Clinical and Biochemical Predictors of Unfavorable. Prognosis for Heart Failure with Preserved. Ejection Fraction and Type 2 Diabetes Mellitus

Biological Markers and Guided Therapy, Vol. 2, 2015, no. 1, 33-48 HIKARI Ltd, www.m-hikari.com http://dx.doi.org/10.12988/bmgt.2015.5103 Clinical and Biochemical Predictors of Unfavorable Prognosis for Heart Failure with Preserved Ejection Fraction and Type 2 Diabetes Mellitus Iurii S. Rudyk 1, Ganna V. Bolotskykh 2,* and Maryna M. Udovychenko 3 1 Department of Clinical Pharmacology and Pharmacotherapy Government Institution L.T. Malaya Therapy National Institute of National Academy of Medical Sciences of Ukraine Postysheva avenue 2-a, 61039 Kharkiv, Ukraine * Corresponding author 2 Department of Clinical Pharmacology and Pharmacotherapy, Government Institution L.T.Malaya Therapy National Institute of National Academy of Medical Sciences of Ukraine, Postysheva avenue 2-a, 61039 Kharkiv, Ukraine 3 Department of Clinical Pharmacology and Pharmacotherapy, Government Institution L.T.Malaya Therapy National Institute of National Academy of Medical Sciences of Ukraine, Postysheva avenue 2-a, 61039 Kharkiv, Ukraine Copyright 2015 Iurii S. Rudyk et al. This article is distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Abstract Purpose: To identify the most important clinical and laboratory markers that can be associated with an unfavorable course of HFpEF in patients with 2 type DM and without it. Methods. 108 diabetic and non-diabetic patients (57 males and 51 females) with HFpEF (EF 45%), NYHA Classes II and III were examined. Observation period was up to 30 months. The Galectin-3, TNF-alpha, NT-pro- BNP, CRP and insulin levels were measured in serum using ELISA according to manufacturer's instructions. For revealing associations between estimated endpoints occurrences and investigated factors a Cox proportional hazard analysis was performed. Discrimination refers to a model s ability to correctly distinguish

34 Iurii S. Rudyk et al. 2 classes of outcomes. All statistical tests were 2-tailed, and p<0.05 was considered statistically significant. Results. Cardiovascular events have been reported in 25 (23.1%) patients. Significant differences between two groups of patients were found in 6 parameters such as loghf duration, lnrscs, logtriglyceride, logtnf-α, loggalectin-3 and lognt-pro-bnp. The most informative parameters for classification of patients in the groups of favorable or unfavorable HFpEF prognosis were level of Galectin-3 (p = 0.001), triglycerides (p = 0.020) and RSCS (p = 0.004). After calculating the coefficients for these parameters CLDF, they have been used for mathematical model creation. The mathematical model obtained using discriminant analysis allows with 84.4% accuracy to include patients in the group of favorable or unfavorable course of HFpEF. Conclusions. It was found, that increased levels of Galectin-3, TNF-α, triglycerides and NT-proBNP were associated with increased 30-months cumulative cardiovascular events number among diabetic and nondiabetic HFpEF patients. The most informative parameters for unfavorable course of HFpEF stratification was concentration of Galectin-3 (p <0.0001), triglycerides (p <0.01) and RSCS scores (p <0, 0,001). Keywords: heart failure with preserved ejection function, diabetes mellitus, prognosis, predictors Introduction Heart failure with preserved ejection fraction (HFpEF) is a syndrome with several underlying etiologic factors and numerous pathophysiological changes that can lead to the development of heterogeneous disease. The main pathophysiological factors causing the development and progression of this disease are inflammation, accumulation of extracellular matrix and fibrosis. These changes make a significant contribution to the development of left ventricular filling abnormalities that is the hallmark of this disease [8, 16, 17]. The most important co-morbidity, associated with an increased risk of heart failure manifestation is type 2 diabetes mellitus (DM). A special feature of heart failure development and progression in patients with type 2 DM is a combination of ischemic and non-ischemic risk factors. According to numerous studies, the presence of metabolic disorders in patients with type 2 diabetes, such as insulin resistance (IR), hyperglycemia, chronic systemic inflammation, excess of free fatty acids (FFA) and free radicals, can lead to the structural and functional heart abnormalities even in the absence of ischemic lesion of heart muscle [6, 10, 24, 25]. Moreover, the combination of "ischemic" and "metabolic" components of myocardial damage in this group of patients is associated not only with more severe structural and functional heart muscle disorders, but also with lesions in such target organs as kidney and liver [9, 19].

