Hypertension is a well-recognized risk factor for cardiovascular disease and is a major contributor to a large percentage of heart failure (HF) cases.^{1, 2, 3} Identifying subclinical left ventricular (LV) systolic dysfunction among hypertensives might be helpful in identifying patients at higher risk of developing heart failure.

Also, given the prevalence of hypertension in the population, identifying patients at high risk for heart failure might permit targeted preventive strategies. Early detection of left ventricular (LV) systolic dysfunction in affected patients is, therefore, a crucial issue.

Conventional echocardiography, however, detects abnormalities in LV systolic function only in the advanced stages of hypertensive heart disease (HHD), when a clear LV remodeling/hypertrophy is evident; diastolic LV dysfunction, in contrast, occurs early and its detection is easier even by conventional techniques.^{4, 5, 6, 7}

Nevertheless, recent research on HF with normal ejection fraction (EF) has demonstrated that isolated diastolic dysfunction is rare, being abnormalities of diastolic performance most frequently associated with a subclinical impairment of systolic function.^{8, 9, 10, 11, 12, 13, 14, 15} In this respect, preclinical alterations of LV systolic function have been demonstrated in patients with essential hypertension and normal EF by tissue Doppler imaging (TDI), strain echocardiography or magnetic resonance tagging.^{16, 17, 18, 19, 20}

It is previously demonstrated that M-mode echocardiography can be used to measure circumferential shortening at the mid-wall (FSmw,%),^{21, 22} a measure of regional circumferential strain. Reduced FSmw is prevalent among hypertensives with hypertrophic remodeling,²¹ and de Simone and coworkers have shown that low FSmw in relation to stress is a marker of poor outcome.²² However, FSmw is related to ventricular geometry—the relative wall thickness(RWT).²¹ Newer methods, such as 2D – speckle imaging, appear to provide a direct, angle and geometry independent measure of circumferential strain (CS).²³

Recently, it has been demonstrated that speckle tracking echocardiography (STE) provides more information than TDI, allowing a non invasive measurement of overall LV strain and twist.^{24, 25, 26, 27} This technique, therefore, can provide new mechanistic insight into systolic dysfunction even in patients without structural cardiac alterations (Stage A of ACC/AHA classification of HF).²⁸

The aim of the current case-control study was to use speckle tracking imaging as well as standard echo techniques to identify subclinical abnormalities in LV function among hypertensives, as compared to normal controls.

To assess the subtle differences in the LV principal strains for characterizing features of subclinical LV dysfunction in patients with systemic hypertension and preserved LV systolic pump function.

Adult subjects of both genders (age between 18-90 years old) attending the SMS Hospital were recruited to participate in this study. Fifty patients with isolated hypertension (42 males, mean age 54.16 ± 8.4 years) and 50 age matched healthy subjects (42 males, mean age 52.17 ± 8.6 years) were studied. Exclusion criteria were: secondary hypertension, angina pectoris, previous myocardial infarction, heart valve disease, HF, atrial fibrillation, chronic obstructive pulmonary disease, diabetes mellitus, and poor image quality on echocardiography.

All subjects underwent a routine physical examination, including body surface area (BSA) calculation using Du Bois and Du Bois formula, followed by a standard transthoracic echocardiogram (TTE). Before TTE, blood pressure was measured using a conventional sphygmomanometer, with the patient in supine position. Three measurements obtained at 5-minute intervals, were averaged. Hypertension was diagnosed in the presence of increased systolic BP (BPs) ≥140 mmHg and/or diastolic BP (BPd) ≥90 mmHg on two consecutive clinic visits and/or if patients were treated with antihypertensive drugs. ²⁹,³⁰

The age-matched control group included normo-tensive healthy subjects free from cardiovascular risk factors.

The protocol was approved by the local research Ethics Committee and a written informed consent was obtained from each subject.

Data were expressed as mean ± SD. Statistical analysis was performed using the SPSS statistical software (SPSS v.17 for Windows, SPSS Inc., Chicago, IL, USA). For any parameter, independent t-test and ANOVA were used to compare variables between different groups. The chi square test and Fisher’s exact test were used for categorical variables. Linear regression analysis, with Pearson’s coefficient, was used to estimate correlation between continuous variables. P values ≤ 0.05 were considered significant

All subjects were in sinus rhythm without signs and/or symptoms of HF (NYHA class I; Table 1). No difference between patients and controls was observed regarding: age, gender and heart rate (P = ns).34 %of hypertensive subject had evidence of concentric remodeling. As expected, most controls had normal LV mass index and geometry. No difference between LVH(–) and LVH (+) patients was found with respect to systolic (152.3±8.89 mmHg vs. 157.2±7.38 mmHg, P = ns) and diastolic blood pressure (97.36±7.27 mmHg vs. 99.24±6.74 mmHg,P = ns).

LVEF was normal both in LVH(–) and in LVH(+) patients; LV wall thickness, RWT were significantly increased in both LVH(–) and LVH(+) patients with respect to controls. As expected, thickness, RWT and LVMI were significantly higher in LVH (+) than in LVH(–) patients (P < 0.001). The RWT in the LVH(+) group was 0.47 ± 0.09%, as an expression of concentric hypertrophy. (Table 2)

Analysis of LV systolic and diastolic function by means of TDI showed different findings according to the LV geometry pattern (Table 3). In the LVH(–) patients (normal geometry), S’ and E’ velocities, as well as the A’ were normal, without any significant difference with respect to controls (S’ = 9.29 ± 0.98 cm/s vs. 9.64 ± 1.05 cm/s,P = ns; E’ = 11.03 ±3.06cm/s vs. 11.24 ±1.73 cm/s,P = ns; A’ 7.03 ± 2.95 vs. 6. 73 ± 1.41, P = ns);whereas in the LVH(+) group (concentric hypertrophy) S’ and E’ velocity were decreased and A’ velocity was increased in comparison to controls (S’ = 7.29± 0.58 cm/s vs. 9.64 ± 1.05 cm/s,P < 0.001; E’ = 6.55 ± 1.45 cm/s vs 11.24 ±1.73cm/s, P < 0.001, A’ = 10.24± 1.88 vs. 6. 73 ± 1.41,P < 0.001).

