- Top of page
Hypertension is a potent risk factor for heart failure.1,2 The main mechanism of heart failure in patients with hypertension is left ventricular (LV) diastolic dysfunction.3 LV diastolic dysfunction associated with hypertension is morphologically characterized by LV wall thickening and increased left atrium (LA) volume. In particular, LA volume is a chronic surrogate of LV filling pressure or LA pressure, and a prognostic marker of various cardiac diseases.4,5
Hypertensive cardiomyopathy with reduced ejection fraction is known as an important cause of reversible cardiomyopathy, but its morphologic characteristics and prevalence remain unclear. At an advanced stage, hypertension induces eccentric LV hypertrophy and LV systolic dysfunction (LVSD). In the general population, a prevalence of asymptomatic LVSD ranges from 0.9% to 14% in patients with hypertension.6–8 According to the Framingham Heart Study, severe LVSD developed in 3% to 6% among hypertensive patients.9 Eccentric pattern of hypertrophy is a particularly strong risk factor for LVSD from the Cardiovascular Health Study.10
Although dilated cardiomyopathy is diagnosed with findings of depressed ejection fraction and dilated LV based on echocardiography, it is very elusive to discern a specific cause of dilated cardiomyopathy, especially when heart failure is first diagnosed. We aimed to investigate whether initial echocardiographic parameters such as left atrial volume (LAV), left ventricular mass (LVM), LV sphericity, right ventricular systolic function, and changes of those parameters at follow-up could give clues to differentiate hypertensive cardiomyopathy with reduced ejection fraction (HTCMREF) from idiopathic dilated cardiomyopathy (idDCM).
- Top of page
We retrospectively studied 18 hypertensive patients (mean age, 62.7 ± 13.2 years, 56% women) who were admitted with severe LVSD (ejection fraction <35%) and heart failure to Kangdong Sacred Heart hospital and recovered to normal ejection fraction (EF) on follow-up echocardiography, referred to as HTCMREF. We excluded patients with ischemic heart disease, valvular heart disease, chronic renal failure, acute myocarditis, tachycardia-induced cardiomyopathy, drug-induced cardiomyopathy, or other causes of reversible cardiomyopathy such as stress cardiomyopathy and alcoholic cardiomyopathy.
We compared clinical, electrocardiographic and echocardiographic parameters between the patients with HTCMREF and 18 age-matched patients with idDCM. idDCM was diagnosed when patients showed dilated LV and low left ventricular ejection fraction (EF <35%) without specific cause or abnormal loading condition (hypertension or valve disease), and no improvement on follow-up echocardiography at least 9 months later. We excluded ischemic cardiomyopathy performing coronary angiography.
LVM and LAV were calculated by a formula using echocardiography, and were indexed to body surface area. To detect the involvement of right ventricular (RV) dysfunction, RV size and RV hypokinesia (RVHK) were evaluated. RV size was measured in diameter at end diastole in the parasternal long-axis view, and RVHK was defined as tricuspid annular plane systolic excursion <15 mm in the apical 4-chamber view. LV sphericity was measured as the ratio between the length (mitral annulus to apex) and diameter (midcavity) of the LV in the apical 4-chamber view. LVM, LAV, and relative wall thickness were calculated as follows11:
LVM (g) = 0.8 (1.04[(diastolic LV dimension [LVDd]] + posterior wall thickness [PWT] + interventricular septal wall thickness [IVSWT])3 − (LVDd)3 ) + 0.6 g
LAV = Π/6 (anteroposterior diameter of LA [LA1] * inferomedial diameter of LA [LA2] * superomedial diameter of LA [LA3]) by prolate ellipsoid formula
Relative wall thickness = (PWT + IVSWT)/LVDd
Laboratory data such as B-type natriuretic peptide, troponin, complete blood count, cholesterol were recorded. Left ventricular hypertrophy (LVH) on electrocardiogram was defined according to Sokolow-Lyon voltage criteria (SV1 + RV5 or RV6 ≥ 35 mm).
Continuous variables were expressed as the mean ± standard deviation. The dichotomous variables were expressed as percents. Statistical comparisons of continuous variables were performed by use of the Student t test. χ2 analysis was conducted to compare categorical variables. A P value <0.05 was considered to be statistically significant. All statistical analyses were performed using of SPSS 12.0 for Windows (SPSS Inc., Chicago, IL).
