Prevalence of, and risk factors for erectile dysfunction in male type 2 diabetic outpatient attendees in Enugu, South East Nigeria.

Context: Erectile dysfunction (ED) is a strong predictor of poor quality of life in men with type 2 Diabetes mellitus (T2DM). Several studies evaluating ED in men with diabetes mellitus have been carried out, but few of these have been done in Nigeria. In Enugu, South East Nigeria, paucity of studies on this subject was observed. Aims: This study aims to determine the prevalence and predictors of ED in men with T2DM attending the diabetes clinics. Settings and Design: A descriptive cross-sectional study of men with T2DM in UNTH and Saint Mary's Hospital, Enugu, was carried out. The systematic sampling method was used to recruit participants. Subjects and
Methods: Data collection from participants and their hospital records was done using semi-structured questionnaire. ED was assessed using the 5 items, international index of erectile function questionnaire. Statistical Analysis Used: Data analysis was done using SPSS version 20 and results presented as texts and tables. P value was set at <0.05.
Results: A total of 325 participants with mean age of 57.8 ± 13.2 years were involved out of which 94.7% had ED. The proportion of participants with ED had increased with its severity. Predictors of ED included poor glycemic control, longer duration of diabetes, overweight/obesity, and older age. Poor ED health-seeking behavior and treatment were noted. Conclusions: The prevalence of ED is high. Lifestyle interventions targeted at improving glycemic control and weight loss may reduce the burden of this complication. We recommend objective ED screening using standard but brief instruments as part of routine evaluation of men with T2DM.

INTRODUCTION

According to the National Institutes of Health Consensus Development Panel on Impotence, erectile dysfunction (ED) is defined as the inability to achieve and maintain an erection sufficient to permit satisfactory sexual intercourse.[1] It is classified into organic and psychogenic subtypes,[2] of which the organic subdivision is often caused by a variety of factors including diabetes mellitus (DM), hypertension, cardiovascular diseases, and hyperlipidemia.[2]

In the United States, the Massachusetts Male Aging Study had shown that the risk of ED is increased by age, lower education, diabetes, heart disease, and hypertension.[3] Researchers in West and East Africa have demonstrated that the prevalence of ED in men with DM is high.[45678]

Globally, there is evidence of a marked increase in the prevalence of type 2 diabetes mellitus (T2DM) in the developing world in the past two decades.[91011] This trend has been supported by a systematic review and meta-analysis of diabetes studies in Nigeria which indicated that the age-adjusted prevalence rates of T2DM in Nigeria among persons aged 20–79 years increased from 2% (95% CI 1.9%–2.1%) in 1990 to 5.7% (95% CI 5.5–5.8%) in 2015, accounting for over 874,000 and 4.7 million cases, respectively.[12]

ED is a strong predictor of poor quality of life in men with DM.[13] Indeed ED has been shown to be an indicator of subclinical cardiovascular disease and overall health, and physicians should refer these men for the early attention of cardiologists, urologists, and other specialists to provide holistic care and better outcomes.[141516]

Previously, T2DM has been shown to be a risk factor for ED, in community screening studies,[17] primary clinic attendees,[518] and studies in men living with diabetes.[6] There is a paucity of studies on the risk of ED in T2DM is South East Nigeria.[1920] We aimed to determine the prevalence of ED in men with T2DM attending the diabetes clinics of University of Nigeria Teaching Hospital (UNTH), Enugu and Saint Mary's Hospital both in South Eastern Nigeria.

We also aimed to assess the effect of age, duration of DM, fasting blood glucose (FBG), glycosylated hemoglobin (HBA1c), and body mass index (BMI) on the presence of ED.

SUBJECTS AND METHODS

Study setting

Our study setting was Enugu, the capital city of Enugu State in South East Nigeria. Data were collected from male attendees of endocrinology (diabetes) clinics of UNTH and Saint Mary's Hospital (SM'sH). The former is a public hospital and premier tertiary/teaching hospital in the region while the latter is a private secondary level facility offering services including Diabetology.

Enugu had an estimated population of 3,800,000 people as at the 2006 census,[21] and receives patients from the neighboring states of Abia, Anambra, Imo, Ebonyi, Rivers, Benue and Kogi States with a combined population of over 20 million people.[22]

Design

This was a descriptive cross-sectional study of male attendees of the Diabetes Clinics of UNTH and (SM'sH).

Sample size

This was calculated using the formula, Z1-α/22 P(1-P)/d2, where Z1-α/22 = 1.96, P = expected proportion in a population based on a previous study,[6] d = absolute error or precision (0.05). Substituting in the formula given, 1.962 × 0.74(1-0.74)/ 0.052 = 296 participants (minimum).

