Background
The aim of this study was to determine right ventricular (RV) and right atrial (RA) deformation assessed by two-dimensional echocardiographic and three-dimensional echocardiographic (3DE) imaging in patients with prediabetes and type 2 diabetes mellitus.
Methods
This cross-sectional study included 47 untreated normotensive subjects with prediabetes, 57 recently diagnosed normotensive patients with diabetes, and 54 healthy controls of similar sex and age distributions. All subjects underwent laboratory analyses and complete two-dimensional echocardiographic and 3DE examinations.
Results
Three-dimensional echocardiographic RV end-diastolic volume index gradually decreased from controls across patients with diabetes to those with diabetes (69 ± 10 vs 63 ± 8 vs 58 ± 8 mL/m 2 , P < .001), whereas 3DE RV end-systolic volume index was higher in controls compared with patients with diabetes and those with diabetes (25 ± 4 vs 23 ± 4 vs 22 ± 4 mL/m 2 , P < .001). However, there was no difference in 3DE RV ejection fraction among the three groups (63 ± 4% vs 62 ± 4% vs 61 ± 5%, P = .063). RV and RA global strain and systolic and early diastolic strain rates were decreased in patients with prediabetes and in those with diabetes compared with controls, whereas RV and RA late diastolic strain rates were increased in these patients. Multivariate regression analysis showed that RV global strain was associated with glycated hemoglobin, independent of left ventricular parameters.
Conclusions
RV and RA myocardial deformation and function obtained by 3DE and two-dimensional echocardiographic strain, even in normal ranges, were decreased in patients with prediabetes and in those with diabetes compared with controls. The long-term parameter of glucose control was correlated with the right heart mechanics.
Diabetes mellitus is a well-known cardiovascular risk factor that induces a wide range of cardiac damage, from diastolic dysfunction and cardiac hypertrophy to coronary artery disease. Most studies, even the latest, focus on left ventricular (LV) functional and mechanical remodeling in diabetes, and only a limited number of studies have investigated right ventricular (RV) structure and function in type 2 diabetes mellitus. Our study group previously published a study that compared RV structure and function in subjects with the metabolic syndrome, diabetes, and arterial hypertension. However, only a few investigations have studied RV myocardial mechanics or three-dimensional RV structure and function assessed by cardiac magnetic resonance (CMR) in patients with type 2 diabetes mellitus. Additionally, a study evaluating RV remodeling in patients with prediabetes has not been published so far.
The introduction of new imaging techniques, primarily three-dimensional echocardiographic (3DE) and two-dimensional echocardiographic (2DE) speckle-tracking, enables detailed insight into RV deformation and function, which could not be detected by conventional echocardiography because of the inaccessibility of the right ventricle. Studies have shown that the accuracy and reproducibility of these particular techniques are comparable with those of CMR, which is considered the current “gold standard” for RV imaging. To our knowledge, this is the first study to comprehensively investigate RV function and mechanics using 3DE and 2DE speckle-tracking imaging in subjects with prediabetes and patients with recently diagnosed type 2 diabetes mellitus.
We sought to investigate RV function and mechanics in individuals with prediabetes and diabetes using 3DE and 2DE strain and to evaluate the association between glucose regulation and RV function and deformation.
Methods
The present cross-sectional study included 47 normotensive subjects (blood pressure < 140/90 mm Hg measured on several separate occasions) with prediabetes, 57 normotensive patients untreated for type 2 diabetes, and 54 controls of similar age and sex distributions. The study was conducted between September 2012 and January 2014. The diagnosis of prediabetes and diabetes was based on current recommendations. Prediabetes was diagnosed if the fasting blood glucose level was between 5.6 and 6.9 mmol/L, glycated hemoglobin (HbA 1c ) was between 5.7% and 6.4%, and the blood glucose level on an oral glucose tolerance test (OGTT) after 2 hours was between 7.6 and 11.1 mmol/L. Diabetes was diagnosed if the blood glucose level was ≥7 mmol/L, HbA 1c was ≥6.5%, and the OGTT blood glucose level after 2 hours was ≥11.1 mmol/L. Subjects were classified into one of these three categories (control, prediabetes, or diabetes) if two or more criteria were satisfied.
Subjects with symptoms or signs of cardiovascular disease (arterial hypertension, angina pectoris, heart failure, myocardial infarction, significant valvular disease, atrial fibrillation, congenital heart disease), obesity (body mass index [BMI] ≥ 30 kg/m 2 ), asthma, chronic obstructive lung disease, neoplastic disease, cirrhosis of the liver, or kidney failure were excluded from the study.
