The vascular erectile dysfunction (ED) is an inability to achieve or maintain an adequate erection for sexual intercourse due to a primary disorder of the penile artery system [1]. The most common cause of vascular damage is atherosclerosis, which is usually systemic in nature. When the first symptoms of ED appear, the violation of vessels of a larger caliber, in particular, the coronary and carotid arteries, is subclinical in nature, which indicates the possibility of fatal cardiovascular complications in the absence of the necessary examination and treatment [2]. The described theory of successive vascular lesions based on their diameter was first presented by Montorsi et al. in 2003 [3]. According to this theory, it becomes clear why ED is more often preceded by the development of various forms of clinically pronounced coronary heart disease and appears 2–3 years before the manifestation of cardiac symptoms [4].

In 2015, 422.70 million cases of cardiovascular diseases and 17.92 million deaths from these diseases were registered worldwide [5]. The prevalence of ED according to the results of an epidemiological study in the Russian Federation reaches 90% in the population of men under 75 years of age, and 71.3% of the respondents have a mild degree of disorders [6], indicating an incipient, but still reversible, lesion of the vascular bed of the penis with a probable subclinical lesion of the visceral vessels. A useful mechanism to reduce the prevalence and mortality of cardiovascular diseases and their complications is to introduce an algorithm into routine practice to assess the state of erectile function as an early and sensitive marker of systemic vascular damage [2].

The erectile function of patients can be evaluated both on the basis of subjective sensations of patients [7] and by objective measurement of nocturnal penile tumescence [8]. Subjective assessment of ED is carried out using various questionnaires, the most common of which is the International Index of Erectile Function (IIEF-15) [7], which has limited capabilities in differentiating psychogenic and organic disorders, as well as in identifying subclinical ED forms. An objective assessment of ED is devoid of such shortcomings due to the lack of cortical control over erection during sleep, which makes it possible to distinguish ED of organic and psychogenic genesis, as well as to diagnose disorders even before the first symptoms appear. The RigiScan® Plus Rigidity Assessment System hardware device («Gesiva Medical» LLC, Eden Prairie, MN, USA) is used in foreign countries for early objective diagnosis of ED [8], while the Russian analog of the device is the Androscan MIT hardware complex [1].

According to the American Heart Association, there are seven main factors that affect the state of the vascular wall: smoking, blood pressure (BP), total cholesterol, glucose, body mass index (BMI), physical activity, and diet [9]. The lesion of the vascular wall most often manifests itself clinically as vasculogenic ED [5]. Consequently, the evaluation of patients for the presence of the above-mentioned risk factors can not only help in the treatment of ED but also significantly increase the chances of early detection of cardiological pathology [3].

In this regard, the purpose of this particular study was to assess the influence of risk factors for cardiovascular diseases on the severity of ED of vascular genesis.

Study sample. The study included 40 patients who were examined on an outpatient basis at Moscow State University for ED. The criteria for inclusion in the study were the age from 25 to 60 years and the desire to be examined to improve the quality of erection. The exclusion criteria were the presence of sexually transmitted diseases, inflammatory diseases of the genitals and lower urinary tract, the presence of mental illnesses, and pronounced neurological disorders. All patients were examined, including collection of anamnesis for the presence of hypertension (HT) and smoking, measurement of BP, BMI, study of lipid profiles (high-density lipoproteins — HDL, low-density lipoproteins — LDL, triglycerides, cholesterol) and glycemic (HbA1c), determination of the level of C-reactive protein. Erectile function was assessed using two methods: the IIEF-15 questionnaire (erectile domain) and a device that registers penile tumescences — Androscan MIT («Minimally Invasive Technologies» LLC, Moscow, Russian Federation). In order to interpret the indicators, the criteria developed by Kamalov and Chaly were used. Therefore, normal erectile function was determined in the case of a relative increase in penis diameter (RIn) of more than 30% and the duration of nocturnal penile tumescence (NPT) with RIn of more than 30% for more than 60 minutes. ED grade 1 was exposed in the case of RIn greater than 30% with the duration of the NPT with RIn greater than 30% from 10 to 60 minutes (option 1), in the case of RIn greater than 30% and the duration of the NPT with RIn more than 20% — more than 10 minutes (option 2), in the case of RIn greater than 20% to 30% with the duration of the NPT with RIn greater than 20% — more than 60 minutes. ED grade 2 was observed with RIn from 20% to 30% with a duration of NPT with RIn of more than 30% from 10 to 60 minutes. ED grade 3 was exhibited in the case of RIn less than 20% or the duration of the NPT less than 10 minutes with any RIn [10].

