Translate this page into:
Sex steroid hormones and cardiovascular risks: A cross-sectional comparative study of pre-diabetic men
*Corresponding author: Dr. Chijioke Stanley Anyigor-Ogah, Department of Family Medicine, Alex-Ekwueme Federal University Teaching Hospital, Abakaliki, Ebonyi State, Nigeria. ogahstanly90@yahoo.com
-
Received: ,
Accepted: ,
How to cite this article: Anyigor-Ogah CS. Sex steroid hormones and cardiovascular risks: A cross-sectional comparative study of pre-diabetic men. Ann Natl Acad Med Sci (India). doi: 10.25259/ANAMS_51_2025
Abstract
Objectives
Metabolic disturbances, including insulin resistance and overweight, are significantly linked to pre-diabetes. These are believed to impact the serum levels of sex steroid hormones. The alteration in both glucose and sex steroid hormone metabolism is an important aspect in the pathogenesis of cardiovascular diseases (CVD). Studies on these associations are scarce in Africa. This research evaluated the correlation between sex steroid hormones and cardiovascular risks amongst pre-diabetic men and compared the same to other degrees of glycemia.
Material and Methods
It was a cross-sectional comparative research. A total of 191 adult men (75 with prediabetes (PD), 58 with diabetes, and 58 with normal glucose), who presented at the study center within the study period, were recruited. The recruitment was done by a systematic random sampling method, using interviewer-administered designed questionnaires. The mean and standard deviations, as well as numbers and percentages, were estimated for the numerical and categorical variables, respectively. The Fisher’s test was used in comparing the numerical variables, while the categorical variables were compared via the Chi-square test. The level of independent association among the dependent and independent variables was analyzed using logistic regression. The strength of these associations was tested using the odds ratio (OR) at a confidence interval (CI) of 95%. The statistically significant level was considered at a p <0.05.
Results
The associations between the age, weight, body mass index (BMI), waist circumference (WC), hip circumference (HC), waist-hip ratio (WHR), systolic blood pressure (SBP), diastolic blood pressure (DBP), and the glycemic levels were statistically significant. The patient with PD was more likely to be obese with a higher CVD risk. However, the BMI, WC, HC, WHR, SBP, and DBP were independent risk factors for the development of PD on the elimination of cofounders. There was age-dependent significantly lower free testosterone (FT) among PD compared to normal glucose tolerance (NGT) at all BMI levels, and significantly lower total testosterone (TT) and FT values among PD compared to other groups as the WHR rises.
Conclusion
The association between major CVD risks and some abnormal reproductive hormones with PD shows the need to screen all men within the reproductive age group, regular re-evaluation of all identified cases, lifestyle changes aimed at weight reduction, regular exercise, healthy eating habit, and intensifying the campaign on the dangers of PD, as it is always done for diabetes. These hormone abnormalities may worsen the incidence of PD, and by extension, diabetes mellitus (DM).
Keywords
Cardiovascular disease
Glucose
Hormone
Pre-diabetes
Sex steroid
INTRODUCTION
It is postulated that the alteration in the metabolism of sex steroid hormones may be key in the modulation of the link that exists between impaired glucose tolerance and the development of chronic diseases.1 However, the association of prediabetes (PD) with the sex-steroid hormones is still elusive.1 Data on the study of the relationship between sex steroid hormones and cardiovascular disease (CVD) risk among the population under study in sub-Saharan Africa is scarce. This research comparatively evaluated the link that sex steroid hormones have with CVD risks amongst subjects with PD with a view to finding out if CVD risk parameters alter sex steroid hormones as well as holistically incorporating it into male factors in infertility or sub-fertility.
There are various postulates on the mechanism of insulin resistance in obesity. Hepatic accumulation of lipids is likely a major mechanism connected with obesity-related resistance to insulin and non-insulin-dependent diabetes mellitus (DM).2 Rise in liver secretion of various lipoproteins leads to sustained accumulation of these particles in the plasma and accretion of extremely lipolyzed cholesterol-rich dense lipoprotein particles.3 Studies have shown that a decrease in steroid hormone levels correlates with increased BMI, non-obese DM, hypertension, and metabolic syndrome.4,5 Low testosterone levels among men are also related to high cardiovascular disease risk, as they cause dyslipidemia, central obesity, glucose intolerance, and DM.6
In a cross-sectional research, Ho and colleagues (2013) demonstrated an association between PD and an increased risk of subnormal total testosterone when compared to those with normoglycemia after adjusting for age, BMI, and waist circumference (WC) as well as metabolic syndrome.7 The link between PD and testosterone deficiency was said to apply to men over a wider age range. Since steroid hormone-binding globulin (SHBG) binds tightly to testosterone, the low total testosterone in PD could be a consequence of low SHBG.7 In a cohort study among Polish men, the average total testosterone (TT) and calculated free testosterone (FT) levels were markedly low among the patients with PD, compared with patients with normoglycemia. Low testosterone level became an independent risk for PD.8 In another study, patients with PD had lower TT and calculated FT levels with higher estradiol and SHBG levels than normal people. Pre-diabetic men also had negative correlations with BMI, WC, and total testosterone.9 In a study of Chinese men, significantly lower serum SHBG and total testosterone levels were recorded among the patients with diabetes, while the pre-diabetics only had significantly lower serum SHBG. The reduced levels of these hormones correlated with rising proportions of PD and diabetes.10 The causative link between reduced levels of testosterone and diabetes, which correlated with advancing age, obesity, and metabolic syndrome, may be bidirectional or multidirectional.11
MATERIAL AND METHODS
The study design
This study was cross-sectional and analytical. It lasted over 4 months.
