“Stroke-Stop” Formula: A Tool for Risk Index Determination in Development of Acute Cerebrovascular Disease in Patients with Asymptomatic Internal Carotid Artery Stenosis

Ivan Kopolovets Vychodoslovensky Ustav Srdcovych a Cievnych Chorob AS Peter Berek (  berekp67@gmail.com ) Vychodoslovensky Ustav Srdcovych a Cievnych Chorob AS Peter Stefanic Vychodoslovensky Ustav Srdcovych a Cievnych Chorob AS Dmytro Lotnyk Cornell University Rastislav Mucha Slovenska akademia vied Zdenka Hertelyova Univerzita Pavla Jozefa Safarika v Kosiciach Lekarska fakulta Stefan Toth Vychodoslovensky Ustav Srdcovych a Cievnych Chorob AS Nadiya Boyko Uzhorodskij Nacionalnij Universitet Medicnij Fakultet Vladimir Sihotsky Vychodoslovensky Ustav Srdcovych a Cievnych Chorob AS


Background
Extracranial carotid artery disease is considered a risk factor for developing acute cerebrovascular diseases [1]. Despite a large number of scienti c researches dedicated to prevention, diagnosis and treatment of stroke, the problem is still relevant and remains unresolved [2].
In patients with symptomatic ICA stenosis > 50%, the average annual risk of recurrent stroke is 37-40%, while in patients with asymptomatic ICA stenosis > 70%, the risk of developing stroke is 3-6%. At the same time, the risk of developing postoperative complications is from 2 to 5% [4,5]. It indicates the fact that performing carotid endarterectomy in patients with symptomatic ICA stenosis allows reducing the risk of stroke development by approximately 9-10 times, while in patients with asymptomatic ICA stenosis -by 1.5-2 times only [6]. Therefore, the criteria for selecting patients for surgical prevention of acute cerebrovascular disease among patients with asymptomatic ICA stenosis are broadly discussed nowadays [7,8]. There have been several studies, where in addition to atherosclerotic ICA stenosis, the morphological structure of the atherosclerotic plaque was taken into consideration [9][10][11]. In recent years, researchers have widely investigated vascular in ammatory markers as well as the relationship between their concentration and the risk of developing atherosclerotic complications [12]. Speci c biomarkers are crucial tool since enabling the early prevention of non-symptomatic stages of diseases and also support prognostic/predicted and patient-centered treatment. The useful biomarker has to correlate with clinical parameters, such as speci c symptoms, clinical signs and validated diagnostic tests [13]. One of the in ammatory markers correlating to atherosclerosis complications with high speci city is the lipoprotein-associated phospholipase A2 (Lp-PLA2) [14,15].
Duplex ultrasonography is one of the main screening methods for detection of atherosclerotic ICA stenosis [16,17]. It allows detecting the degree of ICA stenosis as well as evaluating the structure of the atherosclerotic plaque [18]. It is the degree of stenosis that serves as the main indication for performing carotid endarterectomy [5]. However, according to recent reports, approximately 20-25% of symptomatic patients have ICA stenosis < 70% [19]. This means that one criterion alone, namely the degree of ICA stenosis is not enough to determine the risk of stroke development in asymptomatic patients since this category of patients (20-25%) is not under active supervision by physicians.
Together, these data con rm that the problem of early detection of patients at the highest risk of developing stroke remains relevant. Symptomatic patients were de ned as patients who had suffered an ischemic stroke or transient ischemic attack (TIA) within last 6 months, while asymptomatic patients were de ned as those who did not suffer any stroke or TIA within last 6 months.
The average patients' age was 69 ± 7.5 years. Depending on the clinical course, the patients were divided into two groups: Group I included 30 (43% of all 70 tested) patients with symptomatic ICA stenosis; among them, 20 (66.6% of Group I) patients had a history of ischemic stroke, and 10 (33.3% of Group I) patients had a history of transient ischemic attack (TIA); Group II included 40 (57% of all 70 tested) patients with asymptomatic ICA stenosis.
There was no statistically signi cant difference in age and sex between the groups.
The control group included 20 (10 men and 10 women) individuals; without any statistically signi cant difference in age and sex. The average age was 38 ± 5.2 years. The study characteristics of the current participants (excluding negative controls) are presented in the Table 1.

