Do scores on WOMAC pain and function subscales vary with different definitions of knee osteoarthritis?

Abstract

Listen

Translate article

Purpose: The MRI definition for knee osteoarthritis, developed in 2011, has not yet been validated in other populations. In previous work, we showed that if this MRI definition of tibiofemoral (TF) OA (TFOAMRI) is applied, more cases of knee OA are detected than with the radiographic Kellgren and Lawrence grading (K&L). With a better content validity and at least equal construct validity, we concluded that the TFOAMRI is more sensitive in detecting structural knee OA. However, it is unknown whether women defined with TFOAMRI differ in pain and disability from those who are not, or those who have a radiographic K&L grade ≥2. Furthermore, with the available MRI definitions, a distinction between patellofemoral (PF) OA and TFOA can be made, and the debated contribution to pain and disability by PFOA can be assessed. Therefore the aim of the present study was to investigate if women with knee OA defined with PF- or TFOAMRI report different pain and function scores measured with The Western Ontario and McMaster Universities Arthritis Index (WOMAC), or report different scores than women with knee OA defined by K&L-grading. Methods: Of 891 females (aged 45-60) from a random subpopulation of the Rotterdam Study, radiographs and MRI of both knees were assessed for knee OA; radiographs with the K&L-grading (K&L≥2 was defined as OA) and MRIs with a comprehensive semi-quantitative scoring system. Based on these scored features we applied the proposed MRI definition. We distinguished a PFOAMRI-definition from a TFOAMRI-definition. All women filled in the WOMAC questionnaire. With multivariable regression analysis we tested if the definitions (K&L≥2, PFOAMRI or TFOAMRI) reported different WOMAC pain and function scores independently from each other. Analyses were adjusted for BMI, age and bilaterality (if women had knee OA in one or both knees). Results: Data of 871 women were analyzed. Of 20 women data was missing due to insufficient quality of images (radiographs or MRIs). Table 1 shows the mean and the standard deviation (sd) of age, BMI, WOMAC pain and function scores per definition. 21 women met the K&L≥2 and TFOAMRI definition in one or both knees; 3 women met the K&L≥2 and PFOAMRI definitions; 35 women met the TF- and PFOAMRI definitions; 17 women met all three definitions of knee OA in one or both knees. All three definition contributed significantly (p < 0.001) to higher WOMAC pain scores (K&L≥2: β = 1.61; (95% confidence interval (95% CI) 0.79–2.44), PFOAMRI: β = 1.32 (95% CI: 0.69–1.95) or TFOAMRI: β = 1.15 (95% CI: 0.52–1.77)) and to WOMAC function score (K&L≥2: β = 5.21 (95% CI:2.56–7.86), p < 0.001; TFOAMRI (β = 2.75 (95% CI:0.74–4.75), p = 0.007; PFOAMRI: β = 4.06 (2.02–6.10), p < 0.001). Conclusions: The MRI definitions show differences in WOMAC pain and function scores between women with and without knee OA. Those women with all definitions positive had the highest pain and disability scores, and those with alone K&L≥2 the lowest. The TF- and PFOAMRI definitions, but also the K&L-definition, all contributed significantly and independently from each other, to the higher pain and disability scores.Table 1Mean (sd) of age, BMI, WOMAC pain (0-20) and function subscale (0-68)Women met the definition (n)Age mean (sd)BMI mean (sd)Uni-/bilateral n/nWOMAC pain mean (sd)WOMAC function mean (sd)No OA67654.6 (3.8)26.3 (4.3)-1.0 (2.4)3.0 (7.5)K&L≥26156.2 (3.3)30.1 (6.3)25/364.2 (4.7)12.9 (16.1)- Only K&L≥21756.1 (3.8)27.1 (3.3)11/61.8 (2.3)5.7 (8.2)- K&L≥2 + TFOAMRI2155.7 (3.4)30.1 (7.2)7/143.8 (4.9)11.8 (17.6)- K&L≥2 + PFOAMRI357.5 (1.3)34.8 (7.7)0/35.7 (6.4)21.3 (25.3)TFOAMRI12556.6 (3.3)29.4 (6.2)69/563.5 (4.7)10.3 (14.7)- Only TFOAMRI5155.9 (3.6)27.7 (6.2)42/92.2 (3.4)7.0 (11.2)PFOAMRI10656.9 (2.9)30.1 (4.1)59/473.5 (4.6)10.6 (14.4)- Only PFOAMRI5156.1 (2.9)29.2 (4.1)40/112.3 (3.4)8.2 (11.7)- TF- + PFOAMRI3557.7 (2.8)29. (5.8)15/203.5 (5.1)8.9 (14.0)K&L≥21757.3 (2.6)32.4 (6.6)4/136.7 (5.2)19.2 (17.6) Open table in a new tab