Inter-rater reliability of quantifying mechanical and thermal sensitivity in a musculoskeletal trauma population

Middlebrook, Nicola ORCID: https://orcid.org/0000-0003-2154-5723, Heneghan, Nicola, Rushton, Alison and Falla, Deborah (2020) Inter-rater reliability of quantifying mechanical and thermal sensitivity in a musculoskeletal trauma population. In: Physiotherapy UK.

Preview

Published Version Available under License In Copyright. Download (223kB) | Preview

Abstract

Purpose: In the United Kingdom, approximately 20,000 people experience a musculoskeletal injury annually following a traumatic incident. Sub-optimal recovery is common, with persistent pain commonly reported. Quantitative sensory testing (QST), a psychophysical method to quantify pain, is useful in predicting outcome in populations such as whiplash and osteoarthritis. Early assessment of sensory changes could be beneficial in predicting outcome within a musculoskeletal trauma population, however no study has evaluated the reliability of QST measures in a musculoskeletal trauma population. This study aims to establish the inter-rater reliability of multiple forms of QST including temporal summation, thermal and pressure pain thresholds in an acute musculoskeletal trauma population. Methods: People with acute musculoskeletal trauma were recruited from a major trauma centre hospital in the United Kingdom. Two QST sessions (one per rater) were conducted on the same day with a minimum of two hours between sessions. Measurements were taken at a local site and remote site to the primary injury. The site, rater and modality was randomised. Statistical analyses included intra-class correlation coefficients (ICC 3,2), 95% confidence intervals (CI), standard error of measurement (SEM) and Bland Altman for limits of agreement between raters. Results: Twenty-five participants (70% men) with a mean age (SD) of 44.8 (19.32) years were recruited. Five participants were excluded from analyses (2 were discharged before the full data set could be collected, 1 declined in mental capacity between sessions, and 2 were unable to press the button when pain threshold had been reached). A significant number of participants reached the safety limit of the thermal analyzer for cold pain thresholds (CPT), therefore ICC's and SEM were not calculated for that measure due to low statistical power. Heat Pain Threshold (HPT) Inter-rater reliability was rated good for both testing sites: ICC (CI) local = 0.86 (0.65-0.95), remote = 0.81 (0.49-0.93). SEM was no more than 10C for both raters. Pressure Pain Threshold (PPT) Inter-rater reliability was rated good for both testing sites: ICC (CI) local site = 0.78 (0.43-0.91), remote site = 0.76 (0.43-0.91). SEM was 6-7 Newtons for both raters. Temporal Summation (TS) Inter-rater reliability was rated as moderate: ICC (CI) local = 0.57 (-0.08-0.84), remote = 0.62 (-0.06-0.86). Conclusion(s): Good inter-rater reliability was achieved for HPT and PPT whereas moderate reliability was achieved for TS. Some participants were unable to tolerate the testing procedure for TS, therefore further development of the TS testing protocol is warranted for this population. Due to equipment limitations, statistical analysis of CPT data was not possible. Implications: Measurements of HPT, PPT and TS can be performed in a reliable manner in an acute musculoskeletal trauma population. This study forms the foundation for future work to evaluate changes in sensory function over time following musculoskeletal trauma and the value of QST in predicting outcome. Funding acknowledgements: NIHR - Surgical Reconstruction and Microbiology Research Centre (SRMRC)