Anthropometric measures (height and weight), and laboratory analyses (fasting glucose level, HbA 1c , OGTT blood glucose level after 2 hours, total cholesterol, triglycerides, high-density lipoprotein and low-density lipoprotein levels, creatinine, and C-reactive protein) were taken in all subjects included in the study. Fasting venous blood samples were drawn between 8 and 9 am , and the patients had not had any caloric intake for ≥10 hours. We were not able to determine the presence of microalbuminuria in all patients. Participants did not take any medications before inclusion in the study. BMI and body surface area (BSA) were calculated for each patient. The study was approved by the local ethics committee, and informed consent was obtained from all the participants.
Echocardiography
The echocardiographic examinations were performed by using a 2.5-MHz transducer with harmonic capability and a Vivid 7 ultrasound machine (GE Vingmed Ultrasound AS, Horten, Norway) and a 3V matrix probe for 3D data set acquisitions.
All 2DE parameters were obtained as the average values of three consecutive cardiac cycles. LV end-diastolic and end-systolic diameters, posterior wall thickness, and interventricular septal thickness were determined according to current recommendations. LV ejection fraction was estimated by using the biplane Simpson method. LV mass was calculated by using the Penn formula and indexed to BSA.
Left atrial (LA) maximal volume was measured according to the biplane area-length method in four- and two-chamber views and indexed to BSA.
Transmitral Doppler inflow velocities were obtained in the apical four-chamber view. Pulsed Doppler measurements included transmitral early diastolic peak flow velocity (E), late diastolic flow velocity (A), their ratio (E/A), and E-velocity deceleration time. Doppler tissue imaging was used to obtain LV myocardial velocities in the apical four-chamber view, with a sample volume placed at the septal segment of the mitral annulus during early and late diastole (e′ and a′). The average of the peak early diastolic relaxation velocity (e′) of the septal and lateral mitral annulus was calculated, and the E/e′ ratio was computed for each participant.
Right Ventricle and Atrium
RV internal diameter was measured in the parasternal long-axis view. RV thickness was measured in the subcostal view. Right atrial (RA) maximal volume was obtained in the four-chamber view during ventricular end-systole.
Tricuspid flow velocities were assessed by pulsed-wave Doppler in the apical four-chamber view. The following parameters were determined: early diastolic peak flow velocity (E t ), late diastolic flow velocity (A t ), and their ratio (E/A) t . Doppler tissue imaging was used to obtain RV myocardial velocities in the apical four-chamber view, with the sample volume placed at the lateral segment of the tricuspid annulus during early diastole (e′ t ) and systole (s t ). Tricuspid (E/e′) t ratio was determined by using previously estimated Doppler values.
RV global systolic function was assessed as the tricuspid annular plane systolic excursion. RV systolic blood pressure was assessed in patients with minimal or mild tricuspid regurgitation. For the determination of RV systolic blood pressure, we used the Bernoulli equation for tricuspid regurgitation velocity using RA pressure. RA pressure was estimated from inferior vena cava diameter and collapsibility, and it was usually between 3 and 5 mm Hg in our study population.
Two-Dimensional Strain and Strain Rate Analysis
Two-dimensional echocardiographic strain imaging was performed by using three consecutive cardiac cycles. The frame rate ranged between 50 and 70 Hz. Commercially available software 2DE Q-analysis (EchoPAC version 110.1.2; GE Vingmed Ultrasound AS) was used for the 2DE strain analysis.
The 2DE longitudinal strain of the left ventricle was calculated by averaging all values of the regional peak longitudinal strain obtained from 17 segments in two-chamber, apical long-axis, and four-chamber apical views.
Two-dimensional echocardiographic strain imaging of the right ventricle was performed in the apical four-chamber view. The variables used for evaluation of RV systolic function and contractility were the longitudinal peak and systolic strain rate, respectively. The parameters of early myocardial relaxation and late ventricular filling were estimated by early and late diastolic strain rate. Peak longitudinal strain and systolic and diastolic strain rates for the RV lateral wall, interventricular septum, and global right ventricle were determined separately. The software provided values of longitudinal strain and strain rate for six segments: three segments of the septum and three segments of the RV free wall (proximal, medial, and apical segments for each wall; Figures 1 A and 1C ). Global longitudinal RV strain and strain rates were calculated as the average values of all six segments and were provided by the software ( Figures 1 B and 1D). Longitudinal strain and strain rates of the interventricular septum were computed as average values of three segments that belonged to septum, whereas longitudinal strain and strain rates of the RV lateral wall were calculated as the averages of three segments that formed the lateral wall ( Figures 1 A and 1C). The values of septal and lateral RV wall strain and strain rates were calculated manually.