- Top of page
In the patients with HTCMREF, LV ejection fraction improved from 28.8 ± 4.9 to 52.3 ± 8.8%, and LV end-diastolic dimension decreased from 64.1 ± 7.7 to 52.3 ± 6.4 mm during a mean follow-up of 574 days with medications for systolic heart failure (Figure 1, part A). There were no significant differences in age, sex, and body mass index between the 2 groups (Table 1). Systolic blood pressure was higher in patients with HTCMREF than those with idDCM but not statistically significant between the 2 groups. Of the patients with HTCMREF, 8 patients (44%) had no history of hypertension, and 5 of those patients showed initially normal blood pressure but revealed high blood pressure along with the recovery of LVSD after medical treatment for systolic heart failure. On electrocardiogram, 14 (78%) of the HTCMREF patients and 12 (67%) of the idDCM patients showed LVH findings (Table 1).
Figure 1. Serial change of left ventricular ejection fraction (LVEF), left ventricular mass index (LVMI), and left atrial volume index (LAVI) in patients with hypertensive cardiomyopathy with reduced EF and idiopathic dilated cardiomyopathy. A significant improvement in (A) LVEF, (B) LVMI, and (C) LAVI was observed in hypertensive cardiomyopathy with reduced EF.
Download figure to PowerPoint
Table 1. Comparison of Baseline Characteristics Between Hypertensive Cardiomyopathy With Reduced Ejection Fraction and Idiopathic Dilated Cardiomyopathy
| ||HTCMREF [N = 18]||idDCM [N = 18]||P Value|
|Age, y||62.7 ± 13.2||57.2 ± 16.8||NS|
|Female, n (%)||10 (55.6)||10 (55.6)||NS|
|BMI, kg/m2||24.9 ± 5.3||22.1 ± 3.2||0.073|
|DM, n (%)||3 (16.7)||5 (27.8)||NS|
|Smoking, n (%)||9 (50)||6 (33.3)||NS|
|SBP, mm Hg||148.9 ± 38.5||128.3 ± 20.4||0.056|
|DBP, mm Hg||86.1 ± 14.2||83.9 ± 15.7||NS|
|Heart rate, bpm||91.8 ± 20.9||87.2 ± 27.3||NS|
|Cholesterol, mg/dL||192.0 ± 35.2||171.2 ± 56.7||NS|
|Creatinine, mg/dL||1.05 ± 0.28||1.05 ± 0.29||NS|
|Hemoglobin, g/dL||12.6 ± 2.6||13.0 ± 1.77||NS|
|Follow-up duration, d||574 ± 418||1067 ± 702||NS|
|Use of medications at discharge, n (%)|
|ACE-I or ARB||9 (50)||13 (72)||NS|
|β-blocker||10 (55)||5 (27.7)||NS|
|Spironolactone||13 (72)||10 (55)||NS|
|Statin||9 (50)||8 (44.4)||NS|
|Electrocardiographic variables, n (%)|
|LVH||7 (38.8)||7 (38.8)||NS|
|LVH with ST change||7 (38.8)||5 (27.7)||NS|
Echocardiographic Comparison Between Patients With HTCMREF and idDCM
In HTCMREF, LV wall on M-mode was thicker (septum, 10.8 ± 1.7 vs 9.4 ± 1.9 mm; posterior wall, 10.9 ± 1.4 vs 9.7 ± 1.9 mm, P < 0.05) than in idDCM (Table 2). LA anteroposterior diameter was larger in HTCMREF, but LAV index (LAVI) showed no significant difference between the 2 groups. At follow-up, LAVI in HTCMREF decreased significantly from 31.9 ± 8.3 mL/m2 to 21.0 ± 8.9 mL/m2 (P < 0.001) (table 3), as opposed to a significant increase in idDCM (from 28.5 ± 10.9 mL/m2 to 31.9 ± 8.3 mL/m2, P < 0.001) (Figure 1, part C). Mitral regurgitation (MR) is known as an important contributor to LA enlargement. On initial echocardiography, MR of more than moderate severity was found in 5 HTCMREF patients and 4 idDCM patients. On follow-up test, all HTCMREF patients with significant MR showed improvement in severity, but MR severity in idDCM patients changed variably.
