Impact of Type 2 Diabetes on Five-year Clinical Outcomes Following Successful Percutaneous Coronary Intervention on Chronic Total Occlusions: a Propensity Matched Cohort Analysis


 Background: Despite substantial improvement in chronic total occlusions (CTO) revascularization technique, the long-term clinical outcomes in diabetic patients with revascularized CTO remain controversial. Our study aimed to investigate the five-year cardiovascular survival for patients with or without type 2 diabetes mellitus (DM) who underwent successful percutaneous coronary intervention (PCI) for CTO. Methods: Data of the current analysis derived from a large single-center, prospective and observational cohort study, including 10,724 patients who underwent PCI in 2013 at Fuwai Hospital. Baseline, angiographic and follow-up data were collected. The primary endpoint was major adverse cardiac and cerebrovascular events (MACCE), which consisted of death, recurrent myocardial infarction (MI), stroke and target vessel revascularization (TVR). The secondary endpoint was all-cause mortality. Cox regression analysis and propensity-score matching was performed to balance the baseline confounders. Results: A total of 719 consecutive patients with ≥ 1 successful CTO-PCI were stratified into diabetic (n=316, 43.9%) and non-diabetic (n=403, 56.1%) group. During a median follow-up of 5 years, the risk of MACCE (adjusted hazard ratio [HR] 1.47, 95% confidence interval [CI] 1.08-2.00, P = 0.013) was significantly higher in the diabetic group than in the non-diabetic group, whereas the adjusted risk of all-cause mortality (HR 2.37, 95% CI 0.94-5.98, P = 0.068) was similar. In the propensity score matched population, there were no significant differences in the risk of MACCE (HR 1.27, 95% CI 0.92-1.75, P = 0.155) and all-cause mortality (HR 2.56, 95% CI 0.91-7.24, P = 0.076) between groups. Subgroup analysis revealed a consistent effect on five-year MACCE across various subgroups.Conclusions: In patients who received successful CTO-PCI, non-diabetic patients were related to better long-term survival benefit in terms of MACCE. Further randomized studies are warranted to confirm these findings.


Introduction
Chronic total occlusion (CTO) occurs in approximately 15-25% of patients with coronary artery disease (CAD) undergoing diagnostic coronary angiography [1,2]. Due to the development of interventional devices and dedicated techniques, percutaneous coronary intervention (PCI) for CTO has achieved high technical success rates with a low risk for procedural complications, especially in tertiary medical centers. Current guidelines have regarded revascularization for CTO as the IIa B recommendation [3]. Considerable evidence suggest that successful CTO-PCI is related to a better improvement of symptoms, quality of life, and ventricular function compared to optimal medical treatment alone and unsuccessful CTO-PCI [4][5][6], whereas the bene t in terms of improving patient survival was not signi cant [7,8]. The bene cial effect of CTO-PCI on long-term prognosis is still controversial [2,9], especially for the special group of people with diabetes. Type 2 Diabetes mellitus (DM) is a well-established CAD risk equivalent and is associated with a greater atherosclerotic burden, such as multivessel disease, heavily calci ed coronary lesions, diffuse and small vessel CAD [10,11]. Previous studies have reported that patients with DM have an elevated incidence of CTO (approximately 30-40%) [12,13]. In addition, CTO patients with DM are related to longer and more technically challenging occluded lesions, with lower success rates compared with that in non-DM [14]. Besides, non-DM patients were more likely to fare better after CTO-PCI for up to 3 years compared to their DM counterparts [15]. However, to the best of our knowledge, no previous study has focused on longer term impact of successful recanalisation for CTO lesions in patients with versus without DM. Therefore, we conducted a prospective, observational and real-world study to investigate ve-year clinical outcomes in type 2 diabetic and non-diabetic patients after successful CTO-PCI.

