Our study enrolled a prospective cohort corresponding to diabetic individuals with prior established CVEs, who were at high risk for recurrent ischemic CVEs in the circumstance of following standard secondary prevention strategies recommended by the current guidelines [20, 21]. Data, for the first time, clearly confirmed that Lp(a) was an independent predictor for recurrent CVEs in T2DM patients with prior CVEs. When stratified by HBA1c levels (< 7.0%, or ≥ 7.0%), this association were significant in both HBA1c status independent of the level of the other risk factors. More importantly, in the overall cohort, the addition of Lp(a) to the model improved the risk prediction for recurrent CVEs. Thus, the present study strongly implied that Lp(a) might be a useful marker for further risk stratification in patients with T2DM after they suffered a first CVE.
The prevalence of T2DM has been increasing dramatically over the past few decades, with projections of an even greater growth over coming decades [22, 23]. Convincing evidence indicated that CAD is a common comorbidity in patients with T2DM and has been considered as a CAD risk equivalent based on multiple guidelines [24]. Currently, several clinical investigations indicated that despite aggressive multidisciplinary efforts have been made including revascularization and intensive management of LDL-C, glucose, blood pressure, and thrombotic risk, patients surviving an ACS event are at increased risk of recurrent CVEs, and this risk is further increased in patients with T2DM [25], raising the question of whether the treatment regimens are less effective in these patients. For decades, it has been well elucidated that abnormal lipid metabolism largely contributes to the additional cardiovascular risk for T2DM patients [26]. Therefore, the management of multiple risk factors especially lipid is of great significance for the prognosis. The recent guidelines have clearly recommended the target value of LDL-C [27], nonetheless, residual cardiovascular risk remains high for T2DM patients with a prior CVE compared with non-diabetic patients. Thus, it is essential to search additional modifiable lipid disorders to further improve the prognosis of these patients. Therefore, we consecutively recruited 2,284 T2DM patients with prior CVEs and followed up for 7,613 patient-years, attempting to seek plausible residual risk in terms of lipid disorders.
At the clinical level, elevated Lp(a) has been the least studied among all lipid disorders. Plasma concentrations of Lp(a) are mainly (90%) determined by the LPA gene, without significant dietary or environmental influences [28]. The association of Lp(a) with risk of CAD as well as mortality, which is independent of traditional cardiovascular risk factors, has been rapidly aware in series of studies [29–30]. Lp(a) has been determined as the strongest independent genetic risk factor of CVD and a causal role has been demonstrated by Mendelian randomization [31]. The Copenhagen City Heart Study demonstrated that compared to subjects with Lp(a) levels below 5 mg/dL, those with Lp(a) between 30 and 76 mg/dL had a 1.6-fold increased risk for incident MI. This risk increased to 1.90 for individuals with Lp(a) between 77 and 117 mg/dL and to 2.60 for individuals with Lp(a) concentrations above 117 mg/dL [8]. However, the data mainly based on investigations of apparently healthy participants in the general population rather than patients with a prior CVE. At the same time, among limited existing investigations related to patients with established CAD, inconsistent results were also observed. A recent cohort study support that in patients with stable CAD and chronic total occlusion, increased Lp(a) confers greater risk for poor coronary collateralization when TC, LDL-C or non-HDL-C are elevated especially in patients with T2DM [32]. On the contrary, Schwartz GG, et al. enrolled 969 patients who experienced a recent ACS and treated with statins, Lp(a) concentration was not associated with adverse CVEs [16]. Therefore, studies concerning the prognosis of Lp(a) in patients with a prior CVE are of worth in the real-world, particularly in patients with T2DM.
Consequently, in our study, we observed that Lp(a) levels were significantly higher in patients suffered recurrent CVEs. Of note, our current data also demonstrated that the event-free survival rate was dramatically lower in medium and high Lp(a) groups. Significantly, compared with patients with low Lp(a) levels, those with high Lp(a) had a 1.996-fold higher risk of recurrent CVEs (95% CI 1.266–3.148) after adjusting for other variables including LDL-C, HBA1c, and so forth. Furthermore, when divided the population into two groups by HBA1c status, the predictive value of Lp(a) in risk of recurrent CVEs remains significant independent of the glucose control level (HBA1c < 7.0%, HR(95% CI): 1.914(1.007–3.640); HBA1c ≥ 7.0%, HR(95% CI): 2.174(1.132–4.174)). Finally, the C-statistic was significantly improved by 0.029 when added Lp(a) to the Cox model. Although the results were inconsistent with the study conducted by Schwartz GG [16], the different of enrolled population may partly explain the disparity. As far as we know, it is the first large study involved patients with T2DM and a first CVE, which included the composite of MI, stroke, peripheral arterial disease, PCI, and CABG, instead of ACS or other specific status. Hence, the present study supported the opinion that Lp(a) was an independent predictor for recurrent CVEs in T2DM patients with prior CVEs in the stain era.
Nevertheless, our study had several limitations. First of all, this is a study among Chinese population with T2DM and prior CVEs, and whether the data applied to other populations need to be testified. Secondly, the Lp(a) concentrations were only measured at baseline, and the alterations of the biomarkers may also be clinically significant during the follow-up period. Moreover, the method of Lp(a) measurement used in the study might be influenced by the apo(a) size due to the numbers of the KIV type 2 domain. Variations of apo(a) size between assay calibrators and patients’ samples might overestimate or underestimate the real concentration of Lp(a). Finally, as this was an observational study, further investigations are needed to clarify the underlying mechanism of the associations.