The applied equipment. For an objective assessment of ED, the Androscan MIT device was used, as it is an easily used, small, autonomous, wireless registration device with a processor, memory and autonomous power supply. The registration of the penis diameter after activation of the sensor occurs every 10 seconds for 12 hours. After the study is completed, the device is connected to the reader through a wireless communication channel to read the accumulated data. The reader is connected to a personal computer using a standard USB cable. The number of studies that can be performed using a single sensor (after standard processing) is 20.

Methods of statistical analysis. The data was collected by means of Microsoft Office Excel 365 (Microsoft Corp., Redmond, WA, USA). Statistical data processing was performed using Statistica 12 software (StatSoft Inc., Tulsa, OK, USA) and IBM SPSS Statistics 26 (SPSS: An IBM Company, IBM SPSS Corp., Armonk, NY, USA).

The clinical parameters of the sample for continuous values are presented as averages±standard deviation (SD), for discrete values – as percentages of occurrence (%). The assessment of the normality of the data distribution was carried out using the Kolmogorov-Smirnov and Shapiro-Wilk criteria. ANOVA analysis of variance and the Kruskal-Wallis criterion, respectively, were used to assess the differences between normally and abnormally distributed continuous clinical characteristics depending on the grade of ED. The evaluation of differences in the frequency of qualitative indicators was carried out using the Pearson consensus criterion (χ2 Pearson). Depending on the contribution of clinical characteristics (predictors) to belonging to a certain class of ED, their rank was evaluated on a scale from 0 to 100. Classification trees using the discriminant one-dimensional branching method applying categorical and ordinal predictors (clinical characteristics) were used as a method to predict and classify patients depending on the degree of ED. The differences were considered statistically significant at p < 0.05.

The study included 40 patients aged 33 to 60 years (the average age was 50.0±8.35 years). The clinical characteristics of the sample studied are presented in Table 1.

The comparison data of the main parameters of patients depending on the degree of ED according to androscan data are presented in Table 2. It was found that all the studied clinical characteristics, except for diastolic blood pressure (DBP), differed significantly in patients with different degrees of ED (p < 0.05). At the same time, patients with a more severe degree of ED, compared with milder forms, were characterized by a greater deviation of the studied indicators from the norm. A similar trend was observed for DAD, but the differences were statistically insignificant.

The results of comparing the main clinical characteristics of patients depending on the degree of ED according to IIEF-15 are presented in Table 3. It was found that DBP and HbA1c levels, despite the tendency to increase with increasing severity of erectile function, did not differ significantly for patients with different degrees of ED (p > 0.05). The remaining clinical parameters, similar to the androscan data obtained, differed significantly depending on the grade of ED (p < 0.05). At the same time, patients with a more severe degree of ED, compared with lighter forms, were characterized by a greater deviation of the studied indicators from the norm.

Therefore, Figure 1 shows the rank (significance) of the influence of clinical characteristics of patients on the detection of ED, according to androscanning (A) and IIEF -15 (B). Total cholesterol (100), HDL (89), LDL (81), C-reactive protein (78), BMI (78), triglycerides (74) and HbA1c (63) have the greatest influence in determining the degree of ED using the Androscan MIT device (rank greater than 60). The greatest influence on the determination of the degree of ED using the IIEF-15 questionnaire was only the value of points on the IIEF-15 scale (100), which makes this assessment completely dependent on the subjective feelings of patients without relying on laboratory and instrumental research methods.