Method
Study population A total of 10,724 consecutive patients with CAD who underwent PCI were enrolled between January 2013 and December 2013 in Fu Wai Hospital, National Center for Cardiovascular Diseases, Beijing, China. Notably, we included 1010 (9.42%) patients with at least 1 CTO lesion. CTO lesions were de ned as complete obstruction of a native coronary artery for longer than 3 months with thrombolysis in myocardial infarction (TIMI) ow grade of 0 [16]. Patients who received recanalisation treatment for CTO depended on contemporary practice guidelines, judgment from our team's experienced cardiologists and their own preference [17]. Exclusion criteria included the following: (1) patients who underwent unsuccessful CTO-PCI (n=267); (2) patients lacking both hemoglobin A 1c (HbA 1c ) and fasting plasma glucose (FPG) data (n=9); (3) patients who were diagnosed as acute STEMI within 72h before admission (n=15). Thus, the remaining 316 (43.9%) patients with type 2 DM and 403 (56.1%) patients without DM were enrolled for the nal analysis ( Figure 1). DM was de ned as a FPG of at least 7.0 mmol/L, or glycated HA 1c greater than 6.5% or known diabetes, based on previous medical records of the patients and data of the therapeutic status based on the glucose-lowering therapy [18]. Left ventricular ejection fraction (LVEF) was measured from two-dimensional echocardiography according to modi ed Simpson's rule. Estimated glomerular ltration rate (eGFR) was calculated by the modi ed diet in renal disease equation for Chinese [19]. Data of demographic, clinical and angiographic features were collected from the database and medical records retrospectively, whereas clinical endpoints during follow-up were identi ed prospectively. The study complied with the principles of the Declaration of Helsinki and was approved by the Institutional Ethics Committee at Fu Wai Hospital. All eligible participants gave written informed consent.

PCI procedures
Coronary interventions were performed according to current standard guidelines at the discretion of the operating physician [17]. Before catheterization, unless on chronic P2Y12 inhibitor therapy for > 6 days, selected PCI patients received oral administration of aspirin 300 mg and clopidogrel (loading dose 300 mg) or ticagrelor (loading dose 180 mg) at least 24h. Patients presenting as acute coronary syndrome (ACS) scheduled for PCI received the same dose of aspirin and ticagrelor or clopidogrel (loading dose 300 or 600 mg) as soon as possible. Thereafter, unfractionated heparin (100 U/kg) was administered before PCI, however, the use of glycoprotein IIb/IIIa inhibitors was at the operator's judgment. CTO-PCI was done using bilateral injections, specialized hydrophilic wires, microcatheters and retrograde approach, when available. If both antegrade and retrograde approaches failed, intravascular ultrasound (IVUS) guided wire re-entry technique would be attempted. Standard dual-antiplatelet medication was maintained for at least 12 months after PCI. The PCI procedure was considered successful if residual stenosis < 30% with TIMI ow grade 3 at the end of the procedure was obtained according to visual estimation of the angiograms.

Endpoints and follow-up
The primary clinical outcome was the occurrence of 5-year major adverse cardiac and cerebrovascular events (MACCE) during follow-up, a composite endpoint of death, recurrent myocardial infarction (MI), stroke and target vessel revascularization (TVR). The secondary endpoint was all-cause mortality. Death that could not be attributed to a noncardiac etiology was considered cardiac death. MI was de ned by the Third Universal De nition [20]. TVR was de ned as revascularization for a new lesion on the target vessel either by PCI or by surgery [21].Patients were evaluated at 1, 6 and 12 months postoperatively and annually thereafter for up to 5 years. Clinical follow-up was performed through examination of hospital records, telephone follow-up and outpatient clinical visit by research coordinators.

Statistical analysis
Categorical variables were compared with Chi-square test or Fisher's exact test, where applicable, and data were presented as frequencies and percentages. Continuous variables were tested using Student's ttest and were summarized as the mean ± standard deviation. The cumulative incidence of clinical outcomes was calculated by Kaplan-Meier analysis and compared using log-rank test. Covariates that were signi cant on univariate analysis (P < 0.10) or clinically relevant were included in multivariate models. Cox regression was used to compare adjusted hazard ratios based on age, eGFR, LVEF, prior stroke, prior PCI, prior MI, left anterior descending coronary artery (LAD) involvement and peripheral vascular disease (PVD) (Details available in Additional le, Table S1). Additionally, propensity score matching (PSM) analysis was constructed to adjust for any potential confounder in baseline characteristics between the two groups based on multivariable logistic regression model. The nearest neighbor matching algorithm was used for PSM via a 1: 1 matching protocol. Exploratory subgroup analysis was carried out to assess the effect of glycemic status (DM and Non-DM) on MACCE in speci c patient subsets using the same multivariable model. Cox regression analysis was also conducted to compare the DM group with non-DM group in the risk of MACCE and all-cause mortality during 2 years of follow-up. Two-tailed P value of less than 0.05 was considered as statistically signi cance. The SPSS Version 26.0 (SPSS Inc., Chicago, Illinois, USA) was used for all statistical computations.