The speckle-tracking analysis of the right atrium was done after the endocardial border was manually traced in the four-chamber view. RA peak atrial longitudinal strain was calculated by averaging values observed in all six RA segments ( Figure 2 A). RA peak systolic strain rate was measured at RV systolic phase, while early and late RA strain rates were measured during early RV filling and during the late RV diastolic phase, respectively ( Figure 2 B).
Global longitudinal strain of the left atrium was calculated in the same manner as RA global longitudinal strain but separately for the four- and two-chamber views. The definitive value of 2DE LA longitudinal strain was calculated as the average of the values obtained in both apical views.
In this study, we used several methods to distinguish the influence of the left ventricle on the right heart in our patients. First, we separately determined longitudinal strain of the RV wall and septum. Second, we included parameters of LV structure, function, and mechanics, as well as LA mechanics, in the multivariate regression analysis.
Three-Dimensional Echocardiographic Acquisition
A full-volume acquisition of the right ventricle, required for further analyses, was obtained from an apical approach. Six electrocardiographically gated consecutive beats were acquired during an end-expiratory breath-hold to generate full volumes. All data sets were analyzed by using commercially available software RV TomTec (EchoPAC version 110.1.2). We analyzed RV volumes, RV stroke volume, and RV ejection fraction. The frame rate was between 20 and 30 frames/sec.
Statistical Analysis
The data were analyzed by using SPSS version 21 (SPSS, Inc, Chicago, IL). All parameters were tested for normal distribution using the Kolmogorov-Smirnov test. Continuous variables are presented as mean ± SD and were compared by using analysis of equal variance, as they showed normal distributions. Bonferroni post hoc analysis was used for comparisons between different groups. Chi-square tests were used for comparisons of proportions. Correlations were determined by using Pearson bivariate two-tailed correlation tests. Stepwise multiple regression analyses included age, gender, BMI, 2DE LV mass index, 2DE LV ejection fraction, 2DE LV and LA global longitudinal strain, and all variables with P values ≤ .10 in the correlation analysis. Intra- and interobserver variability for 2DE and 3DE RV parameters was analyzed in 20 randomly selected subjects using the Bland-Altman method. To assess intraobserver variability, one observer (M.T.) evaluated the same studies on two separate occasions, 3 to 4 days apart. For the interobserver variability evaluation, two independent observers (M.T. and V.C.) performed analyses 5 to 7 days apart. P values < .05 were considered statistically significant.
Results
The age and gender distribution, as well as heart rates, smoking prevalence, and blood pressures, were similar among the three groups ( Table 1 ). The fasting glucose level, HbA 1c , and the OGTT glucose level after 2 hours gradually increased from the controls, through the patients with prediabetes, to the patients with diabetes. Similar results were obtained for triglyceride, cholesterol, low-density lipoprotein cholesterol, and C-reactive protein levels ( Table 1 ). Serum creatinine levels were lower in controls than in patients with impaired glucose control, whereas high-density lipoprotein cholesterol levels were higher in controls than in patients with prediabetes and those with diabetes ( Table 1 ).
| Variable | Controls ( n = 54) | Patients with prediabetes ( n = 47) | Patients with diabetes ( n = 57) | P |
|---|---|---|---|---|
| Age (y) | 51 ± 8 | 52 ± 7 | 54 ± 7 | .095 |
| Women | 29 (54%) | 26 (55%) | 28 (49%) | .649 |
| BMI (kg/m 2 ) | 24.2 ± 2.5 | 26.1 ± 2.9 | 27.9 ± 2.8 | <.001 † |
| BSA (m 2 ) | 1.91 ± 0.13 ‡ | 1.95 ± 0.14 § | 2.05 ± 0.14 ‡ , § | <.001 |
| Smoking | 14 (26%) | 15 (32%) | 22 (39%) | .36 |
| Heart rate (beats/min) | 74 ± 7 | 75 ± 8 | 74 ± 9 | .776 |
| Clinic systolic BP (mm Hg) | 122 ± 11 | 125 ± 13 | 127 ± 12 | .091 |
| Clinic diastolic BP (mm Hg) | 73 ± 8 | 75 ± 7 | 76 ± 8 | .119 |
| HbA 1c (%) | 4.6 ± 0.8 | 6.1 ± 0.3 | 8.1 ± 1.4 | <.001 † |
| CRP (mg/L) | 2.17 ± 0.89 | 3.24 ± 1.33 | 4.35 ± 1.9 | <.001 † |
| Serum creatinine (μmol/L) | 58 ± 11 ‡ , || | 64 ± 11 || | 68 ± 12 ‡ | <.001 |
| Triglycerides (mmol/L) | 1.38 ± 0.45 | 1.69 ± 0.54 | 2.23 ± 0.58 | <.001 ∗ |
| HDL (mmol/L) | 1.36 ± 0.18 ‡ , || | 1.25 ± 0.21 || | 1.19 ± 0.2 || | <.001 |
| LDL (mmol/L) | 3.1 ± 0.56 | 3.6 ± 0.6 | 4.4 ± 0.7 | <.001 † |
| Total cholesterol (mmol/L) | 4.7 ± 0.8 | 5.4 ± 1 | 6.2 ± 1.4 | <.001 † |
∗ P < .05 in comparison between any two groups.
