Prevalence, Risk Factors and Predictors of Diabetes-related Sexual Dysfunction

Background: The prevalence and complications of diabetes are currently on the rise, so this study investigated the prevalence, risk factors and predictors of diabetic sexual dysfunction (SD). Methods: The study was cross-sectional multicentred. Patients were randomly selected from the Korle Bu Teaching Hospital, Komfo Anokye Teaching Hospital and Tamale Teaching Hospital. Socio-demographic, medical history, lifestyle and physical characteristics of subjects, as well as sexual dysfunction (SD) characteristics were investigated, using a structured questionnaire. Blood samples were also taken from subjects and analyzed for total cholesterol (TC), triglycerides (TG), high density lipoprotein-cholesterol (HDL-C), low density lipoprotein-cholesterol (LDL-C), serum creatinine and serum urea. Estimated glomerular filtration rate (eGFR) was also determined, using the serum creatinine. Statistics were performed using SPSS version 22. Results: Of the 100 people with diabetes, 31% were males and 69% were females. The mean age was 53.82±13.754 years. It was found that 54.8% of the males and 68.1% of females had diabetic SD. The prevalence of severely abnormal SD was 6.5% and 4.3% in males and females, respectively. In a univariate analysis, none of the independent variables was associated with SD in both men and women. Conclusion: This study has shown that the prevalence of SD is high among diabetics.


BACKGROUND
The prevalence of diabetes is increasing in every country; by 2035 the global prevalence is projected to rise from 8.3% to 10.1% [1]. The prevalence in Africa is projected to rise from 5.1% to 5.3% [1], whilst the prevalence in Ghana which is 3.3% [2] is also expected to rise. As the prevalence of diabetes increases, so does the prevalence of diabetic SD, as more than a third of women with diabetes experience sexual dysfunction [3] and men with diabetes are more likely to have sexual dysfunction than men without diabetes [4].
Diabetes complications lead to disability and death. As the prevalence of diabetes grows in low-and middle-income countries, so too does the impact on both human and economic terms [5]. For instance, sexual dysfunction is associated with poorer quality of life [6]. It also results in loss of physical and emotional intimacy and sometimes leads to divorce. The prevalence of diabetic sexual dysfunction in men stands at 69.3% in Ghana [7], in a study that was limited to men and was conducted only in one site, Tema General Hospital, in the Greater Accra Region of Ghana. Thus, there was the need for a multicenter study for both sexes, on the prevalence, risk factors and predictors of diabetic sexual dysfunction in Ghana to help delineate preventive strategies to lessen the burden of the complications.

SUBJECTS AND METHODS
The study was a cross-sectional multicenter study conducted from 15th June to 30 th  During the physical examination, findings were confirmed with the medical records of the patients or the consultant physician on duty and were reported as present or absent, without further description or grading. The body mass index (BMI) was used to assess the nutritional status of the patients. Height (m) was measured without shoes, using a microtoise (Seca, Germany) and weight (kg) was measured in light clothing, using a uniscale (Seca, Germany). The body mass index (BMI) was determined by dividing the weight (kg) by the square of the height (m 2 ) and was classified as underweight (<18.5 kg/m 2 ), normal (18.5-24.99 kg/m 2 ), overweight (25)(26)(27)(28)(29).99 kg/m 2 ) and obese (≥30 kg/m 2 ). Blood pressure was measured using a digital sphygmomanometer (Omron, Japan). Before blood pressure measurements, every patient rested for at least 10 minutes. High blood pressure was defined as systolic blood pressure (mmHg) ≥ 140 mmHg and/or diastolic blood pressure ≥ 90 mmHg or known hypertensive on treatment.
Sexual dysfunction was measured, using the Golombok Rust Inventory of Sexual Satisfaction (GRISS) questionnaire. The GRISS questionnaire is for the assessment of the existence and severity of sexual problems. The GRISS is a standardized questionnaire and also easy to administer. The reliability of the scales is 0.94 for males and 0.87 for females [9]. Its validity has also been proven [9]. All the questions were answered using a five-point scale (always, usually, occasionally, hardly ever and never).
The male version of the questionnaire gives a total male score, as well as subscales of impotence, non-communication, premature ejaculation, avoidance, infrequency, nonsensuality, and dissatisfaction. The female version also gives a total female score, as well as subscales of anorgasmia, vaginismus, noncommunication, infrequency, female nonsensuality, female avoidance, and female dissatisfaction. Responses were added up to give a total score. The total scores were transformed, using a standard nine point scale. Scores of five or more are considered to indicate sexual dysfunction and scores of eight or more are considered to indicate severe sexual dysfunction [9]. Findings were reported as present or absent.

