Joint space width (JSW in mm)‡ (mean±SD) | |||||
---|---|---|---|---|---|
Genotype or haplotype | Frequency of indicated genotype or haplotype* | Kellgren–Lawrence score >2 (odds ratio (CI))† (N=130) | Test genotype | Reference genotype | p Value§ |
IL1RN rs419598 CC/TC | 0.35 | 0.22 (0.091 to 0.508); p=0.0005 | 3.83±1.91 (n=89) | 3.19±1.88 (n=162) | 0.036 |
IL1RN rs315952 TT | 0.53 | 0.44 (0.21 to 0.92); p=0.0297 | 3.69±1.80 (n=132) | 3.11±2.01 (n=117) | 0.069 |
IL1RN rs9005 AA/GA | 0.45 | 0.15 (0.065 to 0.349); p<0.0001 | 3.86±1.71 (n=114) | 3.04±2.01 (n=135) | 0.0063 |
IL1RN rs419598/rs315952/rs9005 Carriage of Haplotype C,T,A | 0.32 | 0.14 (0.053 to 0.368); p< 0.0001 | 3.99±1.77 (n=80) | 3.14±1.93 (n=169) | 0.0008 |
↵* Frequency within the total patients (n=130 with complete genetic data) with knee osteoarthritis (OA) from two separate populations (NYUHJD; Duke POP Study).
↵† Odds ratio for severe OA, as measured by Kellgren–Lawrence (KL) score, comparing the indicated genotype or haplotype with all other genotypes for that single nucleotide polymorphism. Patients were classified by radiographic severity of knee OA by stratifying the KL scores to compare KL=1 or 2 vs KL=3 or 4. Similar significant associations were found when patients were stratified so that KL=1 was compared with KL=2–4 (data not shown).
‡ The smaller JSW in millimetres for each knee was included in the knee-based analysis (n=251 knees with complete data in 126 patients from the NYUHJD and Duke POP populations). Statistical analysis was adjusted for age, gender and body mass index in a mixed model analysis of variance for intercorrelated data to adjust for the subject effect on two knees.
↵§ Wilcoxon test.
NYUHJD, New York University Hospital for Joint Diseases; POP, Prediction of Osteoarthritis Progression study.