† P < .01 in comparison between any two groups.
‡ Controls versus patients with diabetes ( P < .01).
§ Patients with prediabetes versus those with diabetes ( P < .05).
LV Parameters
LV diameters were similar among the three groups, whereas interventricular septal thickness and LV mass index were increased in patients with diabetes compared with controls and patients with prediabetes ( Table 2 ). LV ejection fractions were similar among the groups, but LA diameters and transmitral E/A ratios differed only between controls and patients with diabetes ( Table 2 ).
| Variable | Controls ( n = 54) | Patients with prediabetes ( n = 47) | Patients with diabetes ( n = 57) | P |
|---|---|---|---|---|
| 2DE LV and LA parameters | ||||
| LA diameter (mm) | 36 ± 4 † | 38 ± 5 | 39 ± 4 † | .001 |
| LA maximal volume/BSA (mL/m 2 ) | 21.3 ± 4.2 | 21.6 ± 4.5 | 22.5 ± 5.1 | .364 |
| LV end-diastolic diameter (mm) | 48 ± 5 | 47 ± 5 | 47 ± 6 | .547 |
| LV end-systolic diameter (mm) | 31 ± 4 | 30 ± 4 | 30 ± 3 | .262 |
| Interventricular septal thickness (mm) | 9.8 ± 1 † | 10.3 ± 1.1 § | 11.2 ± 1.5 † , § | <.001 |
| LV mass/BSA (g/m 2 ) | 97 ± 9 † | 99 ± 10 § | 109 ± 11 † , § | <.001 |
| Ejection fraction (%) | 65 ± 4 | 64 ± 4 | 63 ± 5 | .067 |
| E/A m ratio | 1.26 ± 0.23 † | 1.15 ± 0.28 | 1.06 ± 0.2 † | <.001 |
| e′/a′ m ratio | 1.23 ± 0.27 † | 1.09 ± 0.32 | 0.96 ± 0.25 † | <.001 |
| E/e′ m ratio | 5.9 ± 1.7 | 7.3 ± 2 | 9 ± 1.6 | <.001 ∗ |
| 2DE RV and RA parameters | ||||
| RV diastolic diameter (mm) | 20 ± 4 | 21 ± 4 | 21 ± 5 | .413 |
| RV wall thickness subcostal (mm) | 3.7 ± 1.1 | 3.9 ± 1.2 | 4.1 ± 1.1 | .173 |
| RA maximal volume (mL) | 39 ± 8 || | 42 ± 9 | 44 ± 11 || | .024 |
| RA maximal volume/BSA (mL/m 2 ) | 20 ± 5 | 21 ± 6 | 22 ± 6 | .182 |
| E/A t | 1.33 ± 0.27 † | 1.24 ± 0.22 | 1.13 ± 0.25 † | <.001 |
| E/e′ t | 4.6 ± 1.5 | 5.3 ± 1.4 | 6.2 ± 1.5 | <.001 ∗ |
| s t (cm/sec) | 13.4 ± 2.4 || | 12.6 ± 2.2 | 12.4 ± 1.9 || | .042 |
| TAPSE (mm) | 22 ± 3 | 21 ± 3 | 21 ± 3 | .147 |
| PASP (mm Hg) | 20 ± 4 | 21 ± 5 | 20 ± 5 | .373 |
| 3DE RV parameters | ||||
| RV end-diastolic volume (mL) | 131 ± 19 † | 124 ± 17 | 120 ± 18 † | .006 |
| RV end-diastolic volume/BSA (mL/m 2 ) | 69 ± 10 | 63 ± 8 | 58 ± 8 | <.001 ∗ |
| RV end-systolic volume (mL) | 47 ± 6 | 46 ± 6 | 45 ± 5 | .17 |
| RV end-systolic volume/BSA (mL/m 2 ) | 25 ± 4 † , ‡ | 23 ± 4 ‡ | 22 ± 4 † | <.001 |
| RV stroke volume (mL) | 83 ± 9 † , ¶ | 77 ± 8 ¶ | 74 ± 7 † | <.001 |
| RV stroke volume/BSA (mL/m 2 ) | 44 ± 5 † , ¶ | 39 ± 4 ¶ | 36 ± 4 † | <.001 |
| RV ejection fraction (%) | 63 ± 4 | 62 ± 4 | 61 ± 5 | .063 |
∗ P < .05 in comparison between any two groups.
