Diastolic Heart Failure Versus Diastolic Dysfunction: Difference in Renal Function

Abstract

Background

Despite the common finding of diastolic dysfunction with a preserved ejection fraction on routine echocardiography in elderly patients, it is unknown why some patients with isolated diastolic dysfunction are asymptomatic whereas others develop diastolic heart failure (ie, signs and symptoms of congestive heart failure).

Hypothesis

We hypothesized that renal insufficiency is more common in those patients with diastolic heart failure than those with diastolic dysfunction; it is intrinsic renal insufficiency that determines whether diastolic dysfunction becomes symptomatic.

Methods

We reviewed 686 consecutive transthoracic echocardiograms (TTEs). Patient age, race, weight, and cardiovascular risk factors (hypertension, diabetes, and coronary artery disease) were recorded. We used the Framingham Criteria for Congestive Heart Failure to determine the presence of diastolic heart failure by symptoms, exam findings, and radiological studies. Average creatinine clearance (CrCl), calculated by both total body weight and lean body mass, and estimated glomerular filtration rate (eGFR) were calculated for each group.

Results

Of the 686 TTEs reviewed, 18 patients fulfilled the criteria for diastolic heart failure and 118 patients had asymptomatic diastolic dysfunction. There was no difference in age, race, or gender nor was there any difference in the echocardiographic variables of diastolic function or left ventricular hypertrophy between groups. Multiple regression analysis showed only lower CrCl (44 ± 36 mL/min vs 76 ± 42 mL/min, total body weight, P = 0.0015; and 31 ± 24 mL/min vs 51 ± 27 mL/min, lean body mass, P = 0.0012) and eGFR (44 ± 33 mL/min/M² vs 69 ± 28 mL/min/M², P = 0.0003) were associated with diastolic heart failure. There was no significant difference in the presence of hypertension, diabetes, and coronary artery disease between groups.

Conclusions

The results of this study support the hypothesis that patients with normal left ventricular ejection fractions but diastolic dysfunction develop congestive heart failure because of underlying renal insufficiency. A larger, prospective study is needed to confirm this hypothesis. Copyright © 2010 Wiley Periodicals, Inc.

The authors have no funding, financial relationships, or conflicts of interest to disclose.

Introduction

Heart failure is a significant contributor to morbidity and mortality, with about 3.4 million outpatient visits, 1 million hospitalizations, and 60,000 deaths attributed to the disease each year in the United States.1 Once thought to be caused solely by systolic pump failure manifested by depressed left ventricular ejection fraction (LVEF), it is now known that up to 30%–50% of patients with heart failure have isolated diastolic heart failure with a preserved LVEF.2 It is important to distinguish this clinical syndrome from asymptomatic diastolic dysfunction, which describes mechanical abnormalities of the heart during diastole but might be unaccompanied by signs or symptoms of volume overload or congestive heart failure. Asymptomatic diastolic dysfunction is diagnosed by Doppler echocardiography and can be present in the absence of signs or symptoms of congestive heart failure or systolic dysfunction. The prevalence of diastolic dysfunction is unknown, but it is being increasingly recognized and known to increase with age.3 This abnormality is often noted on routine echocardiography and might be dismissed as unimportant, especially in the elderly. On the other hand, isolated diastolic dysfunction with a preserved LVEF is often observed on echocardiographic evaluation of patients with classic heart failure symptoms, and these patients might be categorized as having diastolic heart failure.

Diastolic dysfunction can be understood as abnormal relaxation and/or increased stiffness of the left ventricle myocardium leading to impaired filling during diastole. A stiff, noncompliant left ventricle with impaired relaxation would result in an elevated left ventricular end diastolic pressure with consequent increased pulmonary venous pressure. It is unclear why similar abnormalities in diastolic properties of the left ventricle cause little or no symptoms in 1 person but overt heart failure in another. Because of the pressure‐volume relationship of a left ventricle with impaired diastolic properties, relatively small increases in ventricular volume might produce a significant increase in left ventricular end diastolic and pulmonary venous pressures with consequent signs and symptoms of congestive heart failure. It would appear then that regulation of volume in diastolic heart failure patients might be a major factor that determines whether or not a patient develops symptoms. In this study we tested the hypothesis that patients with diastolic heart failure have underlying renal insufficiency when compared to patients with asymptomatic diastolic dysfunction, that is, it is intrinsic renal insufficiency that determines whether diastolic dysfunction becomes symptomatic.

