Analysis of Holter Recordings
In all three observation periods (baseline and at every 8 and 12 h) the Holter PVC rates used to determine efficacy of the treatment in each patient were the averages of two recordings. Medians as well as means were used to characterize the central tendency of the PVC rate since an extreme outlier was present on 12-h therapy. Means were used to summarize couplets and VT rates.
To determine whether the reductions in PVCs from baseline observed during the treatment periods were in excess of reductions that might be attributable to day-to-day variability in the PVC rate, the Wilcoxon signed rank test was applied. The variation between the two baseline PVC rates in each patient was used to characterize the day-to-day variability. This study was not designed to take into account the greater variability in PVC rates that may occur over longer periods. Therefore, criteria reported from other studies were used to distinguish a true drug effect from the effect of spontaneous variability over time. The Wilcoxon signed rank test also served for comparison of the end-of-titration responses on the 8-h regimen with those obtained on the 12-h schedule.
To track the time course of arrhythmia suppression during the two different dosage intervals, the number of PVCs occurring every hour after the morning dose were read off from the last two Holter recordings made in each patient during 8-h and 12-h treatment; these data were then grouped to arrive at the medians for each hour.
Mexiletine steady-state plasma concentrations at the end of the dose interval (trough) and 2 h after administration (approximate peak) on the 8- and the 12-h regimens were evaluated. Blood was sampled by venipuncture, centrifuged (2,000 rpm for 10 min) to plasma and frozen until analysis. Mexiletine concentrations were measured by a sensitive and specific HPLC method. They are reported here as means with standard deviations.