Human reproductive function is strictly regulated by the metabolic state. Currently, the prevalence of obesity and diabetes is increasing worldwide (1, 2), and patients are not only at increased risk for cardiovascular events (3–5), but also for complications of the reproductive system. Studies have found that the incidence of reproductive complications in males with obesity and/or type 2 diabetes mellitus (T2DM) can range from 30% to 40%, with patients presenting with hypogonadotropic hypogonadism (6, 7). In contrast, hypogonadism is relatively rare in patients with type 1 diabetes mellitus (T1DM), suggesting that insulin resistance may play an important role (8–10). These patients present with reduced serum total testosterone (TT, recommendations from multiple societies are TT< 12 nmol/L) and clinical symptoms (10, 11), including weakness, decreased libido, erectile and ejaculatory dysfunction, etc. (12, 13). The symptoms have a negative impact on the patient’s work, life, and relationships. Meanwhile McPherson et al. found by studying male mice treated with a high-fat diet that the effects of the abnormal metabolic state on the reproductive system carried over even to their offspring raised on a normal diet (14).

There are several difficulties in the current management of male with T2DM in combination with hypogonadism. For one, the receiving physician tends to focus on renal, neurologic, and retinal complications, and rarely discusses reproductive health and refining tests for TT and other sex hormone levels (15). Secondly, in addition to TT deficiency, clinical symptoms may overlap with other complications of diabetes such as weakness, decreased libido, etc. which may be associated with poor glycemic control, vasculopathy (16), and autonomic neuropathy (17, 18). Third, the causes of hypogonadism are complex and, in clinical practice, often require the exclusion of other diseases. Obesity or diabetes results in secondary hypogonadism. It needs to be differentiated from other secondary causes (e.g., traumatic brain injury, tumors of the hypothalamus or pituitary gland, inflammatory and infectious diseases, etc.), and more importantly, primary conditions need to be excluded (e.g., congenital disorders or traumatic injuries of the testes, medications such as ketoconazole and glucocorticoid, and alcohol abuse) (19). Fourth, the appropriateness of long-term testosterone therapy for males with T2DM in the presence of hypogonadism is unclear, and the available studies have mixed results that have been summarized in review articles (20). There are also significant side effects of testosterone therapy (21, 22), which makes early detection of hypogonadism very important.

Taken together, we hope to identify clinical and or biochemical indices that can be routinely used by physicians to predict hypogonadism and to be able to recommend timely andrological evaluations. There is currently 1 trial conducted in Italy that discussed the relationship between Visceral Adiposity Index (VAI) (23), Triglyceride Glucose Index (TyG) (24), and Lipid Accumulation Product (LAP) (25), three common obesity- and lipid-related indices and TT levels in males with T2DM, and found that all three indices were negatively correlated with TT levels, and corresponding cut-off values were given (26). However, considering the significant differences in adipose tissue distribution between Caucasians and Asians (27), we therefore introduced Chinese visceral adiposity index (CVAI) (27, 28), a metric used to assess visceral adiposity dysfunction in Chinese that has been shown to be associated with obesity and multiple complications of diabetes (29–31), in the assessment of gonadal dysfunction. Similarly, we also analyzed the other three indices. To our knowledge, no studies have evaluated these four indices in Chinese males with T2DM.

To fill this gap, we clarified the strength of association of the four obesity- and lipid-related indices mentioned above in predicting hypogonadism in Chinese people, and gave their predictive cut-off values in identifying hypogonadism and analyzed the diagnostic capability of the indices when they were combined.

This is the first study to investigate the relationship between obesity- and lipid-related indices (VAI, CVAI, LAP and TyG) and hypogonadism in males with T2DM in Chinese population. A total of 958 male patients with T2DM were included in our study, of which 323 patients with hypogonadism, a proportion of 33.72%, was similar to the proportion in previous studies (6, 7). By analyzing this population, we found that all four indices were predictors of low TT and gave the corresponding ORs. At the same time, we made ROC curves for these four indices, confirming that they can be a more effective and reliable predictor of hypogonadism, and giving cut-off values.

Reduced TT is relatively rare in patients with type 1 diabetes, unless the patient is comorbidly obese, but are not uncommon in patients with T2DM (6, 7). A bilinear relationship between T2DM and male hypogonadism has been recognized (35), but the exact physiological mechanisms are not clear (20). It has been shown that prolonged hyperglycemia and more severe insulin resistance (both central and peripheral) affect TT synthesis and secretion (8–10), and low levels of TT are thought to exacerbate insulin resistance and even lead to diabetes (36). In our study, we found that hypogonadism was not associated with age (P = 0.315) and duration of diabetes (P = 0.179), similar to some of the previous studies (26, 37), and some of them different (38, 39). From these studies (all considering TT), age seems to be an ambiguous factor, but a cross-sectional study in Europe that included 3200 people found that TT levels appear to remain essentially stable with age until the age of 70, when elevated LH suggests functional testicular failure, while free testosterone declines progressively in the 40s and sex hormone-binding globulin (SHBG) was elevated (19, 40). Therefore, we consider that, firstly, when considering TT, age below 70 years may not be a significant influencing factor. Secondly, free testosterone, although difficult to obtain data, may be a more accurate index of gonadal function. Also to the best of our knowledge, there are no studies that discuss the relationship between age and these four indices.

In addition, we found that these patients had higher insulin and C-peptide levels, which corresponded to the more severe insulin resistance described above, as confirmed by the difference in HOMA-IR between the two groups (p< 0.001), suggesting to us that hypogonadism may be more related to diabetes control. In addition, there were differences in weight, WC, hip circumference, and BMI between the two groups (all P< 0.001), all of which have been found and repeatedly elucidated in previous studies to be associated with lower TT in males with T2DM (26, 37–39, 41), and there was a significant difference in BMI between the two groups in the present study, which fell in the obese and overweight ranges, respectively, and thus we considered BMI in the subsequent analyses.