Clinical and biochemical predictors of unfavorable prognosis 35 The two most common features associated with type 2 DM such as abdominal obesity (AO) and IR are the integral pathogenetic and prognostic unfavorable factors for HFpEF [11, 14, 20, 22]. These factors make a significant contribution to the development and progression of heart failure, increase the risk of adverse events in HFpEF patients, enhance apoptosis of cardiomyocytes, cause energy balance derangements and create not only structural, but also a "metabolic" cardiac remodeling [3, 11, 18]. The presence of common etiopathogenetic mechanisms in HFpEF manifestation creates a new paradigm of HFpEF and explains the development of structural and functional myocardial lesion. The presence of comorbidities enhance the systemic proinflammatory state. It has been established that the diabetic patients with HFpEF have more pronounced immunoinflammatory activation with increased levels of LP-IgG, LP- IgA and TNF-α. [27]. Conclusions regarding the levels of cytokines in HFpEF can be related to the disturbance in cytokines complex that can include the activation of both inflammatory mediators and profibrotic factors. Recently, the question of determination of potential pathophysiological mechanisms regarding the action and participation of Galectin-3 in the development and progression of cardiovascular disease, particularly heart failure, is very actual. A number of studies involving the heart failure patients have demonstrated an independent predictive property of Galectin - 3 levels on risk of cardiovascular events [4, 15, 23], response to heart failure treatment [7] and shortand long-term prognosis determination of HFpEF [1, 5, 13, 21]. Thus, morphological changes in heart failure myocardium in diabetic patients are characterized by hypertrophy of cardiomyocytes and myocardial fibrosis with increasing number of extracellular matrix in the ventricular interstices wall [2, 18]. (PoorimaIG et. all, 2006). Purpose To identify the most important clinical and laboratory markers that can be associated with an unfavorable course of HFpEF in patients with 2 type DM and without it. Methods Study design and patient population. The baseline clinical characteristics of examined patients are shown in Table 1 and include 108 patients (51 males and 51 females) with HFpEF and left ventricular EF 45%, NYHA Classes II and III with and without type 2 DM. The mean age was 62.71±9.18 years. All patients were Caucasian. The exclusion criteria in our study were: left ventricular EF <45%; type 1 diabetes mellitus, valvular heart disease; recent (up to 10 days) episodes of acute heart failure; acute coronary syndrome within the previous 3 months; inflammatory diseases in the acute stage; increase of thyroid function; cancer. All participants signed written informed consent. The study protocol was approved by the Ethics Committee of the Government Institution L.T. Malaya National Institute of Therapy NAMS of Ukraine and all study procedures complied

36 Iurii S. Rudyk et al. with the ethical standards outlined in the 1975 Declaration of Helsinki, as revised in 1983. All patients received standard medical therapy for HF without any differences in both groups. Diabetic patients received metformin. End points during 30 months in our analysis used for prognosis of HFpEF course were all-cause mortality, heart failure decompensation and development of adverse cardiovascular events. In addition, cumulative fate of patients without adverse events and time to adverse events occurrence were analyzed. Clinical, biochemical and instrumental measurements. Rating scale of clinical state (RSCS) and 6 min walk distance were used for clinical and functional status assessment. The Galectin-3, TNF-alpha, NT-pro-BNP, CRP and insulin levels were measured in serum using ELISA according to manufacturer's instructions. Additionally, the serum glucose, creatinine, cholesterol, triglycerides levels at fasting state were determined. Optical density measurements were performed on a semiautomatic ELISA analyzer «Іmmunochem-2100». HOmeostasis Model Assessment (HOMA) index was calculated is calculated by the formula: IR=fasting glucose fasting insulin/22.5. The upper limit of the HOMA-IR was 2.77 [Wallace, 2004] The ultrasound investigations were performed using Ultrasound's Vivid Three with a 2.5-MHz probe (Japan) and calculated following the American Guidelines of Echocardiography Society. Statistical Analysis. Baseline characteristics are described using medians (Me) and interquartile ranges (P25; P75) or proportions. Categorical variables are expressed as percentages. Mann-Whitney U test for nonparametric data and χ2-test categorical variables were used for groups comparisons. For revealing associations between estimated end-points occurrences and investigated factors a Cox proportional hazard analysis was performed. Probabilities of end point-free survival were obtained by Kaplan-Meier estimates for the 2 groups and then compared by a log-rank test. In the development of mathematical classification model that can predict unfavorable course of HFpEF in diabetic and nondiabetic patients, the discriminant analysis was used. Discrimination refers to a model s ability to correctly distinguish 2 classes of outcomes. Receiver-operator characteristic (ROC) curve analysis was performed to determine the cut-off values that best distinguished the issue (e.g., symptoms, death or combined events) (AUC: Area under the curve). The area under the receiver operating characteristic curve (AUC) summarized the diagnostic discrimination. All statistical tests were 2-tailed, and p < 0.05 was considered statistically significant.

