This study aimed to estimate, in a large group of Algerian adolescents, the prevalence of the metabolic syndrome (MetS), using four definitions (by Cook, De Ferranti, Viner and the IDF), and to test the validity of unique thresholds of waist circumference, waist/height ratio and BMI in screening for the MetS regardless of the definition used.

A total of 1100 adolescent students, aged 12–18 y, were randomly selected from schools and classrooms in the city of Constantine; all had anthropometric measurements taken and 989 had blood tests.

Prevalences of the MetS were: 2.6% for boys and 0.6% for girls by the Cook definition; 4.0% for boys and 2.0% for girls by the De Ferranti definition; 0.7% for boys and 0% for girls by the Viner definition; and 1.3% for boys and 0.5% for girls by the 2007 IDF definition. Prevalences ranged from 3.7% to 13.0% in obese adolescents. Unique thresholds, independent of gender, age and height, of 80cm for waist circumference, 0.50 for waist/height ratio and 25kg/m² for BMI had sensitivities of 72–100%, 67–100% and 72–100%, respectively, and specificities of 74–78%, 74–86% and 74–78%, respectively, depending on the MetS definition used.

The MetS is present in Algerian adolescents and the prevalence is especially high in obese young people. Our thresholds for waist circumference, waist/height ratio and BMI for screening for the MetS should now be tested in other adolescent populations.