Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
J Urol. Author manuscript; available in PMC 2011 November 18.
Published in final edited form as:
PMCID: PMC3220602


Hunter Wessells, M.D.,1 David F. Penson, M.D.,2 Patricia Cleary, M.S.,3 Brandy N. Rutledge, Ph.D.,3 John M. Lachin, Sc.D.,3 Kevin T. McVary, M.D.,4 David S. Schade, M.D.,5 Aruna V. Sarma, Ph.D,6 and the Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications Research Group*



To determine whether intensive glycemic therapy reduces the risk of erectile dysfunction (ED) in men with type 1 diabetes enrolled in the Diabetes Control and Complications Trial (DCCT).


DCCT randomized 761 males with type 1 diabetes to intensive or conventional glycemic therapy in 28 sites between 1983–1989, of whom 366 had diabetes for 1–5 years and no microvascular complications (primary prevention cohort) and 395 for 1–15 years with non-proliferative retinopathy or microablbuminuria (secondary intervention cohort). Subjects were treated until 1993 and followed in the Epidemiology of Diabetes Interventions and Complications (EDIC) study. In 2003, we conducted an ancillary study using a validated assessment of ED in 571 men (80% participation rate); 291 in the primary cohort and 280 in the secondary cohort.


Twenty-three percent of participants reported ED. The prevalence was significantly lower in the intensive versus conventional treatment group in the secondary cohort (12.8% versus 30.8%, p=0.001); but not the primary cohort (17% versus 20.3%, p=0.49). The risk of ED in both primary and secondary cohorts was directly associated with mean HbA1c during DCCT and EDIC combined. Age, peripheral neuropathy, and lower urinary tract symptoms were other risk factors.


A period of intensive therapy significantly reduced the prevalence of ED ten years later among those in the secondary intervention cohort, but not the primary prevention cohort; higher HbA1c was significantly associated with risk in both cohorts. These findings provide further support for early implementation of intensive insulin therapy in young men with type 1 diabetes.

Keywords: Glycemic Control, Diabetes, Erectile Dysfunction, Risk


One in five men over 20 years of age in the United States report ED. Of all co-morbid medical conditions, diabetes mellitus imparts the greatest risk of ED and its onset is associated with a reduction in health related quality of life1 and an increase in depressive symptoms.2 With diabetes, ED begins 10–15 years earlier than in the general population3 and is less responsive to oral pharmacological therapy.4 Although its pathogenesis is not completely understood, ED in diabetes is associated with peripheral neuropathy, nephropathy and retinopathy.5, 6

Longitudinal and cross-sectional studies have demonstrated an association between glycemic control and ED in men with diabetes;710 those with “poor” control having 2–5 fold increased risk of ED compared to those with “good” control. One small clinical trial, in men with type 2 diabetes and ED, showed reductions in HbA1c and blood pressure led to improvements in erectile function.11 However, no controlled clinical study has shown that the risk of ED could be lowered by reductions in glycemia.

The DCCT, a randomized controlled trial comparing conventional and intensive therapy for glycemic control in type 1 diabetes (T1DM), and its observational follow-up, the EDIC Study, convincingly demonstrated that intensive therapy aimed at near normal levels of glycemic control reduces the risk of retinopathy, nephropathy, neuropathy, and cardiovascular disease. 12 As part of the UroEDIC ancillary study of urologic complications, we assessed erectile function to determine whether intensive glycemic therapy during DCCT, and improved glycemic control during DCCT/EDIC, lowered the risk of prevalent ED ten years after the close of the DCCT.



The DCCT randomly assigned 761 males with T1DM to intensive or conventional therapy, with a mean of 6.5 years of treatment during 1983–1993.13 The 378 primary prevention cohort males had no retinopathy or nephropathy and 1–5 years of diabetes. The 383 secondary intervention cohort males had mild to moderate non-proliferative retinopathy, microalbuminuria and 1–15 years of diabetes. Individuals with hypertension, symptomatic ischemic heart disease, or symptomatic peripheral neuropathy requiring treatment were excluded from enrollment in DCCT at baseline.