Table 2. Comparison of Initial Echocardiographic Data Between Hypertensive Cardiomyopathy With Reduced Ejection Fraction and Idiopathic Dilated Cardiomyopathy
| ||HTCMREF [N = 18]||idDCM [N = 18]||P Value|
|LVDd, mm||64.1 ± 7.7||65.5 ± 10.0||NS|
|LVDs, mm||54.2 ± 7.2||57.2 ± 8.6||NS|
|LVEF, %||28.8 ± 4.9||23.9 ± 7.5||0.15|
|IVSWT, mm||10.8 ± 1.7||9.4 ± 1.9||0.026|
|PWT, mm||10.9 ± 1.4||9.7 ± 1.9||0.038|
|LVMI, g/m2||192.1 ± 43.7||177.7 ± 47.5||0.09|
|LA, mm||43.6 ± 5.8||38.9 ± 6.3||0.027|
|LAVI (mL/m2)||31.9 ± 8.3||28.5 ± 10.9||NS|
|RVHK, n (%)||2 (11)||8 (44)||0.026|
|MR (≥moderate), n (%)||5 (27.7)||4 (22.2)||NS|
|LV sphericity index||1.46 ± 0.21||1.53 ± 0.20||0.30|
Table 3. Follow-up Echocardiographic Findings in Hypertensive Cardiomyopathy With Reduced Ejection Fraction
| ||HTCMREF [N = 18]|
|LVDd, mm||52.2 ± 6.4||64.1 ± 7.7||<0.001|
|LVEF, %||52.3 ± 8.8||28.8 ± 4.9||<0.001|
|IVSWT, mm||12.0 ± 1.9||10.8 ± 1.7||0.01|
|PWT, mm||11.9 ± 1.9||10.9 ± 1.4||0.028|
|LVMI, g/m2||151.4 ± 42.1||192.1 ± 43.7||0.001|
|LAVI, mL/m2||21.0 ± 8.9||31.9 ± 8.3||<0.001|
|LV sphericity||1.60 ± 0.30||1.46 ± 0.20||0.023|
|RVHK (%)||0 (0)||2 (11)|| |
There was no significant difference in initial LVM index between HTCMREF and idDCM patients (192.2 ± 43.8 vs 177.7 ± 47.6 g/m2), but only in HTCMREF did LVM index decrease significantly (151.5 ± 42.1 g/m2, P < 0.01) at follow-up (Figure 1, part B;, Table 3). The incidence of RV hypokinesia was higher in idDCM than in HTCMREF patients at initial examination (44% vs 11%, P = 0.026) (Table 2). RV hypokinesia in HTCMREF patients recovered completely on follow-up echocardiography.
Electrocardiography Changes After Medications in Patients With HTCMREF
On initial Electrocardiography (ECG) in patients with HTCMREF, 14 patients (78%, 14/18) revealed LVH in voltage or LVH with secondary ST segment change. On the follow-up examination, we observed a significant regression of LVH on ECG in 8 patients (57%, 8/14), of whom the initial LVH pattern improved from LVH in voltage or LVH with secondary ST segment changes to normal or LVH in voltage, respectively.
We performed a subgroup analysis in which patients with HTCMREF were divided into 2 groups (ECG regression vs nonregression) to clarify whether the regression of LVH on ECG is a sensitive marker for short recovery time, reduction of LV mass, or reduction of LA size. The regression of LVH on ECG was not associated with age, recovery days, initial echocardiographic parameters such as left ventricular ejection fraction (LVEF), LAVI, and LVM index, or the amount of change of 3 echocardiographic parameters (Table 4).
Table 4. Possible Factors in Association With Regression of Left Ventricular Hypertrophy on Electrocardiogram in Patients With Hypertensive Cardiomyopathy With Reduced Ejection Fraction
| ||Regression [N = 8]||No Regression [N = 9]||P Value|
|Age, y||62.2 ± 11.4||62.4 ± 15.8||NS|
|Recovery time, d||638 ± 440||561 ± 421||NS|
|SBP, mm Hg||148.7 ± 36.4||142.2 ± 38.3||NS|
|LVEF, %||27.1 ± 5.7||29.3 ± 3.1||NS|
|LVDd, mm||62.7 ± 5.6||64.8 ± 9.6||NS|
|LV sphericity index||1.47 ± 0.16||1.47 ± 0.25||NS|
|LAVI, mL/m2||29.7 ± 7.2||33.1 ± 9.3||NS|
|LVMI, g/m2||183.2 ± 30.2||192.7 ± 51.2||NS|
|D-LVEF||25.8 ± 9.3||22.2 ± 11.5||NS|
|D-LAVI||−11.1 ± 10.5||−10.9 ± 7.37||NS|
|D-LVMI||−37.9 ± 24.3||−37.8 ± 57.6||NS|
Predicting Factors for Determining Recovery Time of LVSD in Patients With HTCMREF
In patients with HTCMREF, echocardiographic parameters such as LVEF, LV dimension, LAVI, LVM index, and LV sphericity were not associated with recovery time of LVSD. Age, laboratory data, and regression of LVH on ECG were also not associated with recovery time of LVSD.
- Top of page
In our study, we found that the HTCMREF group showed eccentric LVH and higher LA diameter and lower incidence of RV hypokinesia compared to idDCM, but there were no differences in clinical characteristics, incidence of LVH on ECG, and LV sphericity between the 2 groups.