Study population
The study consisted of adult males with PD, diabetes, and normal glucose tolerance (NGT) based on testing, who came for care at the General Out-Patient unit of the study institution within the duration of this study.
Selection criteria
Included in this study were those who were aged 35-75 years, gave consent to participate, were without limb/skeletal deformities, had PD, had type 2 DM, were not on metformin therapy, and had NGT.
Excluded from this study were those who had chronic alcoholism, smoking addiction, prostate gland problems, were on drug treatment, had pre-existing renal or liver diseases, thyroid gland disorders, or testosterone replacement therapy.
Sample size determination
Sample size (n) was calculated using the Fisher’s formulas thus:12 n = Z2pq/d2, where; n is the minimum sample size, Z is the standard normal deviate at 95% confidence level (1.96), p is the prevalence of PD in men from previous study (6.0%),13 q = 1 - p is the fraction of the people with no PD (which was 0.94), while the degree of precision, generally set as 0.05 is regarded as d.
Applying the values as shown above in the formula:
The sample size (n) = 86.67
With a population size of 560 (<10,000), correction of the sample size was done using the formula: Nf = n/[1+(n/N)], where Nf was the corrected sample size; n was the population size (560); N was the sample size (86.67).
Applying the formula, Nf = 75.06 (∼75)
Hence, the sample size was 75
Using the same formulae, population size of 400 and a prevalence (p) = 4.6%.13
The sample size for the patients with diabetes was 58, while the sample size for the NGT group was arbitrarily chosen to be 58.
Sampling technique
The study participants were selected by the systematic random sampling method. The sample interval was 7 for PD and DM subjects, while normal subjects had an interval of 12. All selection numbers that met the exclusion criteria were rejected, and the subsequent number was chosen.
The study instruments
The research data were obtained in three different ways:
Use of a face-to-face interviewer–administered structured questionnaire
The questionnaire was prepared in English. Verbal interpretations were made to those who did not understand English in Igbo and local dialects. The questionnaire was divided into two parts. The first part documented the demographic characteristics such as age, occupation, educational level, and level of income. The second part documented the anthropometric and clinical measurements. The measured anthropometries were: the body weight, height, WC, HC, WHR, as well as the BMI, while the clinical parameter measured the blood pressure. The body weight was measured using a standardized weighing machine to the nearest 0.1 kg, while the height was measured to the nearest 0.1 cm using the stadiometer with the subject in anatomical position. The BMI was calculated as the ratio of weight (kg) to height (m2). The WC was measured with the patient in anatomical position under normal protocol using a non-flexible calibrated tape. This was converted to the nearest 0.1 cm. The HC was measured from the point of the widest diameter of the buttocks, also using a non-flexible calibrated tape while the patient was in anatomical position, and recorded to the nearest 0.1 cm. The WHR was calculated by dividing the WC by the HC. The blood pressure was measured in millimeters of mercury using the sphygmomanometer and stethoscope with the patient in a relaxed sitting position. Two different measurements were made, and the average taken.
Laboratory measurements
About 5 mL of whole blood was collected from each subject. These were put into a covered sterile plain bottle and labelled. The containers were left undisturbed for up to 30 min at room temperature to allow for clotting. The samples were centrifuged within 1 h for 10 min at 1,000-2,000 revolutions to remove the clot. The resulting supernatants (serum) were transferred immediately into another plain bottle with a Pasteur pipette and were maintained at 2-8°C while handling. The serum was subsequently stored at -20°C for analysis of the sex hormones, namely TT, dihydroepiandrostenedione (DHEA), SHBG, luteinizing hormone (LH), follicle-stimulating hormone (FSH), and FT. The TT, LH, FSH, SHBG, and DHEA were measured by chemiluminescent immunometric assays (Immulite 2000; DPC United States, Siemens, Los Angeles, CA). The FT was measured by enzyme-linked immunosorbent assay (BIOTEK ELx 808, USA). Testosterone levels of 8-28 ηmol/L, LH of 2-6 miu/L, FSH of 3 – 10 miu/L, DHEA of 200-37 ng/dL, and SHBG of 10-57 ηmol/L were taken as normal, while FT level of <0.250 ηmol/L (6ηg/dL) was considered low.
Furthermore, capillary venous blood was also collected aseptically from each participant. This followed about 8 h of fasting (about the same time of blood collection for hormonal tests) for a direct fasting glucose or 2-hour postprandial glucose tests using a fine test glucometer kit. The results were read in mg/dL. PD was diagnosed if any of the following criteria were met at fasting glucose of 100-125 mg/dL, 2-hour postprandial glucose of 140-199 mg/dL. The diagnosis of type 2 DM was made in subjects who had a history of diabetes, or if the fasting blood glucose was ≥126 mg/dL, or the 2-hour post-prandial glucose was ≥200 mg/dL.14
The research assistants were nominated and educated by the principal investigator and supervisors. They were trained on the aims of the research, the questionnaire content, data gathering and measurement techniques, and confidentiality. The participation was voluntary; hence, participants could decline participation at any time in the research with no consequence.
