Lipoprotein(a) (Lp(a)) is a complex in human plasma assembled from low-density lipoprotein (LDL) and apolipoprotein(a) (apo(a)). High plasma concentrations of Lp(a) are a risk factor for coronary heart disease (CHD) in particular in patients with concomitant elevation of LDL. We have analysed for elevated Lp(a) levels in patients with familial hypercholesterolaemia (FH), a condition caused by mutations in the LDL receptor (LDLR) gene and characterised by high LDL, xanthomatosis and premature CHD. To avoid possible confusion by the apo(a) gene which is the major quantitative trait locus controlling Lp(a) in the population at large, we used a sib pair approach based on genotype information for both the LDLR and the apo(a) gene. We analysed 367 family members of 30 South African and 30 French Canadian index patients with FH for LDLR mutations and for apo(a) genotype. Three lines of evidence showed a significant effect of FH on Lp(a) levels: (1) Lp(a) values were significantly higher in FH individuals compared to non-FH relatives (p < 0.001), although the distribution of apo(a) alleles was not different in the two groups; (2) comparison of Lp(a) concentrations in 28 sib pairs, identical by descent (i.b.d.) at the apo(a) locus but non-identical for LDLR status, extracted from this large sample demonstrated significantly elevated Lp(a) concentrations in sibs with FH (p < 0.001); (3) single i.b.d. apo(a) alleles were associated with significantly higher Lp(a) concentrations (p < 0.0001) in FH than non-FH family members. Variability in associated Lp(a) levels also depended on FH status and was highest when i.b.d. alleles were present in FH subjects and lowest when present in non-FH individuals. The study demonstrates that sib pair analysis makes it possible to detect the effect of a minor gene in the presence of the effect of a major gene. Given the interactive effect of elevated LDL and high Lp(a) on CHD risk our data suggest that elevated Lp(a) may add to the CHD risk in FH subjects.