Meanwhile, our study found a possible relationship between TT and TG and HDL-C (i.e., TT was positively correlated with HDL-C), which is consistent with many studies (26, 42, 43). This may be one of the reasons why hypogonadism is susceptible to cardiovascular disease complications (44–46), and this risk is certainly more prominent in the group of patients with T2DM. However, basic experiments yielded contradictory results, as Langer et al. found that TT upregulated the expression of scavenger receptor B1 (SR-B1) in HepG2 hepatocytes, leading to a decrease in HDL-C levels, and induced the cholesterol efflux from macrophages and retrograde transport toward the liver, and thus concluded that TT could prevent atherosclerosis (47). Taken together, the potential mechanisms underlying the association between TT and dyslipidemia in males are unclear and contradictory conclusions seem to emerge from clinical and basic studies, so further research is needed to clarify more detailed mechanisms of the relationship under the overall conditions.

VAI, CVAI, TyG, and LAP are common obesity- and lipid-related indices that reflect the distribution of fat and indirectly glyco-metabolic (48, 49). Our study also found that these indices are also sensitive predictors of hypogonadism in Chinese males with T2DM. The test capability (sensitivity and specificity) of VAI, LAP and TyG were significantly better than that of WC, HOMA-IR, and BMI, and the test capability of CVAI was not inferior to that of WC, HOMA-IR, and BMI. We also tried combinations between these four indices, but there was no significant improvement in AUC, sensitivity, and specificity. Therefore, we suggest LAP (AUC = 0.852, Std. Error = 0.014, 95% CI = 0.818 - 0.873, and optimal cutoff value = 59.850) as a commonly used index.

The relationship between low TT and obesity- and lipid-related abnormalities is currently unclear, but possible mechanisms are discussed below: First, previous studies suggested that low TT was due to increased aromatase activity in adipose tissue and increased conversion of TT to E2, which inhibits the hypothalamic-pituitary-gonadal (HPG) axis, but several more recent studies have found a positive correlation between TT and E2 (50), and our study also had similar findings (P< 0.001). And a partial explanation may be given by the study of Brüning et al. They selectively deleted insulin receptors in mice neurons, leading to hypogonadotropic hypogonadism in addition to metabolic symptoms (9). Moreover, it is known from in vitro experiments that insulin promotes the secretion of gonadotropin-releasing hormone (GnRH) from hypothalamic neurons, a physiological effect that is mainly mediated through phosphatidylinositol 3-kinase (PI3K) and the mitogen-activated protein kinases (MAPK) signaling pathways, two classical signaling pathways of insulin (51, 52), which activate the expression of the c-fos and EGR-1 genes, and then the GnRH mRNA levels are elevated (53), and it is clear that the impedance of the signaling in the insulin-resistant state affects the expression of GnRH as well. Thus, when obesity- and lipid-related abnormalities are present, the reduction of TT and E2 suggests the possible presence of insulin resistance in neurons of the HPG axis, whereas the role of aromatase may not be significant. Second, visceral fat is also an active endocrine tissue, increased secretion of adipose-specific cytokines (e.g., leptin, IL-6, and TNF-α) in patients with obesity or abnormal adipose distribution also inhibits gonadotropin secretion, which in turn inhibits HPG axis (54, 55). This explains the lower LH in hypogonadism patients in the present study (P = 0.044) and corresponds to previous studies (56). Third, insulin resistance in abnormal fat distribution and T2DM result in a decrease in sex hormone-binding globulin (SHBG) (57), secondary to a decrease in TT.

However, our study also has limitations. First, this was a retrospective study, so we lacked the testicular examination (to assess volumes through Prader’s orchidometer), and International Index of erectile function-5 (IIEF-5), International Prostatic Symptoms Score (IPSS) and Aging Male Symptom Score (AMSS) questionnaires, and we also did not measure some important obesity-derived adipokines (e.g., Adiponectin, Leptin, Chemerin, and Nesfatin-1), which some studies suggest are associated with obesity and hypogonadism (58–60), as well as SHBG, which is used to calculate free testosterone, a more accurate reflection of actual testosterone levels. In addition, all of us recruited were hospitalized patients, so there may be some bias, such as higher age being one of them. It is also worth noting that a number of other factors can affect TT, such as statins have been found to reduce TT, particularly atorvastatin, which may be due to the fact that statins affect testicular uptake of cholesterol, the raw material for TT synthesis, and statins may lead to a reduction in SHBG (61). We also look forward to subsequent studies that incorporate relevant indices.

VAI, CVAI, TyG, and LAP are sensitive indicators for predicting hypogonadism in Chinese male patients with T2DM, and we give possible cutoff values, where we recommend LAP as a commonly used assessment index, and indices combinations are often unnecessary because they do not increase predictive capability. We expect to facilitate cooperation between different departments (endocrinologists and andrologists) and to contribute to the early detection and early treatment of patients with reproductive system complications.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving humans were approved by Ethics Committee of the Affiliated Hospital of Medical College Qingdao University. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

YY: Conceptualization, Data curation, Formal analysis, Writing – original draft, Writing – review & editing. YYW: Data curation, Formal analysis, Writing – review & editing. LX: Methodology, Project administration, Writing – review & editing. WL: Data curation, Formal analysis, Writing – review & editing. YGW: Conceptualization, Supervision, Writing – review & editing.