P-wave signal-averaged electrocardiogram predicts atrial fibrillation after coronary artery bypass grafting

Ann Noninvasive Electrocardiol. 2002 Jul;7(3):198-203. doi: 10.1111/j.1542-474x.2002.tb00163.x.

Abstract

Background: AF is one of the most common complications after CABG. The aim of the study was to identify the risk factors for postoperative AF.

Methods: Between June and December 2000, 129 consecutive patients (72 men, 47 women; mean age 67 +/- 6 years) underwent preoperative signal-averaged electrocardiogram (SAECG) with assessment of filtered P-wave duration (fPWD) and of the root mean square voltage of the last 10 and 20 ms of atrial depolarization (RMSV10 and RMSV20, respectively) before CABG.

Results: Fifty-six (43%) patients developed one episode of AF lasting > 30 seconds at a mean distance of 2.6 +/- 1.8 days after surgery (group A), while 73 patients remained in sinus rhythm (group B). No differences between the two groups were found in terms of age, sex, P-wave duration on the standard ECG, left atrial dimensions, and operative characteristics. In contrast, group A patients showed a significantly longer fPWD (138 +/- 10 vs 111 +/- 9 ms; P < 0.001) and smaller RMSV10 and RMSV20 (2.8 +/- 1.0 vs 4.3 +/- 1.1 microV, P < 0.001; 4.2 +/- 2.1 vs 6.2 +/- 2.0 microV, P < 0.001). Multivariate analysis indicated only fPWD as an independent predictor of AF (P = 0.009). With a cut-off value of 135 ms for fPWD, the occurrence of AF could be predicted with a sensitivity of 84%, a specificity of 73%, a negative predictive value of 85%, and a positive predictive value of 70%.

Conclusion: Preoperative SAECG is a simple exam that correctly identifies patients at higher risk of AF after CABG. A more widespread use of this technique can be suggested.

MeSH terms

  • Aged
  • Atrial Fibrillation / diagnosis*
  • Atrial Fibrillation / etiology
  • Coronary Artery Bypass / adverse effects*
  • Coronary Artery Bypass / methods
  • Coronary Stenosis / diagnosis
  • Coronary Stenosis / mortality
  • Coronary Stenosis / surgery*
  • Electrocardiography / methods*
  • Female
  • Humans
  • Logistic Models
  • Male
  • Middle Aged
  • Monitoring, Physiologic / methods
  • Multivariate Analysis
  • Postoperative Period
  • Predictive Value of Tests
  • Prospective Studies
  • ROC Curve
  • Risk Assessment
  • Sampling Studies
  • Sensitivity and Specificity
  • Survival Rate