Statistical Analysis
Data entry and analysis were done using SPSS 22 software (IBM, USA). For the univariate analysis, the Pearson correlation (chi-square) or Fisher's exact test was used for categorical variables, whilst student t-test was used for continuous variables. The independent variables that were significant in the univariate analysis were considered for multivariate binary logistic regression analysis so as to control for confounder risk factors. P< 0.05 was considered significant at two tailed tests. Percentages and cross tabulations were used to show respondents' responses.

Ethical Considerations
Permission to conduct the study was obtained from the Committee on Human Research, Publications and Ethics of the Kwame Nkrumah University of Science and Technology (CHRPE/AP/228/15).
The consent of respondents was sought and they were assured of the confidentiality of the information provided.

RESULTS
A total of 100 diabetics attending three teaching hospitals in Ghana were enrolled into the study. The sex distribution of the study participants was 31% males and 69% females. The overall mean age of the diabetics was 53.8±13.8 years.
Difference in prevalence of SD in males and females was not significant (p= 0.257). Sociodemographic characteristics of the subjects are shown in Table 1.

Prevalence and Severity of Sexual Dysfunction in Male Diabetics
Of the 31 male diabetics who completed the sexual dysfunction questionnaire, 17 (54.8%) had sexual dysfunction, out of which 2 (6.5%) had severe sexual dysfunction. Regarding the sexual dysfunction domains, it was observed that premature ejaculation (p= 0.036), noncommunication (p= 0.008), non-sensuality (p=0.049) and dissatisfaction (p=0.011) were significantly correlated with sexual dysfunction, whilst impotence (p=0.092), infrequency (p= 0.124) and avoidance (p= 0.224) were insignificantly correlated with sexual dysfunction (Table 2). Impotence (12.9%), premature ejaculation (12.9%) and non-communication (12.9%) had higher level of severity than the other four domains of sexual dysfunctions as shown in Table 3. In the univariate analysis, none of the potential risk factor was associated with SD in male diabetics (Table 4).

Prevalence and Severity of Sexual Dysfunction in Females
The female participants were assessed using the seven domains for measuring sexual dysfunction which included anorgasmia, vaginismus, noncommunication, infrequency, female avoidance, female non-sensuality and female dissatisfaction, to determine sexual dysfunction prevalence. The prevalence of sexual dysfunction among the female diabetics was 68.1%, out of which 4.3% had severely abnormal sexual dysfunction.
As shown in Table 5, Anorgasmia (p= 0.001), vaginismus (p=0.000), non-communication (p=0.000), female avoidance (p=0.000), female non-sensuality (p=0.032) and female dissatisfaction (p=0.000) were significantly associated with sexual dysfunction whilst infrequency (p=0.089) was insignificantly associated with sexual dysfunction. Noncommunication (13%) had a higher level of severity than the other six domains of sexual dysfunctions as shown in Table 6. In the univariate analysis, none of the potential risk factor was associated with SD in female diabetics (Table 7).