Figure 1 shows the flow of participants through the trial. Of the 761 men enrolled, 746 completed DCCT closeout in 1993. Of these, 720 (97%) enrolled in EDIC in 1994.14 Of the 713 men still active in EDIC year 10, 591 (83%) agreed to participate in UroEDIC. Of these, 571 (80%) men provided data on erectile function and comprise the study cohort for the current analyses. The institutional review board approved the study, and the Federal Government issued a Certificate of Confidentiality.

Figure 1
Flow of male participants through DCCT and EDIC to UroEDIC showing time line of enrollment and number of participants* unavailable owing to inactive study status or death. ED=Erectile Dysfunction

DCCT Intervention and Other Therapies

Intensive therapy was aimed at achieving near normal levels of HbA1c using multiple daily insulin injections or continuous subcutaneous insulin infusion with dose adjustments based on frequent self-monitoring of glucose. Conventional therapy was aimed at maintaining clinical well-being with 1–2 daily insulin injections with no glucose targets.

At the end of DCCT, conventional group subjects were trained in intensive therapy and all subjects were returned to their own physicians for diabetes management with the recommendation to implement intensive therapy. By EDIC year 10, 97% of intensive and 94% of conventional group subjects were implementing intensive therapy, and mean HbA1c in the two groups had equalized.15, 16

Assessment of Erectile Dysfunction

We administered the International Index of Erectile Function (IIEF), a widely used reliable, validated multi-dimensional self report instrument for the evaluation of male sexual dysfunction, at EDIC year 10.17 Because 20% of participants responded “Did not attempt intercourse” on questions 1–5 of the Erectile Function (EF) domain, we created a proxy item to assess erectile function in the entire cohort regardless of sexual activity and presence or absence of a partner. IIEF Question 15 asks participants: “Over the past 4 weeks, how would you rate your confidence that you get and keep your erection?” Those who answered ‘Very Low’ or ‘Low’ were classified as having ED. Those who answered ‘Moderate’, ’High’, or ‘Very High’ were classified as not having ED. Among the men who engaged in sexual intercourse during the preceding four weeks, this definition of ED correlated strongly with EF domain scores (r=0.77, p<0.001) and ED bother (r= 0.80, p<0.001).

Measurement of Other Factors

HbA1c was measured at baseline, quarterly during DCCT, and annually in EDIC. Therefore, mean HbA1c over DCCT/EDIC weighted each DCCT value by 0.25, and EDIC value by 1.16

Peripheral neuropathy was defined during the DCCT by the presence of distal symmetrical polyneuropathy and an abnormal nerve conduction study; or during EDIC by >6 positive responses to the Michigan Neuropathy Screening Instrument (MNSI) or a score >2 on the examination.18 Retinopathy was assessed using fundus photographs that were centrally graded using the Early Treatment Diabetic Retinopathy Study (ETDRS) scale.19 Albumin excretion rate (AER) was measured in half of the cohort annually. Nephropathy was defined as microalbuminuria (AER 40–300 mg/24hr) or albuminuria (AER >300 mg/24hr).

Subclinical atherosclerosis was assessed by the sum of the standardized common and internal carotid artery intima-media thickness on centrally graded carotid ultrasonography20 at EDIC years 1 and 6; and by the presence of coronary artery calcification on centrally graded computed tomography conducted during EDIC year 8.21 Lipid profiles were measured in half of the cohort annually.

We used the American Urological Association Symptom Index (AUASI) to assess urinary symptoms of nocturia, frequency, urgency, weak urinary stream, intermittency, straining, and the sensation of incomplete emptying.22 Lower urinary tract symptoms (LUTS), were defined as an AUASI ≥ 8. Other drug treatments were assessed during DCCT/EDIC with a yearly drug inventory.

Statistical Methods

The Wilcoxon rank-sum test assessed differences between groups in quantitative variables, and the contingency Chi-square test, or Fisher Exact test as appropriate, assessed categorical variables. Effects nominally significant at p ≤ 0.05 are cited. All analyses were performed using SAS version 8.2 (SAS Institute, Cary, NC).