Methods

This was a retrospective chart review performed of all echocardiograms performed at a single tertiary care university. Echocardiograms were interpreted by experienced echocardiographers. Normal systolic function in this study was defined as LVEF ≥ 50%. Outpatient echocardiograms were only included if the patient was evaluated by a physician within 7 days of the study. Inpatient echocardiograms were included and were performed within a day of admission. Diastolic dysfunction was considered present if the following echocardiographic criteria were present4, 5: mitral inflow velocity ratio of early diastolic mitral inflow velocity (E) and that at atrial filling (A) (E/A) < 1.0 and tissue Doppler mitral annulus velocity (E^′/A^′) < 1.0 measured at both the septal and lateral wall. If a “pseudonormal” mitral inflow velocity E/A ratio were present, the presence of tissue Doppler E^′/A^′ < 1.0 confirmed the presence of diastolic dysfunction.6 Significant valvular heart disease was excluded. The primary variable of this study was renal function. This was evaluated using the Cockcroft‐Gault equation where creatinine clearance (CrCl) (mL/min) = (140 − age) × (weight in kg) × (0.85 if female)/ (72 × Cr).7 We calculated the CrCl using both true body weight and ideal body mass, as both have been used in the literature.8 We also calculated estimated glomerular filtration rate (eGFR) using the Modified Diet in Renal Disease equation, where eGFR (mL/min/1.73m2) = 175 × (Cr) − 1.154 × (age) − 0.203 × (0.742 if female) × (1.212 if African American).9 Age, sex, the presence of diabetes or hypertension, signs and symptoms of heart failure, and chest radiographs were also recorded by review of the electronic medical record. In order to objectively determine whether subjects had heart failure, we used the Framingham Criteria for Heart Failure.10 Major criteria for heart failure were paroxysmal nocturnal dyspnea, neck vein distention, rales, radiographic cardiomegaly, acute pulmonary edema, an S3 gallop, increased central venous pressure (>16 cm at the right atrium), hepatojugular reflux, or a > 4.5‐kg weight loss in 5 days in response to treatment. Minor criteria were bilateral ankle edema, nocturnal cough, dyspnea on ordinary exertion, hepatomegaly, pleural effusion, decrease in vital capacity by a third from maximum recorded, and tachycardia (heart rate > 120 beats/minute). Those subjects who had 2 major or 1 major and 2 minor criteria without an alternate explanation for their signs and symptoms were placed in the diastolic heart failure group, whereas those who did not meet these criteria were placed in the diastolic dysfunction group.

Statistical Analysis

The Student t test was used to test statistical differences in means between groups. To test differences in frequency of population and clinical variables between groups the χ² test was used. Multiple regression analysis was used to assess association of population and clinical variables with the presence of heart failure. P values < 0.05 were considered significant (2‐tailed). Means are expressed as mean ± standard deviation.

Results

Eighteen patients fulfilled the criteria of normal LVEF (≥50%) but abnormal diastolic function with Framingham criteria for congestive heart failure. These were compared with 118 patients with normal ejection fractions and diastolic function but no evidence or history of congestive heart failure. The mean age in the heart failure group was 69 ± 12 years and was not different from the diastolic dysfunction group (Table 1). There was also no difference in the racial or gender composition between groups. The mean total creatinine clearance calculated either by total body weight or by lean body mass and GFR were all significantly different in the diastolic heart failure group when compared to the asymptomatic diastolic dysfunction alone group. Table 2 shows that the pertinent echocardiographic variables of the 2 groups were the same. The thickness of the left ventricular posterior free wall and interventricular septum, reflecting the degree of left ventricular hypertrophy, were the same in both groups. The tissue septal and lateral wall tissue Doppler E^′/A^′ ratio, indexes of intrinsic diastolic function that are relatively load‐ and heart rate‐independent,4, 11, 12 were also not different between groups. These results suggest that the degree of left ventricular diastolic dysfunction between the 2 groups was relatively the same.

Table 1.

Demographic andY Clinical Variables of the Diastolic Heart Failure Group Compared With Diastolic Dysfunction Among Group

	Diastolic Heart Failure	Diastolic Dysfunction	P
Age	69 ± 12	68 ± 13	NS
Gender (% male)	44	47	NS
Race (% white)	67	68	NS
Hypertension (%)	77	71	NS
Diabetes (%)	43	44	NS
Coronary artery disease (%)	38	39	NS
Creatine clearance (total body weight, mL/min)	44 ± 36	76 ± 42	0.0015a
Creatine clearance (lean body mass, mL/min)	31 ± 24	51 ± 27	0.0012a
Glomerular filtration rate (mL/min/M2)	44 ± 33	69 ± 28	0.0003a

Abbreviations: NS, not statistically significant (P>0.05).