DISCUSSION
Sexual dysfunction prevalence among male diabetics was found to be 54.8% in the present study, which is lower than that observed from studies done in Ghana by Owiredu et al. [7] where the prevalence was found to be 69.3%. This could be as a result of the fact that the present study was a multicenter study and also investigated a small sample size, as compared to that of Owiredu et al. [7]. However, the prevalence observed in the present study is comparable to that of Unadike et al. [10] in Nigeria, where the prevalence was reported to be 58% among diabetic males. In the present study, 6.5% of the male diabetics had severe sexual dysfunction, as compared to 4.7% in the study by Owiredu et al. [7]. Regarding severity (severely abnormal) of the sexual dysfunction domains, impotence (12.9%), premature ejaculation (12.9%), non-communication (12.9%) and dissatisfaction (12.9%) recorded the highest in the present study (Table 3) as compared to 15.8% recorded for impotence in the study by Owiredu et al. [7]. Furthermore, premature ejaculation, non-sensuality, avoidance and dissatisfaction were significantly related to sexual dysfunction in the present study as compared to infrequency, non-communication, non-sensuality, dissatisfaction, and impotence in the study by Owiredu et al. [7]. Despite having a high prevalence, impotence did not significantly relate to sexual dysfunction in this study. This is due to the fact that the difference in sexual dysfunction prevalence between diabetics with and without impotence was not significant (Table 2). It is worth emphasizing that despite the smaller sample size of our study, its multicentre design makes the results reliable, as the data was derived from different hospitals in Ghana, making it unlikely for a coincidental factor in one hospital affecting the results.
With regard to female sexual dysfunction in diabetics, the present study is the first to investigate the problem among female diabetics in Ghana. Our observed prevalence of 68.1% is comparable to a prevalence of 73.2%, reported by Singh et al. [11] in India. However, the prevalence reported in the present study is higher than prevalence obtained from other countries.  For instance, studies in Jordan, Italy and Kenya reported prevalences of 59.6%, 53.4% and 36% respectively [12,13,14]. The variation in prevalence observed above could probably be due to differences in methodological and population characteristics. Regarding the female sexual dysfunction domains, the most prevalent areas of difficulty were dissatisfaction (92.7%), vaginismus (86%), anorgasmia (82.2%), avoidance (82%), non-sensuality (80%), noncommunication (79.6%) and infrequency (74%). In terms of severity of sexual dysfunction, 4.3% had severe dysfunction and the most severe area was non-communication (13%). The high prevalence of sexual dysfunction in the female diabetics observed in the present study was expected, in that the hospitals of study were referral centres hence receive all serious and chronic illnesses, including diabetes complications. In our setting, sexual issues are treated discreetly and confidentially, so it is serious cases that will compel people to open up to disclose their pent-up sexual frustrations, as seen in referral hospitals.
Among the subjects, BMI did not relate to sexual dysfunction. Similar to the finding of the present study, Vafaeimanesh et al. [15] had shown that obesity is not correlated with sexual dysfunction in women. Vafaeimanesh et al. [15] had also reported there was no link between obesity and sexual dysfunction in men. In contrast, Owiredu et al. [7] reported that greater body weight (obesity) is a predictor of sexual dysfunction in diabetic men. Obesity has been known to cause sexual dysfunction because of its association with dyslipidemia, the main cause of ischemia, but this is not conclusive [16], hence the finding of the present study. The effect of hypertension on sexual dysfunction was not significant in the current study. This finding is consistent with other studies [13,17]. On the contrary, Sharifi et al. [18] and Peter et al. [19] showed an association between hypertension and sexual dysfunction in diabetics, possibly because of different population characteristics [20]. . On the other hand, the association between duration of diabetes and neuropathy has been reported to be very negligible [20] and thus justifies the finding of the present study. In line with the current study, Ziaei-Rad et al. [17] found no relation between duration of diabetes and sexual dysfunction in both genders. This finding was also supported by Esposito et al. [13] and Omidvar et al. [21] who found that duration of diabetes has no correlation with sexual dysfunction in women.
Poor vision was not correlated with diabetic sexual dysfunction in this study. In contrast, Henis et al. [22] identified impaired vision as a predictor of sexual dysfunction in men. Similarly, Ali et al. [12] identified poor vision as a significant risk factor for sexual dysfunction in diabetic women. The diagnostic criteria for poor vision in the previous studies differed from the current study, hence the difference in outcome [23]. For example, poor vision is relative and can vary in duration and extent; this can produce varying effects on sexuality.
Studies by Chernyshova et al. [24] and Copeland et al. [25] showed a relationship between renal dysfunction and sexual dysfunction in diabetics.
On the other hand, renal dysfunction was not associated with sexual dysfunction in both men and women in the present study, possibly because of the different methodology employed [26,27].
Our study has shown that smoking is not a significant risk factor for sexual dysfunction. This finding agrees with a previous study among diabetic men by Mutagaywa et al. [28]. Similarly, Ali et al. [12] and Esposito et al. [13] also found no relationship between smoking and sexual dysfunction in diabetic women. Smoking is known to cause sexual dysfunction via its nicotine content [29], but a study by Premalatha et al. [16] proved otherwise, hence the finding of the present study.
The effect of high lipids concentration in the development of atherosclerosis which often results in sexual dysfunction is yet to be fully justified [16]. This may explain why there was no difference between diabetics with sexual dysfunction and diabetics without sexual dysfunction in terms of dyslipidemia in the present study. This finding has been confirmed by Sharifi et al. [18] and Mutagaywa et al. [28] who revealed that dyslipidemia is not a significant risk factor for sexual dysfunction in diabetic men. Similarly, Ali et al. [12] showed no correlation between dyslipidemia and sexual dysfunction in women. Alcohol intake and sexual dysfunction in diabetics were not found to be correlated in this study. A similar finding was reported by Mutagaywa et al. [28] in a study of diabetic men. Peter et al. [19] also reported no correlation between alcoholism and sexual dysfunction in men. In women, alcohol has been shown to cause sexual dysfunction [30] but no clinical study has been identified linking alcoholism to sexual dysfunction in women. Theoretically, alcohol is said to be linked to the development of neuropathy, a major cause of sexual dysfunction in diabetics, but this is inconclusive [31], hence the finding of the present study.
The present study found no link between age and sexual dysfunction in both genders. This finding was supported by Ziaei-Rad et al. [11] who found no relation between age and sexual dysfunction in both genders. Similar to the present study, Omidvar et al. [21] showed no relationship between age and sexual dysfunction in diabetics. In contrast, it was revealed in a study by Esposito et al. [13] that age and female sexual dysfunction are correlated, possibly because of the different methodology employed [20].

CONCLUSION
The study showed that the prevalence and severity of sexual dysfunction is high among diabetes patients. None of the independent variables is predictive of sexual dysfunction in diabetics.