Clinical outcomes of drug-eluting stents versus bare-metal stents in patients with cardiogenic shock complicating acute myocardial infarction

    Clinical outcomes of drug-eluting stents versus bare-metal stents in patients with cardiogenic shock complicating acute myocardial infarction Dong-Yi Chen, Chun-Tai Mao, Ming-Lung Tsai, Shao-Wei Chen, YuSheng Lin, I-Chang Hsieh, Ming-Jui Hung, Chao-Hung Wang, Ming-Shien Wen, Wen-Jin Cherng, Tien-Hsing Chen PII: DOI: Reference:

S0167-5273(16)30689-1 doi: 10.1016/j.ijcard.2016.04.014 IJCA 22380

To appear in:

International Journal of Cardiology

Received date: Accepted date:

3 February 2016 2 April 2016

Please cite this article as: Chen Dong-Yi, Mao Chun-Tai, Tsai Ming-Lung, Chen ShaoWei, Lin Yu-Sheng, Hsieh I-Chang, Hung Ming-Jui, Wang Chao-Hung, Wen Ming-Shien, Cherng Wen-Jin, Chen Tien-Hsing, Clinical outcomes of drug-eluting stents versus baremetal stents in patients with cardiogenic shock complicating acute myocardial infarction, International Journal of Cardiology (2016), doi: 10.1016/j.ijcard.2016.04.014

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Clinical outcomes of drug-eluting stents versus bare-metal stents in patients with

IP

T

cardiogenic shock complicating acute myocardial infarction

SC R

Authors: Dong-Yi Chen1, MD; Chun-Tai Mao2, MD; Ming-Lung Tsai1, MD; Shao-Wei Chen3, Yu-Sheng Lin4, MD; I-Chang Hsieh1, MD; Ming-Jui Hung2, MD, PhD; Chao-Hung

MA

NU

Wang2, MD; Ming-Shien Wen1, MD; Wen-Jin Cherng2, MD; Tien-Hsing Chen2, MD

Institution and Affiliations:

D

Division of Cardiology, Department of Internal Medicine, Chang Gung Memorial Hospital,

TE

1

2

CE P

Chang Gung University College of Medicine, Taoyuan, Taiwan Division of Cardiology, Department of Internal Medicine, Chang Gung Memorial Hospital,

3

AC

Keelung, Taiwan, and Chang Gung University College of Medicine, Taoyuan, Taiwan Department of Cardiothoracic and Vascular Surgery, Chang Gung Memorial Hospital, Chang

Gung University College of Medicine, Taoyuan, Taiwan 4

Division of Cardiology, Department of Internal Medicine, Chang Gung Memorial Hospital,

Chiayi, Taiwan, and Chang Gung University College of Medicine, Taoyuan, Taiwan

Corresponding author: Tien-Hsing Chen, MD, No.222, Maijin Road, Keelung City 204, Taiwan 1

ACCEPTED MANUSCRIPT

T

Tel: +886-3281200-8116; E-mail: [email protected]

IP

Disclosures The authors report no financial relationships or conflicts of interest regarding the

SC R

content herein.

NU

Key words: drug-eluting stent; bare-metal stent; acute myocardial infarction; cardiogenic

MA

shock.

AC

CE P

TE

D

Short title: Stents in cardiogenic shock patients

2

ACCEPTED MANUSCRIPT Abstract

T

Background

IP

To investigate the cardiovascular (CV) outcomes of drug-eluting stents (DESs) versus

SC R

bare-metal stents (BMSs) in patients with acute myocardial infarction (AMI) complicated by cardiogenic shock (CS).

NU

Methods

MA

Data from the Taiwan National Health Insurance Research Database was analyzed between January 1, 2007 and December 31, 2011. A total of 3,051 AMI patients in CS were selected as

TE

D

the study cohort. Their clinical outcomes were evaluated by comparing 1,017 subjects who

Results

CE P

used DESs to 2,034 matched subjects who used BMSs.

