Clinical significance of left atrial geometry in dilated cardiomyopathy patients: A cardiovascular magnetic resonance study

Clinical significance of left atrial (LA) function and geometry in patients with dilated cardiomyopathy (DCM) remains uncertain.


| INTRODUCTION
Dilated cardiomyopathy (DCM) is a myocardial disease typically diagnosed by impaired left ventricular (LV) ejection fraction and LV dilation. 1 Guideline-directed medical therapy (GDMT) improved the prognosis of patients with DCM. 2 However, some patients with DCM still have poor outcomes. 3,4 It is important to reveal the factors that predict the prognosis to determine appropriate GDMT in patients with DCM.
The roles of the left atrium include modulating LV filling and cardiovascular performance by functioning as a reservoir for pulmonary venous return during ventricular systole, serving as a conduit for pulmonary venous return during early ventricular diastole, and providing a booster pump that augments ventricular filling during late ventricular diastole. 5 Left atrial (LA) enlargement is associated with cardiac events in patients with LV diastolic dysfunction 6 and atrial fibrillation. 7 LA enlargement is also associated with cardiac events in patients with DCM. 8 However, the clinical significance of LA function and geometry in patients with DCM remains uncertain. 9 To our knowledge, although the previous study has found that the LV sphericity index was prognostic predictor in patients with DCM, 10 there is no paper on the sphericity of the left atrium in patients with DCM. We believed that LA function and geometric change, especially spherical change, might be related to prognosis.
Cardiac magnetic resonance (CMR) is a suitable modality for investigating the properties, functions, and geometry. In addition, CMR is the current gold standard modality to assess LA function and geometry. 11,12 Thus, the aim of the present study was to investigate the relationship between the prognosis and LA parameters assessed by CMR in patients with DCM.

| CMR image acquisition
CMR examinations were performed on a 1.5T (GE Healthcare, Milwaukee, WI, USA) or 3.0T (Siemens Healthineers, München, Germany) imaging scanner using a standard protocol. Cine images in short-axis, long-axis, and four-chamber views were acquired using a breath-hold cine steady-state free precession sequence. Late gadolinium enhancement (LGE) images were acquired 10-15 minutes after the intravenous injection of 0.2 mmol/kg gadolinium using a segmented inversion recovery fast gradient-echo sequence (1.5T) or true fast imaging with steady state free procession sequence (3.0T). Inversion times were optimized to null the myocardium. ([length 4ch + length 2ch ]/2). 14 LA emptying fractions (LAEF) were calculated from the measured LA volumes as follows 14,15 :

| CMR image analysis
Active LAEF = LAV pre-ac -LAV min ð Þ LAV pre-ac × 100 The LA sphericity index was computed as the ratio of the maximum LA volume by the volume of a sphere with maximum LA length diameter ( Figure 1). 16

| Clinical measurement and observation
As clinical evaluations at baseline, blood pressure, heart rate, laboratory examination, and echocardiography were measured upon undergoing CMR. Baseline characteristics and CMR were collected at the same time.
The day when CMR was performed was set to zero. On echocardiography, LV volumes were calculated using the biplane method of disk summation (modified Simpson's rule), and the LV ejection fraction was measured as the difference between the end-diastolic volume and endsystolic volume divided by the end-diastolic volume. 17 Patients were treated according to GDMT, including beta-blockers and reninaldosterone system inhibitors. 18 Cardiac events were based on medical records. The primary endpoint was hospitalization for worsening heart failure (HF). Based on the time of CMR, we classified the study cohort into two groups by the presence or absence of subsequent hospitalization for HF. The hard endpoint was implantation ventricular assist device, death due to pump failure or lethal arrhythmia, or sudden cardiac death.

| Statistical analysis
Data are presented as mean ± SD or as frequency (percentage  Patients were categorized based on the median value of LA sphericity index. The Kaplan-Meier curve showed that patients with a high LA sphericity index (≥0.57) had a significantly higher risk of hospitalization for HF than those with a low LA sphericity index (<0.57, Figure 3).
Univariate regression analysis revealed that no variables were significantly associated with LA sphericity index (Supplementary   Table S2); therefore, multivariate analyses were not performed.

| DISCUSSION
The present study demonstrated the clinical significance of LA parameters in patients with DCM. Although there were no significant differences in LA volume and LAEF between patients with hospitalization for HF and those without, high LA sphericity index was independently associated with hospitalization for HF. LA sphericity index might be one of the important markers for prediction of HF hospitalization in patients with DCM.