Analysis of data
Data was analyzed using the Statistical Package for the Social Sciences (IBM SPSS Inc. and Chicago, IL, USA) version 20.0. The mean and standard deviation were estimated for numerical variables, while the number and percentage were estimated for categorical variables. The Fisher’s test was used to compare the numerical variables, while the chi-square test was employed to compare the categorical variables. The association between the dependent and independent variables was determined using the logistic regression analysis. The strength of the associations was tested using the odds ratio (OR) at 95% confidence interval (CI). A p <0.05 was taken to be statistically significant.
RESULTS
A total of 191 questionnaires were used for the study, but 184 subjects participated, giving an overall response rate of 96.3%. The mean ages of the respondents were 49.07±10.99 kg for PD, 55.37±12.11 kg for DM, and 55.18±11.16 kg for NGT.
The socio-demographic characteristics of the respondents and glycemic levels
This study, as represented in Table 1, only showed a statistically significant correlation between age and glycemic levels (p = 0.001). No significant correlation existed between other variables and glycemic levels.
| Socio-demography | NGT (%) | PD (%) | DM (%) | Total (%) | χ2 | p value |
|---|---|---|---|---|---|---|
| Age Group (years): | ||||||
| 35-44 | 12 (6.5) | 26 (14.1) | 16 (8.7) | 54 (29.3) | ||
| 45-54 | 9 (4.9) | 22 (12.0) | 14 (7.6) | 45 (24.5) | ||
| 55-64 | 22 (12.0) | (9.8) | 9 (4.9) | 49 (26.6) | 23.36 | 0.001 |
| 65-74 | 14 (7.6) | 4 (2.2) | 18 (9.8) | 36 (19.6) | ||
| Total | 57 (31.0) | 70 (38.0) | 57 (31.0) | 184 (100.0) | ||
| Occupation: | ||||||
| Farming | 15(8.2) | 10 (5.4) | 17 (9.2) | 42 (22.8) | ||
| Trading | 5 (2.7) | 11 (6.0) | 10 (5.4) | 26 (14.1) | ||
| Civil servants | 25 (13.6) | 40 (21.7) | 24 (13.0) | 89 (48.4) | 9.74 | 1.36 |
| Self-employed | (6.5) | 9 (4.9) | 6 (3.3) | 27 (14.7) | ||
| Total | 57 (31.0) | 70 (38.0) | 57 (31.0) | 184 (100.0) | ||
| Education level: | ||||||
| None | 8 (4.3) | 11 (6.0) | 16 (8.7) | 35 (19.0) | ||
| Primary | 11 (6.0) | 10 (5.4) | 8 (4.3) | 29 (15.8) | ||
| Secondary | 14 (7.6) | 15 (8.2) | 10 (5.4) | 39 (29.2) | 5.47 | 0.49 |
| Tertiary | 24 (13.0) | 34 (18.5) | 23 (12.5) | 81 (44.0) | ||
| Total | 57 (31.0) | 70 (38.0) | 57 (31.0) | 184 (100.0) | ||
| Av. monthly income | ||||||
| <₦ 10,000 | 20 (10.9) | 20 (10.9) | 20 (10.9) | 60 (32.6) | ||
| ₦ 10,000-20,000 | 9 (4.9) | 12(6.5) | 12 (6.5) | 33 (17.9) | ||
| ₦ 21,000-30,000 | 12 (6.5) | 12 (6.5) | 9 (4.9) | 33 (17.9) | 2.66 | 0.85 |
| > ₦ 30,000 | 16 (8.7) | 26 (14.1) | 16 (8.7) | 58 (31.5) | ||
| Total | 57 (31.0) | 70 (38.0) | 57 (31.0) | 184 (100.0) |
p-value <0.05 is considered significant. PD: Prediabetes, NGT: Normal glucose tolerance, DM: Diabetes mellitus., ₦: Naira (Nigerian).
Association between anthropometric variables and glycemic levels
As presented in Table 2, statistically significant differences were observed in the association between the glycemic levels and BMI (p = 0.0001), WC (p = 0.0001), HC (p = 0.0001), WHR (p = 0.0001), SBP (p = 0.0001), and DBP (p = 0.004).
