The report by Cheng et al in the September 1, 2010, issue of The American Journal of Cardiology provides new insights for clinicians and researchers alike on the relation of QRS duration (QRSd) and the subsequent implantation of a permanent pacemaker (PPM). The investigators, using 8,311 subjects from the original and offspring cohorts of the Framingham Heart Study, showed that QRSd is a predictor of PPM implantation. This association was shown in all subjects and in those from the 3 categorical subgroups with QRSd <100, 100 to <120, and ≥120 ms, with and without the inclusion of subjects with bundle branch block and the use of nodal blocking agents.

The importance of the study stems from the database used, which included mostly healthy community subjects followed periodically with electrocardiography and not patients with various confounding illnesses, which could have influenced the need for PPM implantation irrespective of QRSd. However, one wonders whether Cheng et al could “get more mileage” from their study and enhance its impact on decision making by considering exploration of the following issues.

First, the investigators measured QRSd mostly manually to the nearest 10 ms using all 12 electrocardiographic leads, while automation-based measurements carried out by contemporary electrocardiographs measure only certain leads and calculate a “global” QRSd. Also, the investigators state in their discussion that their “analyses were based on the physician-measured QRSd, which approximates what is done in clinical practice”; this reader doubts that the average clinician currently measures QRSd in all 12 electrocardiographic leads but instead uses for clinical decisions the automated QRSd value provided in the report upon recording an electrocardiogram. Accordingly, it would be important to evaluate whether the association between QRSd and PPM implantation holds its validity when only data from patients who had automated QRSd measurements (even limited) are included. Also, the strength of the association between QRSd and PPM implantation should be compared in the cohorts with manual and automated QRSd measurements. If such an assessment shows comparability, it will provide justification for extrapolation of the conclusions of this study to the contemporary milieu, in which automated measurements of QRSd abound.

Second, with the widely observed variation in QRSd on serial electrocardiograms, continuously altering certain patients’ assignment to 1 of the 3 categories of QRSd, one wonders what were the QRSd values representative for the individual patients over long-term follow-up.

Third, Cheng et al may need to explore whether the increase in QRSd and its rate of development using the entire follow-up period, rather than the mere QRSd at a specific follow-up epoch, are better predictors for the need for PPM implantation.

Fourth, the investigators, surprisingly, did not include an analysis of the presence or the development of left and right fascicular block as a potential confounding factor in the association of QRSd and PPM implantation.

Fifth, in the same vein, an analysis of the frontal QRS axis and its change either to the left or the right (irrespective of the satisfaction for criteria of left or right fascicular block) may provide an insight whether this variable is an independent predictor of the subsequent need for PPM implantation.

Finally, given that 57% of the subjects who underwent PPM implantation had sick sinus syndrome as an indication for the procedure, it would be helpful to assess the role of heart rate and its progressive slowing over consecutive follow-up evaluations as independent predictors of the need for PPM implantation; such an analysis should be carried out with and without the inclusion of subjects who received negative chronotropic drugs.