The correlation between erectile function and adiponectin levels in men with late-onset hypogonadism

Abstract Aim: This study investigated the relationship between erectile dysfunction (ED) and adiponectin levels in hypogonadal men. Methods: In this study, 218 patients with hypogonadism (mean age: 65.1 ± 8.3 years) were enrolled. All patients underwent physical examinations, with measurement of body mass index, body fat ratio, and waist circumference. The erectile function was assessed using the sexual health inventory for men (SHIM) scoring system. Blood biochemical profiles such as free testosterone, fasting blood glucose, and lipid profile including adiponectin levels were measured. All patients were divided into two groups based on their SHIM score: normal to moderate ED (SHIM score ≥ 12) and severe ED (SHIM score < 12), and the factors associated with severe ED were determined. Patients with severe ED were divided into two groups based on adiponectin levels (cutoff value of 7.0 μg/mL), and their basic characteristics were compared between these two groups. Results: The severe ED group was older and had higher adiponectin levels. In patients with severe ED, various metabolic parameters were significantly worse in the low adiponectin groups than in the non-low adiponectin group. Conclusions: The risk of developing cardiovascular diseases is extremely high in hypogonadal men with severe ED who had lower serum adiponectin levels.


Introduction
Erectile dysfunction (ED) in men over the age of 50 is often organic, and it is well known that its development is significantly associated with a variety of lifestyle-related disorders such as hypertension, diabetes mellitus (DM), dyslipidemia, obesity, and metabolic syndrome [1,2]. These metabolic parameters are key elements that can damage systemic vascular endothelial cells and are also risk factors for the development of cardiovascular diseases (CVDs). Therefore, ED and CVDs share the same risk factors, and ED in elderly men has been identified as a predictive sign of systemic vascular endothelial dysfunction and a potential predictor of CVDs [3][4][5].
After the age of 40, serum testosterone levels in men decline by 1-2% per year [6]. Testosterone deficiency is linked to ED either directly or indirectly through various metabolic characteristics such as obesity, hypertension, dyslipidemia, and DM [7,8]. Testosterone deficiency is also a significant risk factor for CVDs in elderly men.
Adiponectin is a protein first discovered by Scherer et al. in 1995 and is encoded by a gene that is highly expressed in adipose tissues [9]. It improves insulin sensitivity in the liver and muscles, as well as vascular endothelial cell function, and has anti-inflammatory properties [10][11][12][13]. High serum adiponectin levels have been reported to be associated with a lower risk of CVDs such as coronary artery disease and myocardial infarction in the general population [14,15]. Specifically, elderly men with low testosterone levels and ED often have numerous metabolic variables that are related to decreased adiponectin. However, the relationship between ED and adiponectin has currently received little attention.
In routine clinical practice, many urologists frequently encounter hypogonadal patients with ED who are likely to be at high risk for CVDs. This studyinvestigated the relationship between ED and adiponectin levels in hypogonadal men. Additionally, we aimed to provide insight into therapeutic intervention in the ED based on serum adiponectin levels.

Study subjects
In this retrospective study, from a total of 328 patients diagnosed with hypogonadism at our hospital between 2008 and 2015, 218 cases for which all data are available were enrolled in the analysis. Hypogonadism was diagnosed according to the Japanese criteria, with a serum-free testosterone (FT) level of less than 11.8 pg/mL [16]. The FT levels were measured in serum samples collected at 09:00 and 11:00 using radioimmunoassay with the DPC Free Testosterone kit (Mitsubishi Kagaku, Inc., Tokyo, Japan).
At the initial visit, all patients had their current medical histories checked, including hypertension, DM, and dyslipidemia, and underwent physical examinations, including measurement of body mass index (BMI) and waist circumference. The waist circumference was measured by one investigator at the midpoint between the inferior costal margin and the superior border of the iliac crest on the midaxillary line. The erectile function was evaluated using the sexual health inventory for men (SHIM) scoring system. Additionally, blood biochemical profiles, including adiponectin, total cholesterol (T-Chol), triglyceride (TG), high-density lipoprotein cholesterol (HDL-Chol), fasting blood glucose (FBG), and hemoglobin A1c (HbA1c) levels, were measured. Low-density lipoprotein cholesterol (LDL-Chol) levels were calculated using T-chol, TG, and HDL-Chol values based on the following formula: LDL-Chol ¼ T-Chol -HDL-Chol -(TG/5), and the arteriosclerosis index (AI) was defined as the LDL-Chol/ HDL-Chol ratio [17]. Serum adiponectin levels were measured by ta latex turbidimetric immunoassay (SRL, Inc, Tokyo, Japan).
Patients who had received antiandrogen agents, finasteride, phosphodiesterase-5 inhibitors, and testosterone agents within the previous 6 months, as well as those with severe chronic heart failure (CHF), renal failure, hemodialysis, or uncontrolled malignancies that may affect serum adiponectin or FT levels and erectile function, were excluded from this study.
The study was approved by our Institutional Review Board (approval no. 2022-023) and conducted according to the Ethical Guidelines for Medical and Health Research.

