Impact of Diabetes Mellitus on Percutaneous Coronary Intervention Outcomes: Real-World Lessons From a Large Chinese Single-Center Registry

The mortality of coronary artery disease (CAD) in the Chinese population is historically low compared with Western levels, but the burden has been increasing. ¹ Coronary artery disease is now the major cause of death for both men and women in China. ¹ This is in part due to an aging population, but more importantly due to an increasing incidence of traditional risk factors. Over 60% of Chinese men are current smokers, and relevant to this, China is the largest tobacco producer and consumer in the world. ¹ In addition, hypertensive heart disease, diabetes mellitus (DM), hyperlipidemia, obesity, and air pollution have all reached epidemic proportions. As a consequence, in 2008, the death rates (per 100 000 population) for cardiovascular (CV) diseases in China actually exceeded those in the United States. ¹

The first cardiac intervention in China, a percutaneous transluminal coronary angioplasty (PTCA), was carried out in 1984. ¹ In the following 2 decades, adoption was slow; a total of only 8000 patients underwent PTCA in 1999. ¹ In 2001, emergency percutaneous coronary intervention (PCI) was performed in only 2820 patients, an impressively insignificant number given a total population of over 1.2 billion at that time. ¹ However, use of cardiac catheterization and PCI has grown tremendously over the past 2 decades, especially in urban Chinese cities. Between 2001 and 2011, coronary catheterizations grew from 26 570 cases per annum to 452 784, a 17-fold increase. ² Annual PCI hospitalizations soared 21-fold from 9678 to 208 954. Moreover, Chinese cardiologists have been quick to adopt new practices; the use of the radial approach increased from 3.5% to 79% and of drug-eluting stent (DES) from 18% to 97%. ²

Due to the absence of a nationwide standard disease registration and classification system in China, precise information does not exist regarding the number of events or deaths from CAD for the entire country. Although new technology has been increasingly applied, there are significant differences between Chinese CAD clinical practice, recommended guidelines, clinical outcomes, ³ and the usage patterns of PCI. ⁴ Data on PCI quality and outcomes have also been lacking. In addition, a specific association of problems with DM has increasingly been noted. In a very recent retrospective study, Jiang et al presented data on apparent type 2 DM patients who were diagnosed by the following: an abnormal blood glucose level (≥126 mg/dL) after an overnight fast, an abnormal glycosylated hemoglobin test (≥6.5%), or an abnormal glucose tolerance test (2 hours, ≥200 mg/dL). ⁵ These patients type 2 DM were considered to have a disease predictive of adverse events after implantation of a DES. ⁵ In a meta-analysis by Zheng et al, it was found that increased blood levels of Hemoglobin A_1c were associated with increased major adverse cardiovascular events (MACEs) in patients with type 2 DM (these patients in the meta-analysis meet the criteria for type 2 DM as defined by the authors) following PCI. ⁶ Wang et al report an analysis of the impact of type 2 DM (type 2 DM criteria met by age) on a large cohort of patients (n =10 724) who underwent PCI at a single high-volume center in China. ⁷ Such articles which involve a large Chinese patient base are extremely important and many of the results are discussed subsequently in this editorial. ⁷ Although the manuscript of Wang et al ⁷ has important limitations, it is still possible to glean valuable observations.

Also worthy of comment regarding the Chinese population is the increased incidence of cytochrome P450 C19 (CYP2C19) reduced function (also referred to as loss-of-function) polymorphisms in Asians ⁸ and in Chinese, ^9,10 which can increase CAD risk. The reduced function alleles of CYP2C19 polymorphism, especially the CYP2C19*2 allele, ¹¹ are associated with a decreased effect of clopidogrel (from a decreased conversion of clopidogrel to its active moiety), with a resultant increase in MACE following PCI in clopidogrel-treated patients. ^8,12 These reduced function alleles have a significantly increased prevalence in Chinese patients and this must be taken into account in comparing PCI results, especially since clopidogrel safety is essentially the same when comparing races. ⁸ Also relevant to this editorial is that in addition to the negative effect of the decreased function CYP2C19 polymorphisms, the association of increased body mass index and DM can synergistically contribute to decreased antiplatelet activity of clopidogrel and thereby to MACE following PCI. ¹³

