Article Text
Abstract
Introduction Identification of limb lead misplacements involve pattern recognition of characteristic ECG changes in multiple leads. This study aims to determine if limb lead misplacements can be identified using Lead I alone.
Methods ECGs from a database of confirmed and corrected limb lead misplacements collected over a 10-year period were analysed for the presence of inverted P, QRS and T waves in Lead I. Statistical significance was determined using Fisher’s exact test. The presence of S waves in Lead I was analysed using the Wilcoxon signed rank test.
Results 269 confirmed limb lead misplacements were identified (table 1). Specificity of inverted P-waves, QRS complexes and T-waves in Lead I for limb lead misplacement were 86.7%, 97% and 94.8% respectively, p=0 (table 2). Corresponding sensitivities were 54.7%, 49.6% and 37.2%, respectively. The presence of an S wave in Lead I had a sensitivity of 68.4% and specificity of 78.4%, p<0.05. Interpreting Lead I alone was insufficient for diagnosis of left arm-left leg (55) and left arm-right leg (17) misplacement as comparison with additional leads was required. ECG machines detected 55% of right arm-left arm misplacements, which equates to 18.6% of all limb lead misplacements. Using Lead I alone detected 73.2% and thus outperformed the ECG interpretation software by a factor of 3.9.
Conclusion Inverted P-waves, QRS complexes, or T-waves and/or the presence of S-waves in Lead I were statistically significant for ECG lead misplacement and can confidently be used to ‘rule-in’ the possibility that leads have been misplaced. Their presence should prompt a lead placement check, repeat ECG and comparison with previous ECGs. Due to the low sensitivities, the absence of these findings cannot be used to exclude the diagnosis of ECG limb lead misplacement. All staff responsible for recording and interpreting ECGs should be aware of these findings.