Fig. 4

Example of the application of different QTc corrections to QT and heart rate data measured in an 8-year old boy. All the measurements (not only those made during postural provocations) made in the recordings are shown. The top left panel shows a scatter diagram between uncorrected QT intervals and the corresponding 10-s heart rates, the top right, bottom left, and bottom right panels show scatter diagrams between Bazett, Fridericia, and Framingham corrected QTc intervals and the heart rates, respectively. The top left panel shows that in this child, QT interval was adapting to heart rate changes in a rather shallow manner and that the true QTc value in this child was somewhere between 410 and 420 ms. Red arrow in this panel identifies one particular ECG segment in which an uncorrected QT interval of 354 ms was measured at the heart rate of 121.9 bpm. Since the Bazett correction assumes that the QT interval adapts to heart rate along a square root of the RR intervals (short-dash red line in the top left panel) it leads to QTc of 505 ms. Fridericia and Framingham corrections assume different heart rate adaptations (middle-dashed and long-dashed lines in this panel, respectively) and thus lead to QTc of 448 and 432 ms, respectively. Hence in this particular child, Framingham correction is the closest to the true QT/RR profile but that does not mean that the same formula preference would be found with other children. Only Bazett correction overcorrected the QT/RR profile in all children of our study