The association between the severity of erectile dysfunction and left ventricular diastolic dysfunction in patients with and without cardiovascular disease

Abstract

Purpose

Erectile dysfunction (ED) is considered a microvascular disorder and serves as an indicator for the potential development of cardiovascular disease (CVD). Although left ventricular diastolic dysfunction (LVDD) reflects early myocardial damage caused by microvascular disorders, the association between ED and LVDD remains poorly elucidated.

Materials and Methods

A cross-sectional study was conducted on 123 patients with ED. They underwent RigiScan, and conventional echocardiography, and attempted International Index of Erectile Function (IIEF) questionnaire. ED severity was evaluated by measuring changes in the penile base circumference and duration of penile rigidity (≥70%) during erection. The early diastolic velocity of mitral inflow (E) and early diastolic velocity of the mitral annulus (e′) were measured using echocardiography. The patients were grouped based on the presence of CVD.

Results

Among 123 patients, 29 had CVD and 94 did not. Patients with CVD exhibited more pronounced ED and more severe LVDD. Associations between increased penile circumference with echocardiographic parameters were more prominent in patients with CVD than in those without CVD (ΔTtop and e′ wave, r=0.508 and r=0.282, respectively, p for interaction=0.033; ΔTbase and E/e′ ratio, r=-0.338 and r=-0.293, respectively, p for interaction <0.001). In the multivariate linear regression, the increase of penile base circumference was an independent risk factor for LVDD (e′, B=0.503; E/e′ ratio, B=-1.416, respectively, p<0.001).

Conclusions

ED severity correlated well with LV diastolic dysfunction, particularly in the presence of CVD. This study highlighted the potential role of ED assessment as early indicator of CVD development.

Graphical Abstract

INTRODUCTION

Erectile dysfunction (ED) means the inability to initiate or sustain penile erection which is strong enough for effective vaginal intercourse [1]. It is a widespread illness affecting approximately 150 million males worldwide [2]. Factors that cause ED might be vasculogenic, neurogenic, or psychogenic, with vasculogenic impairments being the most common. It is recognized as a microvascular ailment owing to an early disease of endothelium-dependent vasodilatation [3,4].

The nitric oxide-cyclic guanosine-3,5-monophosphate system is crucial for both the erectile process and maintenance of vascular endothelial function [5]. This system can increase diastolic compliance and shorten the time of contraction; however, it has minimal effect on the myocardial systolic function. Therefore, there is a possibility of asymptomatic coexisting myocardial dysfunction in patients with ED, especially when they have other cardiovascular risk factors [6,7].

The “artery size hypothesis” suggests a physiological mechanism that explains how ED can precede myocardial dysfunction. Despite the systemic nature of atherosclerosis, symptoms rarely appear simultaneously in different parts of the vascular system. This difference is likely due to larger blood vessels having a higher capacity to withstand the same plaque accumulation than smaller vessels. Consequently, the penile artery, with a diameter of 1–2 mm, is more likely to exhibit clinical symptoms earlier than larger vessels, such as the coronary artery [3].

Microvascular dysfunction is closely related to left ventricular diastolic dysfunction (LVDD), which is a well-known predictor of cardiovascular disease (CVD) [8]. Appropriate assessment and diagnosis of dilatation functions are crucial, and echocardiography is an important diagnostic tool for evaluating these cardiac relaxation functions [9,10]. However, the association between ED and LVDD remains poorly elucidated. This study is designed to evaluate the connection between ED and LVDD and to determine whether underlying CVD affects their correlation.

MATERIALS AND METHODS

1. Study population

This cross-sectional study evaluated 123 male patients diagnosed with ED between October 2008 and September 2018 at our hospital. The research included males aged 18 years or older who had engaged in a sexual relationship within the past 6 months and had experienced ED for at least 3 months. Physical examination data, including height, weight, body mass index (BMI), and systolic and diastolic pressures, and medical history, including hypertension, diabetes, dyslipidemia, and atrial fibrillation, were recorded. All patients were given a heart assessment by a cardiologist. CVD was defined as a composite of heart failure, ischemic heart disease, or stroke.

