Improving the precision of QT measurements

Cardiol J. 2011;18(4):401-10.


Background: Accurate and precise QT interval measurement is very important for both regulatory and drug developmental decision making. These measurements are often made using a manual or semi-automated technique, and the associated variability necessitates sample sizes of around 50 to 70 subjects in thorough QT/QTc studies. The purpose of this study was to compare the reproducibility and precision of a semi-automated (SA) method and a high-precision (HPQT) technique for ECG extraction and QT interval measurement on two thorough QT/QTc (TQT) studies conducted in compliance with ICH E14.

Methods: Data from 35 healthy subjects from two different crossover TQT studies on treatment with placebo and moxifloxacin was analyzed. Both methods examined the RR and QT intervals measured in lead II or the lead with the highest quality T-wave on a single beat basis using the QT algorithm included in the COMPAS software package. ECGs were measured at a protocol-specific timepoint.

Results: The effect of moxifloxacin on the QTc interval was highly reproducible in the two studies, and assay sensitivity was met with both methods. Pairwise comparison of QTcF values between methods demonstrated high agreement with no bias, small mean differences (below 1.5 ms) and narrow limits of agreement. HPQT improved the precision of the QTc measurement by 31% in Study I (standard deviation of DQTcF: SA 8.9 ms; HPQT 6.3 ms) and by 15% in Study II (SD: SA 9.7 ms; HPQT 8.3 ms).

Conclusions: The HPQT QT measurement technique detected the effect induced by moxifloxacin with the same accuracy as SA techniques, and with clearly improved precision. More precise QTc measurement has important implications in terms of lowering the likelihood of false positive results and/or reducing the sample size in TQT studies, as well as improving the utility of QT assessment in early clinical development.

Publication types

  • Comparative Study
  • Randomized Controlled Trial
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adolescent
  • Adult
  • Algorithms
  • Automation, Laboratory
  • Aza Compounds / adverse effects*
  • Cross-Over Studies
  • Double-Blind Method
  • Electrocardiography / standards*
  • Female
  • Fluoroquinolones
  • Heart Conduction System / drug effects*
  • Humans
  • Linear Models
  • Male
  • Middle Aged
  • Moxifloxacin
  • Observer Variation
  • Predictive Value of Tests
  • Quinolines / adverse effects*
  • Reproducibility of Results
  • Signal Processing, Computer-Assisted*
  • Time Factors
  • Young Adult


  • Aza Compounds
  • Fluoroquinolones
  • Quinolines
  • Moxifloxacin