This study evaluated the association between serum Lp(a) concentrations and the durability of aortic bioprosthetic valves in a large cohort undergoing second aortic valve intervention. Contrary to the initial hypotheses and pathophysiological parallels drawn from native AS, we found no significant relationship between elevated Lp(a) and bioprosthetic valve degeneration leading to earlier reintervention.

Lp(a) is a mainly genetically determined lipoprotein variant known to carry pro-calcific oxidized phospholipids on its surface and has been strongly implicated in the pathogenesis and progression of native AS through its pro-calcific and pro-inflammatory effects [6, 7, 19, 20]. Elevated Lp(a) contributes to valvular interstitial cell activation, promotes osteogenic transformation, and mediates lipid infiltration and oxidation within valve leaflets [21]. Several clinical studies have demonstrated faster progression of native AS in patients with elevated Lp(a) levels [7, 22]. Notably, a large-scale, clinical trial (Lp(a)–FRONTIERS) has recently been launched to test whether therapeutic lowering of Lp(a) can slow the progression of calcific AS in patients with elevated Lp(a)​ [10]. The initiation of this trial reflects a broad consensus that Lp(a) is a promising target for intervention in aortic valve disease. Its results, may also have implications for patients with bioprosthetic valves by demonstrating the modifiability of Lp(a)-driven calcific processes [11].

SVD of bioprosthetic aortic valves shares many pathological features with native calcific AS. Both conditions are characterized by progressive fibro-calcific remodeling of the valve leaflets, including deposition of lipids, inflammation, and ultimately leaflet calcification [23, 24]. Notably, leaflet calcification plays a central role in disease progression in both native and bioprosthetic valves. However, a recent work by Botezatu et al., using multimodal imaging (echocardiography, CT, and PET) to assess SVD, found no significant association with Lp(a) concentrations over a 2-year follow-up period [13]. Contrary, Farina et al. reported increased SVD in TAVI patients with high Lp(a) compared to low Lp(a) at four years follow-up; however, without significant difference in survival or aortic valve reintervention [11]. In line with that, findings from Bormann et al. showed that Lp(a) levels were not associated with increased calcification prior to TAVI or all cause-mortality at 30 days and 40 months following TAVI [12].

These findings raised questions about whether Lp(a) contributes to bioprosthetic valve failure or whether longer-term clinical outcomes might reveal an effect not captured in the studies conducted to date. However, unlike native valves, bioprostheses are composed of non-viable xenogeneic tissue that lacks endothelial and interstitial cell layers. This fundamental histological difference may explain the diminished responsiveness of bioprosthetic tissue to circulating pro-calcific mediators like Lp(a) [4]. Nevertheless, these findings are hypotheses-generating, and prospective studies with substantially longer follow-up and precise imaging modalities are required to gain further insights into this issue.

Current ESC/EACTS guidelines recommend the use of bioprosthetic valves in patients over 65 years, with increasing application in younger populations due to improvements in valve design and favorable outcomes of transcatheter valve-in-valve procedures [25]. This shift has raised concerns about long-term valve durability, particularly in younger patients with elevated lipoprotein(a) levels. In our hypothesis-generating analysis, we found no evidence for a significant association between elevated Lp(a) and structural valve deterioration requiring reintervention. While these findings do not support intensified imaging or earlier reintervention strategies based solely on Lp(a) status, definitive conclusions cannot be drawn. However, our findings suggest that elevated Lp(a) alone should not influence decision-making in the selection or monitoring of patients receiving aortic bioprosthetic valves.

Despite the negative findings, several open questions remain. It is conceivable that Lp(a)-mediated effects could still manifest in specific prosthesis types or under particular hemodynamic conditions. Moreover, the potential impact of emerging Lp(a)-lowering therapies, such as pelacarsen or olpasiran on native and prosthetic valve calcification remains to be clarified [26, 27]. The Lp(a)-FRONTIERS trial (NCT05646381) and other ongoing studies will be instrumental in evaluating whether pharmacologic Lp(a) reduction translates into clinically meaningful improvements in valve-related outcomes [10] and may contribute to more personalized strategies to delay progression of native aortic valve stenosis.

In addition, prospective studies are needed that systematically investigate the role of Lp(a) in structural valve deterioration after aortic valve replacement. As with native aortic valves, detailed histological and molecular analyses of explanted prostheses will be essential to gain a deeper mechanistic insight into the pathophysiology of bioprosthetic valve degeneration and potential future treatment targets.

While elevated Lp(a) does not appear to be a major determinant of bioprosthetic valve failure in our cohort, it remains a well-established cardiovascular risk factor and warrants appropriate management as part of an overall preventive cardiovascular approach [28, 29].

Several limitations of our study should be acknowledged. First, the retrospective and predominantly observational study design inherently limits the ability to establish causality. Second, there is a potential survivor bias, as the study only included patients who survived long enough to undergo the second aortic valve intervention, which may not accurately reflect outcomes in the broader population. Third, the measurement of Lp(a) was performed at a single time point, specifically at the time of reintervention, which limits insights into potential temporal variations or trends in Lp(a) levels over time. Fourth, while most Lp(a) values were measured in nmol/L, a subset originally reported in mg/dL was converted using a standard factor. Given that these units capture distinct biological attributes (particle number versus mass), such conversion is inherently imprecise and represents a methodological limitation of the present study. Finally, we did not have the statistical power to show any detrimental effects on bioprostesis degeneration that precipitate only at extremely high circulating levels of Lp(a).