Clinical and biochemical predictors of unfavorable prognosis 37 Characteristics Table 1 Characteristics of patients (Ме [LQ; UQ]) Patients with a favorable course of heart failure (n=83) Patients with a unfavorable course of heart failure (n=25) χ 2, р-value Age, years 63.0 [56.0; 66.25] 64.0 [52.5; 73.0] > 0.05 Male, n (%) 42 (44.7%) 15 (60.0%) > 0.05 Heart failure duration, years 5.0 [3.0; 10.0] 7.0 [3.0; 12.5] > 0.05 NYHA functional class of HF NYHA II functional class, n (%) NYHA III functional class, n (%) 59 (62.8 %) 11 (44.0 %) 35 (37.2%) 14 (56.0%) = 0.090 Myocardial infarction in the past, n (%) 14 (14.9%) 1 (4.0%) > 0.05 Stroke/transient ischemic attack in the past, n (%) 14 (14.9%) 9 (36.0%) χ 2 = 5.642. р = 0.018 Atrial fibrillation, n (%) 11 (11.7%) 6 (24.0%) > 0.05 Disturbances of rhythm and conduction, n (%) 31 (33.0%) 10 (40.0%) > 0.05 Obesity, n (%) 46 (63.9%) 11 (84.6%) > 0.05 Ischemic heart disease, functional class II functional class, n (%) III functional class, n (%) 37 (51.4%) 35 (48.6%) 7 (53.8%) 6 (46.2%) Gastrointestinal diseases, n (%) 52 (55.3%) 20 (80.0%) > 0.05 > 0.05 χ 2 = 5.034. р = 0.025 Nonalcoholic fatty liver disease, n (%) 18 (19.1%) 11 (44.0%) χ 2 = 6.617. р = 0.010

38 Iurii S. Rudyk et al. Table 1 (Continued): Characteristics of patients (Ме [LQ; UQ]) Lung diseases, n (%) 8 (8.5%) 2 (8.0%) > 0.05 Thyroid diseases, n (%) 19 (20.2%) 4 (16.0%) > 0.05 Rating scale of clinical state (RSCS), score 3.0 [2.0; 3.0] 4.0 [3.0; 4.5] = 0.002 SBP, mm Hg 167.5 [160.0; 180.0] DBP, mm Hg 100.0 [90.0; 100.0] 160.0 [152.0; 175.0] 100.0 [85.0; 100.0] > 0.05 0.075 HR, beats/min 72.0 [65.5; 82.0] 72.00 [66.0; 86.5] > 0.05 BMI (kg/m 2 ) 31.17 [29.01; 33.68] 32.65 [31.02; 34.48] = 0.053 6 minute walk test, m 329.0 [246.75; 379.25] 290.0 [174.0; 347.5] = 0.056 Results and discussion The average period of observation in our study was 21.77 ± 6.74 months. In general, during 30 months the course of heart failure was considered as unfavorable in 25 (23.1%) patients with HFpEF. Figure 1 Cumulative incidence for CV events in diabetic patients. Log-rank P=0.067