Subjects

Allowing for an attrition rate of 10%, 325 participants were recruited. A systematic sampling method was used in selecting patients for the study. Data were collected from September 2016 to December 2017.

Inclusion criteria

Men with type 2 diabetes who are sexually active and attending the study clinics were considered eligible to participate

Granting of verbal consent to participate in the study.

Exclusion criteria

Past lower urinary tract or urethral/penile surgery

History of pelvic fracture

History of spine injury or surgery

Patients on medications that affect erectile function.

Ethics

The research project protocol was assessed and ethical clearance granted by the UNTH Research Ethics Committee. Informed consent was obtained from all participants.

Procedure

Eligible male T2DM patients attending the outpatient diabetes clinics were approached by the investigators including a trained research assistant.

Probability sampling methods were used. We used a systematic sampling technique where every Kth patient was recruited. The start off point was determined by the use of a table of random numbers.

The data collection instrument was a semi-structured questionnaire which was validated on a smaller group of T2DM patients at another hospital with the aim of ensuring question clarity and consistency.

Explanation of the purpose of the study and questionnaire administration was done in private by the investigators/research assistant after consent for participation was sought and obtained. It was explained that patients were free to decline participation and that no negative consequences to care or treatment would result.

This interaction was done either in English or the patients’ native language or a combination of both, at the discretion of the research assistant based on the perception of understanding/fluency in either.

From those that consented, the following information was collected using the study questionnaire: age, blood pressure, FBG, HbA1c, height, weight, duration of DM, hypertension, and other known medical conditions, current treatment for DM, smoking, alcohol, ED treatment sought, type of ED treatment (if any), and IIEF-5 score.

From the primary data, BMI was calculated. ED was assessed using a five-item version of the International Index of Erectile Function as described by Rosen et al.[23] Within this index, responses regarding confidence in attaining an erection, quality of erection, orgasm, and satisfaction are each graded on a scale of 1–5 (1 = minimal and 5 = maximum).

Based on the previous work,[24] ED was initially grouped into five grades of severity on the basis of the IIEF-5 score, 22–25 (normal erectile function), 17–21 (mild ED), 12–16 (mild-to-moderate ED), 8–11 (moderate ED), and 5–7 (severe ED).

Subsequently, for purposes of analysis, the primary outcome variable measurement ED score was split into a binary variable where an IIEF-5 score of 22–25 was considered normal and 5–21 was considered as having ED.[24]

Data analysis

Data were analyzed using SPSS 20 (IBM SPSS Statistics for Windows, Version 20.0. Armonk, NY, USA). Continuous variables were analyzed using means and standard deviations. Categorical data were analyzed using the Chi-square test with statistical significance level set at P < 0.05. The results were presented in tables. A binary logistic regression analysis was performed using the presence of ED or otherwise as the dichotomous outcome variable; while age, duration of DM, FBG, glycosylated hemoglobin (HBA1c), and BMI were used as the explanatory variable as adapted from Malavige et al.[24] This was run using the Enter method. The statistical significance level was set at P < 0.05.

RESULTS

Demographics/sample characteristics

In all, 325 men participated in the study. Their ages ranged from 28 to 88 years with a mean age of 57.8 ± 13.2 years and a median age of 59 years. The other sample characteristics are shown in Table 1. Other findings were coexisting hypertension 192/325 (59.1%) and smoking history in 97/325 (29.9%).

Table 1

Demographic and clinical characteristics

Parameter	Mean±SD
Age (years)	57.8±13.2
Systolic blood pressure (mmHg)	123±20.45
Diastolic blood pressure (mmHg)	76.3±11.9
Fasting blood glucose (mg/dl)	168.3±85.3
HBA1c (%)	7.9±1.4
Weight (kg)	73.7±11.8
Height (cm)	167±0.78
BMI (kg/m2)	26.2±3.5
DM duration (years)	4.9±1.8
IIEF 1	2.3±1.2
IIEF 2	2.1±1.2
IIEF 3	2.03±1.1
IIEF 4	2.2±1.2
IIEF 5	2.2±1.3
Total IIEF score	10.8±5.6

SD=Standard deviation, BMI=Body mass index, DM=Diabetes mellitus, HBA1c=Glycosylated hemoglobin, IIEF=International index of erectile function

The prevalence of erectile dysfunction and potential predictors

ED, defined as a total IIEF score of ≤21, was present in 306/325 participants (94.7%) [Table 2].