It is noteworthy in the study of Wang et al that Chinese practice patterns largely mirror those in the West. ⁷ Procedural success rates were high (about 98%), and the great majority of patients (>90%) received a DES. ⁷ In fact, 57% received a second-generation DES, defined as a biodegradable polymer or polymer-free DES. More than 95% of patients were discharged on aspirin, clopidogrel, and a statin; 90% were discharged on a β-blocker. It is also clear that Chinese physicians are comfortable treating complex lesions percutaneously. ⁷ Although the average Synergy Between PCI With Taxus and Cardiac Surgery (SYNTAX) ¹⁴ scores reported were modest, the great majority of lesions (>70%) were classified as type B2/C. ⁷ In addition, most of these procedures were carried out in the context of an acute coronary syndrome and the large majority were performed in the left anterior descending artery. ⁷ The patient population has a somewhat lower risk with an average age of 58 to 59 years. Only approximately 1% had an ejection fraction <40%, and only 2% to 3% had compromised renal function. ⁷

In the study by Wang et al, data were prospectively collected on 10 724 consecutive patients who underwent PCI at Fuwai Hospital in Beijing; 2-year clinical outcomes were compared between patients with and without type 2 DM. ⁷ Standard multivariable-adjusted Cox regression analysis and propensity score matching were performed. Interestingly, the investigators observed that although diabetic patients were at higher baseline clinical risks and had worse clinical outcomes in comparison with patients without type 2 DM (driven by both higher rates of repeat revascularization and occurrence of MACE including death and myocardial infarction [MI]), diabetic status did not emerge as an independent predictor of mortality or repeat revascularization after adjusting for other confounders and propensity. Wang et al hypothesize that these results may be due to the use of better technology (ie, newer DES) and the concomitant liberal use of cardioprotective medications. ⁷ They also raise the possibility that Asian patients may not suffer the same clinical consequences from the impact of type 2 DM that Western patients do, although this hypothesis is not uniformly supported by available data. ¹⁵

The greatest limitation of the study is the short-term follow-up of only 2 years. The short-term outcomes of any revascularization strategy will likely be similar for patients, both with and without type 2 DM. This is especially so with second-generation DES, where stent thrombosis rates are exceedingly low (<1%), ¹⁶ and the inherent marked reductions in the incidence of restenosis and revascularization are most likely to be observed in the near term. In fact, even the best choice of chemotactic agent (limus vs taxol eluting) in patients with type 2 DM remains controversial. Patients with type 2 DM frequently have premature, rapidly progressive, multifocal disease leading to long, diffuse lesions in small-caliber coronary arteries. ¹⁶ As a result, these patients have particularly high rates of acute MI and mortality. ¹⁶ When compared with patients without type 2 DM, patients with type 2 DM (type 2 DM criteria met by age) have increased MACE or worse outcomes with PCI in the PTCA era, bare-metal stent era, and the DES era, ¹⁷ with higher rates of in-stent restenosis, stent thrombosis, death, and MI. ¹⁸ Two seminal trials, SYNTAX ¹⁹ and Future Revascularization Evaluation in Patients with Diabetes Mellitus: Optimal Management of Multivessel Disease (FREEDOM), ²⁰ noted superiority of coronary artery bypass grafting over PCI for diabetic patients with multivessel complex CAD. However, both trials were conducted over longer time periods (5 years and 3.8 years, respectively). Therefore, it is entirely possible that had the cohort of Wang et al ⁷ been followed for a longer time, the presence of type 2 DM may well have become an independent predictor for CV events and/or mortality.

Another important limitation of the study is the nonrandomized cohort. These patients were treated per operator discretion and therefore were subject to unavoidable selection and treatment bias. In addition, although the cohort is large, its study remains underpowered to evaluate rare events such as stent thrombosis. Finally, additional limitations are consistent with those appreciated in any post hoc analysis or whenever subgroup data are analyzed retrospectively. Therefore, these results should be received with a healthy dose of skepticism and considered solely hypothesis-generating at this time.

Fortunately, the number of clinical trials enrolling in China has exploded in the past several years. ²¹ The country’s large pool of patients and its sizable and growing device/pharmaceutical market account for much of this growth. Therefore, we may well soon get results from multiple multicenter prospective randomized trials, evaluating important clinical questions, such as the impact of type 2 DM on contemporary PCI in a Chinese cohort.