The correlation between highly sensitive C-reactive protein levels and erectile function among men with late-onset hypogonadism

Abstract

We investigated the correlation between highly sensitive C-reactive protein (hs-CRP) levels and erectile function, and assessed the clinical role of hs-CRP levels in men with late-onset hypogonadism (LOH) syndrome. For 77 participants, we assessed Sexual Health Inventory for men (SHIM) score, Aging Male Symptoms (AMS) score and International Prostate Symptom Score (IPSS). We also evaluated free testosterone (FT), hs-CRP, total cholesterol, triglyceride levels, high density lipoprotein cholesterol, hemoglobin A1c, body mass index, waist size and blood pressure. We attempted to identify parameters correlated with SHIM score and to determine the factors affecting cardiovascular risk based on hs-CRP levels. A Spearman rank correlation test revealed that age, AMS score, IPSS and hs-CRP levels were significantly correlated with SHIM score. Age-adjusted analysis revealed that hs-CRP and IPSS were the independent factors affecting SHIM score (r= −0.304 and −0.322, respectively). Seventeen patients belonged to the moderate to high risk group for cardiovascular disease, whereas the remaining 60 belonged to the low risk group. Age, FT value and SHIM score showed significant differences between the two groups. A multivariate regression analysis demonstrated that SHIM score was an independent factor affecting cardiovascular risk (OR: 0.796; 95%CI: 0.637–0.995).

Keywords:

• Late-onset hypogonadism
• erectile dysfunction
• highly sensitive C-reactive protein
• cardiovascular disease

Introduction

Serum testosterone levels decrease in men with aging at the rate of 1–2% per year after the age of 40. The clinical condition of testosterone decrease with specific symptoms has been widely accepted as late-onset hypogonadism (LOH) syndrome [1]. The Massachusetts Male Aging Study demonstrated that prevalence of LOH syndrome in men between the ages of 40 and 69 ranged from 6.0% to 12.3% and 2.4 million men in the United States were estimated to be affected [2]. Widely recognized clinical signs of LOH syndrome are erectile dysfunction (ED), decreases in muscle mass and strength, anemia, obesity, osteoporosis, depression and deterioration of insulin resistance [3–5]. Therefore, LOH syndrome is a clinical condition that could affect the function of multiple organ systems, and LOH syndrome itself is an important sign of many other potentially serious conditions. Testosterone replacement therapy (TRT) is a widely accepted treatment in the prevention and amelioration of many of the symptoms and conditions associated with LOH syndrome [5].

C-reactive protein (CRP) level is a commonly used marker for systemic inflammation, and can usually be measured with a lower limit of 0.1 mg/dl for diagnosis of acute inflammation. On the other hand, a high-sensitivity CRP (hs-CRP) test can measure lower levels of CRP with a sensitivity of 0.01 mg/dl, and is a useful marker for the detection of chronic low-grade inflammation [6]. Chronic low-grade inflammation is involved in the pathophysiology of a large number of conditions, including dementia, cardiovascular disease (CVD), chronic obstructive pulmonary disease (COPD), osteoporosis and diabetes mellitus [7–11]. Indeed, recent studies showed that higher levels of CRP are associated with chronic heart failure, CVD, asthma and pulmonary hypertension caused by COPD [9–11]. In particular, arteriosclerosis is a chronic endothelial inflammation, and hs-CRP may be a predictor of developing CVD by detecting minimal systemic inflammation [12].

Hs-CRP levels are also a potential marker for penile vascular disease severity and ED [13]. To date, there have been few studies assessing the correlation between hs-CRP value and erectile function. In particular, hs-CRP levels may correlate with clinical symptoms of hypogonadal men who are at risk for CVD. This can be useful information for clinical management of LOH syndrome. In this study, we investigated the correlation between hs-CRP levels and erectile function, and assessed the clinical role of hs-CRP levels in men with LOH syndrome.

Patients and methods

Study subjects and protocols

Seventy-seven patients (age 50 years or older) diagnosed with LOH syndrome were enrolled in this study. A biochemical diagnosis of hypogonadism was made based on Japanese criteria as follows: free testosterone (FT) < 8.5 pg/ml, TRT is the first choice of treatment for hypogonadism; FT 8.5–11.8 pg/ml, TRT is the relative choice [14].

At the initial visit, all patients were tested for FT values and then they completed some questionnaires, such as the Sexual Health Inventory for men (SHIM) score, Aging Male Symptoms (AMS) score and International Prostate Symptom Score (IPSS). FT value was evaluated using blood serum collected between 09:00 and 11:00. Separated serum samples were stored at −20 °C until assay, and FT value was measured by radioimmunoassay using the DPC Free Testosterone Kit (Mitsubishi Kagaku Iatron, Tokyo, Japan).

