PubMed Health. A service of the National Library of Medicine, National Institutes of Health.

Tsertsvadze A, Yazdi F, Fink HA, et al. Diagnosis and Treatment of Erectile Dysfunction. Rockville (MD): Agency for Healthcare Research and Quality (US); 2009 May. (Evidence Reports/Technology Assessments, No. 171.)

2Methods

Key Questions Addressed in This Report

The UO-EPC's evidence report on the diagnosis and treatment of erectile dysfunction (ED) is based on a systematic review of the scientific literature. A technical expert panel was recruited to help refine key questions and provide expertise to the review team during the review process.

The finalized key questions were:

KQ 1. To determine the clinical utility of routine blood tests - testosterone, prolactin, luteinizing hormone (LH), or follicle-stimulating hormone (FSH) - in identifying and treating specific hormonal causes of ED.

KQ 2. To determine the benefits of pharmaceutical treatments for patients with ED.

KQ 3. To determine the harms of pharmaceutical treatments for patients with ED.

The secondary objectives of this evidence report were:

KQ 2-a. To explore how patient-specific characteristics (e.g. specific symptoms/age/comorbid conditions) may affect prognosis and treatment success for ED patients.

KQ 2-b. To determine if the likelihood of treatment success varies by underlying cause of ED.

KQ 3-a. To identify specific harms—nonarteritic ischemic optic neuropathy (NAION), penile fibrosis—of pharmaceutical treatments in patients with ED.

Analytic Framework

Search Strategy

A preliminary MEDLINE® (1966-January Week 3 2006) search identified systematic reviews and guidelines in erectile dysfunction published between 1990 and 2006. Searches for diagnostic and efficacy studies were undertaken in the following databases: MEDLINE® (1966-July Week 3 2006, updated to May Week 5 2007); EMBASE (1980-2006 Week 29, updated to 2007 Week 2); Cochrane CENTRAL (1st Quarter 2006 and 2nd Quarter 2007); PsycINFO® (1985-January 2006, updated to June Week 1 2007); AMED (1985-January 2006, updated to June 2007); and Scopus Feb 8 2006. All databases were searched for efficacy; MEDLINE® and EMBASE were searched for diagnostic studies. The searches were limited to English publications from 1990 and later. MEDLINE® (1966-August Week 5 2006, updated to May Week 5 2007) and EMBASE (1980-2007 Week 8, updated to 2007 Week 22) were searched for reports of visual problems and sleep apnea associated with the use of sildenafil. MEDLINE® (1950-September Week 1 2007) and EMBASE (1980-2007 Week 37) were searched for reports regarding fibrosis associated with penile injections. Search strategies are presented in Appendix A.

Study Eligibility Criteria and Selection Process

KQ 1. The clinical utility of routine blood teststestosterone, prolactin, LH, FSH - in identifying and affecting therapeutic outcomes for treatable causes of ED was examined using reports of measurements of serum testosterone, FSH, LH, prolactin, and/or other hormone levels, (but not gonadotrophin-releasing hormone [GnRH], Inhibin, Activin, or Follistim). It was also examined in reports of the prevalence of reversible hormonal disorders in males with erectile dysfunction. The study selection criteria included the following:

Source: Primary study report published in English

Study design: Any (prevalence studies)

Population: Adults (age ≥ 18 years) diagnosed with ED with or without concurrent endocrinopathy (i.e., hypogonadism, hyperprolactinemia, abnormal levels of LH/FSH)

Intervention (experimental): Hormonal blood tests (i.e., testosterone/prolactin/LH/FSH)

Outcomes: Prevalence of endocrinopathies (i.e., hypogonadism, hyperprolactinemia, abnormal levels of LH/FSH)

KQ 2. Benefits of pharmaceutical treatments (e.g. oral, injections, hormonal, topical, intra-urethral suppositories) in males with ED. To address how patient specific characteristics (e.g. specific symptoms/origin, duration, severity of ED/comorbid conditions) affect prognosis/treatment success for ED patients. Evidence on the following treatment modalities was excluded from this review: Natural health products (e.g. herbals), yohimbine, vacuum constriction devices, and sex or surgical therapies (e.g. penile prosthesis implantation, penile arterial reconstructive surgery). Study selection criteria included the following:

Source: Primary study report published in English

Study design: RCTs (comparative efficacy and harms studies)

Population: Adults (age => 18 years) diagnosed with ED (with or without comorbidities)

Interventions (experimental/control): Oral (PDE-5 inhibitors, sublingual) injections (IC, cream)

Outcomes: Clinically relevant efficacy measures (i.e., scores for the IIEF “EF” domain, IIEF-Q3/Q4, SEP-Q2/Q3, GAQ-Q1, EDITS)

KQ 3. Harms of pharmaceutical treatments (e.g. oral, injections, hormonal, topical, intra-urethral suppositories) in males with ED. Evidence on the following treatment modalities was excluded from this review: Natural health products (e.g. herbals), yohimbine, vacuum constriction devices, and sex or surgical therapies (e.g. penile prosthesis implantation, penile arterial reconstructive surgery). Study selection criteria included the following:

Source: Primary study report published in English

Study design: RCTs (comparative efficacy and harms studies)

Population: Adults (age ≥ 18 years) diagnosed with ED (with or without comorbidities)

Interventions (experimental/control): Oral (PDE-5 inhibitors, sublingual) injections (IC, SC), hormonal (e.g. testosterone), intra-urethral suppositories, CPAP, and/or topical (e.g. patch, cream)

Outcomes: Any adverse events, serious adverse events, withdrawals due to adverse events, and specific adverse events.

