Development and Internal Validation of a Diagnostic Nomogram for Identifying Left Atrial Enlargement in Patients with...

Left atrial enlargement (LAE) is a critical indicator of target organ damage in essential hypertension. However, simple and accessible tools for individualized early screening in routine practice remain limited. This study derived and internally validated a practical nomogram using standard clinical metrics to estimate LAE probability, acknowledging its development within a single retrospective cross-sectional study.

Clinical data from 251 hypertensive patients admitted to our hospital between August 2023 and September 2024 were retrospectively reviewed. Because clinical predictors and echocardiographic LAE were assessed concurrently, a retrospective cross‑sectional design was applied. LAE was defined according to guideline‑based left atrial anteroposterior diameter criteria. Five clinically prespecified candidate variables (age, hypertension duration, atrial fibrillation, hypertension stage, and coronary heart disease) were entered into a multivariable logistic regression model to construct the nomogram. Discriminative performance was quantified by the area under the receiver operating characteristic (ROC) curve (AUC).

The final model included all five variables. Older age (p < 0.001), increased hypertension duration (p = 0.010), and atrial fibrillation (AF) were independently associated with the presence of LAE. Hypertension stage and coronary heart disease (CHD), though not statistically independent, were retained for clinical relevance. The derived nomogram showed high internal discrimination, with an AUC of 0.937.

In the final model, older age (per year increase: OR = 1.15, 95% CI: 1.08-1.24, p < 0.001), increased hypertension duration (per grade increase: OR = 2.00, 95% CI: 1.25-3.57, p = 0.003), AF (OR = 3.70, 95% CI: 1.15-12.12, p = 0.028), and CHD were identified as independent predictors. Hypertension stage did not reach statistical significance (stage 2: OR = 0.78, 95% CI: 0.17-3.85, p = 0.753; stage 3: OR = 1.96, 95% CI: 0.50-9.22, p = 0.340) but was deliberately retained in the final model owing to its established clinical relevance and pathophysiological role in hypertensive atrial remodeling. The tool demonstrated high internal discrimination, with an apparent AUC of 0.937 and an optimism‑corrected AUC of 0.920 after 1000 bootstrap resamples, along with high calibration accuracy (mean absolute error: 0.017).