Left Atrial Distensibility Guiding Management in Advanced Chronic Heart Failure

Background: The use of E/Em (mitral early-diastolic velocity divided by mitral early-diastolic annular velocity) assessing left ventricular filling pressure has some innate defect since it was born to assess the regional function of myocardium. It is rational to infer that ischemic heart disease with regional wall motion defect will affect the measurement of tissue Doppler and, thereupon, influence the accuracy of using regional parameters to estimate global function. Conduction disturbance with bundle branch block is another significant confounder, since it truly influences the data of regional peak Em. Therefore, it is necessary to develop or design a new global parameter to overcome the handicap of tissue Doppler. In our recent studies (1, 2), there are significant logarithmic relationships between LV filling pressure and LA distensibility in patients with acute myocardial infarction received primary percutaneous coronary intervention and in patients with severe mitral regurgitation. In patient received coronary intervention for chronic angina pectoris, the relationship is also similar (submitted to J Am Soc Echocardiogr. under major revision). In our recent ongoing study, LA distensibility is applied to low-risk general population for assessing the risk of stroke, cardiovascular event, mortality and the presence of atrial fibrillation. This time, the investigators try to use this parameter to monitor patients with advanced chronic heart failure (high-risk group) and to adjust patient's management and treatment.

Introduction LA volume provides the significantly prognostic information in the general population and patients with heart disease, including acute myocardial infarction (3), left ventricular dysfunction (4, 5), mitral regurgitation (6), cardiomyopathy (7, 8) and atrial fibrillation (9). Large LA volume, which represents chronic diastolic dysfunction, is associated with poor outcome, regardless of systolic function (10). Thereby, LA volume provides a long-term view of whether or not the patient has the disease of diastolic dysfunction, regardless of whatever loading conditions are present at the time of the examination, as the hemoglobin A1C in diabetes mellitus (10). Until recently, the relation between LA volume and LV filling pressure has been confirmed directly by simultaneous echocardiography-catheterization, which those studies were conduced by our team (1, 2). LA distensibility is associated logarithmically with left ventricular filling pressure in patients with chronic stable angina, acute myocardial infarction, or severe mitral regurgitation. It can be treated as non-invasive Swan-Ganz catheter, which isn't likely to affect by the tissue Doppler confounders of regional wall motion defect and bundle branch block. Since the relationship between LA distensibility and left ventricular filling pressure is logarithmic, LA distensibility would be relatively insensitive in the low-pressure end. Therefore, the investigators conduct an investigation that assesses the risk stratification of cardiovascular risk by LA distensibility in low-risk general population (potential low left ventricular filling pressure) from 2009. This time, LA distensibility will be used to guide the treatment of advanced chronic heart failure.

Methods Study population: Advanced chronic heart failure (HF) will be defined as left ventricular ejection fraction less than 35%, creatinine less than 2 mg/dL, and CHF NY functional class III-IV for more than 3 months. Enrolled patients should be admitted to hospital due to heart failure requiring adjustment of inotropic agents or diuretics within recent 6 months. Three hundred HF cases will be recruited - 100 with sinus rhythm, and 100 with atrial fibrillation served as LA distensibility guided treatment group (guide group) and another 100 patients, either sinus rhythm or atrial fibrillation, served as control group. The management of guide group will be adjusted by LA distensibility, including the adjustments of inotropic agents, diuretics, beta-blocker, ACEI, and AIIB. The control group will be treated by conventional management and traditional echocardiography can be performed as in-charge doctor request. Renal function will be checked 1 time per 3 months. All patients will give written informed consent to participate in the study, and the study will send to the institutional review board for approval.

Conventional echocardiographic and myocardial tissue Doppler measurement: Echocardiography will be performed in each patient initially. LV ejection fraction is calculated using Simpson's method for biplane images. Mitral inflow is assessed by pulsed-wave Doppler echocardiography form the apical 4-chamber view. From the mitral inflow profile, the E-wave velocity, A-wave velocity, and E-deceleration time are measured. Pulsed-wave tissue Doppler imaging (TDI) is performed using spectral pulsed Doppler signal filters, by adjusting the Nyquist limit to 15 - 20 cm/s and using the minimum optimal gain. In the apical 4-chamber view, a 3-mm, a pulsed-wave Doppler sample volume is placed at the level of the mitral annulus over the septal border. Pulsed-wave TDI results are characterized by a myocardial systolic wave (Sm) and 2 diastolic waves: early (Em) and atrial contraction (Am). The pulsed-wave TDI tracing is recorded over 5 cardiac cycles at a sweep speed of 100 mm/s and is used for offline calculations.