Clinical and biochemical predictors of unfavorable prognosis 39 It should be mentioned that the incidence of adverse events in general, in patients with type 2 DM was higher than in the group of patients without it (27.5% vs 16.2%, respectively), but this difference did not reach statistical significance (p = 0.068). Most of the patients in the group of HFpEF unfavorable course had arterial hypertension and a family history of coronary artery disease and/ or diabetes. Patients with unfavorable course of HFpEF significantly more often had a history of Stroke/TIA. The proportion of such patients was 36.0% in the group of unfavorable course comparing with 14.9% in the group of patients with a favorable course of HFpEF (χ2 = 5.642; p <0.05). Interestingly, in our study the prevalence of gastrointestinal diseases and nonalcoholic fatty liver disease in the group of patients with unfavorable course of HFpEF was higher (55% and 80%, respectively). Mechanisms of structural and functional liver abnormalities formation in patients with HFpEF and especially when it is combined with type 2 DM have certain features. One of the most important pathogenic mechanism is phenomenon of IR. Kimura Y. et al [12] demonstrated that all patients with nonalcoholic fatty liver disease have postprandial hyperinsulinemia, which primarily manifested by developing chronic hyperglycemia. On the one hand, utilization of fatty acids in condition of IR is an alternative way to glucose increase, which prevents binding of insulin on hepatocytes and causes hyperinsulinemia, hyperglycemia and insulin resistance potentiates, closing pathogenetic vicious circle. On the other hand, cascade of reactions leading to the proatherogenic lipoprotein fractions synthesis intensification is formed [26]. In our study, it was established that the patients with unfavorable course of HFpEF have higher levels of creatinine, uric acid, TNF-α and Galectin-3 (p = 0.027, p = 0.034, p = 0.045 and p = 0.027, respectively). At the same time significant differences regarding glucose-metabolic parameters between these two groups of patients were not found (Table 2, Fig. 2 and 3). Table 2 presents the results of this analysis. Table 2 Basic biochemical parameters of patients depending on the course of HFpEF (Ме [LQ; UQ]) Characteristics Patients with a favorable course of heart failure (n=83) Patients with a unfavorable course of heart failure (n=25) χ 2, р-value Glucose, mmol/l 6.24 [5.05; 7.94] 6.21 [5.40; 8.79] > 0.05 Total cholesterol, mmol/l 5.39 [4.54; 6.10 5.73 [4.61; 6.75] > 0.05

40 Iurii S. Rudyk et al. Table 2 (Continued): Basic biochemical parameters of patients depending on the course of HFpEF (Ме [LQ; UQ]) Triglycerides, mmol/l 1.65 [1.16; 1.39] 1.91 [1.39; 2.56] = 0.063 LDL cholesterol, mmol/l 3.0 [2.0; 3.0] 4.0 [3.0; 4.5] = 0.069 GFR CKD-EPI, ml/m Creatinine, mcmol/l 69.50 [57.00; 81.00] 86.05 [73.00; 105.00] 61.00 [43.00; 86.00] > 0.05 98.00 [76.00; 126.00] = 0.027 Uric acid, mcmol/l 342.10 [284.50; 387.70] 393.50 [326.75; 469.30] = 0.034 Insulin, mcui/ml 18.71 [10.26; 29.79] 14.92 [8.62; 24.33] > 0.05 НОМА-IR 5.09 [2.44; 10.46] 5.17 [2.23; 11.02] > 0.05 TNF-α pg/ml 4.43 [3.18; 6.16] 6.06 [3.65; 10.55] = 0.045 СRP, mg/l 6.25 [4.85; 7.81] 6.57 [4.55; 9.71] > 0.05 NT-pro-BNP, pg/ml 133.0 [111.56; 274.28] 189.35 [119.92; 413.97] = 0.085 Galectin-3, ng/ml 3.06 [2.61; 3.48] 3.34 [2.84; 4.41] = 0.027

Clinical and biochemical predictors of unfavorable prognosis 41 Figure 2. Boxplot shows a significant difference between medians of serum Galectin-3 and TNF-alpha levels in patients with and without cardiovascular events. (р<0.05) Figure 3. Boxplot shows a difference between medians of serum NT-pro-BNP and CRP levels in patients with and without cardiovascular events. (no significant difference) To evaluate the predictive values of investigated parameters in HFpEF prognosis assessment in patients with type 2 DM and without, the combined end points (all-cause death, heart failure decompensation and development of cardiovascular adverse events) were used. At the first stage of mathematical model of unfavorable course of HFpEF development in diabetic and nondiabetic patients, we determined and selected the most informative factors. Because of the expected asymmetry, logarithmic transformation of parameter settings or construction them in the second degree have been conducted for the most of the variables. In detecting possible collinearity between two variables, in further analysis included the factor with greater value ratio of discriminant ability.