AC

The risk of the primary composite outcome (i.e., death, myocardial infarction, stroke, and coronary revascularization) was significantly lower in the DES group compared with the BMS group [56.1% vs. 66.2%, hazard ratio (HR), 0.74; 95% CI, 0.67–0.81] with a mean follow-up of 1.35 years. The patients who received DESs had a lower risk of coronary revascularization (HR, 0.78; 95% CI, 0.67–0.91) and death (HR, 0.70; 95% CI, 0.62–0.79) than those who used BMSs. However, the risks of myocardial infarction (HR, 0.89; 95% CI, 0.66–1.18), ischemic stroke (HR, 1.02; 95% CI, 0.67–1.53) and major bleeding (HR, 0.80; 95% CI, 0.56–1.14) were similar between the two groups. 3

ACCEPTED MANUSCRIPT Conclusions

T

Among patients with CS complicating AMI, DES implantation significantly reduced the risk

AC

CE P

TE

D

MA

NU

SC R

IP

of percutaneous coronary revascularization and death compared to BMS implantation.

4

ACCEPTED MANUSCRIPT 1. Introduction

T

Cardiogenic shock (CS) is the most common cause of death among patients hospitalized for

IP

acute myocardial infarction (AMI), with an incidence of approximately 2.5% - 8%. (1)

SC R

Despite advances in medical treatment and revascularization techniques, including percutaneous coronary intervention (PCI) or coronary artery bypass grafting (CABG), CS

MA

NU

remains associated with a high mortality ranging from 22% up to 88%.(2-4)

Early revascularization therapy in CS has been shown to be superior to initial medical

TE

D

stabilization with a significant 12.8% absolute risk reduction for 6-month mortality and a

CE P

13.2% risk reduction for 6-year mortality according to the randomized SHOCK (Should We Emergently Revascularize Occluded Coronaries for Cardiogenic Shock trial.(5, 6) A recent

AC

study also reported improved in-hospital mortality among CS patients managed invasively when compared with those managed conservatively. (7) However, whether the use of the drug-eluting stent (DES) is more beneficial compared with the bare-metal stent (BMS) for patients with CS complicating AMI is unclear, with limited and inconsistent results being reported in different studies. (8, 9)

DES has been reported to be more efficacious than BMS in reducing mortality rates (10, 11) or repeat revascularization rates (12-14) among AMI patients; however, these studies enrolled 5

ACCEPTED MANUSCRIPT only a small proportion of patients with CS. Jaguszewski et al.(9) conducted a single center

T

study which suggested a reduced all-cause mortality among CS patients treated with DESs

IP

compared with BMSs. On the other hand, Champion et al.(8) strongly suggested that DES

SC R

should not be the treatment of choice in patients with CS because of a trend toward a higher mortality after DES implantation in this patient population. As a result, there is an ongoing

NU

debate concerning the cardiovascular (CV) benefits of DES vs. BMS treatment in this high

MA

risk group of patients.

TE

D

Given the current controversy over the benefits of DESs compared with BMSs among CS

CE P

patients, a nationwide cohort study was conducted to evaluate the efficacy and safety of DESs vs. BMSs with respect to CV outcomes including mortality, myocardial infarction,

AC

stroke, coronary revascularization, and major bleeding in patients with CS complicating AMI.

2. Methods

2.1 Data source This nationwide population-based cohort study was conducted using the National Health Insurance Research Database (NHIRD) released by the Taiwan National Health Research Institute (NHRI). The data of NHIRD contain registration files and original claim data submitted by medical institutions that seek reimbursement through the NHI program. The 6

ACCEPTED MANUSCRIPT NHIRD has been described in previous studies.(15, 16) Briefly, the NHI program covers the

T

medical needs of 99.9% of the population in Taiwan (about 23.20 million in 2012). All

IP

clinical diagnoses have been recorded according to the International Classification of

SC R

Diseases, Ninth Revision, Clinical Modification [ICD-9-CM] codes. The accuracy of the diagnoses of major diseases in the claims database such as myocardial infarction, chronic

NU

kidney disease, or stroke has been validated.(17-19) The records and information of the

MA

patients were de-identified prior to analysis to ensure patient anonymity. This study was

TE

D

approved by the Ethics Institutional Review Board of Chang Gung Memorial Hospital.

