Exploring the Link Between Left Atrial Strain and Exercise-Induced Pulmonary Hypertension

Abstract

Heart failure (HF) with preserved ejection fraction (HFpEF) accounts for approximately half of all HF cases and represents a growing public health burden with substantial morbidity, mortality, and healthcare costs [1, 2]. With the emergence of available treatment options, such as sodium-glucose co-transporter 2 inhibitors, glucagon-like peptide 1 receptor agonists, and nonsteroidal mineralocorticoid receptor antagonists, there is an increased clinical focus on diagnosing and phenotyping HFpEF [3]. However, current diagnostic algorithms remain limited in sensitivity and specificity [4].

In this context, left atrial (LA) dysfunction is increasingly recognized as a key contributor to HFpEF pathophysiology and a promising target for improving diagnostic accuracy. Several studies have demonstrated its prognostic value in predicting mortality and HF-related hospitalizations in this population [5].

Exercise-induced pulmonary hypertension (EIPH) is believed to be an early mechanism contributing to symptoms in HFpEF and may serve as a precursor to overt pulmonary hypertension [6]. Right heart catheterization (RHC) remains the gold standard for diagnosing and classifying EIPH [7]. However, its invasive nature with associated procedural risks, albeit low, limits feasibility for widespread clinical application. Stress echocardiography has emerged as a noninvasive alternative for estimating markers associated with abnormal exercise hemodynamics, such as EIPH [8]. However, consensus on a definitive EIPH definition or a standardized testing protocol with this modality is still lacking [9, 10]. Furthermore, stress echocardiography requires experienced operators, limiting its availability. As a result, identifying noninvasive markers of EIPH using standard echocardiographic protocols is a desirable goal.

In this issue of the journal, Kinoshita et al. provide additional insights into the potential value of considering LA dysfunction in patients with HFpEF [11]. In a retrospective study of 188 patients undergoing stress echocardiography, they explored the relationship between LA reservoir strain and EIPH. EIPH was defined as a peak tricuspid regurgitation gradient of > 50 mmHg and was observed in 34 (18%) of the patients.

The primary finding of the study is that LA reservoir strain at rest was associated with EIPH, with an optimal cut-off value of 21% to discriminate EIPH for a modest area under the curve of 0.69, corresponding to a sensitivity and specificity of 73.5% and 59.1%, respectively. Importantly, the association between LA reservoir strain and EPIH remained significant in multivariable logistic regression analysis.

Collectively, these findings suggest that impaired LA reservoir strain at rest may allude to the presence of elevated pulmonary pressures during exercise. LA reservoir strain is determined mainly by LV longitudinal systolic descent, LA compliance, and LA size. Since GLS did not differ between patients with versus without EIPH, reduced LA compliance could be the underlying mechanism at play. This seems biologically plausible since the LA needs to adapt dynamically during exercise to receive an increased pulmonary venous return. Indeed, similar findings have previously been noted by Telles et al., who compared resting LA strain to peak exercise pulmonary capillary wedge pressure (PCWP) as assessed via RHC in 71 patients, demonstrating an association between noninvasive and invasive measures of EIPH [12]. Telles et al. proposed a considerably higher cut-off for LA reservoir strain of 33% to identify EIPH. This marked discrepancy may be explained by several factors, including differences in study populations and definitions of EIPH (noninvasive vs. invasive). The most likely explanations, however, rely on two key features. First, age differed between the studies (72 vs. 68 years), which may be influential considering that LA reservoir strain declines with age [13]. Second, the studies utilized different ultrasound vendors (GE Healthcare vs. TomTech), and Kinoshita et al. used nondedicated software, whereas Telles et al. used dedicated LA strain software. Nondedicated software typically employs a wider region of interest, resulting in lower estimates of LA strain values [14]. Regardless of the reason, it highlights the main challenge for LA strain and the reason for why it has yet to be implemented clinically; namely, the vendor-dependency and lack of established practical guidelines. Practically, it also raises the challenge of defining the optimal cut-off needed to raise suspicion of EIPH and a question as to whether this cut-off should be tailored to the age of the patient.

As a secondary objective, Kinoshita et al. investigated whether the proposed LA reservoir strain cutoff was associated with HF events in a subgroup of the primary population (n = 134, 71% of the study population, with 29 events) over a median follow-up period of 1 year. Their findings suggest that reduced LA strain is associated with HF events. While this is important, it is important to recognize that the results were largely driven by diuretic use, an endpoint that is less clinically relevant and less robust. It would also have been interesting to investigate whether the presence of EIPH would modify the association between LA strain and HF events, as a means to support that LA dysfunction-associated EIPH would be particularly detrimental in terms of prognosis.

Overall, the study findings must be interpreted in light of limitations: (1) the retrospective design with nonsystematic data collection and risk of residual confounding, (2) small sample size with few outcomes precluding extensive control for confounders, (3) unspecified indication for stress echocardiography, which limits generalizability, (4) use of nondedicated LA strain software, (5) loss to follow-up (29%) and HF events driven primarily by diuretic usage (69% of events). Despite these limitations, the findings by Kinoshita et al. do seem to support available evidence, suggesting an association between both LA strain and EIPH and LA strain and HF events [5, 12]. These findings add to the growing body of evidence suggesting that LA strain could be useful for the assessment of diastolic function and elevated filling pressure [15]. Accordingly, LA strain has rightfully been regarded as a promising measure to aid in the diagnosis of HFpEF, a condition for which treatment options are continuously evolving.

In conclusion, integrating LA reservoir strain into routine clinical practice may improve diagnostic reasoning, risk stratification, and ultimately support decision-making in patients suspected of HFpEF. While its ability to classify EIPH is only modest, it increases the likelihood of identifying this elusive condition. The findings by Kinoshita et al. add to our understanding of the relationship between LA function and EIPH and further emphasize the potential of LA strain. Still, efforts are needed to standardize LA strain assessment and interpret clinically relevant deteriorations in LA strain. As such, future research should focus on validating these findings in larger prospective cohorts. Studies are also needed to investigate whether guideline-directed therapy can improve LA strain and, in turn, improve clinical outcomes to clarify whether LA strain primarily represents a marker or also a potential treatment target in patients with HF.

T.B.S.: Research grants from Bayer, Novartis, Pfizer, Sanofi Pasteur, GSK, Novo Nordisk, AstraZeneca, Boston Scientific, and GE Healthcare. Consulting fees from Novo Nordisk, IQVIA, Parexel, Amgen, CSL Seqirus, GSK, and Sanofi Pasteur. Lecture fees from AstraZeneca, Bayer, Novartis, Sanofi Pasteur, GE Healthcare, and GSK. The remaining authors do not have any potential conflicts to report.