42 Iurii S. Rudyk et al. Table 3 Univariate logistic regression analysis of predictors of HF analysis Variables р HRCox Asim, 95% CI Inf Sup Obesity 0,036 3,142 1,077 9,165 Stroke/transient ischemic 0,004 3,375 1,487 7,657 attack in the past NAFLD 0,008 2,941 1,333 6,487 lnrscs 0,0001 1,603 1,276 2,015 Log6 min walk distance 0,045 0,065 0,004 0,941 СRP 0,034 1,051 1,004 1,101 TNF-alpha 0,005 1,122 1,036 1,216 Log galectin-3 0,0001 1,261 1,099 1,446 Log NT-pro-BNP 0,003 1,002 1,000 1,005 Cholesterol 0,028 1,522 1,047 2,211 logtriglyceride 0,017 1,616 1,089 2,397 LogCreatinine 0,003 516,947 8,750 30540,83 3 GFR-CKD-EPI 0,036 2,335 1,058 5,153 LogIVSFT 0,0001 65474,186 131,570 3258248 3,30 LogLVPWFS 0,0001 531691,15 5362,464 5,272Е+ 11 Log MW 0,0001 9474,314 90,899 987501,0 70 LogILVMW 0,005 2039,369 9,637 431579,0 04 IVSFT inter ventricular septum fractional thickening LVPWFS left ventricle posterior wall fraction thickening In the development of mathematical classification model that can predict unfavorable course of HFpEF in diabetic and nondiabetic patients, the discriminant analysis was used. Thirty-two quantitative variables were compulsory included at the first stage of discriminant function creation. Significant differences between two groups of patients were found in 6 parameters such as loghf duration, lnrscs, logtriglyceride, logtnf-α, loggalectin-3 and lognt-pro-bnp (Table 4). We calculated and analyzed the coefficients of canonical discriminant function. In further analysis only seven variables, that showed a statistical significance, were included.

Clinical and biochemical predictors of unfavorable prognosis 43 Table 4. The results of discriminant multivariate analysis using the procedure of compulsory inclusion of all variables Wilks' lambda criterion F p-value loggalectin-3 0.862 12.029 0.001 logtriglyceride 0.930 5.611 0.020 InRSCS 0.894 8.880 0.004 lognt-pro-bnp 0.943 4.512 0.037 logtnf-α 0.909 7.514 0.008 loghf duration 0.950 3.949 0.050 Such well known biological markers as fasting blood glucose, glycosylated hemoglobin and total cholesterol were not included in the discriminant analysis, because they didn t demonstrate high potential predictive value in adverse clinical events occurrence. The most informative parameters for classification of patients in the groups of favorable or unfavorable HFpEF prognosis were level of Galectin-3 (p = 0.001), triglycerides (p = 0.020) and RSCS (p = 0.004). After calculating the coefficients for these CLDF parameters, they have been used for mathematical model creation. The mathematical model obtained using discriminant analysis allows with 84.4% accuracy to include patients in the group of favorable or unfavorable course of HFpEF. Results of classification a Prognostic group affiliation Prognosis,00 1,00 All Basic Frequency Favorable 59 1 60 Unfavorable 11 6 17 % Favorable 98.3 1.7 100.0 Unfavorable 64.7 35.3 100.0 a. 84,4% correctly classified basic grouped observations. For independent predictive value verification of the parameters listed above ROC-analysis was conducted.

44 Iurii S. Rudyk et al. Figure 4 The area under the receiver operating characteristic curve (AUC) Area Std. error Asymp. Mean Asymptotic 95% confidence interval Lower limit Upper limit 0.868 0.052 0.000 0.766 0.970 С-statistic is 0.868, indicating a good quality of this model. Mathematical processing of the primary data set confirmed the unique indicator capabilities of Galectin-3, triglycerides and RSCS. These factors have been used in the development of bio-indicated scale for patients with HFpEF stratification in the group of unfavorable course of this disease based on various modifications of factor analysis. Conclusion 1. Increased levels of Galectin-3, TNF-α, triglycerides and NT-proBNP were associated with increased 30-months cumulative cardiovascular events number among diabetic and nondiabetic HFpEF patients.