| Variable | NGT (%) | PD (%) | DM (%) | Total (%) | χ2 | p value |
|---|---|---|---|---|---|---|
| BMI (kg/m2): | ||||||
| 18.5-24.9 | 15 (8.2) | 8 (4.3) | 22 (12.0) | 45 (24.5) | ||
| 25.0-29.9 | 34 (18.5) | 23 (12.5) | 26 (14.1) | 83 (45.1) | ||
| 30.0-34.9 | 7 (3.8) | 30 (16.3) | 7 (3.8) | 44 (23.9) | 38.65 | 0.0001 |
| ≥35 | 1 (0.5) | 9 (4.9) | 2 (1.1) | 12 (6.5) | ||
| Total | 57 (31.0) | 70 (38.0) | 57 (31.0) | 184 (100.0) | ||
| WC (cm): | ||||||
| <94 | 50 (27.2) | 34 (18.5) | 44 (23.9) | 128 (69.6) | ||
| ≥94 | 7 (3.8) | 36 (19.5) | 7 (7.1) | 56 (30.4) | 25.01 | 0.0001 |
| Total | 57 (31.0) | 70 (38.0) | 57 (31.0) | 184 (100.0) | ||
| HC (cm): | ||||||
| >105 | 21 (11.4) | 52 (28.2) | 29 (15.8) | 102 (55.4) | ||
| ≤105 | 36 (19.6) | 18 (9.8) | 28 (15.2) | 82 (44.6) | 18.52 | 0.0001 |
| Total | 57 (31.0) | 70 (38.0) | 57 (31.0) | 184 (100.0) | ||
| WHR: | ||||||
| >0.9 | 9 (4.9) | 52 (28.2) | 6 (3.3) | 67 (36.4) | ||
| ≤0.9 | 48 (26.1) | 18 (9.8) | 51 (27.7) | 117 (63.6) | 70.33 | 0.0001 |
| Total | 57 (31.0) | 70 (38.0) | 57 (31.0) | 184 (100.0) | ||
| SBP (mmHg): | ||||||
| ≤130 | 38 (20.7) | 23 (12.4) | 38 (20.7) | 59 (53.8) | ||
| 131-139 | 5 (2.7) | 9 (4.9) | 7 (3.8) | 21 (11.4) | ||
| 140-159 | 14 (7.6) | 29 (15.8) | 10 (5.4) | 53 (28.8) | 26.00 | 0.0001 |
| ≥160 | 0 (0.0) | 9 (4.9) | 2 (1.1) | 11 (6.0) | ||
| Total | 57 (31.0) | 70 (38.0) | 57 (31.0) | 184 (100.0) | ||
| DBP (mmHg): | ||||||
| <90 | 47 (25.5) | 41 (22.3) | 45 (24.5) | 133 (72.3) | ||
| 90-99 | 9 (4.9) | 19 (10.3) | 11 (6.0) | 39 (21.2) | 15.48 | 0.004 |
| 100-109 | 1 (0.5) | 10 (5.4) | 1 (0.5) | 12 (6.5) | ||
| Total | 57 (31.0) | 70 (38.0) | 57 (31.0) | 184 (100.0) |
p-value <0.05 is considered significant. PD: Prediabetes, NGT: Normal glucose tolerance, DM: Diabetes mellitus, BMI: Body mass index, WC: Waist circumference, HC: Hip circumference, WHR: Waist-hip ratio, SBP: Systolic blood pressure, DBP: Diastolic blood pressure.
Mean reproductive hormone concentration and deficiencies within BMI levels
As presented in Table 3, statistically significant differences existed between the glycemic levels and mean FT at BMI <18.5 kg/m2 (p = 0.0001), BMI 18.5-24.9 kg/m2 (p = 0.001), and BMI ≥25 kg/m2 (p = 0.0001). However, among the overweight respondents, a statistically significant difference existed between the glycemic levels and mean FT deficiency (p = 0.006).
| BMI/Hormone | NGT (Mean±SD) | PD (Mean±SD) | DM (Mean±SD) | Total (Mean±SD) | p value |
|---|---|---|---|---|---|
| BMI <18.5 kg/m2 | 27 | 49 | 43 | 119 | |
| TT (ηmol/L) | 12.92±3.18 | 9.89±1.46 | 10.23±2.32 | 10.70±2.23 | 0.190 |
| TT <8 ηmol/L | 1.52±0.58 | 1.39±0.57 | 1.42±0.63 | 1.43±0.59 | 0.650 |
| FT (ηg/dL) | 6.04±1.13 | 5.55±1.60 | 4.73±1.23 | 5.37±1.46 | 0.0001 |
| FT <6 ηg/dL | 1.33±0.48 | 1.24±0.43 | 1.16±0.37 | 1.24±0.43 | 0.261 |
| BMI 18.5-24.9 kg/m2 | 34 | 53 | 48 | 135 | |
| TT (ηmol/L) | 13.88±3.66 | 10.33±1.63 | 11.43±3.36 | 11.62±2.87 | 0.119 |
| TT <8 ηg/dL | 1.65±0.60 | 1.43±0.57 | 1.50±0.65 | 1.51±0.61 | 0.280 |
| FT (ηg/dL) | 6.25±1.16 | 5.73±1.69 | 5.01±1.46 | 5.60±1.56 | 0.001 |
| FT <6 ηg/dL | 1.47±0.51 | 1.30±0.46 | 1.25±0.44 | 1.33±0.47 | 0.100 |
| BMI 25-29.9 kg/m2 | 44 | 58 | 50 | 152 | |
| TT (ηmol/L) | 14.13±2.86 | 11.51±2.66 | 11.15±2.75 | 12.15±2.80 | 0.133 |
| TT <8 ηmol/L | 1.75±0.58 | 1.52±0.63 | 1.52±0.65 | 1.59±0.62 | 0.116 |
| FT (ηg/dL) | 6.63±1.33 | 5.87±1.71 | 5.01±1.37 | 5.81±1.62 | 0.0001 |
