We studied 344 (mean age 61 ± 10 years, 54% female) patients with the clinical diagnosis of symptomatic CHF, and New York Heart Association (NYHA) functional class I–III, secondary to ischemic or non-ischemic etiology, based on the current definitions (21). Patients were referred to the Clinic of Cardiology, University Clinical Centre of Kosovo, between May 2013 and September 2017. At the time of the study, all patients were on optimum HF medications, optimized at least 2 weeks prior to enrollment. Based on patient's symptoms and renal function: 85% were receiving ACE inhibitors or ARB, 76% beta-blockers, 11% calcium-blockers, 8% digoxin, 54% spironolactone, and 58% diuretics. Of the enrolled patients, 45% had ischemic etiology, 38% hypertensive, and 17% had unknown etiology. Furthermore, 17% of the included patients were in atrial fibrillation. Patients with clinical evidence for cardiac decompensation (NYHA class IV, those with peripheral edema), limited physical activity due to factors other than cardiac symptoms (e.g., arthritis), severe mitral regurgitation, more than mild renal failure (in patients with raised creatinine, the glomerular filtration rate (GFR) was measured and patients with values <60 ml/min/1.73 m² were excluded), chronic obstructive pulmonary disease or those with recent acute coronary syndrome, stroke, or anemia were excluded from the study. Type 2 DM was defined as a fasting blood glucose level ≥7.0 mmol/L, a glycohemoglobin A1c (HbA1c) level ≥6.5%, and/or the need for oral hypoglycemic medications or insulin. All patients gave written informed consent to participate in the study, which was approved by the local Ethics Committee.

Detailed history and clinical assessment were obtained in all patients, in whom routine biochemical tests were also performed, such as hemoglobin, lipid profile, blood glucose level, and kidney function tests. Estimated body mass index (BMI) was calculated from weight and height measurements. Body surface area (BSA) was calculated using the Du Bois formula: BSA (m²) = 0.007184 × (height in cm) ^0.725 × (weight in kg)^0.425 (22). Waist and hip measurements were also made and a waist/hips ratio was calculated.

A single operator performed all echocardiographic examinations using a Philips Intelligent E-33 system with a multi-frequency transducer, and harmonic imaging as appropriate. Images were obtained with the patient in the left lateral decubitus position and during quiet expiration. Measurements of interventricular septal thickness, posterior wall thickness, and LV dimensions were made at end-diastole and end-systole, as recommended by the American Society of Echocardiography (23).

Left ventricular volumes and EF were calculated from the apical 2 and 4 chamber views using the modified Simpson's method. Ventricular long axis motion was studied by placing the M-mode cursor at the lateral and septal angles of the mitral annulus and the lateral angle of the tricuspid annulus. The total amplitude of ventricular long-axis motion was measured as previously described (24) from peak inward to peak outward points. The indices were registered as lateral and septal mitral annular plane systolic excursion (MAPSE) and tricuspid annular plane systolic excursion (TAPSE). LV and right ventricular (RV) long-axis myocardial velocities were also studied using the Doppler myocardial imaging technique. From the apical 4-chamber view, longitudinal velocities were recorded with the sample volume placed at the basal part of LV lateral and septal segments as well as the RV free wall. Systolic (s′) as well as early and late (e′ and a′) diastolic myocardial velocities were measured with the gain optimally adjusted. A mean value of lateral and septal LV velocities was calculated. The left atrial diameter was measured from aortic root recordings with the M-mode cursor positioned at the level of the aortic valve leaflets.

Diastolic LV and RV functions were assessed from filling velocities using spectral pulsed wave Doppler with the sample volume positioned at the tips of the mitral and tricuspid valve leaflets, respectively, during a brief apnea. Peak LV and RV early (E wave) and late (A wave) diastolic velocities were measured and E/A ratios were calculated. E wave deceleration time (DT) was also measured from the peak E wave to the end of its deceleration. The E/e′ ratio was calculated from the trans-mitral E wave and mean lateral and septal segments myocardial e′ wave velocities. The LV filling pattern was considered “restrictive” when the E/A ratio was >2.0, the E wave deceleration time <140 ms, and the left atrium (LA) dilated >40 mm in transverse diameter (25). Total LV filling time was measured from the onset of the E wave to the end of the A wave and ejection time from the onset to the end of the aortic Doppler flow velocity.

