Le Dung Ha1, Aaron F Grober2, Julia Hock3, Matthew Wheeler4, Ayman Elbadawi5, Nishit Biniwale5, Basarat Baig5, Victor Froelicher4. 1. Department of Internal Medicine, Rochester General Hospital, Rochester, NY, United States. Electronic address: ledung.ha@rochesterregional.org. 2. Stanford University School of Medicine, Department of Medicine/Internal Medicine Residency Program, Stanford, CA, United States. 3. German Heart Center Munich, Technical University Munich, Department of Pediatric Cardiology and Congenital Heart Disease, Germany. 4. The Division of Cardiovascular Medicine, Department of Medicine, Stanford School of Medicine, Stanford, CA, United States. 5. Department of Internal Medicine, Rochester General Hospital, Rochester, NY, United States.
Abstract
OBJECTIVE: Clinical utilization of electrocardiography for diagnosis of left atrial abnormalities is hampered by variable P-wave morphologies, multiple empiric criteria, and lack of an imaging "gold standard". Our aim was to determine the prevalence of P-wave patterns and demonstrate which components have associations with cardiovascular death (CVD). METHODS: This is a retrospective analysis of 20,827 veterans <56 years of age who underwent electrocardiograms at a Veteran's Affairs Medical Center from 1987 to 1999, followed for a median duration of 17.8 years for CVD. Receiver Operating Characteristic, Kaplan-Meier and Cox Hazard analyses were applied, the latter with adjustment for age, gender and electrocardiography abnormalities. RESULTS: The mean age was 43.3 ± 8 years, and 888 CVD (4.3%) occurred. A single positive deflection of the P-wave (Pattern 1) was present in 29% for V1 and 81% for V2. A singular negative P-wave (Pattern 2) was present in 4.6% for V1 and 1.6% in V2. A P-wave with an upward component followed by downward component (Pattern 3) was present in 64.5% for V1 and 17.5% for V2. When the downward component in Patterns 2 and/or 3 is at least -100 μV, a significant association is observed with CVD (adjusted hazard ratios [HRs] 2.9-4.1, P < 0.001). Total P-wave duration ≥140 ms was also associated with CVD (adjusted HR 2.2, P < 0.001). CONCLUSIONS: A negative P-wave in V1 or V2 ≤-100 μV, and P-wave with a duration of ≥140 ms, all have independent and significant associations with CVD, with HRs comparable to other electrocardiography abnormalities.
OBJECTIVE: Clinical utilization of electrocardiography for diagnosis of left atrial abnormalities is hampered by variable P-wave morphologies, multiple empiric criteria, and lack of an imaging "gold standard". Our aim was to determine the prevalence of P-wave patterns and demonstrate which components have associations with cardiovascular death (CVD). METHODS: This is a retrospective analysis of 20,827 veterans <56 years of age who underwent electrocardiograms at a Veteran's Affairs Medical Center from 1987 to 1999, followed for a median duration of 17.8 years for CVD. Receiver Operating Characteristic, Kaplan-Meier and Cox Hazard analyses were applied, the latter with adjustment for age, gender and electrocardiography abnormalities. RESULTS: The mean age was 43.3 ± 8 years, and 888 CVD (4.3%) occurred. A single positive deflection of the P-wave (Pattern 1) was present in 29% for V1 and 81% for V2. A singular negative P-wave (Pattern 2) was present in 4.6% for V1 and 1.6% in V2. A P-wave with an upward component followed by downward component (Pattern 3) was present in 64.5% for V1 and 17.5% for V2. When the downward component in Patterns 2 and/or 3 is at least -100 μV, a significant association is observed with CVD (adjusted hazard ratios [HRs] 2.9-4.1, P < 0.001). Total P-wave duration ≥140 ms was also associated with CVD (adjusted HR 2.2, P < 0.001). CONCLUSIONS: A negative P-wave in V1 or V2 ≤-100 μV, and P-wave with a duration of ≥140 ms, all have independent and significant associations with CVD, with HRs comparable to other electrocardiography abnormalities.