| FT <6 ηg/dL | 1.59±0.50 | 1.36±0.48 | 1.28±0.45 | 1.40±0.49 | 0.006 |
| BMI 30-34.9 kg/m2 | 32 | 54 | 45 | 131 | |
| TT (ηmol/L) | 14.19±3.57 | 10.62±1.80 | 11.06±3.46 | 11.64±2.92 | 0.108 |
| TT <8 ηmol/L | 1.59±0.56 | 1.46±0.61 | 1.47±0.66 | 1.50±0.61 | 0.588 |
| FT (ηg/dL) | 6.20±1.13 | 5.76±1.72 | 4.81±1.27 | 5.54±1.54 | 0.0001 |
| FT <6 ηg/dL | 1.44±0.50 | 1.31±0.47 | 1.20±0.40 | 1.31±0.46 | 0.082 |
| BMI ≥35 kg/m2 | 28 | 52 | 43 | 123 | |
| TT (ηmol/L) | 12.88±3.03 | 10.88±2.62 | 10.23±2.32 | 11.11±2.62 | 0.348 |
| TT <8 ηmol/L | 1.54±0.58 | 1.46±0.64 | 1.42±0.63 | 1.46±0.62 | 0.740 |
| FT (ηg/dL) | 6.08±1.13 | 5.67±1.67 | 4.73±1.23 | 5.44±1.50 | 0.0001 |
| FT <6 ηg/dL | 1.36±0.49 | 1.29±0.46 | 1.16±0.37 | 1.26±0.44 | 0.160 |
p-value <0.05 is considered significant. BMI: Body mass index, SD: Standard deviation, PD: Prediabetes, NGT: Normal glucose tolerance, DM: Diabetes mellitus, FT: Free testosterone, TT: Total testosterone.
Mean reproductive hormone concentration and deficiencies within WHR levels
From Table 4, a statistically significant difference was observed at WHR <0.9 in the association between glycemic levels and mean FT (p = 0.0001) and mean FT deficiency (p = 0.007). However, at WHR ≥0.9, a statistically significant difference was observed in the association between glycemic levels and mean TT (p = 0.015), mean TT deficiency (p = 0.020), mean FT (p = 0.0001), and mean FT deficiency (p = 0.001).
| WHR/Hormone | NGT (Mean±SD) | PD (Mean±SD) | DM (Mean±SD) | Total (Mean±SD) | p value |
|---|---|---|---|---|---|
| WHR <0.9 (#) | 33 | 64 | 43 | 140 | |
| TT (ηmol/L) | 13.41±8.11 | 12.68±8.35 | 10.23±7.32 | 12.10±8.04 | 0.171 |
| TT <8 ηg/dL | 1.64±0.60 | 1.61±0.68 | 1.42±0.63 | 1.56±0.65 | 0.240 |
| FT (ηg/dL) | 6.28±1.21 | 6.01±1.74 | 4.73±1.23 | 5.68±1.61 | 0.0001 |
| FT <6 ηg/dL | 1.45±0.51 | 1.42±0.50 | 1.16±0.37 | 1.35±0.48 | 0.007 |
| WHR ≥0.9 (#) | 51 | 55 | 57 | 163 | |
| TT (ηmol/L) | 15.06±8.33 | 10.43±6.63 | 12.70±9.12 | 12.67±8.27 | 0.015 |
| TT <8 ηmol/L) | 1.78±0.54 | 1.45±0.57 | 1.61±0.67 | 1.61±0.61 | 0.020 |
| FT (ηg/dL) | 6.65±1.25 | 5.84±1.75 | 5.28±1.51 | 5.90±1.62 | 0.0001 |
| FT <6 ηg/dL | 1.65±0.48 | 1.33±0.47 | 1.37±0.49 | 1.44±0.50 | 0.001 |
p-value <0.05 is considered significant. N/B: is the total number recorded per group, SD: Standard deviation, WHR: Waist-hip ratio, NGT: Normal glucose tolerance, DM: Diabetes mellitus, FT: Free testosterone, TT: Total testosterone.
Measure of independent association between anthropometry/clinical parameters, and NGT/PD
The relationship between anthropometry and clinical characteristics with NGT and PD had no confounders, hence BMI (p = 0.019, OR = 0.311, 95% CI = 0.117 - 0.828), WC (p = 0.0001, OR = 0.132, 95% CI = 0.053 - 0.332), HC (p = 0.0001, OR = 4.952, 95% CI = 2.317 - 10.585), WHR (p = 0.0001, OR = 15.407, 95% CI = 6.320 - 37.562), SBP (p = 0.001, OR = 0.274, 95% CI = 0.128 - 0.589), and DBP (p = 0.002, OR = 0.098, 95% CI = 0.022 - 0.440) had statistically significant independent association with NGT and PD as shown in Table 5. Similarly, BMI (p = 0.0001, OR = 0.177, 95% CI = 0.069 - 0.455), WC (p = 0.001, OR = 0.279, 95% CI = 0.128 - 0.606), HC (p = 0.007, OR = 2.789, 95% CI = 1.322 - 5.883), WHR (p = 0.0001, OR = 24.556, 95% CI = 9.021 - 66.884), SBP (p = 0.0001, OR = 0.225, 95% CI = 0.102 - 0.496), and DBP (p = 0.003, OR = 0.149, 95% CI = 0.042 - 0.534) had statistically significant independent association with PD and DM.