^aSignificantly different

Table 2.

Echocardiographic Indices of Left Ventricular Diastolic Dysfunction

	Diastolic Heart Failure	Diastolic Dysfunction	P
LVEF (%)	59 ± 5	60 ± 5	NS
LVPWd	1.25 ± 0.33	1.11 ± 0.26	NS
LVIVSd	1.47 ± 0.59	1.22 ± 0.32	NS
Septal E′/A′	0.82 ± 0.36	0.76 ± 0.66	NS
Lateral E′/A′	0.85 ± 0.45	0.77 ± 0.30	NS
E/E′	17.7 ± 9.3	13.4 ± 6.4	NS
LA area	24.9 ± 5.7	18.6 ± 5.2	NS
E/A	1.2 ± 0.5	0.8 ± 0.2	NS

Abbreviations: E/A, ratio of early diastolic mitral inflow velocity (E) and that at atrial filling (A); E/E′, ratio of early diastolic mitral inflow velocity (E) and early diastolic tissue Doppler velocity of septal mitral annulus displacement (E′); LA area, cross section area in cm² of left atrium measured in 4‐chamber view; Lateral E′/A′, tissue Doppler velocity of mitral annulus displacement at early diastole (E′) and at atrial filling (A′) measured at the lateral wall annulus; LVEF, left ventricular ejection fraction; LVIVSd, left ventricular interventricular septum thickness measured at end diastole; LVPWd, left ventricular posterior wall thickness measured at end diastole; NS, not statistically significant (P>0.05); Septal E′/A′, tissue Doppler velocity of mitral annulus displacement at early diastole (E′) and at atrial filling (A′) measured at the septal annulus

Also shown in Table 2 are the echocardiographic and Doppler descriptors of diastolic physiology that are relatively load‐dependent.4, 5, 12 Not unexpected for the volume‐overloaded state of congestive heart failure, the size of the left atrium and E/E^′ ratio, an index that correlates with left atrial pressure,11 were greater in the heart failure group, but this difference was not statistically significant. Although also not statistically significant, the mitral inflow velocity E/A ratio was greater in the heart failure group likely reflecting more frequent pseudonormal patterns, which is consistent with a greater volume‐overloaded state.

Table 3 gives the results of the multiple regression analysis of possible clinical variables that might be associated with heart failure in patients with diastolic dysfunction. Race, gender, the presence of hypertension, diabetes mellitus, and coronary artery disease were not associated with heart failure. Included in the multiple regression analysis was the tissue Doppler septal E^′A^′ ratio determination; this echocardiographic variable of intrinsic diastolic function did not distinguish the 2 groups. Only reduced glomular filtration rate (GFR) was significantly associated with heart failure. Substitution of GFR with creatinine clearance calculated using either total body weight or lean body mass index into the multiple regression analysis also indicated that only renal insufficiency was associated with heart failure. This finding was further confirmed using a Mann‐Whitney U test showing that the distributions of eGFR and CrCl measured by both total body weight and lean body mass for the heart failure group were not the same as those of the diastolic dysfunction alone group (P = 0.0002, P = 0.0001, P = 0.0001, respectively), although there was considerable overlap in the distributions of the 2 groups.

Table 3.

Multiple Regression Analysis of Clinical Variables Associated With Diastolic Heart Failure

Clinical Variable	Regression Coefficient	P
Gender	−0.0213	0.727
Race	−0.0084	0.846
Hypertension	0.0521	0.463
Diabetes	−0.0708	0.296
CAD	−0.0782	0.246
Septal E′/A′	0.016	0.728
eGFR	−0.0031	0.002a
CrCl (tbw)b	−0.0017	0.009a
CrCl (lbm)b	−0.0028	0.007a

Abbreviations: CAD, coronary artery disease; CrCl (lbm), creatinine clearance calculated by lean body mass; CrCl (tbw), creatinine clearance calculated by total body weight; eGFR, glomerular filtration rate; Septal E′/A′, tissue Doppler velocity of mitral annulus displacement at early diastolic (E′) and at atrial filling (A′) measured at septal annulus.