Preoperative ultrasound imaging of the atherosclerotic plaque
The degree of ICA stenosis, as well as the morphological structure of the atherosclerotic plaque, was determined in the preoperative period using ultrasonography. Duplex scanning of the carotid arteries was performed using the Philips HD11XE ultrasound machine with an 8 MHz linear transducer. The degree of stenosis was evaluated according to the "Consensus Panel Gray-Scale and Doppler US Criteria for Diagnosis of ICA Stenosis" through the determination of the peak systolic velocity (PSV) and the end diastolic velocity (EDV) within the ICA [20].
The structure and composition of the atherosclerotic plaque was assessed using echogenicity: hypoechogenic plaque -soft plaque, heterogenic plaque -mixed plaque, hyperechogenic plaque -hard plaque.

Marker of in ammation in patients with carotid artery stenosis
Blood samples of all patients were taken one day before the surgery, and the concentration of speci c markers, namely, Hpx, Lp-PLA2, and IL-4, were determined. The levels of these parameters were determined using the ELISA method (human lipoprotein associated phospholipase A2, Cussabio, USA; Human Hpx, Abcam, UK), and the results were evaluated with the Tukey test. The plasma level of IL-4 was measured using an ELISA kit Human IL-4 Platinum ELISA (eBioscience, San Diego, USA). Blood samples for the measurement of speci c markers were centrifuged after collection at 2.500 rpm for 10 min. Serum samples were then frozen at -80 °C until analysis. Samples were processed using Synergy H4 multiplate reader (BioTek Vermont, USA).

Histopathology
In the postoperative period, histological assessment of the atherosclerotic plaque removed during carotid endarterectomy was performed. The selected atherosclerotic plaque was xed in a 4% solution of neutral formalin; then, decalci cation was carried out, and the material was poured in para n blocks 1 cm in size. 5-µm-thick sections of para n blocks cut by a microtome were studied histologically. The blocks were examined in the longitudinal section, in the most stenotic area. The hematoxylin and eosin staining were used. Histopathological characteristics of the retrieved carotid plaques were reported according to the updated American Heart Association classi cation of advanced atherosclerosis and a previously wellvalidated scoring system published by Lovett et al. (11). For each plaque, the following features were recorded: the rupture of the brous cap, lipid core size, nodular calci cation, neovascularisation, in ammatory in ltrate, in ltration of the brous cap, proportions of brous tissue, intraplaque hemorrhage, presence of foam cells, and surface thrombus. Based on the presence of these features in each plaque, an overall stability rating was given by the histopathologist as "unstable" or "stable." Unstable plaques demonstrated many or all features, and "stable" plaques demonstrated none of them.

Statistical methods
The obtained results were evaluated using descriptive statistics (the frequency, percentage ratio) as well as the one-way analysis of variance (ANOVA) (Minitab Inc. version 11.24, Coventry, UK) and chi-square test (Preacher, K. J. (2001, April). Calculation for the chi-square test: An interactive calculation tool for chisquare tests of goodness of t and independence). The relationship between two variables -the group of symptomatic patients and the group of asymptomatic ones -was evaluated using the one-way analysis of variance (ANOVA) (Minitab Inc. version 11.24, Coventry, UK). The correlative relationship was evaluated using Pearson's correlation coe cient (Pearson correlation test, MINITAB Inc., Coventry, United Kingdom).
The relationship between more than two variables (symptomatic patients, asymptomatic patients, and the control group) was evaluated using the Tukey's HSD test (Minitab Inc. version 11.24, Coventry, UK); the data were considered statistically signi cant at p < 0.05.