Clinical and biochemical predictors of unfavorable prognosis 45 2. The main predictors for unfavorable course of HFpEF (death from cardiovascular causes or hospitalization for cardiovascular reasons) in diabetic and nondiabetic patients during 30 months were RSCS scores (OR = 1.515; 95% CI: 1,202-1,910), the level of galectin-3 (OR = 1.260; 95% CI: 1,095-1,451), TNF-α (OR = 1.125; 95% CI: 1,037-1,221) and triglycerides (OR = 1.226; 95% CI: 1,060-1,417). It was found that the most informative parameters for unfavorable course of HFpEF stratification was concentration of galectin-3 (p <0.0001), triglycerides (p <0.01) and RSCS scores (p <0, 0,001). Conflict of interests: not declared Ethical principles The study protocol was approved by the Ethics Committee of the Government Institution L.T. Malaya National Institute of Therapy N.A.M.S of Ukraine. Before involving, all patients provided written informed consent and this procedure meet all of the requirements governing informed consent of the Ukraine. The study complied with the principles of the Declaration of Helsinki of the World Medical Association (Clinical Research 1996;14:103) Any procedures performed meet with prevailing ethical standards. Funding. No specific grants were received to this research. Study Restrictions The main restriction of our study is the number of patient and duration of observation, but according to the author s opinion it doesn t impact on the study data interpretation. References [1] I. Bošnjak, K. Selthofer-Relatić, Prognostic Value of Galectin-3 in Patients with Heart Failure, Disease Markers, 2015 (2015), 1-6, Article ID 690205. http://dx.doi.org/10.1155/2015/690205 [2] L. Calvier, M. Miana, P. Reboul, et al., Galectin-3 Mediates Aldosterone- Induced Vascular Fibrosis, Arterioscler Thromb. Vasc. Biol., 33 (2013), no. 1, 67-75. http://dx.doi.org/10.1161/atvbaha.112.300569 [3] C. Christoffersen, et al., Cardiac lipid accumulation associated with diastolic dysfunction in obese mice, Endocrinology, 144 (2003), 3483-3490. http://dx.doi.org/10.1210/en.2003-0242 [4] R.A. de Boer, F. Edelmann, A. Cohen-Solal, et al., Galectin-3 in heart failure with preserved ejection fraction, Eur. J. Heart Fail., 15 (2013), 1095-1101.

46 Iurii S. Rudyk et al. http://dx.doi.org/10.1093/eurjhf/hft077 [5] R.A. de Boer, D.J. van Veldhuisen, R.T. Gansevoort, et al., The fibrosis marker galectin-3 and outcome in the general population, J. Intern. Med., 272 (2012), 55-64. http://dx.doi.org/10.1111/j.1365-2796.2011.02476.x [6] E.S. Ford, W.H. Giles, A comparison of the prevalence of the metabolic syndrome using two proposed definitions, Diabetes Care, 26 (2003), 575-581. http://dx.doi.org/10.2337/diacare.26.3.575 [7] L. Gullestad, T. Ueland, J. Kjekshus, et al., Galectin-3 predicts response to statin therapy in the Controlled Rosuvastatin Multinational Trial in Heart Failure CORONA), European Heart Journal, 33 (2012), 2290 2296. http://dx.doi.org/10.1093/eurheartj/ehs077 [8] K. Hogg, K. Swedberg, J. McMurray, Heart failure with preserved left ventricular systolic function: epidemiology, clinical characteristcs, and prognosis, J. Am. Coll. Cardiol., 43 (2004), 317 327. http://dx.doi.org/10.1016/j.jacc.2003.07.046 [9] E. Ingelsson, et al., Novel Metabolic Risk Factors for Heart Failure, J. Am. Coll. Cardiol., 46 (2005), 2054-2060. http://dx.doi.org/10.1016/j.jacc.2005.07.059 [10] E. Ingelsson, J. Sundstrom, J. Arnlov, B. Zethelius, L. Lind, Insulin resistance and risk of congestive heart failure, Journal of the American Medical Association, 294 (2005), 334 341. http://dx.doi.org/10.1001/jama.294.3.334 [11] М.Т. Kearney, Chronic heart failure: A missing component of the metabolic syndrome?, Diab. Vasc. Dis. Res., 6 (2009), 145. http://dx.doi.org/10.1177/1479164109338771 [12] Y. Kimura, et al., Postprandial insulin secretion pattern is associated with histological severity in non-alcoholic fatty liver disease patients without prior known diabetes mellitus, Journal of Gastroenterology and Hepatology, 26 (2011), 517-522. http://dx.doi.org/10.1111/j.1440-1746.2010.06567.x [13] F. Kramer, Galectin-3: clinical utility and prognostic value in patients with heart failure, Research Reports in Clinical Cardiology, 4 (2013), 13-22. http://dx.doi.org/10.2147/rrcc.s28562 [14] J.G. Leichman, et al., Association of plasma free fatty acids and left ventricular diastolic function in patients with clinically severe obesity, Am. J. Clin. Nutr., 84 (2006), 336-341.