Severe LVSD by hypertension has been reported to be uncommon. Severe LVSD (EF <30%) occurred 6% in the Framingham Heart Study.9 The prevalence of asymptomatic LVSD ranges from 0.9% to 14%, depending on the clinical characteristics and LVEF.6–8 Hypertensive cardiomyopathy was difficult to detect in real practice because other comorbidities including ischemic heart disease, uremia, and anemia might contribute in part to aggravation of LVEF. Asymptomatic LVSD correlated with male, smoking and LV mass, and heart rate in 1 study.6 In our study, we could not find predictors for developing LVSD in hypertensive patients because we had no prospective data to compare between hypertensive patients with LVSD and without LVSD; nonetheless, we found eccentric LVH and increased LA dimension in HTCMREF patients when compared with idDCM patients. According to the Cardiovascular Health Study by Drazner et al, 18% of the patients developed reduced EF on an average follow-up period of 4 years in 159 hypertensive patients with LVH.10 However, there has been no robust evidence that LVH is a risk factor for LVSD. In this context, we tried to evaluate whether regression of LVH on ECG is associated with the recovery time of LVSD or initial echocardiographic parameters, but no differences were found between the 2 groups.
We found that the group of HTCMREF patients had a lower incidence of RVD than idDCM. One explanation is that increased afterload by untreated hypertension affect mainly LV in early status, whereas RVD might develop at an advanced stage, due to pulmonary vascular resistance along with secondary pulmonary hypertension. In contrast, idDCM might be caused by direct myocardial damage involving both ventricles. To our knowledge, there has been no study regarding the incidence of RVD in HTCMREF.
Hypertension can lead to a variety of changes in the myocardial structure and coronary vasculature of the heart in 2 ways: directly, by increased hemodynamic load (afterload) on the heart, or indirectly, by associated neurohormonal activation and vascular changes.12–15 In addition to contractile disturbances of cardiomyocyte function and interstitial and perivascular fibrosis, loss is now being considered as 1 of the determinants of the maladaptive processes implicated in the transition from compensated LVH to decompensated LVH. Several experimental studies have demonstrated that apoptosis may play an important role in the transition from compensated LVH to failure in pressure-overloaded hearts.16 In our study, we found LVH on ECG in 14 HTCMREF patients (86%) and 12 (67%) idDCM patients. There is no difference in incidence between 2 the groups. LVH is an adaptive process to increase cardiac workload to preserve systolic ventricular dysfunction through reducing wall tension by increased wall thickening. Although left ventricular mass index could not reach a significant difference between the 2 groups (P = 0.09), the LV wall was thicker in HTCMREF patients at initial echocardiography. Paradoxically, the LV wall thickness of HTCMREF patients seemed to increase at a follow-up echocardiogram, but in fact, LVMI was decreased along with a decrease of LV size, reflecting the finding that regression on ECG was observed only in 47% of HTCMREF patients after recovery of LVEF in our study.
Hypertension is a potent risk factor for heart failure, and also a primary cause of heart failure itself. In most cases, heart failure in hypertension is associated with diastolic dysfunction. In general, diastolic heart failure develops in older patients, women, or those with hypertension. Mechanisms of diastolic heart failure in hypertension are attributed to concentric LVH and increased afterload of LV, which induce LA enlargement and pulmonary congestion. Gandhi et al reported that systolic dysfunction was found in 20/38 in hypertensive patients admitted with acute pulmonary edema, of whom follow-up echocardiography revealed no significant change, thus they concluded that diastolic dysfunction may be a main contributing factor to acute pulmonary edema.3 In our study, LA diameter and LA volume index were increased compared to the normal value (24–28 mL/m2 according to the various studies).17–19 LA size could be a marker of chronic LA pressure or LV filling pressure, and enlarged LA volume was associated with increased incidence of heart failure.20,21 In contrast to Doppler diastolic variables of LV, which are affected by acute hemodynamic changes, LA volume is a more stable parameter, integrating the effects of elevated LV filling pressures from preexisting cardiovascular conditions as well as acute other diseases. In idDCM, LA volume is associated with LV remodeling, diastolic dysfunction, and the degree of MR. LA volume has an independent and incremental prognostic value.
We treated all patients according to Heart Failure Guidelines of European Society of Cardiology. In 5 patients with HTCMREF, blood pressure increased along with improvement of LVEF, despite the fact that initial blood pressure was normal or below normal before medications for systolic heart failure started. Hypertension is frequently under-recognized as an underlying cause of systolic ventricular dysfunction, partly because the failing heart is unable to generate sufficient blood pressure by the time cardiac pumping function deteriorate seriously, thus obscuring the hypertensive nature of the disease.
Our study was a retrospective and observational study that could not escape selection bias, and with such a small number statistical power is lacking. We could not evaluate initial diastolic Doppler parameters because many patients had a fusion of E and A flow with sinus tachycardia during the acute decompensated stage.