Prognostic value of left atrial volume index in patients with rheumatic mitral stenosis

Abstract Background The significance of left atrial volume index (LAVI) for predicting outcomes in patients with mitral stenosis (MS) has been unclear, even though rheumatic MS is known to be associated with left atrium enlargement and functional deterioration. Hypothesis The current study aimed to investigate the prognostic value of LAVI, based on the severity in patients with rheumatic MS. Methods We retrospectively reviewed 611 patients with pure rheumatic MS. The prognostic value of LAVI and the effect of MS severity on the prognostic value of LAVI for events were evaluated. The events were defined as a composite end‐point that included all‐cause death, heart failure admission, mitral valve replacement, percutaneous mitral valvuloplasty, and stroke. Results There were 236 (38.6%) overall events during a median follow‐up of 8 months. The optimal LAVI cutoff for the prognostic threshold was 57 ml/m2. The MS severity had a significant effect on the prognostic value of LAVI. A LAVI >57 ml/m2 was a prognostic value for events in progressive MS (hazard ratio [HR]: 2.40, 95% confidence interval [CI]: 1.41–5.40, p = .004) and in patients with severe MS (HR: 1.70, 95% CI: 1.06–2.74, p = .029), but it was not prognostic in patients with very severe MS (HR: 1.02, 95% CI: 0.56–1.84, p = .955). Conclusions The prognostic value of LAVI varies and is dependent on the MS severity. A LAVI >57 mL/m2 was independently associated with poor outcomes in patients with progressive MS, while this association was minimized in patients with severe MS.


| INTRODUCTION
The left atrial volume index (LAVI) is a known prognostic marker for cardiovascular outcomes in various cardiovascular diseases, including heart failure, hypertrophic cardiomyopathy, and ischemic heart disease, as well as in the general population. [1][2][3][4] The prognostic role of LAVI in valvular heart disease has also been applied in patients with mitral regurgitation (MR), 5,6 aortic stenosis, [7][8][9] and aortic regurgitation. 10 However, the value of LAVI for predicting outcomes in patients with mitral stenosis (MS) has been unclear, even though rheumatic MS is known to be closely associated with left atrium enlargement, stiffening, and functional deterioration. 11 One report has indicated that LAVI did not predict clinical outcomes in patients with MS. 12 We recently found that LAVI can act as a prognostic marker for outcomes in patients with progressive MS, which is defined as MS patients with the mitral valve area (MVA) larger than 1.5 cm 2 . 13 Progressive MS is a less severe stage during disease progression in MS. Therefore, we hypothesized that the prognostic value of LAVI for MS would differ according to the severity of the disease, and this might cause discordant results for the prognostic value of LAVI in patients with MS. We aimed to investigate the effect of MS severity on the prognostic value of LAVI in a large cohort of patients with MS, categorized as progressive MS, severe MS, and very severe MS. We additionally sought to assess associated factors for an enlarged left atrium other than the severity of MS in those patients.

| Study population
We analyzed patients with rheumatic MS who underwent echocardiography between 2006 and 2015 at a tertiary referral center for valvular heart disease in Korea. Exclusion criteria were as follows: patients with >1+ MR, >1+ aortic regurgitation and/or more than mild aortic stenosis, patients with congenital or myopathic lesions that could affect pulmonary artery pressure, patients with a history of prior percutaneous mitral valvuloplasty (PMV), those who had undergone planned mitral valve replacement (MVR) or PMV before the echocardiographic examination, and those who received MVR or PMV within 30 days after the index echocardiography examination. Demographic characteristics including age, sex, anticoagulation, history of prior stroke, and body surface area were confirmed by chart review. Systolic and diastolic blood pressure were measured before the echocardiography. Therefore, in total, 611 patients with pure rheumatic MS were included. This study was approved by the ethical committee of Yonsei University, Severance Hospital, Seoul, Korea. The need to obtain informed consent was waived for the retrospectively obtained data for this non-interventional study.

| Echocardiography
Two-dimensional and Doppler echocardiography were performed according to the American Society of Echocardiography (ASE) guidelines. 14 Left ventricular (LV) end-diastolic dimension (EDD), LV endsystolic dimension (ESD), septal wall thickness, and posterior wall thickness were measured from the M-mode. LV ejection fraction (EF) was calculated using LV EDD and ESD. LV mass was calculated using the formula, according to the ASE guidelines. The LV mass index was defined as an LV mass indexed for body surface area. LV hypertrophy was defined as LV mass index ≥115 g/m 2 for men and ≥95 g/m 2 for women. 14 The left atrial volume was calculated using the biplane area-length method according to ASE guidelines. 14 15 The mean diastolic transmitral pressure gradient was measured from a continuous wave Doppler signal across the mitral valve by tracing its envelope. The calculated systolic pulmonary artery pressure was defined as 4 × (maximum velocity of the tricuspid regurgitant jet) 2 + right atrial pressure. Right atrial pressure was estimated by measuring the inferior vena cava diameter. 16 Stroke volume was calculated using the LV outflow tract diameter and the LV outflow tract flow pulsed-wave Doppler signal. The stroke volume index was defined as stroke volume indexed for body surface area.

