Related Experiment Video
Updated: Jan 16, 2026

Morphological and Functional Assessment of the Right Ventricle Using 3D Echocardiography
Published on: October 28, 2020
Electrocardiographic LVH criteria: Poor diagnostic accuracy even with optimized cutoffs. Insights from MESA study
José Nunes de Alencar1, Sandro Pinelli Felicioni2
1Electrocardiography Unit, Instituto Dante Pazzanese de Cardiologia, São Paulo, Brazil; Research Division, Instituto Dante Pazzanese de Cardiologia, São Paulo, Brazil.
Insights
Electrocardiogram (ECG) voltage criteria show poor accuracy for diagnosing left ventricular hypertrophy (LVH) when compared to cardiac magnetic resonance (CMR). Combining ECG measures does not improve diagnostic performance for LVH.
Area of Science:
- Cardiology
- Medical Imaging
- Diagnostic Performance Evaluation
Background:
- Electrocardiogram (ECG) criteria for left ventricular hypertrophy (LVH) are commonly used to estimate increased left ventricular (LV) mass.
- However, these ECG criteria demonstrate limited sensitivity in accurately diagnosing LVH.
Purpose of the Study:
- To evaluate the diagnostic performance of various ECG voltage criteria for LVH against cardiac magnetic resonance (CMR) in a large population cohort.
- To assess whether combining ECG indices improves LVH detection compared to individual criteria.
Main Methods:
- Analysis of 4849 participants from the Multi-Ethnic Study of Atherosclerosis (MESA) with ECGs and CMR-derived LV mass.
- LVH defined by LV mass indexed to body surface area (LVMi) using established thresholds for men and women.
- Assessment of multiple ECG criteria (Sokolow-Lyon, Cornell voltage, etc.) and combinations using logistic regression.
Main Results:
- LVH prevalence was 30.2%.
- ECG criteria at classic thresholds showed very low sensitivity (3.7-15.2%) but high specificity (>94%).
- Optimized thresholds improved sensitivity (48.9-64.8%) but reduced specificity (54.4-74.3%), with modest diagnostic utility (LR+ <4).
- Combining ECG indices did not improve diagnostic discrimination compared to the best single measure (AUC 0.646).
Conclusions:
- ECG voltage criteria exhibit poor diagnostic performance for CMR-defined LVH, regardless of threshold optimization.
- Aggregating multiple ECG indices does not materially enhance diagnostic accuracy for LVH.
- Findings suggest reframing ECG voltage criteria as an electrical phenotype with prognostic value rather than a reliable diagnostic surrogate for structural hypertrophy.
Background:
Electrocardiographic (ECG) "left ventricular hypertrophy (LVH) criteria" are widely used as surrogates for increased LV mass but show limited sensitivity. We evaluated their diagnostic performance against cardiac magnetic resonance (CMR) in a large population cohort.
Methods:
We analyzed 4849 participants from the Multi-Ethnic Study of Atherosclerosis (MESA) with standardized digital 12‑lead ECGs and CMR-derived LV mass indexed to body surface area (LVMi). LVH was defined as LVMi >91 g/m2 (men) and > 77 g/m2 (women). We assessed Sokolow-Lyon, Cornell voltage, Cornell product, Peguero-Lo Presti, Sokolow-aVL, and MESA-LVH (SV1 + SV2 + RV5). Diagnostic accuracy was calculated for classic ("pre-specified") thresholds and for cutoffs optimized by the Youden index. We also tested whether combining indices via elastic-net penalized logistic regression improved discrimination.
Results:
LVH prevalence was 30.2 %. Across classic thresholds, sensitivities were very low (3.7-15.2 %) with high specificities (>94 %); likelihood ratios were modest (typical LR+ <4; LR- ≈ 0.89-0.96). Youden-optimized cutoffs improved sensitivity (48.9-64.8 %) at the expense of specificity (54.4-74.3 %), yielding weak LR+ (1.42-1.94) and LR- (0.64-0.71). Pooled area under the ROC curve (AUC) for individual indices ranged 0.62-0.66. An elastic-net model combining all indices did not enhance discrimination beyond the best single measure (AUC 0.646).
Conclusions:
In MESA, ECG voltage criteria-whether applied at classic or optimized thresholds-show poor diagnostic performance for CMR-defined LVH. Aggregating indices provides no material gain. These findings support reframing voltage positivity as an amplitude-based electrical phenotype with prognostic value rather than a reliable diagnostic surrogate for structural hypertrophy.
Clinical Trial Registration:
ClinicalTrials.gov Identifier NCT00005487.
Related Concept Videos
Cardiomyopathy III: Hypertrophic Cardiomyopathy
Mitral Stenosis II: Clinical features and Diagnostic Tests
Aortic Regurgitation II: Clinical Features and Diagnostic Tests
Acute Coronary Syndrome III: Diagnostic Studies
Electrocardiogram
Three major waveforms are present in a typical ECG recording: the P wave, the QRS complex, and...