In addition, body mass index (BMI), waist size and blood pressure (BP) were measured in all patients. Waist size was measured at the midpoint between the inferior costal margin and the superior border of the iliac crest on the midaxillary line by a single investigator. BP was measured after resting for >15 min using an automatic sphygmomanometer. Blood biochemical data, including, hs-CRP, total cholesterol (T-Chol), triglyceride (TG), high density lipoprotein cholesterol (HDL-Chol), hemoglobin A1c (HbA1c) and prostate specific antigen (PSA), were also evaluated. Serum hs-CRP levels were analyzed by latex-enhanced nephelometry assay.

The following patients were excluded in this study: patients who had been previously administrated antiandrogen agents, finasteride, phosphodiesterase-5 (PDE-5) inhibitors or testosterone agents within six months; patients with any present uncontrolled malignancies; patients with a past or present history of chronic illness such as tuberculosis, rheumatoid arthritis other autoimmune disorders. In patients with any confirmed inflammatory disorders after initial blood analysis, an additional blood sample was taken after cure of their inflammatory disease for hs-CRP measurement.

The Kanazawa University Hospital Institutional Review Board approved the present study, and all subjects gave informed written consent before participation in the study.

Statistical analysis

Spearman’s correlations were used to analyze any correlation between SHIM score and the following: age, FT value, hs-CRP, AMS score, IPSS, BMI, waist size, BP and blood biochemical data.

In addition, all patients were divided into two groups according to the cardiovascular risk assessment reported based on hs-CRP levels. Patients with ≥0.12 mg/dl hs-CRP levels were assigned to the moderate to high risk group for developing CVD (Group A) and patients with <0.12 mg/dl were assigned to the low risk group (Group B) [15]. The background of each group was compared using a Mann–Whitney U-test. A multivariate logistic analysis was performed to determine any background factors that correlated with low versus moderate to high cardiovascular risk and hs-CRP levels.

All analyses were performed with SPSSTM version 17.0 Medical Model (SPSS Inc., Chicago, IL), and differences were considered significant at p < 0.05.

Results

The mean age of the patients was 65.1 ± 8.3 years (range 52–82 years), and the mean FT value was 7.2 ± 2.2 pg/ml (range 0.9–10.5 pg/ml). The mean hs-CRP levels were 0.070 ± 0.062 mg/dl (range 0.00–0.30).

The correlation coefficients between SHIM score and each parameter are shown in Table 1. The critical factors significantly correlated with SHIM score were age (r= −0.246, p = 0.027), AMS score (r= −0.320, p = 0.006), IPSS (r= −0.350, p = 0.003) and hs-CRP levels (r= −0.229, p = 0.037). Since age showed a significant positive correlation with some factors such as hs-CRP levels, FT, HbA1c and TG values, each factor was controlled for age (Table 2). Age-adjusted analysis revealed that hs-CRP and IPSS were the independent factors affecting SHIM score (r= −0.304 and −0.322, respectively). An inverse correlation with SHIM score was observed in hs-CRP (r= −0.304, p = 0.009) and IPSS (r= −0.322, p = 0.006).

Table 1. The correlations between SHIM score and each parameter based on Spearman rank test.

