Table 3.
Multivariate logistic regression analyses of fracture according to tertile categories of the S1P concentration and TBS
| Adjusted variables | Tertiles of FRAX probabilitya) | Tertiles of TBS and FRAX probabilitya) | ||
|---|---|---|---|---|
|
|
|
|||
| OR (95% CI) | P-value | OR (95% CI) | P-value | |
| S1P concentration | ||||
| T1 (S1P <2.24 μmol/L) | Ref | Ref | ||
| T2 (2.24 μmol/L ≤S1P <3.30 μmol/L) | 3.45 (1.47–8.13) | 0.005 | 3.31 (1.38–7.90) | 0.007 |
| T3 (S1P ≥3.30 μmol/L) | 3.72 (1.56–8.84) | 0.003 | 3.13 (1.28–7.66) | 0.013 |
|
| ||||
| TBS | ||||
| T3 (TBS ≥1.439) | NA | Ref | ||
| T2 (1.386 ≤TBS <1.439) | NA | 1.57 (0.65–3.79) | 0.312 | |
| T1 (TBS <1.386) | NA | 3.73 (1.68–8.26) | <0.001 | |
P<0.05 is statically significant. Bold values indicate statistical significance.
a)
The FRAX probability was calculated using CRFs and FN-BMD by the FRAX. CRFs were sex, age, body mass index, current smoking, alcohol intake (≥3 U/day), other causes of secondary osteoporosis, and parental history of hip fracture.
S1P, sphingosine 1-phosphate; TBS, trabecular bone score; FRAX, fracture risk assessment tool; OR, odds ratio; CI, confidence interval; CRFs, clinical risk factors; FN-BMD, femur neck-bone mineral density; Ref, reference; NA, not available.