How to make calibration less painful – a proposition of an automatic, reliable and time-efficient procedure

Abstract

Background: The description of the calibration in pain studies is scarce. This study provides an in-depth analysis and proposes an automatic pain-threshold calibration that quantifies its correctness, minimizes its duration and invasiveness, and considers validation of linearity between stimuli intensity and Numerical Rating Scale (NRS) values. Methods: 70 healthy volunteers (37 males) underwent a staircase calibration method adjusted for A-delta specific intraepidermal electrical stimulation. The sensation (t) and pain thresholds (T) intensities were calculated using: the traditional threshold method (TM), a linear regression model (LRM), and two variants of a proposed truncated LRM (tLRM). The tLRM analyses data up to an optimal point, chosen according to the maximum goodness-of-fit value (R2). The optimal value of R2 and the convergence based on the regression line gradient formed the base for rejection criteria of failed procedure. Then, we compared the different methods for thresholds assessment as well as mid-painful stimulation (MP), equivalent to 1.5*T or NRS equal to 8, using repeated measures ANOVA. Results: Four participants were rejected according to our criteria. tLRM methods exhibit higher R2 than LRM with 36% less stimuli application. Compared to other methods, tLRM result in significantly higher t and lower T and MP intensities. Last, significantly lower t and T were found for female compared to men, regardless of the calibration method. Conclusions: The proposed method of calibration reduces the number of stimuli delivered and can be easily adapted to other simulation modalities. However, future investigation is recommended using adaptive LRM fitting and verification of reliability.