KQ 3a. The incidence of specific harms such as Nonarteritic Anterior Ischemic Optic Neuropathy (NAION) and penile fibrosis associated with use of PDE-5 inhibitor and injection therapies, respectively. The review included reports of non-RCTs or observational studies. For identification of data on fibrosis related to use of injection therapies, only studies with at least 6 months of followup were included. Study selection criteria included the following:

Source: Primary study report published in English

Study design: Non-RCTs (experimental or observational case-control and cohort studies, case reports and case-series)

Population: Adults (age ≥ 18 years) diagnosed with ED (with or without comorbidities)

Interventions (experimental/control): Oral (PDE-5 inhibitors), injections (IC, SC)

Outcomes: NAION, penile fibrosis

Systematic and narrative reviews, case reports, editorials, commentaries or letters to the editor were excluded for all questions except Q3-a (specific harms). Studies evaluating interventions such as penile implant devices or natural health products used for the treatment of ED were also excluded.

The results of the literature search were uploaded to the software program TrialStat SRS version 4.0 along with screening questions developed by the review team and any supplemental instructions. A calibration exercise was undertaken to pilot and refine the screening process. One reviewer screened bibliographic records (i.e., title, authors, key words, abstract) using broad screening criteria (Appendix B). All potentially relevant records and those records that did not contain enough information to determine eligibility (e.g. no abstract was available) were retained. The reasons for exclusion are noted in the QUOROM flow diagram (Figure 2). Two reviewers independently performed full-text relevance screening. Disagreements were resolved by consensus. Reasons for exclusion were noted (Appendix E).

Figure 1. Analytic Framework for the Diagnosis and Treatment of Erectile Dysfunction.

Figure 1

Analytic Framework for the Diagnosis and Treatment of Erectile Dysfunction.

Figure 2. Modified QUOROM Flow Chart.

Figure 2

Modified QUOROM Flow Chart.

Relevant studies were then evaluated to determine study design and were categorized accordingly for inclusion by question. The level of eligible evidence on efficacy was limited to RCTs, since systematic bias is minimized in RCTs compared with all other study designs (e.g. cross-sectional, retrospective cohort).

Data Abstraction

Two reviewers independently abstracted relevant information from each included study using a data abstraction form developed a priori for this review (Appendix B). One reviewer completed primary extraction, which was then verified by a second reviewer. Conflicts were discussed and resolved by consensus. Abstracted data included study characteristics (e.g. design, sample size, country), population characteristics (e.g. age, comorbidities, severity of ED), name/type of treatment (e.g. sildenafil, testosterone), route of administration (e.g. oral, injection, topical), dose, and the duration of treatment. The following clinically relevant and validated efficacy outcomes were abstracted: absolute endpoint/change (from baseline) in scores for the International Index of Erectile Function “Erectile Function” domain (IIEF- EF), per-patient percentage on Sexual Encounter Profile for Q2 and Q3 (SEP-Q2/Q3), and the proportion of patients with improved erection measured with a Global Assessment (or Efficacy) Question (GAQ-Q1 or GEQ-Q1).(Appendix H)

For harms, reviewers abstracted information on any adverse events: i.e., number of patients who developed at least one adverse event; most frequently encountered specific adverse events; withdrawals due to adverse events; and the incidence of serious adverse events. Additionally, for Q1, prevalence estimates of hypogonadism and hyperprolactinemia in ED populations for each included study were abstracted.

Assessment of Study and Reporting Quality

The quality of prevalence studies measuring serum hormonal levels in ED patients was assessed using a subset of QUADAS items.46 The QUADAS tool consists of 14 items (Appendix B). QUADAS was designed to evaluate the diagnostic accuracy of a test against the reference standard. Since the included studies for this review involved measurements of serum hormone levels, no reference standards were used to assess the diagnostic accuracy of these tests (i.e., levels of testosterone, prolactin, LH and/or FSH). Therefore, the quality assessment of studies was based on a subset of 8 QUADAS items (score range 1–5) that were deemed to be relevant to the present research question.

The Jadad scale was used to assess the methodological and reporting quality of RCTs. (Appendix B). 47 This instrument is designed to assess the reporting of methods used to generate random assignments and double blinding, as well as to determine whether there is a description of dropouts and withdrawals by treatment group (i.e., number and reasons). The scoring ranges from 0 to 5, with higher scores indicating higher quality. An a priori threshold scheme was used for sensitivity analysis: a Jadad total score of ≥3 indicated studies of higher quality. In addition, the adequacy of allocation concealment was assessed using an approach proposed by Schulz and colleagues as: adequate, inadequate, or unclear (Appendix B).48

Synthesis of Evidence

Qualitative data synthesis. Primary and secondary outcomes were summarized qualitatively for each study. The sample size and demographics, setting, funding source, treatment and comparator characteristics (e.g. type, dose, and duration), study quality, and methods of adjustment for confounders (where applicable) were recorded and summarized in the text, and summary tables.

