Quantitation of chronotropic response: Comparison of methods for rate-modulating permanent pacemakers

Abstract

Objectives. The aim of this study was to developarechnique for quantitating chronotropic response.Background. Although the importance of chronotropic response for optimizing cardiac output during exercise is widely recognized, methods for quantitating the rate-modulating behavlor of permanent pacemakers have not been developed. For a method of quantiting chronotropic response to be clinically useful, the rate-modulating characteristics of a pacing system should be defined at the onset of exertion, over a variety of exercise work loads and during recovery.Methods. Three methods for quantitation of rate modulation were assessed in 10 patients during treadmill exercise testing using the chronotropic assessment exercise protocol with expired gas exchange analysis. To compare the observed chronotropic response with a standard, the “expected” heart rate throughout exercise was calculated by using the concept of heart rate reserve as described by Wilkoff. The pacing rate observed during exercise was analyzed with 1) standard linear regression analysis, 2) comparison of observed and expected pacing rates at the midpoint and end of each quartile of exercise, and 3) integration of the area under the rate-response curve with comparison with the area under the expected curve.Results. With use of anormalized scale relating change in heart rate to change in metabolic work load, with values of heart rate and metabolic work load at rest set to 0 and those at maximal exertion set to a value of 1, the mean y intercept for the study group was 0.10 ± 0.20 (range -0.14 to + 0.45), with a mean slope of 0.81 ± 0.25 (range 0.31 to 1–19). The correlation coefficient relating change in heart rate to change in exercise work load was a mean of 0.90 ± 0.09 (range 0.63 to 0.98). Integration of the area under the rate-response curve observed during exercise yielded a mean area that was 101 ± 36% of that expected. When the range of exercise work loads was divided into quartiles, the area under the observed rate-response curve was 151 ± 114% of that expected during the first quartile of exercise, 113 ± 70% during the second, 96 ± 38% during the third and 92 ± 20% during the fourth. The mean area under the curve during recovery was 93 ± 29% of that expected. Although calculation of the observed heart rate as a percent of that expected at the midpoint and end of each quartile of exercise used fewer observations, it provided similar results.Conclusions. Quantitation of the rate-response curve with comparison with the expected heart rate curve provides accurate methods for quantisation of chroeotropic response. Adoption of this method would facilitate comparisons of artificial sensors and provide a framework to address issues of optimal rate modulation.