Exclusion criteria included history of major pelvic surgery, significant central nervous system injuries or spinal cord injuries, uncontrolled diabetes mellitus, significant renal disease, hypothyroidism, hypogonadism, and premature ejaculation. We also excluded individuals who were taking any medications that could potentially affect erectile function. Additionally, those who were currently using daily phosphodiesterase 5 (PDE5) inhibitors or had used on-demand PDE5 inhibitors within the past week were also excluded from the study.

2. Erectile dysfunction evaluation

Erectile function was evaluated with the International Index of Erectile Function (IIEF) score, which is used to evaluate ED in clinical settings. It consists of 6 questions in the erectile function domain. The scores were classified as no (26–30), mild (17–25), moderate (11–16), or severe (0–10) ED. We also used a RigiScan device (Dacomed Corp.), which is a commonly used objective test for monitoring rigidity and tumescence during erection. The device consisted of two circular loops placed around the penile tip and base to record the change in diameter during erection. This study used a provocative test where penile injection of alprostadil 10 µg was used to initiate penile erection. Patients were considered to have normal erectile function if they displayed episodes including the following criteria: penile rigidity ≥70%, an increase in tumescence at the base ≥3 cm and tip ≥2 cm, and an erectile event of ≥10 minutes.

3. Transthoracic echocardiogram

All patients had two-dimensional Doppler echocardiography utilizing a commercially available echocardiographic equipment with an M5Sc transducer (Vivid-E9; Vingmed-General Electric). Two-dimensional echocardiography images were used to quantify heart chambers. The American Society of Echocardiography recommended formula was used to calculate the left atrial (LA) diameter, pulmonary arterial systolic pressure (PASP), left ventricular (LV) mass index (LVMI), and ejection fraction (LVEF). LA diameter and LVMI provides valuable quantitative data that helps evaluate overall cardiac function. PASP is essential to diagnose pulmonary hypertension, and LVEF offers an evaluation of the heart’s ability to effectively pump oxygenated blood throughout the body, reflecting its pumping efficiency [11]. Pulsed-wave Doppler echocardiography was used to measure the mitral inflow velocity during early (E) and late filling (A) in the apical four-chamber view. Additionally, tissue Doppler imaging was used to evaluate the early diastolic mitral annular velocity (e′) of both the lateral and septal mitral annuli [10]. e′ wave and E/e′ ratio are both important echocardiographic parameters for assessment of LVDD [12].

4. Statistical analysis

Data were analyzed using IBM SPSS version 20 software (IBM Corp.). Mean±standard deviation was used to present continuous variables with normal distribution, whereas median and interquartile range were used for data without normal distribution. Categorical variables are reported as percentages. Independent sample t-tests and chi-square tests were utilized to compare two groups. Pearson’s correlation analysis was utilized to investigate the correlation between ED measurements and echocardiographic parameters. Additionally, we conducted a stepwise linear multivariate regression analysis to determine whether the variables were independent risk factors. p<0.05 was indicated to statistical significance.

5. Ethics statement

All analyses followed the guiding principles of the Declaration of Helsinki and were approved by the Institutional Review Board of Korea University Anam Hospital (approval number: 2016AN0167). Since this was a retrospective study, written informed consent was not obtained.

RESULTS

In this study group, 29 patients (23.6%) had an underlying CVD, whereas 94 patients (76.4%) did not. Table 1 displays the demographic, clinical, and medical comorbidity data of the patients with and without CVD. No statistically significant differences were observed between the two groups in terms of age, height, weight, BMI, systolic and diastolic blood pressure, fasting blood glucose, total cholesterol, serum testosterone, hypertension, and atrial fibrillation. However, patients with CVD had a higher incidence of diabetes mellitus (14 patients vs. 23 patients [48.3% vs. 24.5%]) and dyslipidemia (15 patients vs. 16 patients [51.7% vs. 17.0%]), as compared to those without CVD.