| Characteristics | NGT/PD | PD/DM | ||||
|---|---|---|---|---|---|---|
| OR | 95%CI | p value | OR | 95%CI | p value | |
|
BMI (kg/m2): ≤24.9 ≥25 |
0.311 | 0.117-0.828 | 0.019 | 0.177 | 0.069-0.455 | 0.0001 |
|
WC (cm): <94 ≥94 |
0.132 | 0.053-0.332 | 0.0001 | 0.279 | 0.128-0.606 | 0.001 |
|
HC (cm): >105 ≤105 |
4.952 | 2.317-10.585 | 0.0001 | 2.789 | 1.322-5.883 | 0.007 |
|
WHR: >0.9 ≤0.9 |
15.407 | 6.320-37.562 | 0.0001 | 24.556 | 9.021-66.844 | 0.0001 |
|
SBP (mmHg): ≤139 >139 |
0.274 | 0.128-0.589 | 0.001 | 0.225 | 0.102-0.496 | 0.0001 |
|
DBP (mmHg): < 90 ≥90 |
0.098 | 0.022-0.440 | 0.002 | 0.149 | 0.042-0.534 | 0.003 |
p-value <0.05 is considered significant.NGT: Normal glucose tolerance, PD: Prediabetes, DM: Diabetes mellitus, OR: Odds ratio, CI: Confidence interval, BMI: Body mass index, WC: Waist circumference, HC: Hip circumference, WHR: Waist-hip ratio, SBP: Systolic blood pressure, DBP: Diastolic blood pressure
DISCUSSION
There was a statistically significant association between age groups and glycemic levels, with age 35-64 years more likely to have prediabetes [Table 1]. The statistically significant increase in the likelihood of PD among those within the economically vibrant age group could be due to higher income and improved living conditions, with its attendant sedentary lifestyle and unhealthy eating habits. It has been demonstrated that a sedentary lifestyle, higher income, and unhealthy eating habits were highly associated with obesity and attendant impaired glucose tolerance.15,16 This finding is further supported by the work of Ezeala-Adikaibe et al. (2018) and Oguoma et al. (2017), who demonstrated that the highest risk of developing PD occurs at age 40-49 years, but varied in the peak incidence of developing DM, which they recorded in those 55-64 years.17,18
Kautzy-Willer and colleagues opined that increased consumption of fats and physical inactivity with associated insulin resistance in men, were highly associated with PD and a higher tendency of coming down with DM.19
However, as indicated in Table 2, there is a statistically significant association between glycemic levels and BMI categories, WC, HC, WHR, SBP, and DBP, and a CVD risk range of the above variables is statistically and significantly higher in respondents with PD. After adjusting for confounders, all CVD risk parameters independently predicted PD. The probability of having a CVD risk range of these parameters was lower in NGT and DM groups compared to the PD participants, as shown in Table 5. The statistically significant association between PD and the CVD risk parameters could have resulted from unhealthy eating habits, greater muscle bulk, and increased prevalence/activity of brown fats, especially among the age group studied, which results in increased weight, accumulation of cholesterol, and consequent increased cardiovascular disease risk among the patients with prediabetes.20 The deleterious effect of climate change, with increased exposure of individuals to endocrine-disrupting chemicals (EDCs) may play a part.21 Endocrine disruptors act as hormones and build up in adipocytes to cause reactive changes in adipokine levels in a sex-specific fashion and in so doing, alter cellular receptors, gene responses, and other targets leading to increased insulin resistance.21 The hepatic accumulation of lipids is likely a major mechanism associated with obesity-induced insulin resistance.2 Altered fat distribution within the adipose tissue and impaired incretin effect 3 could also explain the level of significance observed.
Other studies support the findings in this study.22,23 Hu et al. (2019) opined that high blood pressure, overweight, obesity, and increased WHR significantly correlated with the development of PD.24 This is in no way different from a cross-sectional study in Ethiopia25 and Nigeria.15,17,18 However, the findings from this study are at variance with the study carried out by Haghighatdoost and colleagues (2017), as well as Bhowmik and colleagues (2015), which demonstrated that the BMI, WC, WHR, and hypertension were statistically and significantly associated with higher risks of DM but not PD.26,27 While our study was a hospital-based cross-sectional study which focused on diseased and apparently healthy populations, Haghighatdoost and colleagues carried out a population-based cohort study which focused only on healthy participants.