CE P

2.2 Study group and cohort definition All patients in the NHIRD who were admitted for AMI (ICD-9-CM code 410) were identified

AC

between January 1, 2007 and December 31, 2011. Only patients with CS who received coronary intervention were included in this study. CS was defined as: (1) the use of dopamine doses > 880 mg; (2) the use of norepinephrine > 88 mg; (3) the combined use of dopamine and norepinephrine; or (4) the use of an intra-aortic pump to stabilize hemodynamics. In the study of Intra-aortic Balloon Support for Myocardial Infarction with Cardiogenic Shock, (20) the median dosage of dopamine was approximately 4.1~4.2 μg/kg per minute and 0.3~0.4 μg/kg per minute of norepinephrine. The definition of catecholamine dosage in the current study was, therefore, approximately 5μg/kg per minute for a 60 kg adult for 2 days 7

ACCEPTED MANUSCRIPT

T

(dopamine) and 0.5μg/kg per minute for a 60 kg adult for 2 days (norepinephrine).

IP

The index hospitalization was defined as the date when the patient was admitted for AMI. The

SC R

follow-up period was defined as the time from the index hospitalization to the date of death, loss to follow-up, or until December 31, 2011, whichever occurred first. AMI patients were

NU

classified into a DES or BMS group according to the type of stent they received. Patients who

MA

used sirolimus, everolimus, zotarolimus, biolimus, tacrolimus, or paclitaxel stents were defined as the DES group while those who received BMS stents were defined as the BMS

CE P

TE

D

group.

Patients were excluded if they met any of the following criteria: (1) they received stent

AC

implantation before index admission; (2) they received more than one type of stent (i.e., both DES and BMS) during the coronary intervention; (3) no stent implantation was performed (Fig. 1).

2.3 Outcomes and covariate measurements The primary outcome of major adverse CV events was a composite of myocardial infarction, coronary revascularization, stroke, and death during the follow-up period. Coronary revascularization was defined as percutaneous coronary revascularization or coronary artery 8

ACCEPTED MANUSCRIPT bypass surgery. Stroke included ischemic, hemorrhagic or unspecified stroke. Death and

T

causes of death were identified according to the registry data of the NHIRD. Secondary

IP

outcomes evaluated included heart failure on admission and major bleeding. The definition of

SC R

major bleeding has been described previously and listed in the Appendix Table.(21)

NU

2.4 Propensity score matching

MA

Propensity score matching (PSM) was conducted to minimize selection bias. The PSM matched each patient who received DESs with two patients who received BMSs according to

TE

D

propensity score using a nearest-neighbor matching algorithm. The propensity score was

CE P

defined as the predicted probability, given the covariates, of being designated as the treated group (DESs) obtained from the logistic regression. We chose the following covariates to

AC

calculate the propensity score: patient’s characteristics (such as age, sex, history of MI, stroke, or peripheral arterial disease); baseline comorbidities (such as diabetes, hypertension, dyslipidemia, coronary artery disease, heart failure, chronic kidney disease, dialysis, atrial fibrillation, gout, chronic obstructive pulmonary disease, or malignancy); angiographic and procedural characteristics [such as number of stented diseased vessels, number of stents implanted, aspiration catheter use, intra-aortic balloon pump (IABP) use, intubation or venoarterial extracorporeal membrane oxygenators (ECMO) use]; medications administered at discharge , as well as the indexed year and month (Table 1). The matching procedure was 9

ACCEPTED MANUSCRIPT

T

performed using SAS Version 9.3 (SAS Institute, Cary, NC).

IP

2.5 Statistical analysis

SC R

The clinical characteristics between the study groups (DES and BMS groups) were compared using the chi-square test for categorical variables and the independent sample t-test for

NU

continuous variables. The risk of time to event for a primary or secondary outcome after the

MA

index hospitalization was compared between study groups using Cox proportional hazard model with adjustment based on the propensity score. The cumulative incidence of the

TE

D

primary composite outcome and its components for each study group was depicted and

CE P

compared using the log-rank test. All data analyses were performed using SPSS 22 (IBM

AC

SPSS, Armonk, NY: IBM Corp).