A multivariable logistic regression model estimated the associations between several covariates and the presence of ED at EDIC year 10 using odds ratios (OR) and 95% confidence intervals. Probabilities of ED by treatment group were calculated separately for primary and secondary cohorts after adjustment for age, HbA1c levels at DCCT eligibility and duration of diabetes. Odds of ED as a function of HbA1c levels were estimated from separate multivariable logistic regression models using the log of DCCT mean HbA1c and time-weighted DCCT/EDIC mean HbA1c nested within primary and secondary cohort and adjusted for age, HbA1c levels at DCCT eligibility, duration of diabetes and duration of treatment/follow-up in the DCCT. The treatment group by cohort interaction was also adjusted for other covariates in a model that included a cohort by covariate interaction, i.e. allowing for different covariate effects within each cohort.

A separate multivariable logistic regression model examined the association of clinical risk factors with ED in the combined (intensive plus conventional) cohort. Variables nominally significant at the p≤0.05 level in bivariate analyses were included in the model. Throughout, nominal p-values are presented with no adjustment for multiple tests of significance.


Figure 1 shows the history of the 761 males who entered the DCCT. Of these, 735 survived and were eligible to participate in the UroEDIC assessment at the EDIC year 10 visit, 713 attended the EDIC year 10 visit, 591 (84.5%) completed the UroEDIC questionnaire, and 571 answered the erectile function question. Of these, 23% fulfilled the criteria for ED (i.e. reported “Very Low” or “Low” confidence).

Characteristics of the 571 men in UroEDIC (Supplemental Table) revealed that at DCCT baseline, diabetes duration was significantly shorter in the conventional vs. intensive treatment group (p=0.03); at EDIC year 10, the proportion of participants with nephropathy, retinopathy, peripheral neuropathy, and hypertension were all lower in the intensive treatment group, reflecting the previously described benefits of intensive therapy.13, 15 Sildenafil citrate use was reported by 30 subjects (12 with ED and 18 without), representing about 5% of those in each treatment group and 9% of those reporting ED.

DCCT Treatment and Erectile Dysfunction

More conventional treatment group participants expressed “Very low” (17.5%) or “Low” (8.9%) confidence in erection than in the intensive treatment group (11.1% and 8.6%, respectively). Overall, ED was present in 26.5% and 19.6% of the former conventional and intensive groups, respectively.

Table 1 presents the effects of DCCT intensive versus conventional treatment, separately within the primary and secondary cohorts, on the prevalence of ED after adjustment for age, HbA1c at DCCT eligibility and duration of diabetes. Within the primary cohort, DCCT treatment group had no significant effect (OR=1.24, 95%CI=0.68, 2.28), with an adjusted prevalence of 20.3% versus 17% in the conventional versus intensive treatment group. Conversely, in the secondary cohort, DCCT treatment group had a substantial effect (OR=0.33, 95%CI=0.18, 0.60) with an adjusted prevalence of 30.8% versus 12.8% in the conventional versus intensive group.

Table 1
Multivariable logistic regression model* examining the difference between DCCT intensive versus conventional glycemic therapy in the prevalence of erectile dysfunction at EDIC year 10 follow-up after trial completion separately within the primary prevention ...

The treatment group effect in the primary cohort differed significantly from that in the secondary cohort (p=0.003). Increasing age and increased DCCT baseline HbA1c also had strong effects on the prevalence of ED. Results were unchanged in analyses including all 30 sildenafil users as having ED, or excluding them entirely.

HbA1c Measures and Erectile Dysfunction

DCCT mean HbA1c levels were substantially higher among those with ED versus those without in the secondary cohort, but with an attenuated difference in the primary cohort. (Table 2) For every 10% higher mean DCCT HbA1c level (e.g. 8.8% versus 8%), the adjusted odds of ED increased by 55% (p<0.0001) in the secondary cohort, and by 21.5% (p=0.04) in the primary cohort. The difference in the HbA1c effect between cohorts approached significance (p=0.07). For every 10% increase in DCCT/EDIC mean HbA1c, the odds of ED increased by 97% in the secondary cohort versus 74% in the primary cohort, p<0.0001 for each.

Table 2
Mean HbA1c levels by erectile dysfunction status at EDIC year 10 by primary and secondary cohort and adjusted odds of erectile dysfunction per 10% higher HbA1c levels.