Clinical and biochemical predictors of unfavorable prognosis 47 [15] D.J. Lok, S.I. Lok, P.W. Bruggink-Andre de la Porte, et al., Galectin-3 is an independent marker for ventricular remodeling and mortality in patients with chronic heart failure, Clin. Res. Cardiol., 102 (2013), 103-110. http://dx.doi.org/10.1007/s00392-012-0500-y [16] J. Meluzín, J. Tomandl, Can Biomarkers Help to Diagnose Early Heart Failure with Preserved Ejection Fraction?, Disease Markers, 2015 (2015), 1-9, Article ID 426045. http://dx.doi.org/10.1155/2015/426045 [17] W. J. Paulus, C. Tschöpe, A Novel Paradigm for Heart Failure With Preserved Ejection Fraction, Journal of the American College of Cardiology, 62 (2013), 263-271. http://dx.doi.org/10.1016/j.jacc.2013.02.092 [18] I.G. Poornima, P. Parikh, R.P. Shannon, Diabetic сardiomyopathy: The search for a unifying hypothesis, Circulat. Res., 98 (2006), 596-605. http://dx.doi.org/10.1161/01.res.0000207406.94146.c2 [19] L.J. Rijzewijk, et al., Myocardial steatosis is an independent predictor of diastolic dysfunction in type 2 diabetes mellitus, J. Am. Coll. Cardiol., 52 (2008), 1793 1799. http://dx.doi.org/10.1016/j.jacc.2008.07.062 [20] F. Satoshi, Insulin Resistance and Heart Failure: Underlying Molecular Mechanisms and Potential Pharmacological Solutions, J. Cardiovasc. Pharmacol., 62 (2013), 379-380. http://dx.doi.org/10.1097/fjc.0b013e3182a18c56 [21] N. Sherwi, S. Merali, K. Wong, Personalizing biomarker strategies in heart failure with galectin-3, Future Cardiology, 8 (2012), 885-894. http://dx.doi.org/10.2217/fca.12.65 [22] N. Suskin, et al., Glucose and insulin abnormalities relate to functional capacity in patients with congestive heart failure, Eur. Heart J., 21 (2000), 1368-1375. http://dx.doi.org/10.1053/euhj.1999.2043 [23] T. Ueland, P. Aukrust, K. Broch, et al., Galectin-3 in heart failure: high levels are associated with all-cause mortality, Int. J. Cardiol., 150 (2011), 361-364. http://dx.doi.org/10.1016/j.ijcard.2011.05.081 [24] C. Voulgari, D. Papadogiannis, N. Tentolouris, Diabetic cardiomyopathy: from the pathophysiology of the cardiac myocytes to current diagnosis and management strategies, Vasc. Health Risk Manag., 6 (2010), 883-903. http://dx.doi.org/10.2147/vhrm.s11681 [25] R.M. Witteles, M.B. Fowler, Insulin-resistant cardiomyopathy: clinical evidence, mechanisms, and treatment options, J. Am. Coll. Cardiol., 51 (2008), 93-102. http://dx.doi.org/10.1016/j.jacc.2007.10.021

48 Iurii S. Rudyk et al. [26] V.T. Ivashkin, Ju.O. Shul'pekova, Nealkogol'nyj steatogepatit, Rus. Med. Zhurnal, 2 (2000). 41-45. [27] S.A. Serik, Proteїnvmіsnі іmunnі kompleksi ta faktor nekrozu puhlin-a pri dіastolіchnіj sercevіj nedstatnostі u hvorih na cukrovij dіabet 2 tipu, Krims'kij Terapevtichnij Zhurnal, 2 (2010), 123-127. Received: October 15, 2015; Published: November 10, 2015