3. Results

3.1 Patient characteristics There were 9,854 AMI patients with CS who received coronary intervention from January 2007 to December 2011. After relevant exclusion and propensity score matching, a total of 3,051 patients with CS complicating AMI were identified for data analysis (Fig. 1).

The average age of the overall cohort was 68.6 years (SD = 13.3 years) and the ages ranged 10

ACCEPTED MANUSCRIPT from 22.7 to 101 years. The mean follow-up period was 1.35 years (SD = 1.42 years) with a

T

maximum of 5.0 years. Of these 3,051 patients, 1,017 (33.3%) received DES implantation

IP

and 2,034 matched comparison patients (67.7%) received BMS implantation. Among those

SC R

who received DESs, 119 (10.6%) received sirolimus-eluting stents, 183 (16.4%) received everolimus-eluting stents, 407 (36.4%) received zotarolimus-eluting stents, 66 (5.9%)

NU

received biolimus-eluting stents, 297 (26.5%) received paclitaxel-eluting stents and 47

MA

patients (4.2%) received mixed type of DESs. No difference in the distribution of the baseline characteristics or comorbidities was noted between study groups after PSM matching (Table

CE P

TE

D

1).

In total, 44.6% of the DES group and 44.8% of the BMS group received IABP, and 35.2% of

AC

the DES group and 33.8% of the BMS group were intubated. There were no significant differences in the use of medication at discharge which included aspirin, clopidogrel, angiotensin-converting enzyme inhibitors (ACEIs)/angiotensin receptor blockers (ARBs), statin, proton-pump inhibitors (PPIs), and calcium channel blockers between the DES and BMS groups. (Table 1).

3.2 Cardiovascular outcomes The risk of the primary composite outcome (i.e., death, myocardial infarction, stroke, and 11

ACCEPTED MANUSCRIPT coronary revascularization) was significantly lower in the DES group (571 patients, 56.1%)

T

compared with the BMS group (1347 patients, 66.2%) (hazard ratio [HR], 0.74; 95% CI,

IP

0.67–0.81) after a mean follow-up period of 1.35 years (Table 2, Fig. 2A). The risk of major

SC R

CV events was significantly lower in the DES group than in the BMS group during the first year of follow-up (P < 0.001), whereas no group difference in risk was observed during the

MA

NU

next year to the end of follow-up (P = 0.875) (Fig. 2B).

In terms of secondary outcomes, the patients who received DESs had a lower risk of coronary

TE

D

revascularization and death with HRs of 0.78 (95% CI, 0.67–0.91) and 0.70 (95% CI,

CE P

0.62–0.79), respectively, compared with those who received BMSs (Table 3). The reduced risks of coronary revascularization and death in the DES group were limited primarily to

AC

within 1 year of follow-up (Fig. 3A, 3B). In contrast, no beneficial effect of DESs on MI and stroke was observed during either the 1 year of follow-up, or the next year to the end of follow-up (Fig. 3C, 3D).

There were no significant differences in the risks of MI (HR, 0.89; 95% CI, 0.66–1.18), stroke (HR, 1.03; 95% CI, 0.71–1.50), heart failure admission (HR, 1.12; 95% CI, 0.86–1.46), or major bleeding (HR, 0.80; 95% CI, 0.56–1.14) between the DES and BMS groups (Table 3). 12

ACCEPTED MANUSCRIPT

T

Subgroup analysis revealed that the beneficial effect of DESs on the primary composite

IP

outcome did not differ in terms of age, previous history of diabetes mellitus, heart failure,

SC R

CKD, dialysis, atrial fibrillation, or IABP support during the index admission (Fig. 4).

NU

4. Discussion

MA

In this large, real-world study evaluating AMI patients with CS undergoing PCI, we found that DESs were associated with a decreased risk of major CV events, particularly less

TE

D

coronary revascularization and lower mortality rate compared with BMSs. Secondary

CE P

outcome analysis demonstrated no significant differences in the risk of major bleeding or

AC

ischemic stroke between the two study groups.

