CR, Bologna M, Giordano D, Giorgi
nier M, Taylor T, Padma-Nathan H: Vardenafil for treat- M, Dolci S, Jannini EA. Type 5 phosphodiesterase expres-
ment of men with erectile dysfunction: Efficacy and safety sion in the human vagina. Urology 60: 191–195, 2002
in a randomized, double-blind, placebo-controlled trial. J 49. Brown DA, Kyle JA, Ferrill MJ. Assessing the clinical effi-
Androl 23: 763–771, 2002 cacy of sildenafil for the treatment of female sexual dys-
46. Aranda P, Ruilope LM, Calvo C, Luque M, Coca A, Gil de function. Ann Pharmacother 43: 1275–1285, 2009
Miguel A: Erectile dysfunction in essential arterial hyper- 50. Peng YS, Chiang CK, Hung KY, Chiang SS,Lu CS, Yang CS,
tension and effects of sildenafil: Results of a Spanish na- Wu KD, Yang CC, Lin RP, Chang CJ, Tsai TJ, Chen WY:
tional study. Am J Hypertens 17: 139 –145, 2004 The association of higher depressive symptoms and sexual
47. Qaseem A, Snow V, Denberg TD, Casey DE Jr, Forciea dysfunction in male haemodialysis patients. Nephrol Dial
MA, Owens DK, Shekelle P; for the Clinical Efficacy Transplant 22: 857– 861, 2007