Erectile Dysfunction Risk Factors at EDIC Year 10

Table 3 compares clinical characteristics of the 132 men with ED to the 439 without ED. In unadjusted analyses, a number of factors were nominally associated with ED including age, DCCT therapy group, measures of HbA1c, presence of microvascular complications, cardiovascular factors, LUTS, and cigarette smoking, among others.

Table 3
Distribution of clinical characteristics by erectile dysfunction status at EDIC year 10 follow-up after trial completion.

In adjusted analyses, increasing age had the strongest effect on prevalence of ED (χ2=29, p<0.0001, 13% greater odds per year of age, 95% CI=8.5, 20%), followed by DCCT/EDIC mean HbA1c (χ2=14.7, p=0.0002, 60% greater odds per 10% higher mean HbA1c, 95% CI=25, 105%). While retinopathy and nephropathy were not significant after adjusting for mean HbA1c, there remained a higher adjusted proportion with ED among those with peripheral neuropathy and LUTS versus not (33 vs. 17%, p=0.02) and (31 vs. 19%, p=0.04), respectively.


In the largest most extensive prospective study of T1DM in the world, men randomized to intensive glycemic therapy during DCCT had a significantly lower prevalence of subsequent ED in the secondary intervention cohort with limited microvascular complications at baseline. Although no difference was noted between treatment groups in the primary prevention cohort of men without diabetic complications at baseline, in both cohorts the risk of ED increased with increasing levels of HbA1c over the average of 17.5 years of the DCCT and EDIC combined. HbA1c at eligibility for DCCT, at the very outset of the trial, also influenced the odds of subsequent ED. Thus, cumulative glycemic exposure over a long period of time is required before diabetes-related ED becomes clinically apparent. To place the findings in context, a mean HbA1c of 8.8% over the course of DCCT and EDIC imparted a 1.5 to 2 fold greater odds of ED compared to a mean HbA1c of 8.0%. This confirms other reports of increased ED risk associated with poor glycemic control.710 The DCCT/EDIC provides the novel additional insight that, in the secondary cohort, lowering of HbA1c via intensive glycemic control reduced the odds of ED by 63%.

An important question is why we only saw the effect of intensive glycemic control in the secondary cohort. The primary cohort contained more adolescents, shorter diabetes duration, lower baseline HbA1c, shorter treatment duration, lower age and diabetes duration at EDIC year 10, and a lower prevalence of ED. Nevertheless, analyses accounting for all of these factors in a cohort by covariate interaction model failed to explain the diminished treatment group difference in prevalence of ED within the primary cohort.

The accumulation of advanced glycation end products (AGEs), a consequence of long-term hyperglycemia, is thought to mediate many diabetic complications.23 AGEs have also been implicated in impairment of neurogenic and endothelium dependent relaxation of the corpus cavernosum.2426 In the case of other diabetic complications, including retinopathy and nephropathy, the term “metabolic memory” has been used to explain the persistent protective effect of intensive therapy well beyond the time frame of the DCCT.1216 Although intensive therapy during DCCT did reduce the incidence of hypertension, the cohort overall is similar to T1D patients in general with a point prevalence of hypertension in excess of 45% at year 10 of EDIC. An important future question is whether the reduced risk of ED associated with intensive therapy in the secondary cohort will persist.

Even though T1DM greatly increases the risk of ED, age is an independent risk factor for ED in the DCCT/EDIC, and in the general population.1, 27 Peripheral neuropathy and LUTS were the only diabetic complications associated with ED in adjusted analyses. It is possible these complications are surrogate markers of autonomic neural dysfunction, an important proposed mechanism of diabetes-associated ED.