Theory
This study is aimed to search for new criteria for assessing the risk of stroke in patients with atherosclerotic ICA stenosis, based on the principles of personalized medicine. Proposed "Stroke-Stop" formula (based on a mathematical calculation of the degree of ICA stenosis, the morphological structure of the atherosclerotic plaque and the level of lipoprotein-associated phospholipase A2 (Lp-PLA2) concentration) determine the risk of stroke development.

Results
The assessment of the atherosclerotic plaque echogenicity according to the ultrasonographic data has revealed that soft atherosclerotic plaque was observed in 19 (27.1%) patients, a mixed atherosclerotic plaque was detected in 29 (41.4%) patients and a hard atherosclerotic plaque was found in 22 (31.5%) patients. The ratio of the atherosclerotic plaque echogenicity between Group I and Group II is presented in Table 2. Sympt -patients with symptomatic ICA stenosis; asympt -patients with asymptomatic ICA stenosis Histological assessment was performed to test the statistical signi cance of the quality of atherosclerotic plaque density interpretation using ultrasound. Ultrasound detected unstable plaques (patients with soft and mixed atherosclerotic plaque) in 62.9% of patients, while histological assessment detected unstable plaques in 64.3% of patients. It indicated the fact that ultrasound interpretation of the atherosclerotic plaque structure is highly informative and could be considered in assessing the risk of developing atherosclerotic complications.
Among vascular in ammatory markers, lipoprotein-associated phospholipase A2 (Lp-PLA2) showed a statistically signi cant difference (P < 0.01) in both the structure of atherosclerotic plaque and its clinical manifestations. In the patients studied, the concentration of Lp-PLA2was increased at 252.7-328.6 ng/ml in correlation with negative controls 205-240 ng/ml.
When studying vascular in ammatory markers, the ratio of Lp-PLA2 concentration was compared between patients of Group I and those of Group II (symptomatic and asymptomatic ICA stenosis). The symptomatic group had a higher mean plasma level of Lp-PLA2 (285.30 ± 2.05 µg/l) than the asymptomatic group (274.35 ± 3.38 µg/l). The difference in the Lp-PLA2 levels between the symptomatic and asymptomatic groups ( Fig. 1a) was signi cant (p < 0.05). No signi cant differences were noted in the plasma levels of Lp-PLA2 between men and women within the two groups.
On the other hand, the level of Lp-PLA2in 15% of asymptomatic patients with soft atherosclerotic plaque was higher than that in symptomatic patients with hard atherosclerotic plaque.
In addition, the dependence of Lp-PLA2 concentration from the structure of the atherosclerotic plaque was evaluated. While assessing the obtained results, a statistical signi cance (p < 0.05) between Lp-PLA2 concentration in patients with soft atherosclerotic plaque, mixed atherosclerotic plaque and patients with hard atherosclerotic plaque was detected (Fig. 1b).
When assessing the results a statistical signi cance was found between the increase in Lp-PLA2 concentration and the morphological structure of the atherosclerotic plaque (p < 0.001).
Histograms in Fig. 2 indicate a number of patients with a different type of hardness based on Lp-PLA2 concentration. Noteworthy, the peaks and tails of histograms overlap. That means that we cannot separate or identify patients using only Lp-PLA2 concentration level. One could notice from Fig. 1a symptomatic patients have Lp-PLA2 > 250 mg/L. In the current paper, we propose to take into account additional parameters such as Systolic to Diastolic Velocities Ratio (SDVR) of IVA and the type of plaque.
In Fig. 3 Lp-PLA2 concentration is represented as a function of SDVR for each type of plaque. The red dots denote the patients with symptomatic stenosis of ICA inside the corresponding types of plaque. For the soft atherosclerotic plaque, red dots are spread almost homogenously. However, for the mixed and hard atherosclerotic plaque, the red dots are located close to the edge of the point locations. The latter fact can be easily explained by the multiplication of two parameters Lp-LPA2 and SDVR. The geometrical outcome of the multiplication is the area of the corresponding rectangle. In Fig. 3, three rectangles are shown for one point belonging to each type of atherosclerotic plaque. So the higher the value of the area, the higher the probability for a patient having symptomatic stenosis of ICA. We propose the area, i.e. Lp-PLA2 * SDVR, as a modi ed parameter to predict symptomatic stenosis of ICA. Moreover, one can introduce the scaling factor (SF) for each type of plaque de ning the high risk at 100. To satisfy the latter assumption, one must have SF = 5 for the soft atherosclerotic plaque, and SF = 10 and SF = 15 for mixed and hard atherosclerotic plaques, correspondingly. The risk factor consequence can be calculated as follows:

RF = area / SF
We need to provide a more e cient way to determine patient status. Using an empiric equation, we are introducing the Risk index as the universal parameter to identify patient status.
From Fig. 4, one can see the separate peaks for symptomatic and asymptomatic patients with probabilities 37% (asymptomatic) and 26% (symptomatic) which correspond to risk indices 60 and 90, respectively. Even though, the tails are overlapping more than the half asymptomatic patients have risk index < 70 while the majority of the symptomatic patients, 80% have risk index > 70. One can underline that the risk index is independent of gender, age of patients and depends only on well-de ned and measured values such as Lp-LPA2 concentration and velocities ratio.
To study the effect of the atherosclerotic plaque structure, the degree of ICA stenosis, and the concentration of in ammatory markers increasing the risk of development of stroke were analyzed. 3D visualization technology revealed that the risk of stroke development is increased in case of the progression of ICA stenosis and elevation of Lp-PLA2 levels (Fig. 5). In patients with soft atherosclerotic plaque, the risk of stroke development was signi cantly higher in contrast to the patients with hard atherosclerotic plaque.
To assess a dominant risk factor for stroke development, all the patients were analyzed according to the structure of the atherosclerotic plaque.
According to the given empirical scale, all the patients with soft atherosclerotic plaque (both symptomatic and asymptomatic ones), ICA stenosis greater than 70% and Lp-PLA2concentration of more than 285 mg/l had the risk of stroke development of > 100 points (Fig. 6a).
Among patients with mixed atherosclerotic plaque, the risk of stroke development with > 100 points was observed in patients with hemodynamically signi cant ICA stenosis (> 80%) and Lp-PLA2concentration of more than 285 mg/l. (Fig. 6b).
Patients with hard atherosclerotic plaque had low or medium risk of stroke development (Fig. 6c).
Based on the results obtained, a mathematical calculation of the major risk factors (the degree of stenosis, the morphological structure of the atherosclerotic plaque and the level of in ammatory marker concentration) has been proposed to calculate the risk index for developing of stroke in patients with asymptomatic ICA stenosis.
The principle of mathematical calculation using the formula "Stroke-Stop" is the following: in patients with asymptomatic atherosclerotic disease of the carotid arteries, the concentration of Lp-PLA 2 is determined using ELISA; both diastolic velocity and systolic velocity of blood ow within the ICA are measured using ultrasound thereby determining the degree of ICA stenosis;the structure of the atherosclerotic plaque is evaluated; and nally, the risk index for development of ischemic stroke using the proposed formula "Stroke-Stop" is calculated.
The numerator represents the ratio of ICA systolic velocity to ICA diastolic velocity multiplied by the indicator of Lp-PLA2 concentration, and the denominator represents the density coe cient of "5" in soft hypoechogenic atherosclerotic plaque, the density coe cient of "10" in mixed atherosclerotic plaque and the density coe cient of "15" in hard hyperechogenic atherosclerotic plaque. If the indicator is 50-70 points, the index of stroke risk is low, if the indicator is from 70 to 100 points, the index of stroke risk is medium, and if the indicator is more than 100 points -the index of stroke risk is high.
The use of the coe cient does not affect changes in the results since it is used as an identi er of the atherosclerotic plaque structure in all the patients. The results of calculating the risk of developing ischemic stroke by the formula "Stroke-Stop" are presented in Table 3. The results of calculating the risk of developing ischemic stroke by the formula "Stroke-Stop" presented in Table 2 clearly show the dependence and the in uence of three risk factors (ICA stenosis, atherosclerotic plaque structure and Lp-PLA2concentration) on stroke development.
An alternative to the mathematical calculation of the risk of stroke development by the formula "Stroke-Stop" is its graphic display in the form of three diagrams, where each represents the structure of the atherosclerotic plaque (Fig. 7).
Red color points represent a high risk of stroke development; yellow color is used to indicate a medium risk of stroke development; blue color indicates a low risk of stroke development.

