How stable are quantitative sensory testing measurements over time? Report on 10-week reliability and agreement of results in healthy volunteers

Abstract

Background

Quantitative sensory testing (QST) is a diagnostic tool for the assessment of the somatosensory system. To establish QST as an outcome measure for clinical trials, the question of how similar the measurements are over time is crucial. Therefore, long-term reliability and limits of agreement of the standardized QST protocol of the German Research Network on Neuropathic Pain were tested.

Methods

QST on the lower back and hand dorsum (dominant hand) were assessed twice in 22 healthy volunteers (10 males and 12 females; mean age: 46.6±13.0 years), with sessions separated by 10.0±2.9 weeks. All measurements were performed by one investigator. To investigate long-term reliability and agreement of QST, differences between the two measurements, correlation coefficients, intraclass correlation coefficients (ICCs), Bland–Altman plots (limits of agreement), and standard error of measurement were used.

Results

Most parameters of the QST were reliable over 10 weeks in healthy volunteers: Almost-perfect ICCs were observed for heat pain threshold (hand) and mechanical pain sensitivity (back). Substantial ICCs were observed for heat pain threshold (back), pressure pain threshold (back), mechanical pain sensitivity (hand), and vibration detection threshold (back and hand). Some QST parameters, such as cold detection threshold, exhibited low ICCs, but also very low variability. Generally, QST measures exhibited narrow limits of agreement in the Bland–Altman plots.

Conclusion

The standardized QST protocol of the German Research Network on Neuropathic Pain is feasible to be used in treatment trials. Moreover, defining a statistically meaningful change is possible, which is a prerequisite for the use of QST in clinical trials as well as in long-term investigations of disease progression.

Keywords:

• QST
• healthy volunteers
• test–retest reliability
• intraclass correlations
• Bland–Altman plot
• limits of agreement
• standard error of measurement
• minimum detectable difference

Acknowledgments

Jan Vollert is acknowledged for his help in statistical analysis of the QST data. We thank all the healthy volunteers for their participation in our study.

Disclosure

The authors report no conflicts of interest in this work.