Visceral fat area and remnant cholesterol in type 2 diabetes: Nonlinear associations and thresholds

Table 3 Threshold effects of visceral fat area on residual cholesterol and the residual cholesterol to C-reactive protein ratio based on two-piecewise linear regression models

	Adjusted β (95%CI)	P value
RC
VFA (model I)	0.09 (0.01-0.17)	0.0330
VFA (model II)
Inflection point	126
VFA < 126	0.24 (0.12-0.35)	< 0.0001
VFA ≥ 126	-0.22 (-0.42 to -0.02)	0.0279
Difference (segment 2-1)	-0.46 (-0.73 to -0.19)	0.0007
Log likelihood ratio		< 0.001
RCII
VFA (model I)	0.10 (-0.01 to 0.21)	0.0784
VFA (model II)
Inflection point	118
VFA < 118	0.22 (0.05-0.39)	0.0105
VFA ≥ 118	-0.10 (-0.33 to 0.14)	0.4102
Difference (segment 2-1)	-0.32 (-0.66 to 0.02)	0.0623
Log likelihood ratio		0.060

Data are regression coefficients (β) with 95%CIs and P values from two-piecewise linear regression models with a single inflection point for visceral fat area (VFA). The inflection points were 126 cm² for residual cholesterol (RC) and 118 cm² for RC-inflammation index, estimated using a likelihood-based approach. The rows labelled “< inflection point” and “≥ inflection point” represent the slopes of VFA below and above the inflection point, respectively; “β difference” denotes the difference between these two slopes. Likelihood ratio test P indicates the P value from the likelihood ratio test comparing the two-piecewise model with a simple linear model. All models were adjusted for age, sex, current smoking, current drinking, glycated haemoglobin, free triiodothyronine, estimated glomerular filtration rate and alanine aminotransferase. RC: Residual cholesterol; RCII: Residual cholesterol-inflammation index; VFA: Visceral fat area.