^aSignificantly different.

^bSubstitution of eGFR with creatinine clearance (CrCl) in the multiple regression analysis

Discussion

A decrease in compliance and diastolic relaxation and an increase in diastolic stiffness of the left ventricle is noted in aging. Echocardiographic indexes of left ventricular diastolic abnormalities are ordinarily adjusted for age. Indeed, evidence of LV diastolic dysfunction is a common finding on routine echocardiography of elderly subjects.13 There is usually little or no clinical consequence of abnormalities in diastolic function in most subjects. However, in some individuals with diastolic abnormalities there are clinical signs and symptoms of congestive heart failure even though the echocardiographic indexes of diastolic dysfunction are not necessarily inappropriate for age and might be similar to those of persons who are asymptomatic. In this investigation we tested the hypothesis that patients with diastolic dysfunction but normal ejection fractions who develop congestive heart failure have underlying intrinsic renal inefficiency. Congestive symptoms develop because of reduced capacity to clear solute and free water, which is in contrast to individuals who might demonstrate equal degrees of diastolic dysfunction on echocardiography but who do not have signs of congestive heart failure. Because comorbid conditions such as hypertension, coronary artery disease, and diabetes mellitus also might produce left ventricular diastolic dysfunction, we determined whether these conditions were more often associated with diastolic heart failure than with asymptomatic diastolic dysfunction alone. On multiple regression analysis, none of these factors were present more often in the heart failure group when compared to the asymptomatic diastolic dysfunction group. Furthermore, race and sex were also not found to be determinant factors. Only indexes of renal insufficiency were found to be major determinants of diastolic heart failure, consistent with our hypothesis.

A patient with a normal LVEF but diastolic dysfunction, with or without concomitant heart failure, might have a reduced resting cardiac output.14 Abnormal left ventricular relaxation compromises left ventricular filling resulting in reduced left ventricular diastolic volume and attendant stroke volume. Consequently, at a heart rate within the normal resting range, cardiac output might be reduced. One could therefore postulate that the reduced renal function observed in patients with diastolic heart failure was attributed to a reduction in renal perfusion and not to intrinsic renal insufficiency. There was no difference in the relative heart rate and load‐independent tissue Doppler indexes of left ventricular diastolic dysfunction (E^′/A^′) and no difference in left ventricular wall thickness between the diastolic heart failure group and the asymptomatic diastolic dysfunction group, suggesting that there was a similar degree of diastolic dysfunction in both groups. As such, patients in the group with asymptomatic diastolic dysfunction might have the same degree of depression of resting cardiac output with corresponding reduction in renal blood flow causing the same degree of renal insufficiency as in the asymptomatic diastolic dysfunction alone group. Therefore, the postulated reduction in renal perfusion attributable to reduced cardiac output in only the group with heart failure and not in the diastolic dysfunction alone group might not explain the significant difference in renal function between the 2 groups. Nevertheless, we cannot exclude the possibility that invasive testing might have detected different aspects or more complex abnormalities in diastolic function in the heart failure group that were not readily demonstrable by noninvasive Doppler echocardiography used in this investigation. Also importantly, there was no difference in the frequency of the major risk factors for renal insufficiency (hypertension and diabetes), and therefore the renal insufficiency in the heart failure group could not be attributed to these comorbid conditions. However, we cannot exclude the possibility that the heart failure group did have reduced renal perfusion because of coexisting atherosclerotic renal artery disease or other forms of renal artery stenosis causing the intrinsic renal insufficiency.

Conclusion

Renal insufficiency is associated with symptoms of heart failure in patients with echocardiographic indicators of diastolic dysfunction but a preserved LVEF. A prospective and larger study is needed to determine if renal insufficiency is a causative mechanism of heart failure or a secondary effect of heart failure. Patients with asymptomatic diastolic dysfunction might benefit from prophylactic medical therapy to prevent renal insufficiency and consequently future heart failure.

Clinical Significance

• Internists and cardiologists struggle with the clinical significance of isolated diastolic dysfunction on echo‐ cardiography.

• Many patients with diastolic dysfunction develop clinically significant heart failure indistinguishable from systolic heart failure.

• Study shows that these patients are most likely to have renal insufficiency even when correcting for age, sex, race, and comorbidities.

• Patients with asymptomatic diastolic dysfunction might benefit from prophylactic medical therapy to prevent renal insufficiency and future heart failure.