Table 2

Classification of erectile dysfunction severity

Degree of ED	Number of participants, n (%)
Severe (1-7)	115 (35.4)
Moderate ED (8-11)	79 (24.3)
Mild moderate (12-16)	65 (20)
Mild (17-21)	47 (14.5)
No ED (22-25)	19 (5.8)
Total	325 (100)

ED=Erectile dysfunction

Erectile dysfunction health-seeking behaviour and drug use

Concerning ED treatment, 258/325 (79.4%) of the respondents had never sought orthodox medical treatment from a doctor regarding ED including their physicians that manage the DM. Concerning ED drug use, 67/325 (20.6%) had used drugs for ED previously, but these were not prescribed by physicians. Sources of ED drugs were herbal/alternative medicine practitioners in 20/325 (6.2%) and 47/325 (14.5%) from chemists/pharmacy shops. Types of drugs used were PDE-5 inhibitors in 47/325 (14.5%).

Logistic regression analysis

A logistic regression analysis was conducted to predict the occurrence of ED using age, FBG, HBA1c, duration of DM, BMI category, and presence of hypertension as explanatory variables using the enter method (Chi-square = 147.714, df = 8, P < .001). Cox and Snell R2 and Nagelkerke R2 were 0.381 and 0.452, respectively. The analysis revealed that the following factors were associated with the risk of developing ED: Age, FBS, HBA1c duration of DM, and BMI [Table 3].

Table 3

Summary of binary logistic regression analysis of erectile dysfunction against potential predictors

Outcome variable	Explanatory variable	OR	CI	P
Erectile dysfunction	Age	1.05	1.03-1.08	0.02
	Fasting blood sugar	1.03	1.01-1.08	0.001
	HBA1c	5.92	4.01-7.07	0.001
	Duration of DM	2.77	1.48-5.23	0.002
	BMI	2.22	1.17-3.36	0.000

DM=Diabetes mellitus, BMI=Body mass index, OR=Odds ratio, CI=Confidence interval, HBA1c=Glycosylated hemoglobin

DISCUSSION

The mean age of our sample population was 57.8 years and was similar to earlier studies done in Nigeria and Korea that reported 56.8 (±2.4) years and 53.8 (±6.65, respectively. This is probably reflective of the trend of DM predominantly affecting people in middle age.[625] This also may be explained by the fact that among the types of DM, Type 2 DM which is seen in adults still remains the most common type.

The prevalence of ED in our series was quite high (94.7%); when categorized according to severity, severe ED occurred in 35.4%, being the highest. It was also observed that the numbers of patients affect rose as the severity of ED increased from mild to severe. This high prevalence trend is similar to earlier studies in Africa that noted prevalence rates from 55.1% to 74%,[568] but our series is much higher than previously reported trends. This, therefore, suggests that ED may be on the increase among men suffering from DM and hence early and regular objective screening needs to be adopted by care providers. Some authors in the Middle East have reported prevalence rates of 83%–86.1%[2627] while their counterparts in India have reported rates up to 90.9%.[28]

These variations in ED prevalence rates may be due to a variety of factors including varying demography, population size, and severity of the disease. Others have suggested that non-elimination or non-correction of psychological factors and selection bias such as when respondents are taken from a referral center that handles possibly complicated cases of DM may increase the number of ED cases observed.[2930]

Predictors of erectile dysfunction

Several variables namely: age, glycemic control (FBG and HbA1c), duration of DM, and overweight/obesity (BMI) were all significant predictors of ED in this study [Table 3].

Age

The effect of age on ED is not surprising as rising age has been shown in several studies to be associated with ED; both in community studies as well as subpopulations with comorbidities.[6717] This is probably attributable to the rising incidence of ailments such as hypertension, diabetes, general organ decline as well as andropause which all increase with advancing age.

Diabetes mellitus duration

Our study revealed the duration of DM to be a significant predictor of the risk of ED. This has been demonstrated previously in many parts of the world including Africa the Middle East and Asia, with these authors observing this in spite of good control of DM.[617252728] These findings are thought to be due to angiopathy, neuropathy, endothelial dysfunction, hypogonadism, and psychological disorders all of which are frequently seen in DM and worsen with increasing duration of disease.[3132] This observation has been recorded by other researchers in cohorts of diabetics with good control,[3233] and may be due to the cumulative effect of relatively small periods of hyperglycemia that may occur despite “good” control.

