Man versus machine: is there an optimal method for QT measurements in thorough QT studies?

Borje Darpo, Marilyn Agin, David J Kazierad, Gary Layton, Gary Muirhead, Peter Gray, Diane K Jorkasky

Journal: Journal of clinical pharmacology 2006;46(6):598-612

PMID: 16707406

Full Text on Ovid Technologies, Inc. View on PubMed

Abstract

Electrocardiographic (ECG) recordings from 3 placebo-controlled thorough QT healthy volunteer studies were used to compare QT intervals obtained by manual measurement with those generated by ECG machines. The effect of the positive control was compared to placebo at each time point for data obtained from both sources. Both manual and automated techniques consistently demonstrated statistically significant prolongation of QTcF with the positive controls. The proportion of outlier values was small for both methods. The pairwise comparison between manual and automated uncorrected QT intervals demonstrated clear differences, with intervals derived from one machine on average 16 to 19 milliseconds shorter and from the other 7 milliseconds longer than the manually measured QT intervals, but these differences disappeared when analyzing QT change from baseline. Both manual and automated, commercially available QT algorithms demonstrated small statistically significant effects on the QTc interval induced by positive controls.

Address: MSc, Clinical Statistics, Clinical R and D, Pfizer Global Research and Development, Ramsgate Road, Sandwich, Kent, CT13 9NJ, United Kingdom.

DOI: 10.1177/0091270006286900

Link outs

Subscription / membership required

Full Text Sources:

MeSH Terms: Adolescent, Adult, Aza Compounds, Control Groups, Cross-Over Studies, Electrocardiography, Female, Fluoroquinolones, Humans, Ketoconazole, Long QT Syndrome, Male, Middle Aged, Moxifloxacin, Quinolines, Reproducibility of Results, Research Design, Signal Processing, Computer-Assisted