Mitral regurgitation severity was assessed by color and continuous wave Doppler and was graded as mild, moderate, or severe according to the relative jet area to that of the LA as well as the flow velocity profile, in line with the recommendations of the American and European Society of Echocardiography (26, 27). Similarly, tricuspid regurgitation was assessed by color Doppler and continuous-wave Doppler. Retrograde trans-tricuspid pressure drop >35 mmHg was taken as evidence for pulmonary hypertension (27, 28). All M-mode and Doppler recordings were made at a fast speed of 100 mm/s with a superimposed ECG (lead II).

Within 24 h of the echocardiographic examination, a 6-MWT was performed on a level hallway surface, administered by a specialized nurse who was blinded to the results of the echocardiogram. According to the method of Gyatt et al. (29), patients were informed of the purpose and protocol of the 6 MWT, which was conducted in a standardized fashion while patients were on their regular medications (30, 31). A 15-m flat, obstacle-free corridor was used and patients were instructed to walk as far as they can, turning 180 degrees after they have reached the end of the corridor, during the allocated time of 6 min. Patients walked unaccompanied so as not to influence their walking speed. At the end of the 6 min, the supervising nurse measured the total distance walked by the patient.

Data are presented as mean ± SD or proportions (% of patients). Continuous data were compared with two-tailed unpaired Student's t-test and discrete data with a chi-square test. Correlations were tested with Pearson's coefficients. Predictors of the 6MWT distance were identified with univariate analysis, and multivariate logistic regression was performed using the step-wise method. A significant difference was defined as p < 0.05 (two-tailed). Patients were divided according to their ability to walk >300 m into good and limited exercise performance groups, and were compared using unpaired Student's t-test. Additionally, patients with HFpEF (LVEF ≥ 40%) were compared with those with HFrEF (LVEF < 40%) using the unpaired t-test. Due to the possible interaction of age with echocardiographic parameters, we used the general linear model to compare age-adjusted mean values of echocardiographic indices between groups.

Patients with HF and T2DM had a higher prevalence of arterial hypertension (p = 0.004), higher waist/hips ratio (p = 0.041), higher creatinine (p = 0.008), urea (p = 0.003), and lower hemoglobin (p = 0.001), and completed the 6-MWT for a shorter distance (<0.001) than those with HF but no T2DM (Table 1; Figure 1). The rest of the clinical indices and echocardiographic parameters were not different between groups (Table 2).

Patients with limited exercise capacity had a higher prevalence of T2DM (p < 0.001), arterial hypertension (p = 0.004), and atrial fibrillation (p = 0.001), a higher waist/hips ratio (p = 0.041), a level of fasting glucose (p < 0.001), and a lower level of hemoglobin (p < 0.001), compared with those with good exercise capacity. In addition, they had larger LA (p = 0.002), reduced lateral MAPSE (p = 0.032), septal MAPSE (p < 0.001), and TAPSE (p < 0.001), compared to patients with good 6-MWT performance. The rest of the clinical and echocardiographic indices were not different between subgroups (Tables 3, 4).

In the univariate analysis model, T2DM (p < 0.001), low hemoglobin level (p < 0.001), atrial fibrillation (p < 0.001), and NYHA class (p = 0.008) predicted limited 6-MWT distance, as did enlarged LA (p = 0.003), increased E wave velocity (p = 0.019), raised E/e′ (p = 0.028), reduced lateral MAPSE (p = 0.033) septal MAPSE (p < 0.001), TAPSE (p < 0.001) and septal a′ and s′ (p = 0.032 and p = 0.041, respectively), and increased E/e′ (p = 0.028). In multivariate analysis [odds ratio (OR) 95% confidence interval (CI)], only diabetes [3.366 (1.907–5.939), p < 0.001], low hemoglobin [0.847 (0.729–0.985), p = 0.031], atrial fibrillation [2.684 (1.273–5.657), p = 0.009], and reduced septal MAPSE [0.308 (0.125–0.759), p = 0.010], independently predicted the limited 6-MWT distance (Table 5).

Type 2 diabetes mellitus (p < 0.001), low hemoglobin (p = 0.022), atrial fibrillation (p = 0.020), NYHA class (p = 0.005), LA (p = 0.03), septal MAPSE (p < 0.001), and lateral MAPSE (p = 0.044) predicted limited 6-MWT distance in HFpEF. In multivariate analysis [OR 95% CI], only diabetes [6.083 (2.613–14.160), p < 0.001], atrial fibrillation [6.092 (1.769–20.979), p = 0.004], and septal MAPSE [0.063 (0.027–0.184), p = 0.002], independently predicted the limited 6-MWT distance in HFpEF (Tables 6, 7).