To determine the clinical utility of routine hormonal blood tests in identifying and affecting therapeutic outcomes for endocrine causes of ED (KQ 1), the reviewers identified relevant studies and synthesized data for two following constructs:

1.

The prevalence of hormonal abnormalities (hypogonadism, hyperprolactinemia, abnormal levels of luteinizing and/or follicle-stimulating hormones) in patients with ED

2.

The efficacy of hormonal therapies in patients with the above-mentioned hormonal abnormalities for improving clinical symptoms of ED.

The two constructs (i.e., prevalence of hormonal abnormalities and efficacy of available hormonal treatments) jointly determine the clinical utility of routine hormonal blood tests. For example, the administration of routine hormonal blood tests might be justified only if the prevalence of hormonal abnormalities in patients with ED was relatively high (i.e., above a pre-specified threshold) and the available hormonal therapies in affecting symptoms of ED in this subgroup of patients were effective.

Thus, the results for KQ 1 are presented in two sub-sections: 1) the prevalence of hormonal abnormalities in ED patients and 2) the efficacy of hormonal therapy in treating ED in patients with hormonal abnormalities (see also the section for KQ 2–3, Hormonal Treatments, for more detailed description of the studies).

Quantitative synthesis. The decision whether to perform statistical pooling of individual studies was based on clinical and methodological judgment. In the case of outcomes for which meta-analysis was deemed appropriate, we extracted quantitative data (e.g. number of subjects in each group, mean, standard deviation) from reports using a standardized data extraction form that included intervention characteristics and outcome variables at baseline and followup intervals.

If relevant data (e.g. standard deviations) were not reported adequately, we attempted to calculate the needed parameters. Trials that did not report complete numerical information for relevant efficacy/harms outcomes (i.e., arm-specific mean endpoint or change in score, standard deviation, or standard error, proportion of patients with an outcome at followup) could not be incorporated in the meta-analyses. Trial reports presenting measures of variability (e.g. standard deviation) only graphically (i.e., no numerical data were available) were not pooled. Crossover trials not reporting numerical data from the pre-crossover phase were not included in meta-analyses

We calculated standard deviations from standard errors or 95 percent confidence intervals.

For continuous outcomes (e.g. mean endpoint/change in the total score of IIEF), the absolute difference between treatment-specific means and corresponding standard deviations were ascertained for each individual study. A generic inverse variance method was used to calculate the response outcomes and corresponding 95 percent confidence intervals for the combined treatment groups.

For dichotomous outcomes (e.g. improvement in erection GAQ), studies were grouped by type of treatment and dose to minimize clinical heterogeneity. The intent-to-treat group or number enrolled at the time of study was used for analyses and, when this information was unavailable, we used the number provided in the report. Pooled relative risks with corresponding 95 percent confidence intervals were generated.

The DerSimonian and Laird random-effects model was used to obtain combined estimates across the studies.49 The degree of statistical heterogeneity was evaluated by using a chi-square test and the I2 statistic.5052 An I2 of less than 25 percent is consistent with low heterogeneity; 25 to 50 percent with moderate heterogeneity; and over 50 percent with high heterogeneity.52 When statistically significant heterogeneity was identified, it was explored through subgroup and sensitivity analyses when appropriate. Sources of heterogeneity include reporting and methodological quality (e.g. methods for randomization, adequacy of allocation concealment, blinding, washout period for crossover trials, data analysis) as well as clinical heterogeneity (e.g. study population, dosing of therapeutic agent, duration of followup). Estimates from the heterogeneous groups must be interpreted with caution, especially when small numbers of trials are included.

We also performed a series of subgroup analyses to explore the consistency of the results.

The meta-analyses are presented as forest plots (Figures 3–76). Publication bias was explored through funnel plots (Figures D1–16, Appendix D) by plotting the relative measures of effect (relative risk) versus a measure of precision of the estimate (1/standard error).51 The visual asymmetry in funnel plots maybe be suggestive of publication bias, although other potential causes for asymmetry exist. The degree of funnel plot asymmetry was measured using the Egger regression test.5355

Figure 3. The mean IIEF “EF domain” score.

Figure 3

The mean IIEF “EF domain” score.

The statistical analyses in this review were performed using Review Manager 4.2 (The Cochrane Collaboration, Oxford, UK, 2006).

PubReader format: click here to try
Cover of Diagnosis and Treatment of Erectile Dysfunction
Diagnosis and Treatment of Erectile Dysfunction.
Evidence Reports/Technology Assessments, No. 171.
Tsertsvadze A, Yazdi F, Fink HA, et al.

PubMed Health Blog...

read all...

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...