Systemic hypertension adversely impacts cardiac structure and function, causing a wide spectrum of alterations, such as myocyte hypertrophy and interstitial fibrosis leading to left ventricular (LV) concentric remodeling, increased LV mass, and systolic and diastolic dysfunction.

Furthermore, an impressive amount of evidence suggests that left atrial enlargement (LAE), identified by assessing the diameter, or more accurately, the volume with transthoracic echocardiography, is a reliable index of a sustained increase in LV filling pressure in the hypertensive setting [1].

LA acts as a volume sensor of the heart, and its dilatation reflects altered dynamics of ventricular relaxation and filling. The close association between LA volume (LAV) and LV diastolic dysfunction has resulted in LAV becoming one of the key diagnostic criteria for LV diastolic dysfunction [2].

LAE is regarded as a strong risk marker for the future development of atrial fibrillation and a robust, independent predictor of heart failure hospitalizations, stroke, and all-cause death. The mechanism accounting for the association between LAE and cardiovascular events is multifactorial and may be attributed to underlying pathologies causing LA morphofunctional alterations and elevated LV filling pressure and to arrhythmias (i.e., atrial fibrillation), often a consequence of LAE.

The unfavorable prognostic significance of LAE has been further confirmed and expanded in recent years by new evidence based on novel echocardiographic techniques (i.e., two- or three-dimensional speckle tracking echocardiography) and magnetic resonance imaging. LAE measured by four- and two-chamber cine cardiac magnetic resonance imaging demonstrated a strong independent association with all-cause mortality in a very large, heterogeneous cohort totaling 10890 patients [3]. Over a median follow-up period of 49 months, mild, moderate, and severe LAE was a significant predictor of death. After adjustment for clinically relevant covariates, moderate (HR = 1.45, 95% CI: 1.1–1.89) and severe (HR = 1.64, 95% CI: 1.29–2.08) but not mild LAE remained independent predictors of all-cause mortality.

Identification of abnormal cardiac phenotypes (i.e., LV hypertrophy and LAE) by echocardiography has a pivotal role in cardiovascular risk stratification and subsequent therapeutic decision making. Methodological aspects related to this technique may affect the precision of cardiac assessment and the correct classification of patients according to subclinical organ damage. Indeed, the accuracy and precision of quantitative echocardiography are related to multiple factors, including operator experience, patient echogenic characteristics, equipment technology, reliable reading protocols, and appropriate methods for normalizing cardiac parameters for the different body sizes of patients based on precise collection of anthropometric variables. For this latter aspect, it has been reported that indexing echocardiographic parameters to self-reported rather than measured anthropometric values may impair the capacity of detecting LAE. The findings of a multicenter Italian survey showed that misreporting weight and height by individuals attending outpatient echocardiographic laboratories resulted in an underestimation of the prevalence of LAE by 4% and that misclassification was greater among elderly individuals than among young and middle-aged adults [4].

The value of LAV varies according to sex, age, body size, and ethnicity. The indexing of LAV allows comparison among individuals with different body sizes; however, the best method of normalization for body size remains open to debate. The American Society of Echocardiography and the European Association of Cardiovascular Imaging (ASE/EACVI) guidelines recommend indexing the LAV for body surface area (BSA) because the use of this approach also accounts for sex differences in LA size [5].

In this respect, it is worth mentioning that the Normal Reference Ranges for Echocardiography (NORRE) study provided a set of normal echocardiographic values obtained from a cohort of 734 healthy subjects (320 men and 414 women) over a wide range of ages [6]. The study showed significantly different LA sizes and volumes between men and women, but after normalization for BSA, LA volumes were no longer different between groups (27.0 ± 7.0 ml/m2 in men, 27.3 ± 7.9 ml/m2 in women). Upper reference values (means ± 2 SD) for LA volumes were 41.9 ml/m2 in men and 41.5 ml/m2 in women using the area-length method. Overall, these data supported the view that for LA volumes, there is a reduced need to check for sex-specific diagnostic cutoff values. In this issue of the journal, the study by Airale et al. [7] adds a new piece of information to the scientific debate related to methods of indexing echocardiographic parameters by specifically focusing on the left atrium in the context of the assessment of hypertension-mediated target organ damage. Before addressing the results of the study, a few aspects of this issue and related topics may warrant some considerations. The Pressioni Monitorate e Loro Associazioni (PAMELA) study showed that LA size exhibited a near-normal distribution in both sexes and the whole population, which means that for several other biological variables, the use of partition values to distinguish abnormality from normality is arbitrary [8]. More importantly, the PAMELA study reported a direct relationship between LA size and many factors (i.e., age, sex, and office, home, and ambulatory BP) and metabolic alterations, which means that its detection should alert the physician to the link between LAE and the presence of cardiovascular risk factors.

