Martine Paquette1, Diane Brisson2, Robert Dufour3, Étienne Khoury2, Daniel Gaudet2, Alexis Baass4. 1. Nutrition, Metabolism and Atherosclerosis Clinic, Montreal Clinical Research Institute, Montreal, Québec, Canada. 2. Department of Medicine, Université de Montréal and ECOGENE-21 Clinical Research Center, Montreal, Québec, Canada. 3. Nutrition, Metabolism and Atherosclerosis Clinic, Montreal Clinical Research Institute, Montreal, Québec, Canada; Department of Nutrition, Université de Montréal, Montreal, Québec, Canada. 4. Nutrition, Metabolism and Atherosclerosis Clinic, Montreal Clinical Research Institute, Montreal, Québec, Canada; Division of Experimental Medicine, Department of Medicine, McGill University, Montreal, Québec, Canada; Division of Medical Biochemistry, Department of Medicine, McGill University, Montreal, Québec, Canada. Electronic address: alexis.baass@ircm.qc.ca.
Abstract
BACKGROUND: Familial hypercholesterolemia (FH) is a disease characterized by increased low-density lipoprotein cholesterol and premature cardiovascular disease (CVD) but there is marked individuality in the occurrence of CVD events. Recently, the Montreal-FH-SCORE (MFHS) has been shown to stratify CVD frequency in FH subjects, but this score has not yet been validated. OBJECTIVE: The aims of the present study were to conduct an independent external validation of the MFHS in a retrospective cohort of heterozygous FH and to identify additional variables that could significantly improve the prediction of prevalent CVD. METHODS: The MFHS calculation is based on 5 variables: age, high-density lipoprotein cholesterol, gender, hypertension, and smoking status. This score was validated in a cohort of 718 adult FH using receiver operating characteristic (ROC) curves analysis to determine the discriminatory ability of the MFHS. The performance of the MFHS was compared to a novel Combined-FH-SCORE in 1388 FH. RESULTS: The MFHS had an excellent discrimination for prevalent CVD events in the validation cohort, with an area under the receiver operating characteristic curve of 0.799 (0.766-0.832, P < .0001). Patients with a high MFHS score presented a significant 8.8-fold increased odd of CVD events compared with patients with a low score (95% confidence interval 5.8-13.3, P < .0001). The addition of lipoprotein(a) to the score did not improve the prediction of CVD events (area under the receiver operating characteristic curve 0.823 vs 0.817, P = .11). CONCLUSION: This study confirmed that the MFHS is a strong predictor of prevalent CVD in FH and that the addition of lipoprotein(a) offers a minor improvement in the discrimination of the score.
BACKGROUND:Familial hypercholesterolemia (FH) is a disease characterized by increased low-density lipoprotein cholesterol and premature cardiovascular disease (CVD) but there is marked individuality in the occurrence of CVD events. Recently, the Montreal-FH-SCORE (MFHS) has been shown to stratify CVD frequency in FH subjects, but this score has not yet been validated. OBJECTIVE: The aims of the present study were to conduct an independent external validation of the MFHS in a retrospective cohort of heterozygous FH and to identify additional variables that could significantly improve the prediction of prevalent CVD. METHODS: The MFHS calculation is based on 5 variables: age, high-density lipoprotein cholesterol, gender, hypertension, and smoking status. This score was validated in a cohort of 718 adult FH using receiver operating characteristic (ROC) curves analysis to determine the discriminatory ability of the MFHS. The performance of the MFHS was compared to a novel Combined-FH-SCORE in 1388 FH. RESULTS: The MFHS had an excellent discrimination for prevalent CVD events in the validation cohort, with an area under the receiver operating characteristic curve of 0.799 (0.766-0.832, P < .0001). Patients with a high MFHS score presented a significant 8.8-fold increased odd of CVD events compared with patients with a low score (95% confidence interval 5.8-13.3, P < .0001). The addition of lipoprotein(a) to the score did not improve the prediction of CVD events (area under the receiver operating characteristic curve 0.823 vs 0.817, P = .11). CONCLUSION: This study confirmed that the MFHS is a strong predictor of prevalent CVD in FH and that the addition of lipoprotein(a) offers a minor improvement in the discrimination of the score.
Authors: Gerald F Watts; Samuel S Gidding; Pedro Mata; Jing Pang; David R Sullivan; Shizuya Yamashita; Frederick J Raal; Raul D Santos; Kausik K Ray Journal: Nat Rev Cardiol Date: 2020-01-23 Impact factor: 32.419