Erectile dysfunction (ED) is a common sexual dysfunction in men, which is defined as the failure to achieve or maintain an erection for satisfactory sexual intercourse (1, 2). The incidence of ED is age-related, and more than 20% of men aged over 40 years were reported to have ED (3). Besides aging, diabetes mellitus (DM) has also been recognized as a major risk factor for ED (4). A previous meta-analysis including 145 studies showed that the overall prevalence of ED in men with DM was 52.5% (5). Accordingly, screening and prevention of ED in men with DM has gained great attention (6).

In recent decades, the intensive care for patients with hyperglycemia has introduced the concept of prediabetes. Currently, prediabetes is defined as an intermediate metabolic state between normoglycemia and DM, which includes impaired fasting glucose (IFG), impaired glucose tolerance (IGT), and mildly raised glycated hemoglobin (HbA1c) (7, 8). Although IGT is consistently defined as a 2 hour plasma glucose concentration of 7.8-11.0 mmol/L during an oral glucose tolerance test, the definitions of IFG are varying according to the World Health Organization (WHO) criteria (fasting plasma glucose [FPG]: 6.1 to 6.9 mmol/L) and the 2003 American Diabetes Association (ADA) guideline criteria (FPG: 5.6-6.9 mmol/L) (9). Moreover, the glycosylated hemoglobin (HbA1c) of 5.7-6.4% and 6.0-6.4% has also been considered as definitions for prediabetes by ADA and National Institute for Health and Care Excellence (NICE) respectively (10, 11). Similar to DM, prediabetes is suggested to be associated with increased risks of cardiovascular diseases, over cancer events, and all-cause mortality (12, 13). However, previous studies evaluating the association between prediabetes and ED showed inconsistent results (14–22). Some studies suggested that prediabetes was associated with higher prevalence of ED as compared to normoglycemia (15, 16, 19, 20, 22), while others did not (14, 17, 18, 21). In this study, we performed a meta-analysis of cohort studies to systematically evaluate the association between prediabetes and ED. The potential influence of different definitions of prediabetes on the association was also explored.

The meta-analysis was performed in accordance with the MOOSE (Meta-analysis of Observational Studies in Epidemiology) (23) and Cochrane’s Handbook (24) guidelines.

In this meta-analysis, by pooling the results of nine observational studies controlling for age, we found that compared to men with normoglycemia, those with prediabetes were associated with higher prevalence of ED. Further subgroup analysis showed that mean age of the male participants in each study may significantly affect the results, which may substantially contribute to the heterogeneity of the included studies. Specifically, the association between prediabetes and ED seemed to be stronger in studies with younger men (mean age < 50 years) than that in studies with older men (mean age ≥ 50 years). In addition, subgroup analyses showed that difference in definitions of prediabetes, diagnostic methods for ED, and controlling of additional variables besides age did not significantly affect the association. Taken together, these results showed that prediabetes is associated with higher prevalence of ED, which is independent of age and may be stronger in younger men.

To the best of our knowledge, this study is the first meta-analysis which summarized the association between prediabetes and ED. The study has several strengths which should be mentioned for better interpretation of the results. First, since advanced aging has been related to both hyperglycemia (29) and ED (30), it is important to know whether the association between prediabetes and ED could be confounded by aging. Accordingly, only studies matched or adjusted for age was included into the meta-analysis, and the results of the meta-analysis indicated that the association between prediabetes and ED was independent of age of the males. In addition, the robustness of the finding was further evaluated by the “leave-one-out” sensitivity analysis, which showed that the overall result of the meta-analysis was not primarily driven by either of the included studies. Moreover, several predefined subgroup analyses were performed to evaluate the influences of study characteristics on the association. Results showed that the association between prediabetes and ED was consistent and independent of definition of prediabetes, study design, methods for diagnosis of ED, and with or without controlling for variables besides age. It has been confirmed that many factors are involved in the pathogenesis of ED, such as cardiometabolic risk factors besides hyperglycemia (31), central weight gain (32), insulin resistance (33), inflammatory cytokines and leptin (34), and low testosterone (35), all of which may confound the possible relationship between prediabetes and ED. Accordingly, large-scale prospective stuides with adequate adjustment of the above factors are needed to determine the possible independent association between prediabetes and ED.