Variables		SHIM	Age	FT	PSA	BMI	Weist	HbA1c	AMS	IPSS	hs-CRP	SBP	DBP	T-Chol	TG	HDL
SHIM	r	1.000	−0.246*	0.084	−0.105	−0.124	−0.042	0.093	−0.320*	−0.350*	−0.229*	−0.103	−0.040	0.100	0.132	0.091
	p values		0.027	0.257	0.208	0.178	0.375	0.242	0.006	0.003	0.037	0.224	0.389	0.249	0.185	0.269
Age	r		1.000	−0.316*	0.224*	−0.051	0.009	0.052	−0.155	0.157	0.287*	0.082	−0.088	0.045	−0.239*	0.187
	p values			0.003	0.025	0.334	0.470	0.332	0.098	0.099	0.006	0.248	0.239	0.367	0.033	0.076
FT	r			1.000	−0.095	−0.128	−0.104	0.166	0.000	−0.089	−0.243*	0.048	0.000	−0.072	0.156	−0.192
	p values				0.205	0.140	0.188	0.080	0.498	0.234	0.017	0.346	0.499	0.293	0.116	0.070
PSA	r				1.000	−0.071	−0.034	−0.053	−0.078	0.043	0.069	−0.032	−0.008	0.013	−0.309*	0.108
	p values					0.276	0.384	0.327	0.260	0.362	0.276	0.397	0.473	0.462	0.008	0.205
BMI	r					1.000	0.260*	−0.159	0.078	−0.069	0.108	0.288*	0.315*	0.138	0.145	−0.004
	p values						0.014	0.096	0.264	0.292	0.182	0.008	0.005	0.147	0.135	0.488
Weist	r						1.000	0.093	−0.019	−0.096	0.235*	−0.030	0.122	−0.053	−0.067	−0.105
	p values							0.221	0.440	0.221	0.021	0.402	0.163	0.347	0.308	0.215
HbA1c	r							1.000	0.096	−0.053	0.008	0.214*	0.052	−0.048	−0.049	−0.179
	p values								0.218	0.337	0.473	0.041	0.342	0.361	0.356	0.090
AMS	r								1.000	0.372*	0.040	−0.121	−0.173	0.015	0.079	0.059
	p values									0.001	0.370	0.169	0.089	0.457	0.281	0.330
IPSS	r									1.000	−0.075	−0.190	−0.167	−0.071	0.019	−0.003
	p values										0.271	0.068	0.101	0.304	0.444	0.491
hs-CRP	r										1.000	−0.150	−0.084	0.003	−0.180	0.163
	p values											0.106	0.249	0.490	0.085	0.107
SBP	r											1.000	0.403*	0.116	−0.134	−0.018
	p values												0.0003	0.193	0.158	0.447
DBP	r												1.000	0.056	−0.005	0.033
	p values													0.341	0.486	0.406
T-Chol	r													1.000	0.378*	0.391*
	p values														0.001	0.001
TG	r														1.000	−0.261*
	p values															0.022
HDL	r															1.000
	p values

SHIM: Sexual Health Inventory for men; FT: free testosterone; PSA: prostate specific antigen; BMI: body mass index; HbA1c: hemoglobin A1c; AMS: Aging Male Symptoms; IPSS: International Prostate Symptom Score; hs-CRP: highly sensitive C-reactive protein; SBP: systolic blood pressure; DBP: diastolic blood pressure; T-chol: total cholesterol; TG: triglyceride; HDL-chol: high density lipoprotein cholesterol.

* Statistically significant (p < 0.05).

Table 2. The correlations between SHIM score and each parameter based on age-adjusted analysis.

		SHIM	FT	PSA	BMI	Weist	HbA1c	AMS	IPSS	hs-CRP	SBP	DBP	T-Chol	TG	HDL
SHIM	r	1.000	0.024	−0.037	−0.183	0.022	0.036	−0.181	−0.322*	−0.304*	−0.041	−0.017	0.141	0.031	0.030
	p values		0.429	0.387	0.087	0.434	0.393	0.111	0.006	0.009	0.382	0.453	0.173	0.419	0.420
FT	r		1.000	0.023	−0.123	−0.163	0.240*	0.000	−0.022	−0.154	0.145	0.025	−0.049	0.085	−0.047
	p values			0.421	0.151	0.082	0.021	0.499	0.429	0.092	0.116	0.420	0.356	0.260	0.361
PSA	r			1.000	−0.063	−0.045	−0.130	−0.002	−0.087	0.108	−0.071	−0.077	−0.049	−0.251*	−0.002
	p values				0.299	0.352	0.138	0.494	0.240	0.177	0.280	0.267	0.356	0.027	0.493
BMI	r				1.000	0.240*	−0.119	0.055	−0.076	0.145	0.208	0.332*	0.215	0.144	0.217*
	p values					0.023	0.166	0.329	0.276	0.113	0.043	0.003	0.051	0.138	0.049
Weist	r					1.000	0.149	−0.026	−0.203	0.146	−0.035	0.096	−0.026	0.216	−0.096
	p values						0.110	0.417	0.051	0.108	0.388	0.222	0.424	0.053	0.240
HbA1c	r						1.000	0.072	−0.082	−0.020	0.278*	0.076	−0.162	−0.086	−0.162
	p values							0.282	0.259	0.434	0.012	0.277	0.115	0.263	0.114
AMS	r							1.000	0.308*	0.143	−0.062	−0.148	−0.014	0.043	0.038
	p values								0.005	0.118	0.313	0.127	0.459	0.375	0.389
IPSS	r								1.000	−0.064	−0.234*	−0.184	−0.091	−0.004	−0.114
	p values									0.303	0.034	0.081	0.257	0.490	0.206
CRP	r									1.000	−0.205*	−0.098	−0.110	−0.148	0.191
	p values										0.044	0.216	0.203	0.132	0.073
SBP	r										1.000	0.431*	0.092	−0.075	−0.076
	p values											<0.001	0.247	0.289	0.288
DBP	r											1.000	0.007	−0.111	−0.036
	p values												0.481	0.212	0.398
T-Chol	r												1.000	0.276*	0.456*
	p values													0.017	<0.001
TG	r													1.000	−0.184
	p values														0.081
HDL	r														1.000
	p values

