Table 2.

Predictive factors for developing persistent transaminitis during MTX treatment for RA

Variables at baseline	Unadjusted HR (95% CI)	P-value	Adjusted HR (95% CI)	P-value
Age per 1 year more	1.00 (0.98, 1.02)	0.98	–	–
Male vs female	0.68 (0.32, 1.44)	0.31	–	–
Weekly dose of MTX per 1.0 mg more	1.11 (0.96, 1.29)	0.16	–	–
RA duration per 1 month more	1.00 (1.00, 1.00)	0.38	–	–
Anti-CCP positive	2.04 (0.50, 8.39)	0.32	–	–
RF positive	0.80 (0.38, 1.67)	0.55	–	–
Steinbrocker’s stages III/IV vs I/II	1.39 (0.72, 2.68)	0.33	–	–
CRP ≥1.5 mg/dl	1.33 (0.73, 2.42)	0.36	–	–
ESR ≥28 mm/h	1.96 (1.00, 3.89)	0.051	–	–
L/S ratio on high-resolution CT
≥1.3 (no fat deposition)	Reference	–	Reference	–
≥1.1 and <1.3 (mild fat deposition)	1.88 (0.95, 3.74)	0.072	1.85 (0.94, 3.67)	0.070
<1.1 (moderate to severe fat deposition)	10.76 (4.75, 24.37)	<0.001	7.69 (3.10, 19.10)	<0.001
Obesity (BMI ≥25 kg/m2)	3.50 (1.94, 6.33)	<0.001	2.68 (1.37, 5.25)	0.004
Hypertension	1.49 (0.83, 2.67)	0.18	–	–
Type 2 diabetes	2.25 (0.87, 5.81)	0.10	2.76 (1.02, 7.48)	0.046
Chronic kidney disease	0.55 (0.20, 1.51)	0.24	–	–
Current/ex-smokers	0.65 (0.33, 1.28)	0.21	–	–
NSAID use	1.51 (0.84, 2.72)	0.17	–	–
Prednisolone use	2.01 (1.11, 3.64)	0.020	1.96 (1.09, 3.51)	0.030
bDMARD use	0.38 (0.05, 2.86)	0.34	–	–

Univariate and multivariate Fine–Gray competing risks regression analyses were conducted to evaluate baseline patient-specific factors that predict the development of persistent transaminitis during MTX treatment. All variables with P-values <0.20 in the univariate Fine–Gray models were introduced into multivariate analysis. Variables that remained in the final multivariate model are shown as significant predictive factors for persistent transaminitis.

bDMARDs: biological DMARDs; HR: hazard ratio; L/S ratio: liver-to-spleen attenuation ratio.