In univariate analysis, low hemoglobin (p = 0.001), reduced TAPSE (p = 0.001), and enlarged LA (p = 0.043) predicted limited 6-MWT distance in patients with HFrEF. In multivariate analysis, only low hemoglobin [0.786 (0.624–0.998), p = 0.049] and reduced TAPSE [0.462 (0.214–0.988), p = 0.041] independently predicted the limited 6-MWT distance in HFrEF (Tables 6, 7).

Heart failure patients with limited exercise capacity had a higher prevalence of T2DM, arterial hypertension, and atrial fibrillation, compared with those with good exercise capacity. They also had a higher waist/hips ratio, lower hemoglobin, a larger LA, and compromised LV and RV long-axis systolic function. Patients with combined HF and T2DM had a higher prevalence of arterial hypertension, a higher waist/hips ratio, more compromised renal function, lower hemoglobin, and a shorter 6-MWT distance, compared with those with HF with no T2DM. Multivariate analysis showed diabetes, lowered hemoglobin level, atrial fibrillation, and reduced LV long-axis function as independent predictors of limited exercise capacity in patients with HF.

While the above common knowledge on the relationship between atherosclerosis risk factors and HF is confirmed in our patients, their impact on predicting exercise capacity differed significantly according to LVEF. In patients with reduced LVEF, low hemoglobin and compromised RV long-axis systolic function were the two independent predictors of exercise capacity. However, in patients with LV preserved EF, T2DM, low hemoglobin level, atrial fibrillation, higher NYHA class, and compromised LV long-axis systolic function were the respective predictors.

Exercise intolerance is the main symptom in patients with HF, regardless of LVEF (32, 33). In these patients, different echocardiographic indices have been shown as important predictors of exercise capacity, particularly raised LV filling pressures (34–42). Such a relationship can be explained on the basis of reduced stroke volume and pulmonary venous hypertension (43). This is, however, only one explanation of exercise intolerance in HF. Our results provide a clearer image as to the potential mechanisms involved in reduced exercise capacity in patients with HF, when they are classified according to EF. HF with reduced EF is commonly caused by ischemic myopathy that involves both ventricles with their impact on cardiac output and kidney function (44). Indeed, our findings confirm that, having shown that low hemoglobin (45) and compromised RV systolic function (46) are the two main predictors of limited exercise. However, in patients with preserved LVEF, the scenario differs having shown that T2DM and atrial fibrillation are two additional predictors of exercise capacity. Atrial fibrillation is a very common finding in HFpEF. Most such patients are known to have long-standing hypertensive LV disease and left atrial enlargement with its known complications (47). Atrial fibrillation loses the atrial systolic filling component of the LV, hence compromising overall stroke volume with its impact on exercise capacity (48). T2DM, on the other hand, enhances the atherosclerosis pathology at both main coronary arteries level as well as microcirculation with resulting subendocardial fibrosis and LV cavity stiffness, raising the filling pressures and eventually causing atrial fibrillation (49). Through the same atherosclerotic pathophysiology, T2DM also impacts the peripheral circulation and, over the years, causes peripheral neuropathy. Although our analysis identified individual independent predictors of the limited exercise capacity in patients with HF, it must be mentioned that the pathomechanisms are closely related, particularly T2DM and atrial fibrillation, and their impact on LV myocardial function with its consequence on left atrial enlargement, atrial fibrillation, and its complications.

The main limitation of our study is that we did not assess the response of echocardiographic measurements to exercise, at the time of symptoms development. However, the main objective of the study was to identify predictors of ordinary walking exercise limitations rather than a heavy exercise in patients with HF. The lack of left atrial pressure invasive measurements is another limitation, but the study was based on conventional Doppler measurements, which have been shown to be reproducible and correlate closely with invasive pressure measurements (50). We did not have myocardial deformation measurements in our cohort, which might have altered the results.

Type 2 diabetes mellitus has a significant impact on exercise intolerance in patients with HFpEFEF. While the cardiac pump function looks better preserved than in patients with HFrEF, the multi-system complications associated with diabetes should be acknowledged, particularly myocardial microcirculation and peripheral arterial disease as well as peripheral neuropathic complications.