| Study endpoint
Patients were followed across a median of 41 months (Interquartile range: 8-84 months) for a composite end-point that included allcause death, inpatient admissions for heart failure, MVR, PMV, and incidence of stroke. The occurrence of any of the clinical events was ascertained by a review of hospital records and by telephone interviews, as necessary.

| Statistics
Demographic characteristics are reported as percentages or as the mean ± SD. The patient groups were compared using chi-square statistics for categorical variables and the Student's t-test for continuous variables. Receiver operating characteristic (ROC) curves were used to determine the sensitivity and specificity of LAVI in predicting the primary outcomes and to determine the optimal cut-off value for continuous variables. Univariable and multivariable Cox proportionalhazards regression models reporting the hazard ratio (HR) and 95% confidence interval (CI) was employed to determine potential useful variables for predicting event-free survival following echocardiography. Variables with statistical significance in univariable analysis were entered into the multivariable Cox proportional hazard model, as well as age and sex. Kaplan-Meier survival curves were employed to plot all events according to the time-to-first event. To determine potential independent associations between variables and LAVI, binary logistic regression was applied. Variables displaying statistical significance in univariable analysis were entered into a multivariable binary logistic regression model, reporting the odds ratio (OR) and 95% CI. A p value <.05 was considered statistically significant. Table 1 demonstrates the baseline characteristics of the study population. There were 207 patients with progressive MS, 281 patients with severe MS, and 123 patients with very severe MS from the overall 611 patients with MS. The mean age was 60 ± 12, and 76.4% of the patients were women. Patients with very severe MS were younger than those with progressive and severe MS (56 ± 12 years vs. 62 ± 12 years vs. 60 ± 12 years). There were no significant differences in the sex between the groups. LV hypertrophy was more common in patients with progressive and severe MS compared to those with very severe MS (37.7% vs. 32.7% vs. 16.2%). The outcomes of the population are shown in Table S1. There were 236 (38.6%) overall events, The optimal cut-off value for predicting overall events according to the ROC curves for the LAVI was 57 ml/m 2 , in which the area under curve (AUC) was 0.657, the sensitivity was 76.2% and specificity was 50.2%. We categorized patients into two LAVI groups as follows: LAVI >57 ml/m 2 and LAVI ≤57 ml/m 2 for a comparison. To confirm optimal LAVI cut-offs at each MS severity grade, there was additional investigation of the optimal cut-off value for predicting overall events according to the ROC curves for LAVI in each of the progressive, severe, and very severe MS groups. The optimal cut-off for LAVI was 51 ml/m 2 for progressive MS (AUC: 0.693, sensitivity = 77.1%, specificity = 57.9%), 52 ml/m 2 for severe MS (AUC: 0.531, sensitivity = 81.2%, specificity = 29.6%), and 58 ml/m 2 for very severe MS (AUC: 0.529; sensitivity = 73.6%, specificity = 26.4%).

| DISCUSSION
The main findings of this study are that (1)   Despite the long-held belief that rheumatic MS is an isolated disease of the stenotic mitral valve, several studies have suggested LV myocardial abnormalities in a subset of patients with rheumatic MS. [21][22][23] We previously reported that LV diastolic dysfunction and increased LV diastolic pressure is a mechanism for the low-gradient phenomenon in subsets of MS, which is commonly found in elderly patients with atrial fibrillation. 22 From the current study, we also found that LV hypertrophy was independently associated with a large LAVI. Because LV hypertrophy is a well-known cause of LV diastolic dysfunction and an increase in LV stiffness, we can speculate that LV diastolic dysfunction in elderly

| Limitations
The main limitation of the current study is that the results of this study were based on retrospective analysis. However, patient medical records and echocardiography were carefully reviewed. We defined the presence of symptoms from clinical records, even though all medical records and echocardiography were reviewed carefully to minimize any bias. The choice to perform surgery or another intervention was made by each attending physician; therefore, it was not standardized and may have been influenced by the patient's LAVI. However, since LAVI is not included in the echocardiographic parameters for determining MS intervention in the current guideline, 15,25 the LAVI would not have had a significant effect on each physician's decision.

| CONCLUSION
The prognostic value of LAVI varies and is dependent upon the MS severity. A LAVI >57 ml/m 2 was independently associated with poor outcomes in patients with progressive MS, while this association weakened in patients with severe MS. The prognostic value of LAVI was not identified in patients with very severe MS. LAVI was independently associated with the presence of LV hypertrophy, suggesting the influence of LV myocardial properties on LAVI in patients with MS.

CONFLICT OF INTEREST
The author declares that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.