In Table 3, the overall mean FT had a statistically significant association with glycemic levels, with the lowest values recorded among the respondents with diabetes when compared with the other groups within a BMI range of 18.5-24.9 kg/m2. However, a statistically significant association existed between the glycemic levels and overall mean FT, with participants who had DM more likely to have lower overall mean FT and higher FT deficiencies. In the obesity range, the overall mean FT remained statistically lower among the respondents who had DM when compared to those with NGT and PD at Grade-I obesity. At higher BMI, the risk of metabolic syndrome increases with a rise in plasma cholesterol. However, one would have expected a higher level of FT among the respondents with DM. This is because most steroid hormones (including testosterone) are synthesized from cholesterol in a tightly regulated fashion. The secretion of testosterone during spermatogenesis is a function of Leydig cells;28 hence, if the Leydig cell functions sub-optimally in a state of high testicular testosterone store, not much will be secreted. In a rodent study, Leydig cells expressed leptin receptors and leptin has been shown to inhibit testosterone secretion, buttressing the role of obesity and leptin in the pathogenesis of low testosterone.29
In an animal study, increased testicular cholesterol concentration in a streptozocin-induced diabetic model was reported with a significant decrease in 3β-hydroxysteroid dehydrogenase and 17β-hydroxysteroid dehydrogenase activities in the testis of such models.29 The possible rise in testicular cholesterol concentration without a corresponding rise in serum FT concentration could be due to down-regulation of the cholesterol transporter protein and the rate-limiting enzyme in steroidogenesis in those with DM, thus leading to the accumulation of cholesterol.29 The peripheral translation of testosterone to estrogen in obese men might diminish the level of luteinizing hormone (LH) surge, and potentially prevent testosterone production, exacerbating the vicious cycle of reduced blood testosterone level and obesity.30
These findings are supported by the work of Mohammed et al. (2018), which demonstrated a causative relationship with decreased free levels of testosterone and non-insulin-dependent DM in a bi-directional or multi-directional fashion and an interrelationship between obesity and metabolic syndrome.31 Tanabe et al. (2015) reported that low serum testosterone might predispose to obesity and metabolic syndrome, with its attendant risk of type 2 DM.32
At the CVD risk level of WHR (≥0.9), mean TT and TT deficiencies were significantly lower in those with PD [Table 4]. The negative correlation between WHR, low serum level of circulating testosterone, and non-insulin-dependent DM among men could be due to improved synthesis of estrogens from androstenedione via aromatization, especially in obese men, which results from adipose tissue-based increased enzyme (aromatase) level.33 A prospective study, Karakas et al. (2018), also documented a significant correlation between type 2 DM, low testosterone level, and increased WHR in men, with a testosterone level that was inversely related to the WHR.34 In Sweden, another population-based prospective study in men also observed a statistically significant relationship between testosterone level, WHR, and the progression of resistance to insulin and non-insulin-dependent DM.35 Though our assertion was similar to theirs, their study was centered on men in the reproductive age group and could not exclude risk factors such as alcoholism and smoking, as was done in our study.
CONCLUSION
The association between major CVD risks and some abnormal reproductive hormones with PD in this study showed the need for screening of all men within the reproductive age group, regular re-evaluation of all identified cases, lifestyle changes aimed at weight reduction, regular exercise, healthy eating habits, and intensifying the campaign on the dangers of PD, as it is always done for diabetes. These hormone abnormalities may worsen the incidence of PD, and by extension, DM.
Acknowledgement
We sincerely, thank the Postgraduate Committee and Members of the Faculty of Basic Medicine and the Department of Human Physiology. We also thank our tutors, and supervisors for their support, and continued advice throughout the duration of this study. The cooperation, and willingness of the participants throughout the study period is recognized.
Authors’ contributions
CSAO: Conceptualization, manuscript preparation.
Ethical approval
The research/study approved by the Institutional Review Board at faculty of Basic Medicine, Ebonyi State University, Abakaliki, number EBSU/REC/BMS/1907/01/001, dated 19th July, 2017.
Declaration of patient consent
The author certify that they have obtained all appropriate patient consent forms. In the form, the patient has given consent for clinical information to be reported in the journal. The patient understands that the patient’s names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
Use of artificial intelligence (AI)-assisted technology for manuscript preparation
The authors confirm that there was no use of artificial intelligence (AI)-assisted technology for assisting in the writing or editing of the manuscript and no images were manipulated using AI.
References
- Pre-diabetes and semen sex steroid hormones among US men. Andrology. 2017;5:49-57.
- [CrossRef] [PubMed] [Google Scholar]
- Free fatty acid-induced hepatic insulin resistance is attenuated following lifestyle intervention in obese individuals with impaired glucose tolerance. J Clin Endocrinol Metab. 2010;95:323-7.
- [CrossRef] [PubMed] [PubMed Central] [Google Scholar]
- Expanding diabetes prevention: Obstacles and potential solutions. Am J Prev Med. 2019;57:853-7.
- [CrossRef] [PubMed] [Google Scholar]
- The effect of paternal diet-induced obesity on sperm function and fertilization in a mouse model. Int J Androl. 2011;34:402-10.
- [CrossRef] [PubMed] [Google Scholar]
- Metabolic syndrome and inflammatory biomarkers: A community-based cross-sectional study at the Framingham Heart Study. Diabetol Metab Syndr. 2012;4:28.
- [CrossRef] [PubMed] [PubMed Central] [Google Scholar]
- Testosterone levels and cardiovascular disease. Heart. 2010;96:1787-8.
- [CrossRef] [PubMed] [Google Scholar]
- Pre-diabets is associated with an increased risk of testosterone deficiency, independent of obesity and metabolic syndrome. PLoS One. 2013;8:e74173.
- [CrossRef] [PubMed] [PubMed Central] [Google Scholar]
- The prevalence of prediabetes in population of Polish men with late-onset hypogonadism. Aging Male. 2014;17:141-6.
- [CrossRef] [PubMed] [Google Scholar]
- Late onset hypogonadism among old and middle age males with pre-diabetes in polish population. Aging Male. 2015;18:16-21.
