Specificity for each of the 46 criteria of the Selvester QRS score for electrocardiographic myocardial scar sizing in left bundle branch block

J Electrocardiol. 2015 Sep-Oct;48(5):769-76. doi: 10.1016/j.jelectrocard.2015.07.018. Epub 2015 Jul 21.

Abstract

Background: The Selvester QRS score consists of a set of electrocardiographic criteria designed to identify, quantify and localize scar in the left ventricle using the morphology of the QRS complex. These criteria were updated in 2009 to expand their use to patients with underlying conduction abnormalities, but these versions have thus far only been validated in small and carefully selected populations.

Aim: To determine the specificity for each of the criteria of the left bundle branch block (LBBB) modified Selvester QRS Score (LB-SS) in a population with strict LBBB and no myocardial scar as verified by cardiovascular magnetic resonance imaging with late gadolinium enhancement (CMR-LGE).

Methods: We identified ninety-nine patients with LBBB without scar on CMR-LGE, who underwent a clinically indicated CMR scan at three different centers. The ECG recording date was any time prior to or <30days after the CMR scan. The LB-SS was applied and specificity for detection of scar in each of the 46 separate criteria was determined.

Results: The specificity ranged between 41% and 100% for the 46 criteria of LB-SS and 27/46 (59%) met ≥95% specificity. The mean±SD specificity was 90%±14%.

Conclusion: Several of the criteria in the LB-SS lack adequate specificity. Elimination or modification of these nonspecific QRS morphology criteria may improve the specificity of the overall LB-SS.

Keywords: Electrocardiography; Left bundle branch block; Myocardial scar; Selvester QRS score; Specificity.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Algorithms*
  • Bundle-Branch Block / classification
  • Bundle-Branch Block / complications
  • Bundle-Branch Block / diagnosis*
  • Cicatrix / classification
  • Cicatrix / complications
  • Cicatrix / diagnosis*
  • Diagnosis, Computer-Assisted / methods*
  • Electrocardiography / methods*
  • Female
  • Humans
  • Male
  • Middle Aged
  • Myocardial Stunning / classification
  • Myocardial Stunning / complications
  • Myocardial Stunning / diagnosis*
  • Observer Variation
  • Reproducibility of Results
  • Sensitivity and Specificity
  • Severity of Illness Index
  • Software Validation