Abstract TP151: 30-Minute ABPM Underdetects White Coat Hypertension Despite Good Diagnostic Accuracy

Résumé fourni par la source

Introduction: Hypertension is the leading modifiable stroke risk factor, yet 24-hour ambulatory blood pressure monitoring (ABPM) access is limited by patient burden and capacity constraints. Shorter 30-minute protocols may expand population-based stroke screening but could misclassify white coat hypertension (WCH), causing unnecessary treatment and inefficient resource allocation. Aims: To compare 30-minute versus 24-hour ABPM for diagnosing hypertension and detecting WCH, and evaluate a tiered approach for stroke prevention in resource-limited settings. Methods: Cross-sectional study of 90 Vietnamese adults from routine care (median age 34 years [IQR 25-55]; 60% male) completing both protocols. Office hypertension was ≥140/90 mmHg. Normal ambulatory BP was <130/80 mmHg (24-hour) and <135/85 mmHg (30-minute). Agreement assessed using Bland-Altman analysis; diagnostic performance using ROC curves; classification consistency using Cohen's kappa. Results: Mean BP by 30-minute monitoring was 136.0±13.1/84.1±11.7 mmHg versus 130.4±12.6/77.5±9.3 mmHg by 24-hour ABPM. Strong correlations existed for SBP (r=0.687, R2=0.479) and DBP (r=0.787, R2=0.620), both p<0.001. Bland-Altman analysis revealed systematic bias with 30-minute ABPM overestimating BP by +5.5 mmHg for SBP (95% LoA: -14.3 to 25.4 mmHg) and +6.7 mmHg for DBP (95% LoA: -7.4 to 20.8 mmHg), indicating substantial individual variability. ROC analysis demonstrated good diagnostic performance: SBP AUC=0.830 with 94.5% sensitivity and 62.9% specificity; DBP AUC=0.786 with 80.0% sensitivity and 77.1% specificity. Overall classification accuracy was 78.9% with moderate agreement (κ=0.544). For WCH detection, 24-hour ABPM identified 27/90 patients (30.0%) versus 20/90 (22.2%) by 30-minute ABPM, representing 7.8% absolute underdetection rate. Conclusions: Thirty-minute ABPM provides accessible screening with acceptable accuracy but systematically overestimates BP and underdetects WCH. For population-level stroke prevention, we recommend a two-step pathway: first-line 30-minute monitoring with interpretive adjustments (subtract 5–7 mmHg), followed by confirmatory 24-hour monitoring for positive screens. This tiered approach prioritizes accurate risk stratification, prevents WCH overtreatment, and enables cost-effective stroke prevention in resource-constrained health systems.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.