- [CrossRef] [PubMed] [Google Scholar]
- Low sex hormone-binding globulin levels is associated with pre-diabetes in Chinese men, independent of total testosterone. PLoS One. 2016;11:e0162004.
- [CrossRef] [PubMed] [PubMed Central] [Google Scholar]
- Systematic literature review of the risk factors, comorbidities and consequences of hypogoradism in men. Andrology. 2014;2:819-34.
- [CrossRef] [PubMed] [Google Scholar]
- How to calculate sample size for different study designs in medical research. Indian J Psychologic Med. 2013;35:121-26.
- [CrossRef] [Google Scholar]
- Prevalence of diabetes and pre-diabets in Oke-Ogun Region of OYO State, Nigeria. J Med Res Health Educ. 2017;1:6.
- [Google Scholar]
- Standards of medical care in diabetes--2012. Diabetes Care. 2012;35 Suppl 1:S11-63.
- [CrossRef] [PubMed] [PubMed Central] [Google Scholar]
- The prevalence and correlates of prediabetes and diabetes among public category workers in Akure, Nigeria. Open Public Health J. 2017;10:167-76.
- [Google Scholar]
- Factors associated with prediabetes in Tehranian men and women: A structured equation modelling. PLoS One. 2017;12:e0188898.
- [CrossRef] [PubMed] [PubMed Central] [Google Scholar]
- Diabetes and pre-diabetes among adults in an urban slum in South East Nigeria. JDM. 2018;08:131-44.
- [CrossRef] [Google Scholar]
- Cardiovascular disease risk factors in a Nigerian population with impaired fasting blood glucose level and diabetes mellitus. BMC Public Health. 2017;17
- [CrossRef] [Google Scholar]
- Women show a closer association between educational level and hypertension or diabetes mellitus than men: A secondary analysis from the Austrian HIS. BMC Public Health. 2016;12:392.
- [Google Scholar]
- Brown adipose tissue activation is inversely related to central obesity and metabolic parameters in adult human. PLoS One. 2015;10:e0123795.
- [CrossRef] [PubMed] [PubMed Central] [Google Scholar]
- Chlorinated persistent organic pollutants, obesity, and type 2 diabetes. Endocr Rev. 2014;35:557-601.
- [CrossRef] [PubMed] [PubMed Central] [Google Scholar]
- Overweight and abdominal obesity as determinants of undiagnosed diabetes and pre-diabetes in Bangladesh. BMC Obes. 2016;3:19.
- [CrossRef] [PubMed] [PubMed Central] [Google Scholar]
- Prevalence of prediabetes and its associated risk factors among rural adults in Tamil Nadu. Arch Med Health Sci. 2015;3:178.
- [CrossRef] [Google Scholar]
- A case-control study on risk factors and their interactions with prediabetes among the elderly in rural communities of Yiyang City, Hunan Province. J Diabetes Res. 2019;2019:1386048.
- [CrossRef] [PubMed] [PubMed Central] [Google Scholar]
- Prevalence of prediabetes and its risk factors among the employees of Ambo University, Oromia Region, Ethiopia. Res Mol Med (RMM). 2017;5:11-20.
- [CrossRef] [Google Scholar]
- Are body mass index and waist circumference significant predictors of diabetes and prediabetes risk: Results from a population based cohort study. World J Diabetes. 2017;8:365-73.
- [CrossRef] [PubMed] [PubMed Central] [Google Scholar]
- Obesity and associated type 2 diabetes and hypertension in factory workers of Bangladesh. BMC Res Notes. 2015;8
- [CrossRef] [Google Scholar]
- Diabetes mellitus-induced male reproductive impairment: The role of natural products: A review. J Appl Pharm Sci. 2017;7:233-42.
- [CrossRef] [Google Scholar]
- Fluoride exposure aggravates the testicular damage and sperm quality in diabetic mice: protective role of ginseng and Banaba. Biol Trace Elem Res. 2017;177:331-44.
- [CrossRef] [PubMed] [Google Scholar]
- Testosterone level in men with type 2 diabetes mellitus and related metabolic effects: A review of current evidence. J Diabetes Investig. 2015;6:112-23.
- [CrossRef] [PubMed] [PubMed Central] [Google Scholar]
- Impact of metabolic syndrome factors on testosterone and SHBG in type 2 diabetes mellitus and metabolic syndrome. J Diabetes Res. 2018;2018:4926789.
- [CrossRef] [PubMed] [PubMed Central] [Google Scholar]
- Total testosterone is the most valuable indicator of metabolic syndrome among various testosterone values in middle-aged Japanese men. Endocr J. 2015;62:123-32.
- [CrossRef] [PubMed] [Google Scholar]
- Low testosterone levels are predictive for incidental atrial fibrillation and ischemic stroke in men but protective in women- Results from the FINRISK Study. Eur J Prev Cardiol. 2018;25:1133-39.
- [CrossRef] [PubMed] [Google Scholar]
- Testosterone levels and type 2 diabetes mellitus- no correlation with age, differential; Predictive value in men and women. Biomolecules. 2018;8:76.
- [CrossRef] [PubMed] [PubMed Central] [Google Scholar]
- The association between serum testosterone and insulin resistance: A longitudinal study. Endocr Connect. 2018;7:1491-500.
- [CrossRef] [PubMed] [PubMed Central] [Google Scholar]