Signal-averaged P-wave duration and risk of paroxysmal atrial fibrillation in hyperthyroidism

Am J Cardiol. 1996 Feb 1;77(4):266-9. doi: 10.1016/s0002-9149(97)89391-5.

Abstract

The onset of atrial fibrillation (AF) in hyperthyroid patients constitutes an unfavorable clinical event associated with high risk of cardiovascular complications, occurring in approximately one fifth of patients. Therefore, it is advantageous to define noninvasive markers that may identify patients at risk. The high-resolution, signal-averaged electrocardiogram was used to evaluate the relation between P-wave duration and occurrence of paroxysmal AF in a group of 50 patients with hyperthyroidism, of whom 24 had a history of paroxysmal AF and 26 did not. Filtered signal-averaged P-wave duration was measured over an average of 300 beats/patient while in sinus rhythm, both at the time of first diagnosis of hyperthyroidism and after restoration of euthyroidism by medical treatment. The 24 patients with paroxysmal AF had significantly greater P-wave duration than the 26 patients without it (135 +/- 7 vs 124 +/- 9 ms; p = 0.001). A P-wave duration cut-off value of 130 ms held specificity, sensitivity, and positive predictive accuracy values of 79%, 85%, and 83%, respectively. Of several variables, multivariate analysis showed P-wave duration to be the only independent variable significantly associated with the occurrence of paroxysmal AF. Thus, the high-resolution signal-averaged electrocardiogram may be a useful noninvasive clinical tool for the identification of electrical instability associated with paroxysmal AF in hyperthyroid patients.

Publication types

  • Clinical Trial

MeSH terms

  • Adult
  • Aged
  • Atrial Fibrillation / diagnosis*
  • Atrial Fibrillation / etiology
  • Electrocardiography*
  • Female
  • Humans
  • Hyperthyroidism / complications*
  • Logistic Models
  • Male
  • Middle Aged
  • Multivariate Analysis
  • Predictive Value of Tests
  • Reproducibility of Results
  • Risk Factors
  • Sensitivity and Specificity
  • Signal Processing, Computer-Assisted*
  • Time Factors