Statistical analysis
All patients were divided according to their SHIM scores into two groups: those with 12 points or more as normal erectile function to moderate ED (Group 1) and those with less than 12 points as severe ED (Group 2). The basic characteristics of the two groups were compared using the Mann-Whitney U test, whereas categorical data were compared using the Chi-squared test. As considering the variables with a p value of less than 0.2 in univariate analysis as confounding factors, the independent factors associated with severe ED were determined using multiple regression analysis. In a separate subgroup analysis, patients with severe ED were divided into two groups based on serum adiponectin levels (cutoff value of 7.0 lg/mL), which is associated with a lower risk of CVDs development in Japanese [14], and the basic characteristics of both groups were compared using the Mann-Whitney U test.
In all analyses, a p value of <0.05 indicates statistical significance. Statistical analyses were conducted using the Statistical Package for the Social Sciences (SPSS) software, version 25 (SPSS Inc., Chicago, IL).

Results
In this study, 218 patients were included, with a mean age of 65.1 ± 8.3 years (range: 42-82 years), a mean FT level of 7.4 ± 2.2 pg/mL (range: 2.2-11.7 pg/mL) (Table 1), and a mean serum adiponectin level of 6.3 ± 4.5 lg/dL (range: 1.1-23.9). Normal erectile  The included patients were divided into two groups based on SHIM severity score, and the patients' basic characteristics in these two groups were compared and revealed that patients in Group 1 were significantly younger than those in Group 2 (p< 0.001) ( Table 2). Additionally, serum adiponectin levels were significantly lower in Group 1 than in Group 2 (p¼ 0.0168). Multiple regression analysis demonstrated that age and adiponectin levels were independent factors related to severe ED, with an odds ratio of 0.933 (95% confidence interval [CI]: 0.885-0.972, p¼ 0.001) and 0.911 (95% CI: 0.838-0.990, p¼ 0.028), respectively (Table 3). Therefore, patients with severe ED had significantly higher serum adiponectin levels.
The subgroup analysis of patients with severe ED based on adiponectin levels revealed that patients with low adiponectin levels of less than 7.0 lg/dL were younger and had higher FT levels (p< 0.05) (Table 4). However, regardless of age and FT level, waist circumference, BMI, body fat ratio, FBG, TG, and HDL-Chol were significantly worse in the low adiponectin group than in the non-low adiponectin group (p < 0.05). Additionally, patients with severe ED and low adiponectin levels had a significantly higher AI (p¼ 0.0274).

