Carbohydrate status in patients with phenylketonuria

Table 2 Correlation (Rho) of carbohydrate metabolism markers with age, Phe tolerance, Phe anual median levels, body mass index, waist circumference and caloric intake in the form of carbohydrates in hyperphenylalaninemia patients

	AGE (years)	Phe tolerance (mg)	Phe median levels (adequate to age)	BMI (Kg/m²)	WC (cm)	Caloric intake in the form of carbohydrates
Basal Insulin (mIU/L)	0.463 p = 8.5e⁻⁵	−0.164 n.s.	0.38 p = 0.003	0.728 p = 1.5e⁻¹²	0.557 p = 1.4e⁻⁵	0.468 p = 0.006
HOMA IR	0.461 p = 9.4e⁻⁵	− 0.201 n.s.	0.353 p = 0.008	0.69 p = 8.1e⁻¹¹	0.508 p = 1.5e⁻⁴	0.423 p = 0.02
QUICKI	−0.461 p = 9.4e⁻⁵	0.201 n.s.	−0.353 p = 0.008	− 0.69 p = 8.1e⁻¹¹	−0.508 p = 1.5e⁻⁴	−0.423 p = 0.02
Basal Fructosamine (μM)	0.364 p = 0.006	−0.041 n.s	0.152 n.s.	0.177 n.s.	0.124 n.s.	−0.011 n.s.
Basal Grelin (pg/mL)	−0.231 n.s.	−0.373 n.s.	− 0.209 n.s.	−0.464 p = 0.034	− 0.348 n.s.	−0.245 n.s.
IGF1 (mg/mL)	0.128 n.s	0.0899 n.s.	0.087 n.s.	0.116 n.s.	0.059 n.s.	0.114 n.s.

BMI body mass index, WC waist circumference, HOMA IR homeostasis model assessment insulin resistance, QUICKI quantitative insulin sensitivity check index
Using a stepwise regression method based on AIC (Aikake Information Criterion), we found that the variable that most influenced insulin was BMI (linear regression model: adjusted R² = 0.356; model, p- = 1.591e-09; BMI, p = 1.59e-09); the variable that most influenced HOMA-IR was BMI (linear regression model: adjusted R² = 0.3262; model, p = 1.036e-08; BMI, p = 1.04e-08); and the variables that most influenced BMI were phenylalanine tolerance and age (logistic regression model: null deviance = 110.674 on 82 degrees of fredom, residual deviance = 88.154 on 80 degrees of freedom; Phe tolerance, p = 0.00592; age, p = 0.00642)

ISSN: 1750-1172