A meta-analysis of 15 studies in a pooled hypertensive population of 10,141 patients documented that echocardiographic LAE is frequently detected in treated and untreated hypertensive patients, with an overall prevalence of ~30% [9]. In a sex-specific analysis based on >80% of the total sample, LAE prevalence was found to be nonsignificantly higher in women than in men (OR = 1.23 95% CI: 0.83–1.83, p = 0.30).

In a multicenter study involving 2170 hypertensive individuals (mean age, 62 years; 53% men) referred by practitioners to 17 outpatient echocardiographic laboratories across Italy, prevalence rates of LAE in both sexes showed different trends depending on the diagnostic criteria used [10]. LAE according to absolute LA diameter was present in 39.5% of men and 35.5% of women (p = ns), whereas a significant opposite trend was found by using the value normalized to BSA (14.3% versus 26.9%, respectively; p < 0.01). The more pronounced LA remodeling in women was associated with a constellation of risk factors such as older age, higher systolic and pulse pressures, and higher prevalence of obesity and LVH.

In this line of research, the paper by Airale et al. tested the difference in LAE prevalence by using two different diagnostic criteria: (I) an isometric and non-sex-specific criterion (i.e., LA volume >34 ml/m2) endorsed by the ASE/EACVI guidelines, and (II) an allometric and sex-specific criterion (i.e., >18.5 ml/h2 in males and >16.5/h2 in females) cited in the 2018 European Hypertension Society/European Cardiology Society (ESH/ESC) guidelines. For this purpose, 441 essential hypertensive patients (mean age, 60 ± 14 years; 52% men; 26% obese and 10% diabetic) were enrolled in the study. Approximately one-third of the patients had LVH (24% men and 32% women), most of the concentric type. In the whole population, the prevalence of LAE was twice as high when using allometric sex-specific indexation rather than BSA (50.6% vs. 23.4%, p < 0.001). The corresponding figures were 42.8% versus 23.1% in men and 59.0% versus 23.6% in women. According to the absence or presence of LAE, the patients were further classified into three groups: (A) normal LA size (n = 218; mean age, 55 ± 13 years; 40% women), (B) LAE defined by allometric indexation (n = 120; mean age, 64 ± 14 years; 63% women), and (C) LAE defined by both allometric and isometric indexation (n = 103; mean age, 67 ± 15 years; 49% women). The prevalence of LVH increased progressively from group A (14%) to group B (34%) and group C (50%). Of note, a deterioration of diastolic function parameters, as assessed by the ratio of early (E) peak mitral inflow velocity to mitral (septal and lateral) annular velocity (E/e’ ratio), occurred in parallel with the increase in LV structural alterations. Consequently, the findings by Airale et al. suggest that indexing for BSA markedly underestimates the prevalence of LAE because in >25% of a large number of untreated and treated hypertensive patients classified as being free of abnormalities in LA size, indexation for height2 allowed us to identify LAE, indicating unrecognized target organ damage of adverse prognostic significance and thus leading to the reclassification of these patients as high-risk individuals. Interestingly, the underestimation of the risk of LAE was strongly prevalent in the female sex, thus emphasizing the appropriateness of using different sex-specific cutoffs (as widely accepted for the diagnostic criteria of LVH).

The key argument provided by the authors supporting the clinical value of LAE, identified only by height2, relies on its association with LV remodeling expressed by increased LV mass and prevalence of LVH, as well as an impairment in diastolic dysfunction. The strength of this argument is, however, limited by the fact that patients diagnosed with LAE by height2 were older (+9 years) and more likely to be obese and had higher pulse pressure than their counterparts with normal LAVs. In fact, these differences, not adjusted in the statistical analysis, may explain per se the greater involvement of subclinical cardiac organ damage in this subset of patients. Regardless of the limits recognized by the authors themselves, this study has the merit of shedding light on a topic that has been neglected so far, represented by the need to use sex-specific diagnostic criteria in phenotyping an essential component of hypertension-mediated organ damage such as LAE. From this perspective, further cross-sectional and longitudinal studies are needed, aimed at defining (I) the normal values of the LAV indexed by height2 in the context of large cohorts representative of the general population and (II) its independent prognostic value.