**2.3 Biochemical analysis**

The methods of dosage and the normal values of the different biological parameters are shown in table 2.

Cardiovascular Risk in Tunisian Patients with Bipolar I Disorder 325

Statistical analyses were performed using SPSS 17.0 (SPSS, Chicago, IL, USA). Quantitative variables were presented as mean ± SD and comparisons were performed using the Student's t test. Qualitative variable comparisons were performed using the Chi-squared test (χ2) and Fisher's exact test (when n < 5). Comparisons between patients and controls in biological parameters were performed using analysis of variance (ANOVA) after adjustment for potential confounder factors. Odd ratios (ORs) and their 95% condence interval (CI) were calculated and adjusted for potential confounder factors by binary logistic regression. The statistical significance level was set at *p* < 0.05. All variables with a *p* value < 0.25 between the two studied groups (patients and controls) were considered as potential

Table 3 shows the comparisons of biological variables between bipolar I patients and controls.

*Triglycerides* (mmol/L) 1.95 ± 1.55 1.23 ± 0.81 **< 0.001 < 0.001**  *Cholesterol* (mmol/L) 4.42 ± 0.99 4.37 ± 1.26 0.707 0.856 *c-LDL* (mmol/L) 2.14 ± 1.10 2.37 ± 1.38 0.118 0.047

Men (85/102) 1.04 ± 0.37 0.98 ± 0.29 0.192 0.017 Women (45/73) 1.17 ± 0.36 1.21 ± 0.48 0.542 0.702 *ApoA1* (g/L) 1.20 ± 0.23 1.40 ± 0.67 **< 0.001 0.028**  *ApoB* (g/L) 0.82 ± 0.28 0.83 ± 0.24 0.784 0.777 *ApoB/Apo A1* 0.71 ± 0.26 0.65 ± 0.25 0.086 0.314 *Lp(a)* (mg/L) 243 ± 223 87 ± 129 **< 0.001 < 0.001**  *Homocysteine* (µmol/L) 15.8 ± 8.9 11.5 ± 5.0 **< 0.001 < 0.001**  *Vitamin B12* (ng/L) 356 ± 198 360 ± 190 0.837 0.819 *Folate* (µg/L) 3.3 ± 0.9 5.1 ± 2.8 **< 0.001 <0.001** 

Men (85/102) 311 ± 99 250 ± 107 **0.001 0.005**  Women (45/73) 246 ± 97 197 ± 73 **0.012** 0.408 \* Lipid profile parameters, folatemia, vitamin B12 and uric acid were adjusted for gender, BMI, cigarette

\*Hcys was adjusted for gender, BMI, cigarette smoking, alcoholic beverages, diabetes, hypertension,

Compared with controls, patients had significantly higher triglycerides (1.95 ± 1.55 *Vs* 1.23 ± 0.81 mmol/L; p < 0.001), Lp(a) (243 ± 223 *Vs* 87 ± 129 mg/L; p < 0.001), homocysteine levels (15.8 ± 8.9 *Vs* 11.5 ± 5.0 µmol/L; p < 0.001) and uric acid (311 ± 99 *Vs* 250 ± 107 µmol/L; p = 0.001 in men; 246 ± 97 *Vs* 197 ± 73 µmol/L; p *=* 0.012 in women), and significantly lower ApoA1 (1.20 ± 0.23 *Vs* 1.40 ± 0.67 g/L; p < 0.001) and folate (3.3 ± 0.9 *Vs* 5.1 ± 2.8 µg/L; p < 0.001) levels. After adjustment for potential confounder factors, these differences remained significant for all of these parameters except for uric acid which is remained

Table 3. Comparisons of biological variables between bipolar I patients and controls.

**Controls** 

**(n = 175) p p\***

**(n = 130)** 

**2.7 Statistical analysis** 

**3. Results** 

*c-HDL* (mmol/L)

*Uric acid* (µmol/L*)*

folatemia and vitamin B12

confounder factors for this analysis.

**Biological variables Patients** 

smoking, alcoholic beverages, diabetes and hypertension

significantly higher only for men (table 3).


Table 2. Methods of dosage of the studied parameters.
