Elsevier

Heart Rhythm

Volume 5, Issue 1, January 2008, Pages 89-98
Heart Rhythm

Original-clinical
New algorithm using only lead aVR for differential diagnosis of wide QRS complex tachycardia

https://doi.org/10.1016/j.hrthm.2007.09.020Get rights and content

Background

We recently reported an ECG algorithm for differential diagnosis of regular wide QRS complex tachycardias that was superior to the Brugada algorithm.

Objective

The purpose of this study was to further simplify the algorithm by omitting the complicated morphologic criteria and restricting the analysis to lead aVR.

Methods

In this study, 483 wide QRS complex tachycardias [351 ventricular tachycardias (VTs), 112 supraventricular tachycardias (SVTs), 20 preexcited tachycardias] from 313 patients with proven diagnoses were prospectively analyzed by two of the authors blinded to the diagnosis. Lead aVR was analyzed for (1) presence of an initial R wave, (2) width of an initial r or q wave >40 ms, (3) notching on the initial downstroke of a predominantly negative QRS complex, and (4) ventricular activation–velocity ratio (vi/vt), the vertical excursion (in millivolts) recorded during the initial (vi) and terminal (vt) 40 ms of the QRS complex. When any of criteria 1 to 3 was present, VT was diagnosed; when absent, the next criterion was analyzed. In step 4, vi/vt >1 suggested SVT, and vi/vt ≤1 suggested VT.

Results

The accuracy of the new aVR algorithm and our previous algorithm was superior to that of the Brugada algorithm (P = .002 and P = .007, respectively). The aVR algorithm and our previous algorithm had greater sensitivity (P <.001 and P = .001, respectively) and negative predictive value for diagnosing VT and greater specificity (P <.001 and P = .001, respectively) and positive predictive value for diagnosing SVT compared with the Brugada criteria.

Conclusion

The simplified aVR algorithm classified wide QRS complex tachycardias with the same accuracy as standard criteria and our previous algorithm and was superior to the Brugada algorithm.

Introduction

Despite the publication of numerous algorithms and criteria, relatively few improvements have been made in the ECG differential diagnosis of wide QRS complex tachycardia since the landmark 1978, 1988, and 1991 publications of Wellens et al,1 Kindwall et al,2 and Brugada et al.3 Thus, making an accurate rapid diagnosis in patients with wide QRS complex tachycardia remains a significant clinical problem. Using all reported traditional ECG criteria1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 may yield an accurate diagnosis in approximately 90% of wide QRS complex tachycardias.5, 6, 9 However, many of these criteria are complicated, or they are not applicable in a large proportion of cases and thus are not useful in an urgent setting. Brugada et al3 stated that their algorithm had higher sensitivity and specificity than the traditional criteria, but they did not provide a true head-to-head comparison with other methods used in their study. We recently proposed a new simplified four-step decision treelike algorithm to distinguish between regular monomorphic wide QRS complex tachycardias caused by supraventricular ventricular tachycardia (SVT) and ventricular tachycardia (VT), which proved to be superior to the Brugada algorithm.15 Two new criteria were incorporated in the recent algorithm, which required knowledge only of typical bundle branch and fascicular block patterns, and eliminated some of the complicated morphologic criteria used in the final (fourth) step of the otherwise simple Brugada algorithm. However, in most cases, application of our recent algorithm required more time than did the Brugada algorithm. Therefore, we sought to further simplify and improve our recent algorithm. To this end, another four-step, decision treelike algorithm was devised that completely eliminated the complicated morphologic criteria and limited analysis to a single lead (aVR) while preserving our two previous new criteria. Lead aVR was chosen because we hypothesized that it might be more sensitive than the other leads in differentiating wide QRS complex tachycardias because, in normal sinus rhythm and SVT, the ventricular activation wavefront proceeds in a direction away from aVR, typically yielding a QS complex in aVR. In the present study, the overall test accuracy, sensitivity, specificity, and predictive values of the new aVR algorithm were compared with those of our previous algorithm as well as the Brugada algorithm.

Section snippets

Methods

A different set of patients from that used to test the already established algorithm was used to devise the algorithm. To devise an optimal algorithm, we retrospectively used 103 wide QRS complex tachycardias available from the database of Indiana University. The recordings were obtained from 69 patients with proven electrophysiologic diagnoses in whom wide QRS complex tachycardia was induced during electrophysiologic study. To test the established algorithm, 483 regular wide QRS complex

Patient characteristics

The patient groups differed in that the preexcited tachycardia and SVT groups had younger patients, more female patients, fewer patients with a history of prior myocardial infarction or dilated cardiomyopathy, and far more patients without structural heart disease than the VT group (Table 1).

Overall test accuracy

Results are given in Table 2. The new aVR algorithm was not applicable in only 1 (0.2%) of 483 wide QRS complex tachycardias because of an isoelectric lead aVR in this tracing. In order to compare the

Major findings

The new aVR algorithm is based solely on the principle of differences in the direction and velocity of the initial and terminal ventricular activation during wide QRS complex tachycardia due to VT and SVT. Our study shows that both our new aVR algorithm and the previous algorithm devised for differential diagnosis of wide QRS complex tachycardias have superior overall test accuracy and greater sensitivity and negative predictive value in VT diagnosis compared with the Brugada algorithm. This

Conclusion

Our new aVR algorithm proved to be a reasonably rapid, easy, and accurate means for obtaining the correct diagnosis in the differential diagnosis of wide QRS complex tachycardias. We found the algorithm be superior to the relatively simple Brugada algorithm and at least as accurate as the more complicated traditional morphologic criteria. Therefore, our new algorithm seems to be well suited for application in typically stressful clinical circumstances in which wide QRS complex tachycardia is

Cited by (194)

View all citing articles on Scopus

This manuscript was processed by a guest editor.

View full text