On the other hand, STE revealed relevant changes in LV mechanics even in patients with normal geometry (LVH(–)): these subjects, characterized by normal LV function on TDI, had an impaired systolic LS (–16.3±2.01 % vs. –19.31±1.87 %, P = <0.0001) and CS (–17.31±2.28 % vs. –18.98±1.56 %, P = <0.0001), with normal RS (19.22±1.90 % vs. 18.88±1.54 %, P = 0.38) when compared to healthy subjects (Figure 1). Moreover, in the LVH(+) patients STE revealed lower LS (–12.91 ±1.67 % vs. – 19.31±1.87 ,CS -14.62 ±1.216%, vs -18.98±1.56 P<0.001, Figure 1) and higher RS (20.38±1.56 % vs. 18.88±1.54 %, P = 0.003), (Figure 1).

Figure 1. Global longitudinal strain (panel A), circumferential strain (panel B), and radial strain (panel C) by STE in controls, LVH(–) and LVH(+) patients A. Longitudinal strain is lower both in LVH(–) (P =< 0.0001) and even more in LVH(+) patients (P < 0.0001), compared to controls; furthermore, it is lower in LVH(+) than in LVH(–) group (P = <0.0001). B. Circumferential strain is lower both in LVH(–) (P = <0.0001) and LVH(+) patients (P = <0.0001), in comparison with controls, and it is significantly decreased (P =< 0.0001) in LVH(+) than in LVH(–) patients. C. Radial strain is normal in LVH(–) (P = ns) and higher in LVH(+) patients (P = 0.003), in comparison with controls, and it is insignificantly higher (P = 0.05) in LVH(+) than in LVH(–) group. All these findings suggest that progressive adaptative changes occur in LV mechanics from patients with normal geometry to those with concentric hypertrophy.

The following relationships have been observed:

(1) Systolic blood pressure was inversely related with LS (r = –0.73, P < 0.0001); CS (r = –0.61, P < 0.0001); and positively with RS (r = 0.22, P =0.02). (2) The systolic blood pressure was also inversely related with S’ (r = –0.75, P < 0.0001); E’ (r = –0.65, P < 0.0001); and positively related with A’ (r = 0.55, P < 0.0001).Similar correlation were found with diastotic blood pressure.(3) In hypertensives we found a significant inverse relation between LV mass and indices of global LV systolic function, including both LS (r = –0.1; p<0.001) and CS ( r = – 0.59; p<0.0001)

It is well known, and confirmed by this study, that a preclinical systolic dysfunction occurs in hypertensive patients with LV hypertrophy.^{35, 36}

We have observed, in addition, an impairment of longitudinal and circumferential systolic function even in patients with normal LV geometry. In this respect, STE appears more sensitive than both conventional echocardiography and TDI in identifying a reduction of intrinsic myocardial contractility, evident in hypertensive patients even long before LV hypertrophy becomes detectable. This finding highlights the advantages of 2D strain over TDI in early identifying LV systolic function impairment.

These early changes of myocardial contractility may be secondary to hemodynamic and/or biochemical alterations. The increased end-systolic wall stress, an expression of afterload, as well as the loading conditions, are likely to play a role in determining the longitudinal dysfunction in HHD. On the other hand, the chronic increase of endsystolic wall stress may promote subendocardial synthesis of collagen, thereby contributing to reduction of longitudinal deformation. With regard to the biochemical changes, myocardial fibrosis is one of the factors responsible for myocardial function deterioration in hypertension ³⁷,³⁸ in affected patients with normal EF, impaired LS and increased LV torsion are directly related with serum levels of tissue inhibitor of metalloproteinase-1 matrix, a marker of myocardial fibrosis.³⁹ This suggests that abnormal collagen turnover and myocardial fibrosis may contribute to early LV contractile dysfunction.^{40, 41} On the other hand, in this study we observed that RS is preserved in hypertensive patients without structural LV changes.

LV hypertrophy is a common response to increased and prolonged afterload and is an independent predictor of cardiovascular morbidity and mortality.^{42, 43, 44, 45, 46} LV hypertrophy is associated with a variety of alterations including: fibrosis, impaired coronary artery vasodilatory capacity, depressed LV wall mechanics, and abnormal LV diastolic filling pattern.^{47, 48, 49, 50, 51} Subendocardium is susceptible to the deleterious effects of both interstitial fibrosis and hypoperfusion, to the extent that in HHD impairment of the subendocardial function results in abnormal longitudinal function.⁵² This is also suggested by a recent study based on 3D wall motion tracking analysis, demonstrating that myocardial strain pattern depends on LV adaptative responses to high blood pressure.⁵³ Our results show that in a relatively late stage of HHD, corresponding to stage B of HF, additional LV changes occur, such as increase of RS, suggesting an involvement of mid-myocardial layers. The finding is consistent with the results of a recent study, demonstrating that in hypertension LV longitudinal systolic function progressively deteriorates.⁵⁴