Discussion
It was established that the carotid endarterectomy is the effective treatment to reduce the risk of subsequent stroke in symptomatic patients with carotid stenosis [6]. Moreover, the risk-bene t ratio prefers surgery for about 70% symptomatic stenosis. In correlation with symptomatic stenosis, it is the unusual behavior for the asymptomatic lesions [7]. Therefore, the stenosis degree stand-alone might not be su cient to predict the risk of a stroke. Clearly, additional markers are required to characterize more precisely the patients who would bene t the most from a surgery.
The development of new diagnostic technologies is contributing to the transition to a multi-parameter systematic model that allows the formation of a personalized approach to the diagnosis and prevention of stroke.
The main criteria, evaluated in our study, were the degree of ICA stenosis, the morphological structure of the atherosclerotic plaque and the level of Lp-PLA2 concentration.
We propose to further study unique biomarker Lp-PLA2 -an enzyme produced by in ammatory cells and hydrolyzes oxidized phospholipids in LDL, which in connection with individual patient plague stability may lead to earlier detection of atherosclerosis progression / manifestation. Lipoprotein-associated phospholipase A2 is also known as platelet-activating factor acetylhydrolase (PAF-AH). In the blood it is mainly connected with low density lypoprotein (LDL, near 80%) and only less than 20% of this enzyme is associated with high density lypoprotein (HDL) [14].
The combination of three factors -stenosis, ulceration of the atherosclerotic plaque and the in ammatory process within it -is one of the leading mechanisms of embologenicity as well as the development of stroke [21,22]. This mechanism is not taken into consideration by the classical approach to determining the indications for carotid endarterectomy where the main criterion is the degree of ICA stenosis (50% in symptomatic patients and 70% in asymptomatic patients).
Alongside with the degree of stenosis, plaque morphology can provide crucial information to predict the stroke risk. The recent ultrasound studies showed a higher risk of cerebrovascular events for hypo-or anechogenic plaques compared to echogenic ones [23].
Among symptomatic patients, unstable carotid plaques were found in 76.6% of cases, while stable plaques were detected in 23.4% of cases only. The difference was statistically signi cant (p < 0.0001).
Ultrasonographic assessment of tissue characteristics is performed according to the overall distribution of grey tones (overall brightness). There are anechogenic (dark) and hyperechogenic (bright) plaques [22].
To provide more objective description, there were developed more detailed classi cations; however, an interobserver agreement was weak and there was observed a low correlation with the histopathological ndings [24,25].
Several classi cations of plaque echogenicity have been reported in the literature. However, echogenicity on plaque character should be standardized against three reference structures: owing blood for anechogenic, sternocleidomastoid muscle for isoechogenic, and the adjacent transverse apophysis of the cervical vertebrae for hyperechogenicity [26,27].
Unstable carotid plaques are associated with increased risk of stroke not only in symptomatic but also in asymptomatic patients. A meta-analysis of eight prospective studies, with a total of 7,557 patients, observing patients with asymptomatic carotid stenosis found that patients with unstable, echolucent plaques had a 2.31-fold increased risk of stroke compared to patients with stable plaque based on ultrasound assessment [22,28].
Although, according to the study ACST-1, carotid plaque echolucency assessment offered no predictive value for stroke risk [29].
In our research, 25% of patients with asymptomatic ICA stenosis had unstable atherosclerotic plaque.