To classify patients by ED grades, based on androscan data, depending on the value of the predictors studied, a classification tree was obtained with two branches and four terminal vertices (Figure 2). The branching condition of the root vertex (node 0) is the level of total cholesterol. It follows that with a cholesterol level of ≤ 7.775 mmol/l (node 1), 100% of the patients have grade 1 ED or normal erectile function (86.7% and 13.3%, respectively). The condition for the branching of node 1 is the LDL level. If the LDL level is less than 2.865 mmol/L (node 3) and total cholesterol is less than 7.775 mmol/ L, then 100% of patients have a normal erectile function, while at the LDL level ≥ 2.865 mmol/l (node 4) — ED grade 1. With a cholesterol level of more than 7.775 mmol/l (node 2), 100% of patients suffer from ED grades 2 and 3 (76% and 24%, respectively), while with cholesterol of more than 9 mmol/l (node 6), 100% of patients suffer from ED grade 3, and with cholesterol < 9 mmol/l (node 5) they suffer from ED grade 2. There were no classification errors in the prediction of the ED grades, which in this case indicates the good significance of the mathematical forecast.

In the next stage of the study, classification trees for the degree of ED were constructed, according to the IIEF-15 questionnaire, with two branches and three terminal vertices (Figure 3). The only condition for all the branches obtained is the number of IIEF-15 points. The precision of this assessment was 100% for all grades of ED, which in this case states the good significance of the mathematical forecast, but makes this assessment completely dependent on the subjective feelings of patients without relying on laboratory and instrumental research methods.

ED is a widespread problem and a threat to the demographic situation both in the Russian Federation and around the world. The risk of ED increases with age, which was confirmed during the Massachusetts Study on Aging Men, in which ED was detected in about 50% of patients over 50 years of age, 60% at the age of 60 years and 70% at the age of 70 years [11]. Thus, the generally accepted risk factor for the development of ED is the age of the male population, which was confirmed by the results of Russian and foreign colleagues. According to the work by Iacono et al. (2012), ED, estimated by measuring nocturnal penile tumescences using the RigiScan test, was more common among men aged 65 years and older. The risk of development of ED for men in this age group was found to be 7.5 times higher (CI = 5.1 – 9.9) than for patients younger than 65 years, which was associated with the development of cavernous fibrosis due to excessive deposition of adipose tissue on the background of metabolic disorders (including elevated uric acid levels) [12], smoking, cardiovascular diseases [13]. Similar data were obtained by Russian colleagues when evaluating erectile function using the IIEF-15 questionnaire.

The severity of erectile function was shown to be positively correlated with the age of the patients interviewed. Thus, for men under 50 years of age, severe ED occurs less than in 3% of cases, while in the age group of 71 – 80 years — in 56.5% of cases. In parallel with the increase in severe ED with age, there was a tendency to decrease the proportion of patients with mild forms of ED [14]. The results of this study combine the data obtained by Russian and foreign colleagues. The severity of erectile disorders evaluated using both IIEF-15 and the Androscan MIT apparatus depends on the age of the patients. According to objective and subjective data, severe forms of ED are more common among older patients than among younger patients.

The results of the IIEF-15 questionnaire do not always correlate with the results of an objective assessment of erectile function [15]. According to Tokatli et al. (2006), based on the results of the IIEF-15 questionnaire, 87% of the patients (78 patients) had ED, but 55 of them had normal erectile function when measuring nocturnal penile tumescences. Taking the evaluation of ED using the RigiScan apparatus as the gold standard of diagnostics and the method of ED control, researchers found that the sensitivity, specificity, and prognostic value of positive and negative results were 100%, 17.9%, 29.4%, and 100%, respectively [16]. This indicates that there is no tendency to overestimate the subjective evaluation of erectile function, which was also observed in the work by Melman et al. (2006) [17] and confirmed in the present study, according to which all surveyed patients had ED of varying severity, while no ED was detected in 2 patients (5%) during androscanning.

Hyperlipidemia has been proven to be one of the most significant risk factors for ED and cardiovascular diseases. According to Saltzman et al. (2004), correction of the level of total cholesterol, LDL, and triglycerides against the background of hypolipidemic therapy with atorvastatin is associated with an improvement in penile rigidity during nocturnal penile tumescences measured using the RigiScan apparatus. At the same time, an increased lipid level is associated with a more severe course of ED [18]. Similar results were obtained in the work by El-Sisi et al. (2013). Administration of atorvastatin led to a significant improvement in erectile function, assessed both by the results of the IIEF-15 questionnaire (an increase in IIEF-15 scores by 53.1%) and by measuring nocturnal penile tumescences (an increase in the number of erections with sufficient rigidity of the glans penis and base by 16.6% and 17.2%, respectively). Furthermore, in the studied group, a 57.9% decrease in C-reactive protein was observed synchronously with an improvement in erection, as well as an increase in the concentration of the main vasodilator of the cavernous vessels, nitric oxide, indicating an improvement in the functional state of the endothelium in the background of therapy [19]. The data obtained by foreign colleagues were confirmed in this study conducted using the Androscan MIT hardware complex. Hyperlipidemic disorders associated primarily with elevated levels of LDL, triglycerides, and cholesterol are associated with ED, and more pronounced metabolic disorders are associated with more severe ED, assessed by both subjective and objective methods. A similar trend was shown for the level of C-reactive protein, which, despite normal values in all studied groups, tended to the upper limit of the norm with an increase in the degree of ED, according to IIEF-15 and androscanning.