There are several limitations to this study. First, 25% of the original cohort failed to participate in the UroEDIC study due to losses-to-follow-up over time and non-response. Second, the absence of data on serum testosterone is a potential limitation of the study, although hypogonadism is not prevalent in T1DM28 and its contribution to ED remains controversial.29 Third, the use of a single item as opposed to the full EF domain of the IIEF may have reduced our ability to detect milder degrees of ED severity, as might our dichotomous definition of ED that excluded “moderate” responses. Patients who used sildenafil were counted as having ED in the primary analyses only if they reported “very low” or “low” confidence in erection. This also may have underestimated ED prevalence in the study population. However, the results of the study were unchanged when all subjects using sildenafil were counted as having ED, or if they were excluded from the analysis. Finally, ED was assessed at a single point in time in the history of diabetes for these subjects. Characterization of baseline ED status and assessment over a period of years to describe the evolution of ED and the growth (or waning) of treatment group effects would be optimal.30


Men with diabetes who received intensive glycemic therapy during the DCCT had a significantly lower prevalence of ED ten years later in EDIC, principally among those enrolled into the secondary intervention cohort. The prevalence of ED was also strongly associated with the levels of HbA1c during the DCCT and later in EDIC. These results from an extremely well characterized longitudinal cohort study provide further strong support for early implementation of intensive therapy in young men with established T1DM and limited microvascular complications, for whom tight glycemic control could reduce the burden of ED.

Supplementary Material

Supplemental Table


Characteristics of the study population at DCCT baseline and at EDIC year 10 follow-up after trial completion by intensive vs. conventional glycemic therapy


A complete list of participants in the DCCT/EDIC research group can be found in Archives of Ophthalmology, 2008;126(12):1713.

Contributors of free or discounted supplies and/or equipment: Lifescan, Roche, Aventis, Eli Lilly, OmniPod, Can-Am, B-D, Animas, Medtronic, Medtronic Minimed, Bayer (donation one time in 2008), Omron.

Funding/Support: The DCCT/EDIC project is supported by contracts with the Division of Diabetes, Endocrinology and Metabolic Diseases of the National Institute of Diabetes and Digestive and Kidney Diseases, National Eye Institute, National Institute of Neurological Disorders and Stroke, the General Clinical Research Centers Program and the Clinical and Translational Science Awards Program, National Center for Research Resources, and by Genentech through a Cooperative Research and Development Agreement with the National Institute of Diabetes and Digestive and Kidney Diseases.