Experimental studies have shown a key role of in ammation in destabilization and rupture of the atherosclerotic plaque [9,11]. Speci c vascular markers may serve as one of the criteria for assessing in ammation and destabilization of atherosclerotic plaque. In scienti c journals, many articles on the comparison of different vascular markers were published [12,13]. However, Lp-PLA2 is currently considered an independent biomarker for stroke, as well as coronary artery disease and peripheral arterial occlusive disease [15].
The JUPITER trial con rmed that patients with high Lp-PLA2 activity had more than twofold higher risk of developing cardiovascular events compared to those with low Lp-PLA2 activity [30].
Our study showed that patients with symptomatic stenosis of the internal carotid artery had signi cantly higher plasma levels of Lp-PLA2 compared to patients with asymptomatic stenosis.
When assessing the results, there was found a statistical signi cance between the increase in Lp-PLA2 concentration and the morphological structure of the atherosclerotic plaque.
Our results strongly con rm the role of Lp-PLA2 in the pathophysiology and clinical presentation of an unstable carotid plaque. Similar to our ndings, some studies have reported the association of increased plasma levels of Lp-PLA2 in patients with unstable atherosclerotic plaque [31,32].
The obtained results indicated the fact that in patients with soft atherosclerotic plaque, the level of Lp-PLA2 concentration was statistically higher compared to patients with hard atherosclerotic plaque. In addition, in patients with soft atherosclerotic plaque, the level of Lp-PLA2 increased to the level of Lp-PLA2 concentration observed in symptomatic patients.
Therefore, we can state that soft atherosclerotic plaque, as well as an increased concentration of Lp-PLA2, is a risk factor for developing stroke.
The innovative approach proposed, clinically veri ed individually detected formula for determining the risk of stroke development. It considers three main risk factors: the degree of ICA stenosis, atherosclerotic plaque structure and Lp-PLA2 concentration.
According to the results of our study, the consideration of several factors increases the accuracy of calculating the risk of stroke development for particular individuals, which is the basis for risk strati cation algorithms.
The use of the proposed method for mathematical calculation of the risk index for stroke development using the formula "Stroke-Stop" may serve as an auxiliary criterion at the stage of determining and selecting treatment tactics for patients with ICA stenosis greater than 70%, and nally to apply predictive and prognostic patient-speci c treatment of atherosclerosis supporting the shift from reactive medicine to predictive, preventive, and personalized medicine.
In addition, thus proposed formula accompanied and accomplished with lipid individual pro le can be applied alternatively to typical post symptomatic treatment of atheroslerosis. HDL-associated Lp-PLA2 may substantially contribute to the HDL antiatherogenic activity, and could be additionally applicable for the prediction of the e cacy of prescribed medication. Correspondingly, our Stop-Stroke formula is recommended for implementation for personalized clinical application of therapies. It is anticipated that, ultimately, this change in diagnosis and therapy will help in the future design and development of new, more selective and effective therapies for each individual patient.

Conclusions
The The study was approved by the Ethics Committee of the East Slovak Institute of Cardiovascular Diseases (07/12/2015). All the patients, including negative controls, were informed and gave written consent to participate in the study.

Consent for publication
All the patients, including negative controls, were informed and gave written consent to participate in the study.
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests. Ratio of ICA stenosis, atherosclerotic plaque structure and Lp-PLA2 concentration 1 -soft atherosclerotic plaque, 2 -mixed atherosclerotic plaque, 3 -hard atherosclerotic plaque