Glycemic control

Poor glycemic control (FBG and HBA1c) as demonstrated in this study was found to significantly predict ED. This trend as has been demonstrated severally by other workers who have shown a relationship between poor glycemic control and ED.[34353637] Glycemic control measured by HbA1c seems to be the most significant predictor as shown by its high odds ratio (OR) of 5.92 in our study and 7.12 by Ugwu et al.[19] Given that the adequacy of DM control is a key factor in the prevention of microvascular (particularly), and neuropathic complications of ED, it is noteworthy that the mean FBG in our series was comparatively higher than those of other authors who have conducted similar studies.[30] Similarly, the mean HBA1c level in our series was 7.9%, previous authors have shown that these parameters tend to increase the risk of ED occurrence.[383940] This potentially offers a window of opportunity for reduction of risk of ED as well as those of cardiovascular diseases that have a similar etiology by ensuring better glycemic control. This may be achieved through better counseling, use of support groups, and involvement of close family members. In addition, the incidence of associated hypogonadism in DM is elevated by poor glycemic control, and it has also been shown that therapeutic benefit is derived by improving the glycemic control.[3141]

Overweight/obesity

Overweight/Obesity determined by BMI categories was found to be a very significant predictor of ED (OR 2.22). This is association has been demonstrated previously, Hassan et al.[42] showed that obesity as measured by BMI was a significant predictor of ED in a cohort of Saudi men with type 2 diabetes. Some authors in Pakistan, India and Sweden, respectively, have also made similar observations regarding the effect of BMI on the risk of ED[35364344] and provide a possible window for therapeutic intervention through lifestyle interventions such as exercise and dietary modification.[45]

Hypertension

Hypertension was not observed to increase the risk of ED in our series, this is contrary to the findings of some other workers.[457] This disparity may be related to the fact that these were all community-based studies that assessed the prevalence of ED in an unselected population of men. Indeed, Oyelade et al.[4] showed that when initially on univariate analysis, hypertension, and other factors were found to have significant unadjusted odds associations with ED, these became non-significant on binary logistic regression save for DM. Some other previous studies have made similar observations to ours.[619] We suggest that in the population under scrutiny hypertension may not exert a significant effect on the occurrence of ED as mean blood pressure indices suggested well-controlled blood pressure. This may possibly be due to control of complications of hypertension or limited effect/contribution to the pathogenesis of ED in this population.

Erectile dysfunction health-seeking behavior and drug use

In this study, it was instructive to note that majority (79.4%) of respondents had never raised the issue or sought treatment for ED from a doctor including those managing the DM. This worrisome trend has been observed in other climes such as Brazil, Asia, and Australia,[464748] where the majority of patients had not sought or received treatment for ED (<10% had received treatment). The reasons for this low treatment seeking rate as may be related to patient barriers to help-seeking included shame/embarrassment, culturally inappropriate services, and lack of awareness.[4849] These findings support a possible role for direct questioning or screening for ED in this at-risk population, which should to be done in private in an empathetic manner so as to gain the confidence of the patients.

Regarding ED drug treatment, only 20% of the respondents had used ED drugs previously, and these were PDE-5 inhibitors obtained over the counter without prescriptions in 14.5% and herbal/alternative medications in the remainder. This pattern of seeking informal portals to access care or medication has been observed by other workers,[495051] who noted that though the majority of patients would want treatment, only a small minority would actually voice out the request to the doctor in a routine clinic setting. Clearly, a large unmet need for ED diagnosis and treatment exists as has been reported elsewhere.[52] Possible solutions may lie in developing culturally acceptable methods of history taking and greater use of written information material and questionnaires before interaction with the physician to aid detection.[53]

CONCLUSIONS

We conclude that among this population of male participants living with T2DM that the burden of ED is not just very high but may be on the rise. The predictors of the occurrence of ED included age, duration of DM, poor glycemic control, and overweight/obesity.

Of these, the greatest impact was made by poor glycemic control (HbA1c), long duration of DM, and overweight/obesity (BMI). Two of these factors, glycemic control and weight abnormalities offer opportunities for preventive intervention by the way of lifestyle modification.

ED health-seeking behavior in the formal clinic setting was poor and the minority of patients who that had sought and received treatment, obtained such from informal sources.

Limitations

This study was limited by the cross-sectional nature of the study, reliance on hospital records for laboratory values and absence of longitudinal follow-up to allow time trends assessment.

Recommendations

We recommend objective ED screening using standard brief instruments as part of the initial and continued management of men with T2DM, more efforts at early achievement of glycemic goals through lifestyle interventions, holding of ED-related conversations in a culturally appropriate manner in private, and where possible that the patients be referred to an ED clinic for urological input.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.