| LA geometric change
The geometric change of LA is associated with various factors such as LV function, degree of mitral regurgitation, and histological changes. 12 The change of LA geometry into a sphere was considered to be related to the anatomical position, 19 imbalance wall stress, 20 histological changes in the atrial wall. 21 Furthermore, an elevated inflammatory response, such as C-reactive protein, 22 interleukins, and cytokines, 23 was shown to be involved in the geometric change of LA. In a mouse model, pressure overload was associated with LA dilatation, increased LA mass, loss of myofibrillar content, atrial cardiomyocyte hypertrophy, and atrial fibrosis. 24 In the present study, it was considered that the geometric change of LA was caused not by specific factors but by various factors such as pressure overload, volume overload, and myocardial fibrosis.
In the present study, a high LA sphericity index, which means toward a more spherical shape, was associated with poor cardiac prognosis. The sphere transformation of LA might be caused by the stagnation of blood flow, leading to the exacerbation of HF. 25 Moreover, it was shown that LA changes such as dilatation and failure were accompanied by LA endocrine failure and LA regulatory failure contributing to neurohumoral overactivity, vasoconstriction, and volume overload. Thus, these LA changes are thought to be associated with poor clinical outcomes. 26 In the hard endpoint, there was no significant difference in LA sphericity index. The reason of this might be due to small number of hard outcome events and short follow period.
In the HF hospitalization group, LAV tended to be high and LAEF tended to be low. However, there were no significant differences between the HF hospitalization group and the Non-HF hospitalization group. It could not be denied that these results were due to the small sample size. However, although the population and outcomes were different, some papers have demonstrated that LA geometric parameter had a significant difference even if there was no significant difference in LAV according to cardiac events. 27,28 Therefore, the LA geometric parameter might be a more sensitive indicator of changes in the left atrium than LAV.
Few papers have focused on LA parameters in patients with reduced ejection fraction 29 and DCM patients. 8 In these studies, LA volumes were measured by echocardiography. LAEF and LA geometry were not evaluated. Compared to echocardiography, CMR imaging could measure LA volume, function, and geometry more accurately.
LA measurement of echocardiography sometimes underestimates LA volumes. 11,12 Therefore, we think that our method using CMR could evaluate LA function and geometry in more detail compared with the previous studies using echocardiography assessment.
A previous study showed that a high LA sphericity index was associated with the recurrence of atrial fibrillation after cardiac ablation. 16 We have newly found that LA sphericity index was also associated with cardiac events in patients with DCM. LA sphericity index could be an essential predictive marker for cardiac events. However, there are unclear points about the mechanism by which a sphere transformation of LA leads to a poor prognosis. We think that further research is needed.

| The usefulness of CMR
In previous meta-analyses, the presence of LGE has been associated with poor prognosis in patients with DCM. 30 In the present study, however, the presence of LGE was not a predictor for cardiac outcomes. A similar result has been obtained in other studies. 31,32 The reason of discordant results was unclear. The differences in cohorts, the number of events, and observation period might be related to the reasons.
LA sphericity index was calculated by a simple method using routine cine CMR without gadolinium contrast. Therefore, the LA sphericity index can provide incremental prognostic value for all patients who undergo CMR without additional cost and scan time.

| Clinical implications
There are various parameters for predicting the prognosis of patients with DCM, and LA sphericity index could be a new parameter associated with cardiac outcome. Moreover, LA sphericity index can be measured by appending one simple analysis to routine CMR.

| Study limitations
Several limitations need to be acknowledged in the present study.
First, this was a single-center study with a limited number of patients, which possibly resulted in a patient selection bias and lower statistical power. Second, only patients undergoing CMR were included, which way also poses a risk for bias. Third, genetic tests were not performed in our cohorts. Forth, the fact that the data were collected with the use of different devices and operators was a limitation of this study.

| CONCLUSIONS
A high LA sphericity index was an independent predictor of hospitalization for HF. LA sphericity index could be one of the new parameters for predicting the prognosis in patients with DCM.

ACKNOWLEDGMENTS
We thank Hirofumi Hata, Department of Radiology, Kitasato University Hospital, for technical assistance with the analysis of cardiac magnetic resonance.

CONFLICT OF INTEREST
The authors declare no potential conflict of interest.

DATA AVAILABILITY STATEMENT
The datasets generated and analyzed in the current study are not publicly available due to a request from our ethical committee but are available from the corresponding author on reasonable request.