HDL, unlike cholesterol and LDL, have a protective effect on the state of the cardiovascular system. A similar relationship was established for HDLs and erectile function. Elevated HDL levels, according to the data of Xu et al. (2021), were associated with high rigidity of the glans and the base of the penis during monitoring of nocturnal penile tumescences by the RigiScan apparatus [20]. A positive relationship between IIEF-15 scores and HDL levels was demonstrated in a study by Molina-Vega et al. (2020), again indicating the beneficial effect of HDL on erectile function [21]. In this study, the level of HDL was associated with the severity of erectile disorders detected both by the results of the IIEF-15 questionnaire and by the results of androscanning, which is also consistent with the data obtained by foreign colleagues. Furthermore, when assessing the significance rank of the predictors, the HDL level was recognized as one of the most significant markers of ED detected by measuring nocturnal penile tumescences, while for subjective evaluation of erectile function, only the scores of the IIEF-15 questionnaire were the most significant predictor.

Patients with hyperlipidemic disorders are generally characterized by a hypersthenic physique with excessive subcutaneous fat deposition, which in turn, like hyperlipidemia, affects the deterioration of erectile function and also leads to the development of cardiovascular diseases. Thus, according to the work by Corona et al. (2011), increased BMI is associated with a progressive decrease in the quality of nocturnal penile tumescences measured using the RigiScan hardware [22]. According to a similar study by Andersson et al. (2015), BMI had a significant negative correlation with the number of nocturnal erections, their duration, and the rigidity of the head and base, estimated using the RigiScan test [23]. According to the results of a Chinese study by Lu et al. (2021), among patients with increased BMI, the ED (detected by means of the IIEF-15 questionnaire) is 1.2 times more common than among patients with normal weight [24]. Similar results were obtained for the Russian population of patients interviewed for erectile disorders in the study by Korneyev et al. (2016). For patients with a BMI of more than 25 kg/m2, the risk of developing ED is 1.21 times higher than for patients with lower body weight. The grade of ED was positively correlated with the BMI of the surveyed patients [14]. According to the results of this study, all the examined patients with ED suffered from obesity and the severity of obesity increased proportionally to the increase in the severity of erectile disorders both in their subjective and objective assessment, which is consistent with the data of the world literature.

It has been established that concomitant diseases caused or causing endothelial function disorders are also associated with ED and the severity of these disorders. Already in 1999, in a study by McMahon and Touma, it was shown that patients with cardiovascular diseases, such as HT and coronary heart disease, are 5 times more likely to have erectile disorders detected during monitoring of nocturnal penile tumescences [25]. Similar results were obtained later. Thus, according to the results of a study by Russian colleagues led by Korneyev (2016), it was found that the severity of ED, according to a subjective assessment of erectile function using the IIEF-15 questionnaire, was associated with the presence of diabetes mellitus (DM) and HT. At the same time, ED was 4.71 times more common in patients with DM than among men without glycemic profile disorders, while the risk of developing ED in the presence of HT increased 4.38 times [14]. Patients with DM and elevated levels of HbA1c according to the results of the RigiScan test also have lower indicators, similar to the results of the IIEF-15 questionnaire. In a study by Andersson et al. (2016), it was found that the number of nocturnal erections, the duration of erections, and the rigidity of the head and base in patients with DM were significantly lower compared to the control group of patients with ED without metabolic disorders [23]. The available literature data are fully consistent with the results of this study. Therefore, the level of glycated hemoglobin increases proportionally to the increase in the severity of ED detected by subjective and objective methods. The presence of HT was more often associated with moderate forms of ED. Mild and severe ED were evenly distributed in the study sample, which may be due to the small sample size, as well as the outpatient nature of the examination of the patients, which does not give a complete picture of the severity of increases in BP during the day.

