Left Ventricular Remodeling in Diabetic Patients with and without Hypertension

Left ventricular (LV) hypertrophy either defined by echocardiographic or electrocardiographic criteria has been shown to be a strongly independent risk factor for cardiovascular morbidity and mortality [1-3]. Hypertension and obesity have been regarded as the most common cause of increased LV mass in general population [4,5]. Diabetes Mellitus (DM) is a well known and important risk factor for cardiovascular mortality and morbidity. Several epidemiological investigators described the association between LV hypertrophy and impaired glucose tolerance [6,7]. The results of the studies that evaluated the relation between DM and LV mass are inconsistent [8-11]. 2-dimensional echocardiography (2DE) is the primary noninvasive diagnostic modality for the calculation of LV mass because it is cost-effective and offers real-time, high spatial and temporal resolution imaging for initial evaluation and follows up [12].


Introduction
Left ventricular (LV) hypertrophy either defined by echocardiographic or electrocardiographic criteria has been shown to be a strongly independent risk factor for cardiovascular morbidity and mortality [1][2][3]. Hypertension and obesity have been regarded as the most common cause of increased LV mass in general population [4,5]. Diabetes Mellitus (DM) is a well known and important risk factor for cardiovascular mortality and morbidity. Several epidemiological investigators described the association between LV hypertrophy and impaired glucose tolerance [6,7]. The results of the studies that evaluated the relation between DM and LV mass are inconsistent [8][9][10][11]. 2-dimensional echocardiography (2DE) is the primary noninvasive diagnostic modality for the calculation of LV mass because it is cost-effective and offers real-time, high spatial and temporal resolution imaging for initial evaluation and follows up [12].
The aim of our study was to evaluate the effect of DM on LV dimensions, wall thickness and mass in both normotensive and hypertensive patients.

Subjects and Methods
Two hundred consecutive patients (mean age 56.5 ± 12.5 years, 69.7% males) ( Table 1) were referred to cardiology clinic for the first time for cardiac assessment. Depending on the history, the patients were divided into 4 groups:

Clinical examination
Detailed history and through clinical examination were performed. Hypertension was defined and graded according to ESC and ESH guidelines [13]. DM was defined according to European society of cardiology and European Association for the study of Diabetes [14].

Transthoracic 2DE examination
2DE was performed to all subjects with a commercially available ultrasound system (Philips Sonos 5500 with 3.5 MHz probe, Best, The Netherlands). Examination was performed by echo-technician while the patient in the left lateral decubitus position using both apical and parasternal views. The following M-mode parameters were measured: 1) LV systolic (LVSD) and diastolic (LVDD) internal dimensions, 2) Thickness of interventricular septum (IVS) and LV posterior wall (PW) at diastole, 3) LV fractional shortening (FS) by the standard equation: (LVDD -LVSD)/ LVDD, 3) LV mass using the corrected formula [15]: LV mass (g) = 0.8 {1.04 x (LVDD + IVS + PW) 3 -(LVDD) 3 } + 0.6, 4) LV mass index calculated by: LV mass/ body surface area and 5) left atrial (LA) dimension. The interpretation and analysis of echo studies were performed blindly with the study hypothesis and the type of patient group.

Statistical analysis
All data were presented as mean ± SD. Independent sample t-test was used for comparison between the three patients groups and normal group. Comparison between the groups was performed by analysis of variance (ANOVA) test. Frequencies are expressed by percentage with 95% confidence intervals. A single and multiple linear regression analysis with stepwise elimination were performed. LV mass was considered as the dependent variable and the independent variables included in the model were gender, age and the presence or absence of hypertension and DM. The level of significance was set to P <0.05.

Baseline criteria
There were no significant differences between all groups regarding age, sex distribution and body surface area. Comparison between males and females showed no significant difference in LVDD, LVSD, LV wall thickness, LV mass and mass index except for age as the females were older than males. (Table 1) Changes in LV dimensions, wall thickness and mass ( Table 2, Table 3).

Effect of DM in normotensive patients
Compared to normal's, Group I patients showed significant increase of LVDD and LVSD (5.7 ± 0.7cm, 4.2 ± 1.0 cm vs 4.7 ± 0.4cm, 3.0 ± 0.5cm respectively; p<0.0001). Values of LV wall thickness (IVS and PW) and mass were significantly higher in Group I than Group IV (all p<0.0001). The prevalence of increased LV wall thickness and mass was 45%.

DM versus hypertension
In patients with hypertension only (Group II), thickness of IVS and PW were significantly higher than in patients with DM only (Group I) (1.5 ± 0.2 cm and 1.3 ± 0.2 cm vs.1.3 ± 0.2cm and 1.2 ± 0.1cm; p<0.001 respectively). LV mass and mass index were significantly higher also in Group II than in Group I (462.1 ± 80.8 g vs.402.6 ± 96.5 g; p= 0.001 and 325.2 ± 71.9 g/m 2 vs. 280.3 ± 71.5 g/m 2 ; p=0.002 respectively). LVDD and LVSD showed no significant difference between both patient groups. The prevalence of increased LV wall thickness and mass was 55%.

