Improving the Congruence between Knee Radiographs and Clinical/Experimental Pain: The Importance of Dispositional Traits

Abstract

Knee osteoarthritis (OA) is a common progressive joint disease with increasing prevalence. There are inconsistent findings on the concordance between radiograph derived Kellgren-Lawrence (KL) scores and pain measures. We have previously reported that dispositional traits predict sensory sensitivity and psychosocial functioning. This study aims to determine whether consideration of a measure of dispositional trait, affect balance style (ABS), improves the congruence between KL scores and clinical and experimental pain. Adults between 45 and 85 years with knee pain with or at risk for osteoarthritis and who self-identified as non-Hispanic Black (NHB) or non-Hispanic white (NHW) were recruited and enrolled in the study. Participants completed a health assessment and quantitative sensory testing (QST). Data collected included sociodemographics, knee radiographs, clinical pain and function, and affect. Punctate temporal summation and pressure pain threshold were assessed on the most painful knee. ABS was determined and protective traits (health/low) and vulnerable traits (reactive/depressive) identified. Unadjusted and adjusted regression analyses with and without dispositional trait groups were completed using SAS (9.4). A total of 218 individuals, 57.8 ± 8.0 years of age, 63.6% women, and 48.2% NHB were included. Participants with vulnerable dispositional traits reported higher clinical pain (p <0.0001) and lower experimental pain thresholds (p<0.05) than those with protective traits regardless of KL score. Dispositional traits and KL scores were predictive of all clinical pain measures (p<0.0001) and knee punctate temporal summation (p<0.0001) in regression models adjusted by age, sex, and ethnicity/race. Improvements in R2 were noted across all models with inclusion of dispositional traits. Previous findings between knee radiographs and clinical and experimental pain are mixed; consideration of dispositional traits may clarify incongruences. Further investigation is needed to determine if dispositional traits may serve as a key factor to elucidate the convoluted relationships between pathology and clinical pain experience. Grant support from NIH/NIA Grants R01AG054370 (KTS); R37AG033906 (RBF) and UF CTSA Grant UL1TR001427 and UAB CTSA Grant UL1TR001417 from the NIH Center for Advancing Translational Sciences.