Abstract Introduction Elevated lipoprotein(a) Lp(a) is an established, independent risk factor for coronary artery disease (CAD), and has also been shown to be associated with worse outcomes amongst patients with established CAD. However, association status of baseline Lp(a) levels with prognosis of percutaneous coronary intervention (PCI), especially with respect to incidence of in-stent restenosis (ISR) remain contentious. Purpose We conducted a systematic review and study-level meta-analysis to evaluate the relationship between pre-PCI Lp(a) levels and the incidence of ISR. Methods Online databases of PubMed, EMBASE, MEDLINE, Scopus, and Web of Knowledge were systematically searched for relevant articles published until June 30th, 2025. Qualifying studies had to have: (i) an angiographically determined ISR group, (ii) an angiographically or clinically determined no-ISR group, and (iii) pre-PCI Lp(a) levels assessment for both groups. Data was extracted from included studies, transformed (if needed) and synthesized for summary effect measures of standardized mean difference (SMD), mean difference (MD), and incidence rate ratio (IRR). Study level results were pooled using Z-test employing appropriate effects for analysis. Analysis was repeated after stratifying the studies into planned and clinically-driven repeat angiogram subgroups. Begg’s funnel plots and Egger’s test were used to detect publication bias. Results From the 307 unique records retrieved, 24 studies were included for SMD based analysis (n= 4,596), 13 for MD based analysis (n= 2,016), whilst 17 for IRR based analysis (n= 9,633). Cases with ISR overall, were found to have significantly higher pre-PCI Lp(a) levels, demonstrating small-moderate effect size (SMD= 0.25), whilst the mean difference in Lp(a) levels between the ISR and no-ISR was found to be 5.87 mg/dl (SMD= 0.25, 95% CI= 0.11, 0.38; p= 0.0003 and MD= 5.87, 95% CI= 1.78, 9.96 mg/dl; p= 0.005 respectively). In addition, the risk of ISR was demonstrated to be 37% higher in patients with clinically elevated pre-PCI Lp(a) levels as compared to the rest (IRR= 1.37, 95% CI= 1.14, 1.64; p= 0.0007). As expected, our derived associations were generally found to be stronger in the clinically-driven repeat angiogram subgroup as compared to the planned repeat angiogram subgroup. (Figure 1.) There was no evidence of publication bias amongst the analysed study groups. Our leave-one-out sensitivity analyses further confirmed the robustness of our findings. Conclusions The present systematic review and meta-analysis confirms the association between elevated pre-procedural Lp(a) levels and angiographically confirmed ISR. Our findings underscore the potential of Lp(a) as a biomarker for risk stratification at the time of index PCI, helping to identify patients who may benefit from closer clinical monitoring and intensified secondary prevention strategies.For image description, please refer to the figure legend and surrounding text.
Rai et al. (Sun,) studied this question.