1. Saigal CS, Wessells H, Pace J, et al. Predictors and prevalence of erectile dysfunction in a racially diverse population. Arch Intern Med. 2006;166:207. [PubMed]
2. De Berardis G, Franciosi M, Belfiglio M, et al. Erectile dysfunction and quality of life in type 2 diabetic patients: a serious problem too often overlooked. Diabetes Care. 2002;25:284. [PubMed]
3. Costabile RA. Optimizing treatment for diabetes mellitus induced erectile dysfunction. J Urol. 2003;170:S35. [PubMed]
4. Fonseca V, Seftel A, Denne J, et al. Impact of diabetes mellitus on the severity of erectile dysfunction and response to treatment: analysis of data from tadalafil clinical trials. Diabetologia. 2004;47:1914. [PubMed]
5. Brown JS, Wessells H, Chancellor MB, et al. Urologic complications of diabetes. Diabetes Care. 2005;28:177. [PubMed]
6. Chitaley K. Type 1 and Type 2 diabetic-erectile dysfunction: same diagnosis (ICD-9), different disease? J Sex Med. 2009;6 (Suppl 3):262. [PubMed]
7. Kalter-Leibovici O, Wainstein J, Ziv A, et al. Clinical, socioeconomic, and lifestyle parameters associated with erectile dysfunction among diabetic men. Diabetes Care. 2005;28:1739. [PubMed]
8. Klein R, Klein BE, Moss SE. Ten-year incidence of self-reported erectile dysfunction in people with long-term type 1 diabetes. J Diabetes Complications. 2005;19:35. [PubMed]
9. McCulloch DK, Campbell IW, Wu FC, et al. The prevalence of diabetic impotence. Diabetologia. 1980;18:279. [PubMed]
10. Romeo JH, Seftel AD, Madhun ZT, et al. Sexual function in men with diabetes type 2: association with glycemic control. J Urol. 2000;163:788. [PubMed]
11. Khatana SA, Taveira TH, Miner MM, et al. Does cardiovascular risk reduction alleviate erectile dysfunction in men with type II diabetes mellitus? Int J Impot Res. 2008;20:501. [PubMed]
12. Nathan DM, Zinman B, Cleary PA, et al. Modern-day clinical course of type 1 diabetes mellitus after 30 years’ duration: the diabetes control and complications trial/epidemiology of diabetes interventions and complications and Pittsburgh epidemiology of diabetes complications experience (1983–2005) Arch Intern Med. 2009;169:1307. [PMC free article] [PubMed]
13. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. The Diabetes Control and Complications Trial Research Group. N Engl J Med. 1993;329:977. [PubMed]
14. Epidemiology of Diabetes Interventions Complications (EDIC) Design implementation and preliminary results of a long-term follow-up of the Diabetes Control and Complications Trial cohort. Diabetes Care. 1999;22:99. [PMC free article] [PubMed]
15. Effect of intensive therapy on the microvascular complications of type 1 diabetes mellitus. JAMA. 2002;287:2563. [PMC free article] [PubMed]
16. Nathan DM, Cleary PA, Backlund JY, et al. Intensive diabetes treatment and cardiovascular disease in patients with type 1 diabetes. N Engl J Med. 2005;353:2643. [PMC free article] [PubMed]
17. Rosen RC, Riley A, Wagner G, et al. The international index of erectile function (IIEF): a multidimensional scale for assessment of erectile dysfunction. Urology. 1997;49:822. [PubMed]
18. Feldman EL, Stevens MJ, Thomas PK, et al. A practical two-step quantitative clinical and electrophysiological assessment for the diagnosis and staging of diabetic neuropathy. Diabetes Care. 1994;17:1281. [PubMed]
19. Early Treatment Diabetic Retinopathy Study design and baseline patient characteristics. ETDRS report number 7. Ophthalmology. 1991;98:741. [PubMed]
20. Nathan DM, Lachin J, Cleary P, et al. Intensive diabetes therapy and carotid intima-media thickness in type 1 diabetes mellitus. N Engl J Med. 2003;348:2294. [PMC free article] [PubMed]
21. Cleary PA, Orchard TJ, Genuth S, et al. The effect of intensive glycemic treatment on coronary artery calcification in type 1 diabetic participants of the Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications (DCCT/EDIC) Study. Diabetes. 2006;55:3556. [PMC free article] [PubMed]
22. Barry MJ, Fowler FJ, Jr, O’Leary MP, et al. The American Urological Association symptom index for benign prostatic hyperplasia. The Measurement Committee of the American Urological Association. J Urol. 1992;148:1549. [PubMed]
23. Brownlee M. Negative consequences of glycation. Metabolism. 2000;49:9. [PubMed]
24. Cartledge JJ, Eardley I, Morrison JF. Advanced glycation end-products are responsible for the impairment of corpus cavernosal smooth muscle relaxation seen in diabetes. BJU Int. 2001;87:402. [PubMed]
25. Musicki B, Kramer MF, Becker RE, et al. Inactivation of phosphorylated endothelial nitric oxide synthase (Ser-1177) by O-GlcNAc in diabetes-associated erectile dysfunction. Proc Natl Acad Sci U S A. 2005;102:11870. [PubMed]
26. Seftel AD, Vaziri ND, Ni Z, et al. Advanced glycation end products in human penis: elevation in diabetic tissue, site of deposition, and possible effect through iNOS or eNOS. Urology. 1997;50:1016. [PubMed]
27. Bacon CG, Mittleman MA, Kawachi I, et al. Sexual function in men older than 50 years of age: results from the health professionals follow-up study. Ann Intern Med. 2003;139:161.28. [PubMed]
28. Tomar R, Dhindsa S, Chaudhuri A, et al. Contrasting testosterone concentrations in type 1 and type 2 diabetes. Diabetes Care. 2006;29:1120. [PubMed]
29. Kupelian V, Shabsigh R, Travison TG, et al. Is there a relationship between sex hormones and erectile dysfunction? Results from the Massachusetts Male Aging Study. J Urol. 2006;176:2584. [PubMed]
30. Travison TG, Shabsigh R, Araujo AB, et al. The natural progression and remission of erectile dysfunction: results from the Massachusetts Male Aging Study. J Urol. 2007;177:241. [PubMed]