Homocysteine and risk of periprocedural myocardial infarction in patients undergoing coronary stenting

Background: Despite improvements in pharmacological and mechanical devices, the risk of periprocedural myocardial infarction (PMI) is still high, particularly in prothrombotic conditions. Hyperhomocysteinemia has been associated with enhanced platelet function, impaired endothelial function and prothrombotic status, thus increasing the risk of cardiovascular events. No study has, so far, investigated the relationship between homocysteine levels and the risk of periprocedural MI in patients undergoing percutaneous coronary intervention (PCI), and this is therefore the aim of the current study. Methods: In 1150 patients undergoing PCI, homocysteinemia was assessed at admission. Cardiac biomarkers were measured at intervals from 8 to 48 h after PCI. Periprocedural myonecrosis was defined by a troponin I increase to three times the upper limit of normal (ULN) or by 50% if elevated at the time of the procedure. PMI was defined as a CK-MB increase to three times the ULN or of 50% if elevated at the time of the procedure. Results: We grouped patients according to tertile values of homocysteine. Higher homocysteine levels were associated with older age (P<0.001), male sex (P=0.02), arterial hypertension (P=0.007), diabetes (P=0.04), renal failure (P<0.001), higher creatinine levels (P=0.01), previous MI (P=0.02), previous PCI (P=0.04) and previous cerebrovascular accidents (P=0.01). Homocysteine was associated with lower ejection fraction (P<0.001), treatment with angiotensin-receptor blockers (P<0.001), nitrates (P=0.008) and diuretics (P<0.001) and acetylsalicylic acid (P=0.01). Homocysteine levels were directly related with the extent of coronary disease (P=0.04) and coronary calcifications (P<0.001) but inversely with type C lesions (P=0.001), TIMI 3 flow pre-PCI (P=0.02), stenosis severity (P=0.01) and thrombus (P=0.004). In addition, they are associated with higher rates of balloon predilatation (P=0.02), lower use of thrombectomy (P=0.01) and periprocedural administration of GPIIbIIIa inhibitors (P=0.02). Ageing, male sex, diabetes, renal failure, creatinine levels, diuretics use, coronary calcifications and type C lesions were independently related to homocysteine. Homocysteine did not affect the risk of PMI [adjusted odds ratio (OR) 1.14 (0.91-1.42), PU0.26], or periprocedural myonecrosis [adjusted OR 1.17 (0.98-1.39), P=0.08]. Similar results were found after propensity score adjustment [adjusted OR 1.19 (0.95-1.48), P=0.14 for PMI and adjusted OR 1.18 (0.99-1.4), P=0.07 for myonecrosis] and at subgroup analysis in higher risk subsets of patients. Conclusion: In patients undergoing PCI, the risk of PMI is not influenced by hyperhomocysteinemia.