Table 1.
Baseline characteristics of the study population before and after propensity score matching for statin use according to baseline BML in MRI
| HN+ participant with no/minimal BMLs | HN+ participant with moderate/severe BMLs | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Characteristic | Before matching | Propensity score-matched | Before matching | Propensity score-matched | ||||||||
| Statin (−) | Statin (+) | SMD | Statin (−) | Statin (+) | SMD | Statin (−) | Statin (+) | SMD | Statin (−) | Statin (+) | SMD | |
| No. of knees | 339 | 175 | 166 | 166 | 386 | 203 | 190 | 190 | ||||
| Variables in the P.S matching (Potential confounders) | ||||||||||||
| Age (year) [mean (SD)] | 60.99 (8.69) | 63.82 (7.87) | 0.342 | 63.66 (8.41) | 63.61 (7.93) | 0.006 | 62.91 (8.65 | 64.91 (7.75) | 0.244 | 63.94 (7.54) | 64.75 (7.80) | 0.098 |
| No. of women [N (%)] | 239 (70.5) | 109 (62.3) | 0.175 | 110 (66.3) | 104 (62.7) | 0.076 | 251 (65.0) | 125 (61.6) | 0.072 | 120 (63.2) | 118(62.1) | 0.022 |
| BMI (kg/m 2 ) [mean (SD)] | 28.10 (4.69) | 28.40 (3.84) | 0.069 | 28.15 (3.94) | 28.31 (3.91) | 0.04 | 28.92 (4.46) | 30.05 (4.19) | 0.261 | 29.50 (4.67) | 29.72 (3.97) | 0.049 |
| Statin CVD indications except dyslipidemia[N(%)]♦ | 76 (22.4) | 52 (29.7) | 0.167 | 48 (28.9) | 50 (30.1) | 0.026 | 119 (30.8)66 (17.1) | 89 (43.8) | 0.272 | 76 (40.0) | 78 (41.1) | 0.021 |
| Alcohol use, ≥ 1/week [N(%)] | 167 (49.3) | 75 (42.9) | 0.129 | 76 (45.8) | 75 (45.2) | 0.012 | 174(45.1) | 83 (40.9) | 0.085 | 78 (41.1) | 81 (42.6) | 0.032 |
| Smoking, current or past [N (%)] | 151 (44.5) | 92 (52.6) | 0.161 | 84 (50.6) | 87 (52.4) | 0.036 | 170 (44.0) | 101 (49.8) | 0.115 | 90 (47.4) | 91 (47.9) | 0.011 |
| PASE score [mean (SD)] | 166.32 (79.40) | 155.45 (70.78) | 0.145 | 156.54 (74.01) | 155.16 (72.30) | 0.019 | 164.61 (82.81) | 150.55 (72.49) | 0.181 | 152.93 (78.22) | 152.84 (73.36) | 0.001 |
| Race, non-white [N (%)]† | 42 (12.4) | 19 (10.9) | 0.048 | 17 (10.2) | 17 (10.2) | 0.001 | 66 (17.1) | 34 (16.7) | 0.009 | 32 (16.8) | 32 (16.8) | 0.001 |
| Other variables not in the P.S matching | ||||||||||||
| History of knee injury, [N(%] | 80 (23.6) | 36 (20.6) | 0.07 | 37 (22.3) | 35 (21.1) | 0.03 | 121 (31.3) | 72 (35.5) | 0.09 | 59 (31.1) | 65 (34.2) | 0.07 |
| Statin type | – | – | – | – | ||||||||
| Atorvastatin | – | 79 (60.3) | – | 73 (59.3) | – | 82 (48.2) | – | 77 (49.0) | ||||
| Fluvastatin | – | 2(1.5) | – | 2(1.6) | – | 3(1.8) | – | 2(1.3) | ||||
| Lovastatin | – | 6 (4.6) | – | 6(4.9) | – | 9(5.3) | – | 7 (4.5) | ||||
| Pravastatin | – | 9 (6.9) | – | 8 (6.5) | – | 14 (8.2) | – | 14 (8.9) | ||||
| Rosuvastatin | – | 6 (4.6) | – | 6 (4.9) | – | 7(4.1) | – | 6(3.8) | ||||
| Simvastatin | – | 29 (22.1) | – | 28 (22.8) | – | 55 (32.4) | – | 51 (32.5) | ||||
| Statin use duration, years, [mean (SD)] | 0.00 (0.00) | 3.92 (2.05) | 2.7 | 0.00 (0.00) | 3.90 (2.06) | 2.68 | 0.00 (0.00) | 4.23 (2.06) | 2.9 | 0.00 (0.00) | 4.18(2.08) | 2.85 |