SHIM: Sexual Health Inventory for men; FT: free testosterone; PSA: prostate specific antigen; BMI: body mass index; HbA1c: hemoglobin A1c; AMS: Aging Male Symptoms; IPSS: International Prostate Symptom Score; hs-CRP: highly sensitive C-reactive protein; SBP: systolic blood pressure; DBP: diastolic blood pressure; T-chol: total cholesterol; TG: triglyceride; HDL-chol: high density lipoprotein cholesterol.

* Statistically significant (p < 0.05).

In addition, all patients were divided into two cardiovascular risk groups based on hs-CRP levels. Of 77 participants, 17 (22%) patients belonged to the moderate to high risk group (Group A), whereas the remaining 60 (78%) had a low risk for cardiovascular events (Group B). A comparison of background parameters between Groups A and B demonstrated that age, FT value and SHIM scores were significantly different between the two groups (Table 3). In a multivariate regression analysis, SHIM score was an independent factor affecting cardiovascular risk with an odd’s ratio of 0.796 (95% confidence interval [CI], 0.637–0.995) (Table 4). No other factors showed a significant correlation with cardiovascular risk.

Table 3. Comparison in each patients' background between Groups A and B.

	Group A	Group B
	(n = 17)	(n = 60)	p values
hs-CRP value (mg/dl)	0.176 ± 0.059	0.049 ± 0.035
Age (years)	69.3 ± 6.8	64.2 ± 8.3	0.0273
FT value (pg/ml)	5.8 ± 2.4	7.5 ± 2.0	0.00581
PSA value (mg/dl)	0.90 ± 0.55	0.78 ± 0.41	0.176
BMI	24.9 ± 3.7	23.6 ± 3.1	0.111
Weist size (cm)	90.0 ± 14.5	87.5 ± 9.4	0.231
SBP (mmHg)	121 ± 11	131 ± 15	0.102
DBP (mmHg)	73 ± 11	76 ± 11	0.293
AMS score	43.0 ± 5.7	38.5 ± 10.7	0.0921
IPSS	10.9 ± 9.9	9.3 ± 7.0	0.271
SHIM score	7.1 ± 3.9	12.1 ± 5.7	0.0165
T-chol (mg/dl)	170 ± 50	180 ± 30	0.222
TG (mg/dl)	87 ± 35	124 ± 103	0.147
HDL-chol (mg/dl)	65 ± 32	54 ± 12	0.0914
HbA1c (%)	6.2 ± 0.7	6.4 ± 1.1	0.139

hs-CRP: highly sensitive C-reactive protein; FT: free testosterone; PSA: prostate specific antigen; BMI: body mass index; SBP: systolic blood pressure; DBP: diastolic blood pressure; AMS: Aging Male Symptoms; IPSS: International Prostate Symptom Score; SHIM: Sexual Health Inventory for men; T-chol: total cholesterol; TG: triglyceride; HDL-chol: high density lipoprotein cholesterol; HbA1c: hemoglobin A1c.

Table 4. A multivariate regression analysis for factors affecting CVD risk based on hs-CRP levels.

Variables	Odd's ratio	95% confidence interval	p values
Age (years)	0.953	0.810–1.121	0.56
FT value (pg/ml)	0.770	0.288–2.058	0.602
AMS score	0.954	0.850–1.072	0.432
SHIM score	0.796	0.637–0.995	0.045
HDL-chol (mg/dl)	0.999	0.984–1.015	0.91

CVD: cardiovascular disease; hs-CRP: highly sensitive C-reactive protein; FT: free testosterone; AMS: Aging Male Symptoms; SHIM: Sexual Health Inventory for men; HDL-chol: high density lipoprotein cholesterol.