Effect of DM in hypertensive patients
Compared to normal's, Values of LVDD, LVSD, wall thickness (IVS and PW) and mass were significantly higher in Group III than Group IV (all p<0.0001).Comparison between Group I and Group III showed no significant difference in all values except for IVS thickness which was significantly higher in Group III than Group I (1.3 ± 0.2cm vs 1.4 ± 0.2cm; p<0.006). The prevalence of increased LV wall thickness and mass was significantly higher in Group III than both Group I and Group II (75% vs. 45% and 55%). (Table 4 (A and B) Multiple linear regression analysis was performed through including age, gender, presence or absence of hypertension and DM and considering LV mass as a dependent variable. As shown in (Table 4-A), age, hypertension and DM were independently associated with an increased LV mass (P: 0.04, 0.06 and 0.007 respectively). (Table 4-B) showed that when single regression analysis was performed including only hypertension and DM, both were significant statistically association with higher LV mass (P: 0.002 and 0.0001 respectively).

LV function and LA dimension
Compared to normals, all patients in Group I, II and III showed significant reduction in LVFS (28.6 ± 7.0 vs 35.7 ± 7.1; p<0.0001). LA was significantly larger in all patient groups than in normal's (4.2 ± 0.5 cm vs 3.6 ± 0.4 cm; p<0.0001). Comparison between the three patient groups (I, II and III) showed no significant differences in LVFS and LA dimension.

Discussion
It is widely acknowledged that increased LV mass is thought to increase cardiovascular risk through a series of unfavorable metabolic, functional and structural cardiac changes [16]. This explained the facts that an increased LV mass is a premier risk factor for cardiac events e.g. myocardial infarction and heart failure [1,2,17].
Accumulating data from experimental, pathological, epidemiological and clinical studies have shown that DM affects the cardiac function (systolic and diastolic) and structure independent of hypertension, coronary artery disease or any other known risk factors [18][19][20][21].
The current study aimed to evaluate the effect of DM on LV internal dimensions, systolic function and LV mass using conventional echocardiography. The study included 3 equal patients groups (DM without hypertension, Hypertension without diabetes and DM with hypertension) and described the changes in echo parameters in each group in comparison with the other groups and with normal individuals.
Struthers and Morris reported that LV hypertrophy was present in 30% of patients with type 2 DM independent of blood pressure or use of antihypertensive medication [22]. The prevalence of LVH increases with the severity of hypertension ranging from 38% to 72% [11]. Our results showed that the prevalence of increased LV wall thickness and mass in patients with DM (Group I) was comparable to those with hypertension (Group II) (45% and 55% respectively). In patients with both hypertension and DM (Group III), the prevalence became significantly higher (75%). Compared to normal individuals, LV mass was significantly higher in all 3 patient groups. When the 3 patient groups were compared with each other, no significant differences in LV wall thickness, systolic function and LV mass were detected.
Chinali [23] demonstrated the high prevalence of other cardiac abnormalities included LA dilatation, reduced midwall shortening and diastolic dysfunction in patients with metabolic syndromes. The prevalence was independent on demographics and individual components of metabolic syndrome [23].
De Simone [24] evaluated 12-year incident heart failure in 2740 participants without prevalent cardiovascular or severe kidney disease, at the time of the first exam of the Strong Heart Study cohort. They concluded that type 2 DM is a potent, independent risk factor for heart failure. Risk of heart failure in diabetic patients cannot be fully explained by incident MI and coexisting cardiovascular risk factors [24]. In the current study, DM was associated with lower LV-FS values than in normals regardless of associated hypertension or not. This finding is not restricted to specific age or sex groups. The current study showed an increased LA size in both diabetic patient groups (normotensive and hypertensive). It was explained as secondary effect to impaired LV systolic and diastolic dysfunction or as an impact of increased LV mass [25]. The cardiac abnormities are probably attributed to direct effect of DM on myocytes including replacement fibrosis caused by focal myocyte necrosis and increased interstitial fibrosis [18]. Many studies described the utility of tissue  Doppler imaging and speckle tracking for early detection of LV systolic and diastolic functions pre-clinically and even before the alteration of conventional 2D echocardiography [26][27][28].
Many studies [7,19] described that the association between DM and cardiac abnormalities is more evident in women than men while other studies [6] excluded it. Our results showed that this association is equal in both women and men. This may be attributed to the small percentage of females included in the study.
Our findings may have clinical impact in treating and following up of patients with DM. It will be of great value for the treating physician to assess the parameters of LV hypertrophy and systolic function with the start of treatment and during follow up of diabetic patients with or without associated hypertension.

Study limitation
The study had many limitations including: 1) The number of patients with DM included is small that made the analysis of subgroups is not obtainable. For example differences between type 1 and 2 DM and differences between controlled and uncontrolled DM.
2) The study used the m-mode formula for calculation of LV mass while its calculation by 3-dimensional echocardiography was not performed. The 3D calculation is feasible and accurate; however both techniques (2D and 3D) measurements were moderately correlated [29].
3) The study used FS and EF as indicators of LV systolic function, while the tissue Doppler indices (tissue velocity, strain and strain rate) and speckle tracking were not included.
4) The study described the cardiac abnormalities with the first visit and no follow up data are available. However, to assess the progression or regression of these cardiac abnormalities in response to treatment, a large follow up study is recommended.

Conclusion
The prevalence and severity of increased LV wall thickness and mass in diabetic patients is independent of hypertension. Impaired LV systolic function was associated with DM which explains the increased risk for heart failure.