Discussion
This study demonstrated that age and serum adiponectin levels were significantly correlated with severe ED in hypogonadal men, with patients with severe ED being older and having higher adiponectin levels. It is widely accepted that aging is the most common risk factor for ED development and severity. However, the relationship between ED and adiponectin levels has received little attention. To the best of our knowledge, this is the first report to demonstrate a negative correlation between serum adiponectin levels and ED severity.
Previously, only three studies have investigated the relationship between adiponectin and ED. One study that investigated adipokines, hormones, and cardiovascular risk factors in patients with ED found that adiponectin levels were lower in patients with ED than in those without [18], which is contradictory to our results. Another study including 17 nondiabetic and 15 diabetic patients with ED and 21 healthy controls revealed a negative correlation between adiponectin levels and DM, with no correlation between adiponectin and erectile function being detected [19]. Alternatively, one study evaluated adiponectin levels in DM patients with and without ED, and found that erectile function did not correlate with adiponectin [20]. These contradictory results of the adiponectin-ED correlation may be attributable to the heterogeneity of the study population.  The most common cause of ED in hypogonadal men is vascular endothelial dysfunction because of arteriosclerosis. Metabolic factors such as hypertension, DM, dyslipidemia, and obesity, which are common in hypogonadal men, are well established to be risk factors for CVDs. These metabolic factors are also significantly associated with ED [21], and hence, ED can be considered a potential predictor of CVDs. Generally, adiponectin secretion decreases with these metabolic factors, implying that patients with ED are likely to have lower serum adiponectin levels.
Surprisingly, this study found that serum adiponectin levels were significantly higher in hypogonadal men with severe ED. However, the reason for this elevation remains unknown. Adiponectin has been reported to be negatively related to CVDs in various study populations, including those on hemodialysis and those with hypertension and CHF [22][23][24][25]. Even though patients with chronic kidney disease (CKD) often have potential risks for insulin resistance and CVDs, their adiponectin levels are generally high [26,27]. Indeed, a previous study on 85 patients on hemodialysis found that elevated adiponectin levels independently predicted total mortality over a 3-year follow-up period [22]. Additionally, another study found that elevated adiponectin levels increased the risk of CVD death in patients with Stage 3 and 4 CKD [23]. By contrast, a subgroup analysis based on ATTEMPT-CVD research data from 1228 hypertensive patients revealed that the higher adiponectin group was significantly more likely to develop CVDs and renal function deterioration [24]. Additionally, higher adiponectin levels have been linked to negative outcomes in patients with CHF [25]. It has been speculated that adiponectin is highly secreted as a compensatory hormone in patients with CKD to relieve vascular endothelial cell damage and suppress systemic vascular inflammation [22,23]. In patients with CHF, adiponectin expression in skeletal muscle is increased approximately fivefold, whereas its receptor expression is diminished, resulting in increased adiponectin levels [28]. These findings suggest that adiponectin can be substantially secreted in specific clinical conditions such as ED and hypogonadism to compensate for vascular inflammation and arteriosclerosis. Therefore, our paradoxical correlation between adiponectin and ED could be explained by compensatory upregulation of adiponectin in response to the patients' metabolic backgrounds. However, further studies are warranted to support this hypothesis.
Furthermore, Magnussen et al. conducted a randomized control study in hypogonadal men with type 2 DM to investigate an effect of testosterone administration on adiponectin. This previous study demonstrated that testosterone reduced adiponectin levels independent of changes in fat distribution [29]. An experimental data showed that testosterone caused a reduction of total circulating adiponectin levels by selectively inhibiting high-molecular weight form adiponectin secretion from adipocytes [30]. Since the study population was comprised with hypogonadal men, a low testosterone environment may have facilitated the compensatory secretion of adiponectin.
Conversely, in the subgroup analysis based on adiponectin levels in patients with severe ED, the lower adiponectin group had significantly worse metabolic factors, despite being younger and having higher FT levels. Particularly, the presence of many metabolic factors in patients with severe ED was expected to reduce serum adiponectin levels, which is consistent with many previous studies [31,32]. Therefore, we hypothesized that the compensatory secretion of adiponectin may be lost in patients with lower adiponectin levels (<7.0 lg/dL) and severe ED because of several metabolic factors. Additionally, AI was significantly higher in these patients. Lower serum adiponectin levels in patients with severe ED and hypogonadism may pose an exceedingly high risk for CVDs development because of a loss of compensatory adiponectin secretion. Therefore, active medical interventions should be considered in hypogonadal men with severe ED and low adiponectin levels.
There are some limitations to this study. First, the study population only included patients with hypogonadism, who frequently had lifestyle-related diseases, metabolic syndrome, and ED. These factors have a significant impact on serum adiponectin levels. Particularly, normal erectile function (SHIM � 22) was found in only nine cases (4.1%), and the majority of participants had ED. Second, erectile function was determined subjectively based only on a self-administered questionnaire, rather than tools such as nocturnal penile tumescence and the prostaglandin assay. Furthermore, there was a lack of data about objective arteriosclerosis evaluations such as ankle-brachial index or pulse wave velocity to assess CVD risk in this study. Therefore, further studies involving a larger population, including healthy men, as well as objective erectile and arteriosclerosis evaluations, are required to reach more conclusive findings. Third, hypogonadism was diagnosed based on FT levels measured using a radioimmunoassay kit, as opposed to the international diagnostic criteria that use total testosterone (TT) or bioavailable TT. A large epidemiological study in Japan demonstrated that TT levels did not decrease with age, whereas FT levels gradually decreased. In addition, a highly significant correlation (r ¼ 0.706) between calculated FT and analogue FT values have been confirmed among Japanese men [33]. Therefore, FT is commonly used to diagnose hypogonadism in Japan [16].

Conclusions
Serum adiponectin levels were significantly higher in hypogonadal men with severe ED than in those with normal erectile function or moderate ED. Nevertheless, various metabolic factors were significantly worse in patients with severe ED who had lower adiponectin levels. Therefore, these patients may be at a high risk of CVDs development.