Evaluation of individual heart rate correction formulas
| Individual correlation coefficients QTc (correction with mean α) v RR | |||||||||
|---|---|---|---|---|---|---|---|---|---|
| Optimum parameter α | p Value (numbers of subjects) | ||||||||
| Mean (SD) | Range | Range | 10−2–10−3 | 10−3–10−4 | 10−4–10−5 | 10−5–10−6 | 10−6–10−7 | < 10−7 | |
| For each generic heart rate correction formula, the table summarises the values of correction parameters α optimised in individual subjects (that is, parameters that lead to the correlation between QTc and RR intervals being zero) as well as the performance of heart rate correction based on the mean value of parameter α among all subjects. For example, in the third line of the table, when a parabolic heart rate correction QTc = QT/RRα was optimised for individual subjects, the individually optimum parameters α ranged from 0.2336 to 0.4856 and their mean was 0.3715. When the heart rate correction QTc = QT/RR0.3715, corresponding to the mean optimum parameter α, was applied to individual subjects, the individual correlation coefficients between QTc and RR intervals ranged from –0.7130 to 0.5765. These correlation coefficients were different from 0 in two subjects with a p value between 10−2 and 10−3 and in four subjects with a p value between 10−3 and 10−4, etc, and in 33 subjects with p value < 10−7. | |||||||||
| (A) Linear | 0.1713 (0.1764) | 0.0928 to 0.2577 | –0.8469 to 0.6485 | 3 | 2 | 1 | 2 | 4 | 38 |
| (B) Hyperbolic | 0.1090 (0.1266) | 0.0721 to 0.1507 | –0.5216 to 0.6239 | 2 | 8 | 6 | 0 | 6 | 28 |
| (C) Parabolic | 0.3715 (0.3830) | 0.2336 to 0.4856 | –0.7130 to 0.5765 | 2 | 4 | 2 | 1 | 8 | 33 |
| (D) Logarithmic | 0.1378 (0.1663) | 0.0884 to 0.1929 | –0.6936 to 0.6518 | 6 | 2 | 0 | 3 | 3 | 36 |
| (E) Shifted logarithmic | 0.2485 (0.2623) | 0.1356 to 0.3741 | –0.8442 to 0.6755 | 3 | 1 | 1 | 3 | 4 | 38 |
| (F) Exponential | 0.3878 (0.4035) | 0.2439 to 0.5427 | –0.7157 to 0.6529 | 3 | 1 | 2 | 1 | 6 | 37 |