Table 1. Baseline demographic characteristics of erectile dysfunction patients according to the presence of cardiovascular disease (CVD).

Parameter	Erectile dysfunction patients		p-value
	Without CVD (n=94)	With CVD (n=29)
Age (y)	58.9±9.9	58.5±9.7	0.832
Height (cm)	167.9±5.7	169.2±6.4	0.294
Weight (kg)	70.1±11.0	71.5±9.5	0.543
Body mass index (kg/m²)	24.8±3.3	25.0±3.0	0.836
Systolic blood pressure (mmHg)	128.7±14.1	137.6±22.5	0.144
Diastolic blood pressure (mmHg)	78.5±13.6	83.9±14.5	0.266
Fasting blood glucose (mg/dL)	93.0±9.8	91.7±13.5	0.708
Total cholesterol (mg/dL)	168.6±30.8	185.0±37.6	0.443
Serum testosterone (ng/mL)	4.80±0.89	4.68±0.93	0.649
Hypertension	45 (47.9)	17 (58.6)	0.312
Diabetes	23 (24.5)	14 (48.3)	0.015
Dyslipidemia	16 (17.0)	15 (51.7)	<0.001
Atrial fibrillation	8 (8.5)	2 (6.9)	0.781

Values are presented as mean±standard deviation or number (%).

Table 2 presents the ED measurements and echocardiographic parameters of patients with and without CVD. We found that non-CVD patients had a greater increase in penile circumference while experiencing an erection both on the penile base (3.55±1.15 cm vs. 2.88±0.97 cm) and penile top (2.50±0.83 cm vs. 2.04±0.49 cm), according to the RigiScan test.

Table 2. Erectile dysfunction assessments and echocardiographic parameters of erectile dysfunction patients according to the presence of cardiovascular disease (CVD).

Parameter			Erectile dysfunction patients		p-value
			Without CVD (n=94)	With CVD (n=29)
Erectile dysfunction assessment
	IIEF score		7.69±4.25	5.34±3.70	0.124
		Mild (score 17–25)	21.66±5.53 (n=39)	20.47±4.38 (n=11)	0.264
		Moderate (score 11–16)	13.17±3.78 (n=25)	12.10±2.96 (n=6)	0.225
		Severe (score 0–10)	4.26±1.25 (n=72)	4.31±1.50 (n=16)	0.419
	ΔTbase (cm)		3.55±1.15	2.88±0.97	0.006
	ΔTtop (cm)		2.50±0.83	2.04±0.49	0.005
	Flaccid penile base		5.27±0.75	5.13±0.90	0.285
	Flaccid penile top		4.15±0.52	4.33±0.41	0.347
Echocardiographic parameters
	e' wave (cm/s)		6.55±1.71	5.80±2.00	0.041
	E/e' ratio		9.02±2.86	13.15±8.08	0.011
	LA diameter (cm)		37.19±5.87	41.22±7.49	0.003
	LV mass index (mL/m2)		116.17±21.76	139.41±40.44	0.006
	LVEF (%)		56.78±5.94	53.31±9.64	0.071
	Estimated PA systolic pressure (mmHg)		29.17±5.94	30.43±8.26	0.376

Values are presented as mean±standard deviation.

IIEF, International Index of Erectile Function; ΔTbase, increase in penile base circumference; ΔTtop, increase in penile top circumference; e′, early diastolic mitral annular velocity; E, mitral inflow velocity during early filling; LA, left atrial; LV, left ventricular; LVEF, left ventricular ejection fraction; PA, pulmonary artery.

Regarding echocardiographic parameters, we observed that CVD patients exhibited worsened e′ wave (p=0.041) and E/e′ ratio (p=0.011), as well as increased LA diameter (p=0.003) and LVMI (p=0.006). However, regarding the IIEF score, estimated pulmonary artery systolic pressure, and LVEF, there was no discernible difference between the two groups.

