Background The association of structural changes of osteoarthritis (OA) on imaging and clinical symptoms has seen contradictory results in previous studies.
Objectives We performed a comparison of ultrasound features of knee OA with clinical symptoms among members of the Newcastle Thousand Families birth cohort.
Methods Participants from the cohort aged 63 (born in May-June 1947), had both knees scanned by a trained musculoskeletal sonographer. Ultrasound protocols were derived from EULAR guidelines. Knee osteophytes (yes/no), minimum femoral cartilage thickness in right knee (cm) and effusion >4 mm (yes/no) were the pathologies identified. These data were analysed in relation to pain, stiffness and dysfunction in the lower limbs as reported by participants using the WOMAC questionnaire. Each of the three WOMAC subscales was subdivided in to four categories of severity. Logistic and linear regression was used to calculate the association of features of knee OA with clinical symptoms. Adjustment for potential confounders such as BMI, sex and presence of knee effusion were also performed.
Results 311 participants were scanned; 55% women, mean BMI was 27.9. Prevalence of knee osteophytes was 30%, mean right knee minimum cartilage thickness was 1.47 cm; prevalence of knee effusions was 22%. Those in the highest category of pain had an OR of 4.42 for osteophytes (95% CI 2.17, 8.98) when compared to those without knee pain. Similarly, those with severe stiffness had an OR of 4.21 (95% CI 2.01, 8.83) and those with physical dysfunction had an OR of 4.15 (95% CI 1.96, 8.80) for knee osteophytes when compared to those with no symptoms. These estimates were reduced in magnitude but remained statistically significant after adjustment for BMI and sex. Minimum cartilage thickness was associated with pain (adjusted co-efficient -0.11; 95% CI -0.20, -0.01) and physical dysfunction (adjusted co-efficient -0.13; 95% CI -0.24, -0.02) but not stiffness; when comparing those in the most severe symptom category with those without symptoms. Knee effusion had no association with any of the three subscales of the WOMAC questionnaire.
Conclusions This is the first study to compare ultrasound features of OA with clinical symptoms in a population based cohort. The presence of knee osteophytes had a positive association with pain, stiffness and physical dysfunction in the lower limbs. Femoral cartilage thickness in the knee had an inverse association with pain and physical dysfunction. However, there was no association of knee effusion with pain, stiffness or dysfunction. The associations of ultrasound detected osteophytes and cartilage thickness with clinical symptoms improve our understanding of the relationship between symptoms and structural changes in knee OA. This also demonstrates the potential utility of ultrasound in prospective population based epidemiological studies of OA as well as in clinical practice.
Disclosure of Interest None Declared