Interestingly, results of the subgroup analysis showed that the association between prediabetes and ED seemed to be stronger in studies with younger men than that with older men. The potential reasons for the results remain unknown. However, compared to older male participants, younger males were less likely to have various comorbidities that were associated with pathogenesis of ED, such as metabolic syndrome, hypertension, obesity, coronary artery disease, and multiple concurrent medications (36). Accordingly, compared to older male participants who usually have more comorbidity, the adverse influence of prediabetes and hyperglycemia on the pathogenesis and deterioration of ED may be more remarkable in younger men. This is important because ED has traditionally been considered a disease of old age; however, contemporary evidence suggests a growing incidence of ED in men younger (36). Indeed, a previous systematic review of large multinational studies has estimated the prevalence of ED in young men to be as high as 30% (36). Another study showed that age at first presentation for ED significantly decreased over the past decade, which highlighted the importance of careful assessment of ED even at younger age groups (37). Our study expanded these findings by showing that prediabetes in younger men may at risk for ED. From the perspective of prevention, these results highlight the possible importance of screening ED in men with prediabetes, particularly in young male population with prediabetes.

The potential pathophysiological basis for the association between prediabetes and ED may be similar to that between DM and ED. Persistent hyperglycemia is associated with systematic inflammation and endothelial dysfunction, which represents the common denominator leading to vascular ED (38, 39). Besides, hyperglycemia is associated with oxidative stress and autonomic and peripheral neuropathies, both of which were involved in the pathogenesis of ED via impairment of sensory impulses from the penis and parasympathetic activity for relaxation of the smooth muscle of the corpus cavernosum (40, 41). Moreover, prediabetes has also been related with reduced testosterone level, which has been recognized as a hormonal risk factor for ED (42). Similarly, prediabetes is shown to be associated with an increased risk of testosterone deficiency, independent of obesity and other cardiometabolic factors (43). Accordingly, screening for low testosterone may be considered in people with prediabetes and symptoms of ED and should be mandatory if ED is diagnosed. Interestingly, a recent study showed that patients with undiagnosed prediabetes were associated with lower rates of response to phosphodiesterase type 5 inhibitors (PDE5i) than normoglycemic men. These findings suggest that even milder forms of glucose impairment are associated with a worse efficacy of PDE5i in men with ED. Future studies are warranted to determine the exact and specific mechanisms underlying the association between prediabetes and ED.

Our study also has limitations. Firstly, all of the included observational studies were cross-sectional, and no longitudinal data was available for the association between prediabetes and incidence of ED. Accordingly, results of the meta-analysis should be validated in large-scale prospective cohort studies. Moreover, the meta-analysis was based on data from study level rather than data of individual patient. Therefore, results of subgroup analysis should be interpreted with caution. In addition, although age was controlled for all of the included studies, we could not exclude the possibility that other factors may confound the association between prediabetes and ED. In this regard, large-scale prospective cohort studies with careful adjustment of possible confounding factors are needed. Finally, a causative relationship between prediabetes and ED should be derived based on the finding of the meta-analysis because this is a meta-analysis of observational studies.

Taken together, results of the meta-analysis showed that prediabetes is associated with higher prevalence of ED, which may be independent of age of the males and may be stronger in young men. Although these findings should be validated in large-scale prospective cohort studies, these results may highlight the possible importance of screening ED in men with prediabetes, particularly in young male population with prediabetes.

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author.

MJ, SY and CW designed the study. MJ and SY performed literature search, data extraction, and quality evaluation. MJ, BW, and LY performed statistical analyses and interpreted the results. MJ wrote the manuscript. All authors reviewed and revised the manuscript, and approved the manuscript for submission.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.