† Controls versus patients with diabetes ( P < .01).
‡ Controls versus patients with prediabetes ( P < .05).
§ Prediabetes versus patients with diabetes ( P < .01).
|| Controls versus patients with diabetes ( P < .05).
Longitudinal strain of the left ventricle in the healthy controls was higher than in patients with prediabetes and those with diabetes. There was no difference in LV longitudinal strain between the latter two groups of participants. On the other hand, LA longitudinal strain was significantly reduced only in patients with diabetics compared with controls, while patients with prediabetes did not differ from either controls or patients with diabetes ( Table 3 ).
| Variable | Controls ( n = 54) | Patients with prediabetes ( n = 47) | Patients with diabetes ( n = 57) | P |
|---|---|---|---|---|
| 2DE LV and LA strain | ||||
| Longitudinal LV strain (%) | −21 ± 2 † , ‡ | −19.6 ± 1.8 ‡ | −19 ± 1.6 † | <.001 |
| Longitudinal LA strain (%) | 38 ± 4 † | 36 ± 3 | 35 ± 5 † | .001 |
| 2DE RV strain and strain rates | ||||
| Longitudinal RV strain (%) | ||||
| Global RV | −27 ± 5 † | −25 ± 4 | −24 ± 4 † | .002 |
| Lateral wall | −32 ± 6 † , ‡ | −29 ± 5 ‡ | −28 ± 5 † | <.001 |
| Septum | −21 ± 5 § | −20 ± 4 | −19 ± 3 § | .039 |
| RV systolic strain rate (sec −1 ) | ||||
| Global RV | −1.73 ± 0.33 † , ‡ | −1.55 ± 0.27 ‡ | −1.42 ± 0.3 † | <.001 |
| Lateral wall | −1.83 ± 0.37 | −1.67 ± 0.34 | −1.5 ± 0.28 | <.001 ∗ |
| Septum | −1.65 ± 0.26 † , || | −1.42 ± 0.22 || | −1.32 ± 0.2 † | <.001 |
| RV early diastolic strain rate (sec −1 ) | ||||
| Global RV | 1.83 ± 0.38 † , ‡ | 1.67 ± 0.35 ‡ | 1.54 ± 0.32 † | <.001 |
| Lateral wall | 1.94 ± 0.32 † , || | 1.76 ± 0.3 || | 1.66 ± 0.25 † | <.001 |
| Septum | 1.7 ± 0.29 † , ‡ | 1.55 ± 0.27 ‡ | 1.43 ± 0.22 † | <.001 |
| RV late diastolic strain rate (sec −1 ) | ||||
| Global RV | 1.46 ± 0.28 † | 1.6 ± 0.3 | 1.71 ± 0.36 † | <.001 |
| Lateral wall | 1.58 ± 0.31 † | 1.73 ± 0.34 | 1.89 ± 0.39 † | <.001 |
| Septum | 1.38 ± 0.23 † | 1.49 ± 0.28 | 1.57 ± 0.32 † | .005 |
| 2DE RA strain and strain rates | ||||
| Longitudinal RA strain (%) | 43 ± 5 † , ‡ | 39 ± 4 ‡ | 37 ± 4 † | .004 |
| RA systolic strain rate (sec −1 ) | 2.05 ± 0.36 | 1.86 ± 0.32 | 1.7 ± 0.29 | <.001 ∗ |
| RA early diastolic strain rate (sec −1 ) | −2.13 ± 0.32 † , || | −1.92 ± 0.28 || | −1.83 ± 0.25 † | <.001 |
| RA late diastolic strain rate (sec −1 ) | −1.7 ± 0.25 † , ‡ | −1.87 ± 0.3 ‡ | −1.96 ± 0.33 † | <.001 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