1. Correlation analysis

In all patients with ED, ΔTbase and ΔTtop both showed a positive correlation with the e′ wave (r=0.406, p<0.001 and r=0.340, p<0.001, respectively) and a negative correlation with the E/e′ ratio (r=-0.327, p<0.001 and r=-0.317, p<0.001, respectively). The IIEF score showed no significant correlation with these two echocardiographic factors (Fig. 1).

Fig. 1. Correlation of ΔTbase with e′ wave (A) and E/e′ ratio (B), and ΔTtop with e′ wave (C) and E/e′ ratio (D) in erectile dysfunction patients. ΔTbase, increase in penile base circumference; ΔTtop, increase in penile top circumference; e′, early diastolic mitral annular velocity; E, mitral inflow velocity during early filling.

The correlation between E/e′ ratio and ΔTbase was more pronounced in CVD patients (r=-0.338) compared to non-CVD patients (r=-0.293) (p for interaction <0.001). Similarly, the association between e′ wave and ΔTtop was stronger in CVD patients (r=0.508) than in non-CVD patients (r=0.282) (p for interaction=0.033) (Fig. 2).

Fig. 2. Correlation of ΔTbase with e′ wave (A) and E/e′ ratio (B), and ΔTtop with e′ wave (C) and E/e′ ratio (D) in with or without cardiovascular disease (CVD) patients. ΔTbase, increase in penile base circumference; ΔTtop, increase in penile top circumference; e′, early diastolic mitral annular velocity; E, mitral inflow velocity during early filling. a:p-value for the presence of a significant difference in the correlation coefficient (r).

There were no significant differences between the CVD and non-CVD groups regarding the correlation between ΔTbase and e′ wave, as well as ΔTtop and E/e′ ratio (Fig. 2).

2. Regression analysis

After conducting multivariate stepwise linear regression analysis and adjusting for age, BMI, systolic and diastolic pressure, ΔTbase, ΔTtop, and IIEF score, we observed that ΔTbase was an independent risk factor for an increased e′ wave (B=0.503; confidence interval [CI] 0.246 to 0.761; p<0.001), and decreased E/e′ ratio (B=-1.416; CI -2.152 to -0.680; p<0.001). Furthermore, we observed age as a risk factor for lowering e′ wave (B=-0.057; CI -0.087 to -0.027, p<0.001), but the coefficient was minimal (Table 3).

Table 3. Multivariate stepwise linear regression analysis results for determination of predictors of e′ wave and E/e′ ratio in erectile dysfunction patients.

Parameter		B±SE	β	95% CI		p-value
				Lower	Upper
e' wave
	ΔTbase	0.503±0.130	0.318	0.246	0.761	<0.001
	Age	-0.057±0.015	-0.312	-0.087	-0.027	<0.001
		R2=0.255, adjusted R2=0.242
E/e' ratio
	ΔTbase	-1.416±0.372	-0.327	-2.152	-0.68	<0.001
		R2=0.107, adjusted R2=0.100

Regression analysis was performed in parameters among ΔTbase, ΔTop, International Index of Erectile Function score, age, body mass index, systolic blood pressure, and diastolic blood pressure.

e′, early diastolic mitral annular velocity; E, mitral inflow velocity during early filling; ΔTbase, increase in penile base circumference; B, unstandardized coefficients; SE, standard error; β, standardized coefficients; CI, confidence interval.

DISCUSSION

In this study, we observed a notable impairment in LV diastolic function ED patients. Furthermore, the severity of ED was found to be strongly related to the e′ wave and E/e′ ratio, both of which are reliable indicators of diastolic function. Moreover, our analysis revealed that ΔTbase, an objective evaluation factor for ED, was independently identified as a major risk factor for developing LVDD.