Baseline patient characteristics
The prevalence of CTO was 9.42% in the total population. Success rate of CTO-PCI was 73.6%. Among a total of 719 selected patients with at least 1 successful CTO-PCI at least in our prospective and observational cohort, 316 (43.9%) patients had DM and 69 (21.8%) were dependent on insulin (Fig. 1).
The baseline demographic and treatment characteristics of the patients with and without DM are shown in Table 1. Notably, compared with non-diabetic patients, those in the DM group were older, had more females in gender distribution and exhibited a higher percentage of current smoking, hypertension, hyperlipidemia, lower LVEF, lower eGFR, prior PCI, prior MI and prior coronary artery bypass grafting (CABG). Previous stroke and higher level of low-density lipoprotein cholesterol were more common in the non-diabetic group. We did not observe a signi cant difference in the choice of baseline medication.
Angiographic and procedural characteristics of the patients are shown in Table 2. Patients in the diabetes group more often had LAD involvement, one CTO lesion, severe calci cation, angulation > 45° and multivessel disease. However, SYNTAX score and J-CTO score between the two groups were similar. After performing propensity score matching for the enrolled patients, 289 matched pairs of patients were created and we did not nd considerable differences in the baseline clinical and lesion characteristics between the two matched groups (Table 1,2).   (Fig. 2). Through multivariate analysis, we found that the MACCE risk was signi cantly higher in the diabetic patients compared to the nondiabetic patients (adjusted HR 1.47, 95% CI 1.08-2.00, P = 0.013). However, the occurrence of all-cause mortality (adjusted HR 2.37, 95% CI 0.94-5.98, P = 0.068) was not signi cantly different between the diabetic and non-diabetic groups (Table 3). and multivariable analyses showed that the risk for the primary and secondary clinical outcomes was similar between the two matched group after PSM (Table 4). Additionally, after adjustment of underlying confounding factors using the same method of previous Cox regression analysis, we did not nd signi cant difference between the two groups in the riske of MACCE (adjusted HR 1.37, 95% CI 0.93-2.03, P = 0.106) and all-cause mortality (adjusted HR 1.14, 95% CI 0.28-4.63, P = 0.849) at 2 years (Details available in Additional le, Table S2).
Post-hoc subgroup analysis showed no signi cant interactions following MACCE between those covariates (age, sex, hypertension, hyperlipidemia, LVEF and SYNTAX score, all P for interaction > 0.05) and patients' glycemic status (Fig. 3).

