Urinary pentosidine does not predict cartilage loss among subjects with symptomatic knee OA: the BOKS Study

Abstract

Age-related changes in articular cartilage are likely to play a role in the etiology of osteoarthritis (OA). One of the major changes in the extracellular matrix of cartilage is the age-related accumulation of advanced glycation end products (AGEs). Pentosidine, an AGE crosslink, is one of the few characterized AGEs and is considered an adequate marker for the many AGEs that are formed in vivo. We used data from a longitudinal observation study to determine if urinary pentosidine could serve as a marker to predict cartilage loss. We conducted a prospective analysis of data from the Boston Osteoarthritis of the Knee Study (BOKS); a completed natural history study of knee OA. All subjects in the study met American College of Rheumatology (ACR) criteria for knee OA. Knee magnetic resonance (MR) images were scored for cartilage in 14 plates of the knee using the Whole Organ Magnetic Resonance Imaging Score (WORMS) semiquantitative grading scheme. Within the BOKS population, a nested sample of 127 subjects (39% of the whole sample) who had both baseline pentosidine and longitudinal magnetic resonance imaging (MRI) measurements (MRIs performed at baseline and 30 months later) was assessed. Urinary pentosidine was assayed and normalized to creatinine to account for differences in urine concentrations. We analyzed the data using three different methods to assess if baseline measures of pentosidine predicted subsequent cartilage loss on MRI. These were (1) analysis 1: logistic regression with the outcome cartilage loss in any plate; (2) analysis 2: proportional odds model where the outcome was defined as 0=no cartilage loss, 1=cartilage loss in one plate, 2=cartilage loss in two plates, and 3=cartilage loss in at least three plates; and (3) analysis 3: Poisson regression with the outcome the number of plates with cartilage loss. All analyses were adjusted for age, sex and Body Mass Index (BMI). At baseline the mean (standard deviation) age was 67 (9) years and 54% were male. The results for the three analytic steps are as follows: Analysis 1: the odds ratio for cartilage loss is 1.01 (95% confidence interval (CI) 0.93-1.09) with 1 unit increase in pentosidine. Analysis 2: the odds ratio for more cartilage loss is 0.99 (95% CI 0.92-1.06) with 1 unit increase in pentosidine. Analysis 3: the relative number of plates with cartilage loss decreased was 1.00 (95% CI 0.95-1.03) with a 1 unit increase in pentosidine. Urinary pentosidine does not predict knee cartilage loss. Previous studies have suggested that local content within cartilage of AGEs is elevated in persons at high risk for progression. Our data suggest that these changes are not measurable systemically. Alternatively, urinary pentosidine levels reflect cartilage degradation in all joints (thus whole body cartilage breakdown) and may therefore not relate to OA severity in a single knee joint.