Measurements of LA volume: All LA volume measurements will be calculated from apical 4- and 2-chamber views using the biplane area-length method (11). The LA volumes are measured at 3 points: 1) immediately before the mitral valve opening (maximal LA volume or Volmax); 2) at onset of the P-wave on electrocardiography (pre-atrial contraction volume or Volp); and 3) at mitral valve closure (minimal LA volume or Volmin). The LA distensibility was calculated as (Volmax - Volmin) / Volmin. The LA ejection fraction is calculated as (Volp - Volmin) / Volp. In all patients, LA volumes are indexed to body surface area (BSA) (12). For patients with atrial fibrillation, the average of 5 LA distensibilities will be used for clinical assessment.

Study Design: This study is a prospective, observational, and open label investigation. The guide group will be followed 1 time per 2 week at first 3 months, then 1 time per month later, and then 1 time per 3 months. Each time, the guide group should receive echocardiography, including the measurement of LA distensibility. LA distensibility-guided therapy is enabled in 2 ways. Prescriptions should be adjusted according to overall left ventricular filling pressure estimated by LA distensibility. The specified treatments after adjustment by the current LA distensibility are also allowed. Investigators individually adjust medications for each patient according to the ranges of LA distensibility. The specified therapy is based on 5 LA distensibility ranges (very low, low, optimal, high, and very high). Each range is associated with a prescription for medication dosing, sodium and fluid intake, or activity level. Although there are no specific prescribing rules ad dynamic prescribing is at the discretion of the investigators, the general aim is to reduce or eliminate diuretic doses for high or very high LA distensibility and increase diuretic or vasodilator doses for low or very low LA distensibility. The necessity of hospitalization for heart failure will be adjusted by 2 cardiovascular specialists and all patients admitted for heart failure will be managed by the same one cardiovascular specialist. The total duration of follow-up will be 2 years. Patients will be encouraged to take additional visit if they have worsening symptoms. For life-threatening heart failure, intravenous nitroprusside drip under continuous A-line monitor, percutaneous coronary intervention, Swan-Ganz catheter insertion, intra-aortic balloon pump, and ECMO can be done according to the order of in-charge doctor. The primary end-point will be all-cause mortality. The second end-points will be heart failure with hospitalization and the duration of each hospitalization. Additionally, the ratios of medication changes in 2-year follow-up, including diuretics, inotropic agents, beta blockers, ACEI, and AIIB, will be assessed. Analysis will also be performed to estimate the trends of heart function (either systolic or diastolic) and renal function during 2-year follow-up according to whether guiding by LA distensibility is done or not.

Interobserver variability: In the first 50 enrolled cases, Volmax, Volmin, and Volp will be measured by 2 independent observers. Interobserver variability is calculated as the difference between the values obtained by the 2 observers divided by the mean. Interobserver difference and variability of Volmax, Volmin, and Volp will be assessed.

Statistical analysis: The SPSS software (version 12) will be used for all statistical analyses. All continuous variables are presented as means ± standard deviation. Analysis of variance and post hoc test (Scheffe F-test) for unpaired data are used to evaluate the significance of differences between groups. A p vale of < 0.05 is considered statistically significant. Comparison of clinical characteristics is performed by chi-square analysis for categorical variables. Bivariate analysis, simple correlation and linear regression are used as appropriate. Target ACEI, AIIB, and beta blocker doses are defined as > 50% of maximal American Heart Association/American College of Cardiology heart failure guideline doses. LA distensibility control will be empirically defined if the frequency of LA distensibility < 90% (according to the logarithmic regression curve in acute myocardial infarction patients, LA distensibility < 90% indicated left ventricular filling pressure more than 15 mmHg in our prior study) is less than 10% for 6 consecutive months. ROC curve analysis is performed to assess the sensitivity and specificity of the cut-off points of LA distensibility when predicting hospitalization for heart failure and the duration of hospital stay. Kaplan-Meier curve will be performed to assess the cumulative event-free rate according to subgroup analysis (guide group with sinus rhythm, guide group with atrial fibrillation and control group). To evaluate the effect of covariates on mortality and heart failure with hospitalization, relative risk and 95% confidence intervals will be calculated as hazard ratios derived from the Cox proportional-hazards model.

Solutions to the Anticipated Difficulty The problem is to unify echocardiographic measurement of left atrial border in sonographers participating in this study. Therefore, the investigators will perform those measurement in 50 consecutive patients scheduled for routine echocardiography and discuss all cases to reduce the technique's error of measurement.

Anticipated results:

1. LA distensibility is useful for guiding the management of advanced chronic heart failure.
2. The treatment adjusted by LA distensibility provides more advanced information for clinical physician. Therefore, it potentially influences patient's outcome. The hypothesis will be confirmed by comparison of the primary and second end-points between treatment group and control group.
3. Although patient with sinus rhythm will be better to assess LA distensibility, the average of 5 LA distensibilities in patients with atrial fibrillation is possibly as useful as in patients with sinus rhythm. Subgroup analysis of treatment group according to whether sinus rhythm or atrial fibrillation will be done to assess the feasibility of LA distensibility.