Discussions

ED is prevalent among elderly men, and the Massachusetts Male Aging studied of 1290 men aged 40–70 years showed that 52% of men had some degree of ED [2]. Although there are several different causes of ED, including depression, testosterone deficiency, nerve problems and some medications, the most noteworthy cause is a blood endothelial disorder associated with arteriosclerosis. On the other hand, modifiable risk factors for CVD, including aging, cigarette smoking, hypertension, hyperlipidemia, DM, obesity, lack of physical exercise and poor diet, are shared with ED [16]. Some previous studies demonstrated that increased severity of ED within LOH patients correlated with an increased waist circumference, hyperglycemia, hypertriglyceridemia, hyperlipidemia and a history of diabetes mellitus [17,18]. Therefore, ED has been widely recognized as a predictive sign of systemic vascular endothelial dysfunctions and is a potential predictor of CVD [19–21].

A meta-analysis of 45 558 participants from seven cohort studies showed that pooled adjusted relative risk for incidence of cardiovascular events among ED subjects compared with non-ED subjects was estimated at 1.47 (95%CI, 1.29–1.66) [21]. Furthermore, patients with CVD had a significantly higher incidence of ED, and ED usually occurred two to three years before the development of CVD or heart attack [22–24]. In other words, ED can be a sentinel marker for the presence of silent vascular disease in asymptomatic men.

In the present study, an age-adjusted analysis revealed that SHIM score was inversely correlated with hs-CRP levels. Indeed, some previous studies have shown an affirmative correlation between hs-CRP and ED [25–27]. A previous study examined penile vascular disease severity by penile Doppler ultrasonography and hs-CRP levels among 137 men with ED, demonstrated that hs-CRP levels were significantly higher in patients with ED, and were furthermore associated with penile arterial disease severity among men with ED without clinically apparent CVD [25]. Another report showed that 74% of 121 men with ED had moderate to high CVD risk based on hs-CRP levels, and that testosterone deficiency could especially enhance risk of CVD [26]. On the other hand, Lee et al. showed a significant correlation between serum hs-CRP levels, the degree of ED, and responsiveness to tadalafil; hs-CRP levels were significant higher in the tadalafil-nonresponders [27]. These data suggest that hs-CRP levels can reflect erectile function and predict cardiovascular risk in men. Furthermore, one previous study also described that TRT for hypogonadal men contributed to a continuous improvement of body weight, serum lipids, glucose, HbA1c and BP, and resulted in a significant decrease in CRP levels [28].

In our data set, any other factors related to CVD risk, including total-Cho, HDL-chol, TG and HbA1c values were not correlated with SHIM score. In our study group, these values were modified to normal ranges controlled by medications. Some limitations of the present study include a small number of participants and some population-selection biases; however, hs-CRP levels may be useful marker of ED and its severity even in populations with these selection biases.

Furthermore, SHIM scores were also negatively correlated with IPSS. Indeed, many previous studies have shown that lower urinary tracts symptoms (LUTS) are closely associated with ED, and the severity of LUTS was correlated with the magnitude of ED [29–31]. PDE-5 is also present in the urinary bladder and prostate and regulates the smooth muscle tone of the bladder and prostatic urethra [31,32]. PDE-5 inhibitors have been commonly used for the treatment of male LUTS worldwide. Interestingly, it has been reported that pelvic ischemia or decreased bladder blood flow, caused by aging and arterial sclerosis, may be associated with the development of ED and LUTS [33,34]. In this study, we found no significant correlation between hs-CRP levels and IPSS. Our study population consisted of patients with mild LUTS. Therefore, further studies including a large number of patients with LUTS/overactive bladder (OAB) are required to clarify whether hs-CRP levels are a potential marker for bladder blood flow problems or pelvic ischemia.

Finally, we tried to identify factors affecting cardiovascular risk based on hs-CRP levels. SHIM score was an independent factor affecting cardiovascular risk. Indeed, a previous study described that many ED patients had high cardiovascular risk based on hs-CRP levels, and that especially low testosterone levels and obesity could increase risk [26]. Furthermore, low testosterone itself is associated with CAD [3,35]. Wickramatilake et al. showed that testosterone was lower, while hs-CRP level was higher, in patients with CVD compared to controls [36]. Since our population consisted of patients with LOH syndrome, no significant correlation between hs-CRP levels and FT value were found. However, our participants initially had a single risk (low testosterone) for developing CVD events. The present study demonstrated that SHIM score was inversely correlated with hs-CRP value and was an independent factor affecting CVD risk. This suggests that severity of ED among men with LOH syndrome is an important predictor of CVD events. Therefore, we recommend measuring hs-CRP levels while managing the patients with severe ED and LOH syndromes.

Declaration of interest

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this article.