Discussion
We assessed the 5-year cardiovascular survival of successful CTO-PCI patients with or without DM in a large-scale, prospective and real-world cohort population. Notably, we con rmed the following: (1) Diabetic patients with successful recanalization for CTO lesions are highly prone to lower LVEF, compared with non-diabetic patients.
(2) Non-diabetic patients were related to better long-term survival bene t in terms of MACCE for the treatment of successful CTO-PCI.
With substantial and signi cant improvement in interventional devices and techniques, CTO-PCI has emerged as an effective revascularization strategy with high success rates for diabetic patients.
Moreover, it is well-established that DM represents an important risk equivalent of CTO and an independent factor for increased MACE after CTO-PCI [22,23]. Sanguineti et al. reported that DM was a signi cant predictor of cardiac mortality in CTO patients [24]. Additionally, Yan et al. found that both successful CTO-PCI and CTO-CABG of right coronary artery in diabetic patients showed signi cant reduction of all-cause death (HR 0.445, 95% CI 0.278-0.714) during long-term follow-up [25]. Recently, Guo et.al also reported that in DM group, successful CTO-PCI reduced MACE risk (HR 0.61, 95% CI 0.42-0.87, P = 0.005) compared to optimal medical therapy alone [26]. Likewise, Tsai et al. also found that DM was associated with poor prognosis in patients with CTO lesions compared with non-DM [15]. Moreover, this study also showed that successful CTO-PCI was independently associated with reduced risks of allcause death and adverse cardiovascular events only in DM population, but not in non-DM patients, which was consistent with the nding of Guo and co-workers [26]. These evidences highlighted the unfavorable role of DM in CTO patients and the importance of complete recanalization of CTO patients with DM.
Contrary to the results of previous ndings, subgroup analysis of the randomized COURAGE trial demonstrated that there was no obvious difference in the incidence of adverse events between the medical therapy group and the PCI group in DM patients with stable coronary disease [27]. This difference may be explained by the high rate (approximately 30%) of crossover from medication to revascularization during the follow-up period, which may underestimate the actual effect of successful CTO-PCI.
Considerable evidence has demonstrated that the existence of DM has a detrimental effect on glucose and lipid metabolism, endothelial function and angiogenesis, leading to premature development and progression of coronary artery atherosclerosis, inadequate collateral development and harmful clinical outcomes [28][29][30]. Previous studies have showed that well-established collateral circulation after CTO is crucial to supply the downstream perfusion area, alleviate myocardial damage, reduce infarct size and eventually improve LVEF [31,32]. Our study found that DM patients with successful CTO-PCI were more likely to have lower LVEF, which may be related to poor coronary collateral circulation. However, recently,  [34]. Likewise, consistent with Guo and co-workers [26], our study also reported that the rates of MACCE after successful CTO-PCI were higher in diabetic patients than in non-diabetic patients. In contrast, Ruiz Garcia et al. reported that in patients who underwent successful revascularization of CTO comparable rate of MACE was observed between the diabetic and non-diabetic patients in the drug-eluting stent era [35]. Although this was a prospective randomized clinical study, the atypical de nition of CTO (occlusion longer than 2 weeks), the small sample size of its enrolled patients (75 diabetic and 132 non-diabetic patients) and the modest follow-up period of 1 year restricted the accuracy of the results. In our study, we also found that the prevalence of 2-year (shorter term) clinical outcomes was comparable between the diabetic patients and non-diabetic patients, which was consistent with the ndings of Ruiz Garcia and co-workers. Thus, it is necessary to evaluate longer term prognosis for diabetic patients undergoing successful CTO-PCI.
To date, there is a paucity of data on the bene t of successful PCI for CTOs in diabetic and non-diabetic patients on long-term survival. Our ndings can be explained by the following mechanisms. First, patients in DM group were more likely to have complex clinical characteristics, like lower LVEF, which had a detrimental effect on cardiac function [10,14]. Second, DM, as a greater risk factor for adverse cardiovascular outcomes, alters glucose and lipid metabolism and in uences vascular endothelial function and angiogenesis [28,29,33]. Third, poor collateral circulation and microcirculation in diabetic patients may also partially account for the worse long-term prognosis, compared with non-diabetic patients [31,33,36].
Our study had some inevitable limitations. First, it was a single-center, prospective and observational study. Although we performed propensity score matching to reduce potential selection bias and minimize the confounding factors, unadjusted confounders still existed. Second, our real-world cohort comprised of CAD patients, who had been consecutively enrolled in our study and underwent PCI. This was not a specialized CTO cohort, so we are short in sample size of CTO-PCI patients. Third, there was a lack of speci c information in our database, such as coronary collateral scoring. Fourth, our center was a tertiary medical hospital which performed high volume of CTO-PCI and had many experienced cardiologists.
Generalizability might be limited in less experienced center with lower number of CTO-PCI cases.

Conclusions
The present study suggests that diabetic patients with successful CTO-PCI encountered more long-term adverse clinical outcomes, based on their complex lesions and co-morbidities. After a successful CTO-PCI, non-diabetic patients were associated with better long-term survival bene t in terms of MACCE. These ndings may provide clinical insight into treatment option for unselected patients with diabetes.
Further randomized controlled trials with longer term follow-up are required to validate our results.

Declarations
Ethics approval and consent to participate The study was conducted in accordance with the Declaration of Helsinki and was approved by the local ethics committee of the Fuwai hospital's Research Ethics Committee. The Institutional Review Board approved the study protocol and all of the participants provided written informed consent.

Consent for publication
Page 15/21 The manuscript was approved by all authors for publication.

Availability of data and materials
Due to ethical restrictions related to the consent given by subjects at the time of study commencement, our datasets are available from the corresponding author upon reasonable request after permission of the Institutional Review Board of State Key Laboratory of Cardiovascular Disease, Fuwai Hospital, National Center for Cardiovascular Diseases.

Competing interests
The authors declare that they have no con ict of interest.