Revista Brasileira de Medicina do Esporte
Print version ISSN 1517-8692
On-line version ISSN 1806-9940
LAZZOLI, José Kawazoe; CASTRO, Claudia Lúcia Barros de; NOBREGA, Antonio Claudio Lucas da and ARAUJO, Claudio Gil Soares de. Accuracy of vagotonia criteria in the 12-lead resting electrocardiogram: analysis with ROC curves. Rev Bras Med Esporte [online]. 2002, vol.8, n.2, pp.50-58. ISSN 1517-8692. http://dx.doi.org/10.1590/S1517-86922002000200004.
OBJECTIVE: To determine sensitivity (Sens), specificity (Spec), positive (PPV) and negative (NPV) predictive value, and accuracy of different electrocardiographic criteria for vagotonia. METHODS: Seventy-four Olympic athletes (51 male and 23 female; age = 24 ± 5 years) underwent an evaluation consisting of a 12-lead resting electrocardiogram (ECG), and an assessment of cardiac vagal tone (CVT) by the 4-second cycling exercise testing (4-sET), with the B/C index utilized as the gold standard. The 4-sET is a well-standardized and valid protocol for CVT assessment based on the mechanism of exercise-onset tachycardia regardless of the sympathetic component. Using ROC (receiver operating characteristic) curves, the authors selected the cut-point with the best sensitivity (Sens)/specificity (Spec) ratio for each ECG variable, and calculated the positive (PPV) and negative predictive values (NPV), and accuracy (AC). The authors also correlated B/C index with ECG variables. RESULTS: B/C index showed a poor association with ECG variables, with the exception of R-R interval (r = 0.353; P = 0.004), that also presented a significant ROC curve (c2= 0.863; P = 0.002), whereas the R-R interval ³ 990 ms criterion showed the following results: Sens - 100%; Spec - 73%; VPP 20%; VPN 100%; AC 75%. The ECG variables that presented cut-points with accuracy > 80% were: T-wave amplitude in lead V5 ³ 8.0 mm; T-wave amplitude in lead V6 ³ 7.0 mm; Q-wave amplitude in V5 ³ 7.0 mm; R-wave amplitude in V4 ³ 50.0 mm; presence of J-point elevation in two or more leads; presence of S-T segment elevation in five or more leads, and presence of fast precordial transition. DISCUSSION AND CONCLUSION: The performance of ECG criteria considering PPV, NPV, and AC was generally very poor, with the exception of the R-R interval. The combination of other criteria such as T-wave amplitude in leads V5 and V6, R-wave amplitude in V4, and presence of J-point and S-T segment elevation in several leads should help predicting vagotonia.
Keywords : Cardiac vagal tone; Electrocardiography; ROC curves; 4-s cycling test.