Development and Validation of a Time-Varying Correction Factor for QT Interval Assessment in Drug-Resistant Tuberculosis Patients

Academic Background and Research Motivation

During the treatment of active tuberculosis (Tuberculosis, TB), patients often experience tachycardia (Tachycardia), which typically diminishes as they recover. Elevated heart rate (Heart Rate, HR) can affect the measurement results of the QT interval (QT Interval). Using standard correction factors (Correction Factors, CFs), such as Fridericia’s formula (QTcf), may lead to suboptimal correction, thereby impacting the accurate assessment of QT prolongation (QT Prolongation). Olivero et al. proposed a correction factor suitable for pre-treatment TB patients, but these constant correction factors do not address the changes in the QT-HR relationship during treatment.

Therefore, this study aims to develop a time-varying correction factor (Time-Varying Correction Factor, TCF) that captures the natural changes in the QT-HR correlation during TB treatment, improving the accuracy of QT interval assessment and providing a more reliable basis for drug effect analysis in clinical trials.

Paper Source

This paper was authored by Thanakorn Vongjarudech, Anne-Gaëlle Dosne, Bart Remmerie, and others from institutions including Uppsala University, Janssen R&D, Vanderbilt University Medical Center, etc. It was published in the International Journal of Antimicrobial Agents, received on May 17, 2024, and accepted on January 31, 2025.

Research Workflow and Key Results

Research Workflow

Heart Rate Model Construction

To describe the changes in heart rate over time, researchers first built a heart rate model. This model included several components: 1. Baseline Heart Rate: Reflects the heart rate of untreated TB patients. 2. Time-on-Treatment Effect: Assumes that heart rate gradually approaches a lower asymptote, reflecting the recovery process. 3. Circadian Rhythm: Models 24-hour and 12-hour cycles using a cosine function. 4. Drug Effects: Investigates the impact of bedaquiline metabolite M2 on heart rate using an Emax model.

Time-Varying Correction Factor Development

Based on the heart rate model, researchers developed a time-varying correction factor (QTctbt). This factor starts from Olliaro’s pre-treatment correction factor (0.4081) and gradually transitions to the standard correction factor (0.33), controlled by the heart rate change rate parameter (tprog). Specifically, the formula for the correction factor changing over time is:

[ cf(t) = 0.4081 - (0.4081 - 0.33) \times \left(1 - e^{-\frac{t}{tprog}}\right) ]

where ( t ) is the time after the start of treatment (weeks), and ( tprog ) is the half-life of heart rate recovery.

Key Research Results

Performance of Heart Rate Model

The final heart rate model successfully described the changes in heart rate over time. The typical baseline heart rate was 78.2 bpm, recovering to 73.1 bpm with a half-life of 7.74 weeks. Additionally, the model captured 24-hour and 12-hour circadian rhythm changes and the mild inhibitory effect of M2 on heart rate (maximum inhibition of 17.9%).

Evaluation of Correction Factor Performance

Through linear regression analysis, researchers compared the performance of different correction factors (QTcb, QTcf, QTco, and QTctbt) at various time points. Results showed that the time-varying correction factor (QTctbt) effectively maintained QT independence from heart rate throughout the entire treatment period, with slopes close to zero, indicating good correction. In contrast, other correction factors exhibited inconsistent performance, leading to overcorrection or undercorrection at different time points.

Conclusion and Significance

The time-varying correction factor developed in this study significantly improves the accuracy of QT interval assessment in TB patients, reducing the risk of misdiagnosis due to heart rate changes and avoiding unnecessary treatment discontinuation. This method enhances the reliability of drug safety evaluations and supports safer, more informed treatment decisions, especially when dealing with drugs that pose a risk of QT prolongation.

Research Highlights

1. Addressing Limitations of Existing Correction Factors: Traditional correction factors may perform poorly under elevated heart rates, while the time-varying correction factor dynamically adapts to heart rate changes, providing more accurate QT interval assessments.
2. Innovative Methodology: This is the first introduction of a time-varying correction factor, integrating heart rate models and drug effects to ensure scientific and practical validity.
3. Broad Applicability: This method is applicable not only in clinical trials but also in routine clinical practice, offering healthcare providers a practical tool to better assess and manage QT intervals in TB patients.

Additional Valuable Information

Researchers also conducted sensitivity analyses to verify the stability of the time-varying correction factor across different heart rate recovery rates. Additionally, they provided an online calculator (https://thanakornv.github.io/qtctbt/) for easy application of the method. Future research will further validate this method’s applicability in more diverse patient populations and treatment regimens.

---

This study provides a novel and reliable method for assessing QT intervals in TB patients, promising to improve drug safety evaluations and clinical treatment decisions, with significant scientific and practical value.