Although ED severity is typically evaluated using the IIEF, we attempted to supplement it with the objective measurement of RigiScan [13]. In our study, provocation was initiated with a penile injection of alprostadil, which is commonly used in the clinical setting for ED treatment. The provocative RigiScan test does not have a specific cut-off value indicated, in comparison to the nocturnal tumescence test, in which ΔTbase ≥3 cm and ΔTtop ≥2 cm is considered be normal [14,15]. Therefore, we evaluated the degree of increase in tumescence and its correlation with echocardiographic factors rather than using a cutoff value.

Atherosclerosis is a systemic disease in which major vascular areas are affected to the same arterial lesions. Therefore, patients with arterial ED are more likely to have CVD regardless of cardiac symptoms [16]. However, the symptoms do not typically manifest simultaneously. This difference is attributed to the circumference of the arteries supplying the different vascular areas. Larger vessels are believed to tolerate the same amount of vascular wall proliferation better than smaller vessels, which may account for these differences [3,17].

Systolic blood pressure and pulse pressure both rise as arterial stiffness rises. This leads to increased oxygen demand in the cardiac muscles, resulting in hypertrophy of left ventricle, increased cardiac wall pressure, and myocardial ischemia, ultimately impairing LV diastolic function [18]. This mechanism is thought to be strongly linked to the heart failure with preserved ejection fraction [19].

LVDD is known as a powerful predictor of CVD, and it was evaluated using echocardiography in this study. Two key parameters, the e′ wave and E/e′ ratio, play a significant role in this assessment [12]. The e′ wave reflects myocardial relaxation, and its decreased velocity indicates impaired relaxation. The E/e′ ratio, which is calculated by dividing the transmitral E-wave velocity by the e′ wave velocity, estimates LV filling pressures. Elevated E/e′ ratios suggest impaired diastolic function [10,20].

In our study, the ΔTbase and ΔTtop evaluated by RigiScan showed a positive correlation with the e′ wave and a negative correlation with the E/e′ ratio. This association was even more pronounced in patients with CVD regarding the correlation between ΔTbase and E/e′ ratio, as well as ΔTtop and e′ wave. Through multivariate stepwise linear regression analysis, ED was identified as an independent risk factor for LVDD. This suggest that as the severity of ED worsens, so does LVDD.

The reason behind this can be explained by the “artery size hypothesis.” According to this theory, we can infer that arterial ED is more prevalent in patients with CVD because atherosclerosis affecting the penile artery occurs before its manifestation in the coronary artery. Nevertheless, in the non-CVD group, excluding neurogenic or psychogenic factors as potential causes of ED was challenging. Hence, the correlation with LVDD is more significant among patients with CVD who exhibit a relatively higher prevalence of arterial ED.

Our study findings align with prior research that has demonstrated a correlation between ED and LVDD [17,21], and this correlation has also been recently affirmed through a systematic review [22]. However, previous studies predominantly relied on the assessment of ED through only the IIEF scores, which could be a subjective patient assessment. In an effort to enhance the validity of our analysis, we incorporated objective quantitative measurements, utilizing RigiScan—a distinct modality not extensively utilized in earlier research.

Our study has some limitations. It had a retrospective design that divided patients showing ED into groups based on the presence of CVDs. However, this requires validation in randomized clinical trials. The number of patients included in the study was reasonable; however, to detect the actual relevance, it is necessary to enhance the validity of the research by including a larger number of patients. Also, while the RigiScan is a dependable tool for distinguishing between organic and psychogenic ED, there was a lack of literature demonstrating a correlation between this and the degree of ED. Finally, the notable absence of penile duplex sonography, a dependable diagnostic modality, underscores the necessity for its inclusion in future studies to enhance results.

CONCLUSIONS

The severity of ED is closely associated with echocardiographic indicators of LV diastolic function, with a particularly stronger correlation observed in patients with CVD. Furthermore, utilizing RigiScan in the evaluation of ED, the circumferential change of the penile base has been confirmed as an independent risk factor. This study emphasized the potential role of ED evaluation as a precursor to the CVD development.