When comparing the erectile function of smoking and non-smoking patients using the RigiScan test in a study by Elhanbly et al. (2004), it was found that, despite the younger age of smoking patients, the indicators of measuring nocturnal penile tumescence, such as rigidity and volume of the head and base during erection, were disturbed to a greater extent than in older non-smoking patients, which confirmed the organic nature of ED against the background of smoking [26]. Similar data were first obtained in 1992 by Hirshkowitz et al. It has been shown that the duration of nocturnal penile tumescence with sufficient penile rigidity is negatively correlated with the number of cigarettes smoked per day. In patients who smoke more than 40 cigarettes a day, the duration of erections is usually minimal [27]. Among the patients with ED identified by the results of the IIEF-15 questionnaire, smoking is much more common than among patients without ED (38% and 22.8%, respectively) [24]. In addition, the violations that occur up to a certain point are reversible. Therefore, after quitting smoking, an increase in IIEF-15 scores was observed in patients of all age groups, with the exception of patients with a pack-year index of more than 101, indicating the irreversibility of cavernous tissue disorders with prolonged exposure to tobacco smoke in the body [28]. In the present study, smoking patients were characterized by moderate to severe ED, while mild ED was much less common, which is consistent with data from the available literature. The predominance of ED grade 2 among smoking patients, both according to IIEF-15 and according to androscanning, may be due to a short duration of smoking, which did not have time to lead to irreversible severe ED; however, this analysis was not carried out in this study.

Within the framework of the study, the function of ED predictors was evaluated for the first time, identified both using the IIEF-15 questionnaire and using the Androscan MIT apparatus. It was found that for an objective assessment of the presence of ED and its severity, the main role was played by the parameters of the well-being of the cardiovascular system, such as the level of cholesterol, HDL, LDL, C-reactive protein, triglycerides, BMI, HbA1c, and the most predictive function is performed by lipid metabolism indicators. Similar studies were conducted earlier [23]; however, the factors affecting erectile function were put in the same row, without showing the superiority of any one predictor over others. For ED, exposed using the IIEF-15 questionnaire, only the value of the questionnaire scores was recognized as the key predictor without relying on clinical and laboratory data. Thus, when diagnosing ED on the basis of subjective data, it is not always clear which disorders need to be corrected, while ED, detected during the registration of nocturnal penile tumescences, requires a thorough assessment of the state of the cardiovascular system with primary correction of the lipid profile, which plays the most significant role in the occurrence of the organic form of ED.

The risk factors described above lead to the development of endothelial dysfunction, which is known to be the main etiological factor of cardiovascular disease and ED [1]. Violations of the vasodilating ability of the endothelium under the influence of exogenous and endogenous factors lead to primary systemic damage to small caliber vessels, followed by damage to large arteries, which formed the basis of the so-called arterial diameter theory or Montorsi theory [3]. Thus, with the development of ED recorded using the Androscan MIT hardware complex, it is necessary to pay attention to the state of the cardiovascular system, realizing that reversible measurements have also begun in the coronary vessels. Conversely, when hyperlipidemic disorders are detected, based on the constructed classification trees, it is possible to judge the severity of erectile disorders and predict one or another degree of ED before receiving androscan data.

Limitations. The main limitation of this work is the sample size, which does not allow using all the possibilities of discriminative analysis and reliably evaluating the constructed classification trees; however, the proposed method is a kind of prerequisite for creating objective criteria for classifying degrees of ED based on androscanning data, considering the somatic status.

The severity of vascular ED, assessed using objective and subjective diagnostic methods, depends on age, the severity of lipid and carbohydrate metabolism disorders, smoking, as well as the presence of cardiovascular diseases. Dyslipidemia is the most significant predictor of the detection of ED with the help of andro scanning, according to the severity of which it is possible to judge the severity of erectile disorders. The subjective assessment of the severity of ED using the IIEF-15 questionnaire is not based on clinical data and the somatic state of the patient, which does not allow a reliable evaluation of the severity of vascular disorders from one questionnaire. Only an integrated approach to the diagnosis of ED can assess the true causes of the disorder and develop tactics for further treatment.