| Number of affected subregions with BML | 0.07 | 0.05 | 0.2 | 0.32 | ||||||||
| 0 | 206 (60.8) | 101 (57.7) | 95 (57.2) | 94 (56.6) | 75 (19.4) | 39 (19.2) | 37 (19.5) | 37 (19.5) | ||||
| 1 | 83 (24.5) | 44 (25.1) | 45 (27.1) | 43 (25.9) | 24 (6.2) | 18 (8.9) | 7 (3.7) | 17 (8.9) | ||||
| 2 | 50 (14.7) | 30 (17.1) | 26 (15.7) | 29 (17.5) | 51 (13.2) | 34 (16.7) | 22(11.6) | 34 (17.9) | ||||
| 3 | 0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 99 (25.6) | 55 (27.1) | 56 (29.5) | 51 (26.8) | ||||
| 4 | 0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 64 (16.6) | 29 (14.3) | 32 (16.8) | 25 (13.2) | ||||
| ≥5 | 0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 73 (18.9) | 28 (13.8) | 36(18.9) | 26 (13.7) | ||||
| Maximum BML grade in knee | 0.16 | 0.03 | 0.02 | 0.09 | ||||||||
| 0 | 148 (43.7) | 63 (36.0) | 63 (38.0) | 61 (36.7) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) | ||||
| 1 | 191 (56.3) | 112(64.0) | 103 (62.0) | 105 (63.3) | 58 (15.0) | 30 (14.8) | 34 (17.9) | 28 (14.7) | ||||
| 2 | 0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 217 (56.2) | 116(57.1) | 107 (56.3) | 109 (57.4) | ||||
| 3 | 0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 111(28.8) | 57 (28.1) | 49 (25.8) | 53 (27.9) | ||||
| Baseline KL grade | 0.21 | 0.235 | 0.062 | 0.166 | ||||||||
| Grade 0 | 111(32.7) | 49(28.0) | 53 (31.9) | 45 (27.1) | 54 (14.0) | 28 (13.8) | 22(11.6) | 28 (14.7) | ||||
| Grade 1 | 102 (30.1) | 68 (38.9) | 50(30.1) | 64 (38.6) | 92 (23.8) | 50 (24.6) | 43 (22.6) | 48 (25.3) | ||||
| Grade 2 | 94 (27.7) | 39 (22.3) | 49 (29.5) | 38 (22.9) | 137 (35.5) | 76 (37.4) | 65 (34.2) | 67 (35.3) | ||||
| Grade 3 | 32 (9.4) | 19 (10.9) | 14 (8.4) | 19(11.4) | 103 (26.7) | 49(24.1) | 60(31.6) | 47 (24.7) | ||||
Data are presented as numbers of knees. Statin (+) and statin (−) corresponds to statin users and non-users, respectively. Knees with both ≤2 knee subregions with BMLs and maximum BML score ≤ 1 were regarded with no/minimal BML involvement, while knees either having > 2 knee subregions with BMLs or maximum BML score > 1 were considered with moderate/severe BML involvement. BMI body mass index, BML Bone marrow lesion, HN Heberden’s node, PASE physical activity scale for the elderly, SMD standardized mean difference, SD standard deviation, N number of knees
A significant difference for SMD was defined as ≥0.1
Statin CVD indications except dyslipidemia were indicated as the presence of either history of coronary artery disease, cerebrovascular accident, diabetes (any stage of diabetes vs. no medical history of diabetes), or hypertension in clinical examination systolic blood pressure ≥140 mm Hg or diastolic blood pressure ≥ 90 mm Hg at OAI visit clinical examination)
Race of participants was categorized as white and non-white considering the small number of participants in each non-white race group