Higher Lp(a) in children with clinical presentation of FH but no genetic mutation

Introduction and methods

Background

Familial hypercholesterolemia (FH) is caused by mutations in the genes LDLR, APOB, or PCSK9 that lead to impaired circulatory clearance of LDL-c and, subsequently, very high LDL-c levels from birth onwards [1]. Depending on the diagnostic tool used, the clinical diagnosis of FH in adults relies on (extremely) high LDL-c levels, clinical features of FH (such as tendon xanthomas, arcus cornealis, and a history of premature CVD) and/or affected first-degree relatives.

The preferred method for diagnosing FH in adults and children is genetic testing [2-4]. However, in many children with a clinical presentation of FH, no FH-causing mutation can be identified [5,6], which suggests that other factors may lead to this clinical presentation. As it turns out, a substantial part of the clinical FH diagnoses in adults can be explained by high levels of Lp(a) [7-9]. The reason for this is the LDL-like core of Lp(a), as conventional LDL-c assays measure the cholesterol content of both LDL particles and Lp(a) particles [10-12].

Aim of the study

The authors sought to examine whether high Lp(a) levels can lead to a clinical presentation of FH in children by comparing Lp(a) levels in children with a clinical presentation of FH in whom no mutation was detected with those in children with an FH-causing mutation and their unaffected siblings.

Methods

In this cross-sectional study, 2721 children (mean age: 10.1 years; range: 0.7–17.9 ) were included who were referred to the pediatric lipid clinic of the Amsterdam University Medical Centers in Amsterdam, the Netherlands between June 1989 and January 2020 for a tentative diagnosis of FH based on a family history of premature CVD (i.e., any CV event <60 years of age), family members with FH, or high LDL-c levels in the child or its family members. As part of routine care, a lipid profile, including LDL-c and Lp(a) levels, was obtained and DNA analysis to detect FH mutations was performed. For the analyses, log-transformed Lp(a) levels were used.

Children were divided into the following 4 groups:

• 1931 children with definite FH (mutation detected) (71%);
• 290 unaffected siblings/normolipidemic controls with definite non-FH (mutation excluded) (11%);
• 108 children with probable FH (clinical presentation of FH (i.e. LDL-c >5 mmol/L or 4–5 mmol/L with family history of premature CVD); no mutation detected) (4%);
• 392 children with probable non-FH (no clinical presentation of FH; mutation not excluded) (14%).

Main results

• In the entire study population, the geometric mean Lp(a) level at first visit was 11.5 mg/dL (95%CI: 11.0–12.1; range: 0.2–171.0). In total, 665 children (24.4%) had Lp(a) levels >30 mg/dL and 331 (12.2%) had Lp(a) levels >50 mg/dL.
• In children with probable FH, the geometric mean Lp(a) level was 15.9 mg/dL (95%CI: 12.3–20.6), which was significantly higher than that in children with definite FH (11.5 mg/dL; 95%CI: 10.9–12.1; P=0.017) and those with definite non-FH (9.8 mg/dL; 95%CI: 8.4–11.3; P=0.002). In children with probable non-FH, the geometric mean Lp(a) level was 12.7 mg/dL (95%CI: 11.1–14.6).
• High Lp(a) levels (>50 mg/dL) were more frequently seen in children with probable FH compared with children with definite FH (31.5% vs. 10.2%; P<0.0001). Of note, 9.7% of the unaffected siblings also showed high Lp(a) levels.
• Mean (95%CI) LDL-c levels corrected for Lp(a)-c were 5.2 mmol/L (5.1–5.2) for children with definite FH, 4.2 mmol/L (4.0–4.3) for those with probable FH, 2.7 mmol/L (2.7–2.8) for those with probable non-FH, and 2.4 mmol/L (2.3–2.5) for those with definite non-FH (P<0.001).

Conclusion

References

Find this article online at Eur Heart J.