Our results revealed a considerably high mortality in CS patients receiving PCIs, with nearly two in five (41.1%) patients dying within 3 years with an annualized death rate of 30.2%. The primary composite CV event rate was > 60% during a mean follow-up period of 1.35 years. Despite these high adverse event rates, we found a significant reduction in risks of both major CV events and all-cause mortality associated with the use of DESs compared with BMSs in CS patients. This resulted in a number needed to treat of 9.9 for major CV events and 9.3 for all-cause mortality. Importantly, we did not detect significant safety outcome differences such 13

ACCEPTED MANUSCRIPT as major bleeding, GI bleeding, or hemorrhagic stroke between the two different stent groups.

T

These results provide significant information that could be useful in helping clinicians

SC R

IP

determine whether a DES or BMS is preferred in treating patients with CS complicating AMI.

Controversies still exist regarding the efficacy and safety of DES compared with BMS in AMI

NU

patients with CS. Our study suggests that the reduction in major CV events is likely clinically

MA

relevant as there was a significantly lower CV event rate among the DES group after 3-months of follow-up and the benefit of risk reduction persisted at one-year of follow-up and until the

TE

D

end of the study. These findings are consistent with a previous study conducted by Jaguszewski

CE P

et al.(9) based on a smaller population of CS patients. They demonstrated that treatment with DESs compared with BMSs was associated with improved clinical outcomes, including a

AC

reduction in all-cause mortality both at 30 days and during long-term follow-up.(9)

In contrast, the superiority of DES regarding cardiovascular outcomes of mortality in patients with CS was not reported in the study by Champion et al.(8) Furthermore, they suggested that DES should not be the treatment of choice in this patient population.(8) The exact mechanism for the disparity in results between studies remains unclear. However, unlike our study (which included 5.1% of hemodialysis patients and 4.2% of ECMO use patients), 43% of the patients in the study by Champion et al.(8) received hemodialysis treatment and the use of assist 14

ACCEPTED MANUSCRIPT devices accounted for 10% of their study population. Dual antiplatelet medication was given

T

to 94% of our study population; in contrast, in the study by Champion et al.(8), the

IP

antiplatelet therapy had to be stopped for a few days in nearly half of their patients because of

SC R

major hemorrhages diagnosed in 59% of the patients. Over-all mortality was 41.1% in our study compared with approximately 50% in the study by Champion et al.(8) Therefore, the

NU

different severities of critically ill patients in that study may have affected their results and

MA

offset potential differences between the study groups.

TE

D

Our study demonstrated a significantly lower coronary revascularization at all course

CE P

follow-up with DESs implantation compared with BMSs in AMI patients with CS. The substantial reduction in coronary revascularization rate has been documented with respect to

AC

the efficacy of DESs in real world reports.(22) Notably, from the analysis in our cohort, the benefit of DESs regarding a reduction in the need for coronary revascularization was primarily contributed by the effect within one year. Our analysis also revealed a lower cumulative incidence of major CV events and death in the DES group within one year but a similar cumulative incidence from one year to end of the follow-up. We believed that DESs were more effective than BMSs in reducing the need for coronary revascularization at one year follow-up which was associated with lower major CV events and death within first year and the beneficial effect persisted until the end of the study. 15

ACCEPTED MANUSCRIPT

T

Interestingly, although the rate of coronary revascularization was lower in the DES group

IP

compared with the BMS group during the first year, this benefit was lost after one year, with

SC R

a higher rate of revascularization in the DES group during the second year to the end of study. This crossover of rates of revascularization after the first year has also been reported for

NU

patients in the Ontario’s registry and is consistent with findings from other studies evaluating

MA

the effectiveness of DES in patients with AMI.(11, 23) The precise mechanism underlying this finding remains unclear. Previous studies have reported late complications such as very

TE

D

late stent thrombosis and late catch-up phenomenon in patients after DES implantation

CE P

although it is not possible for us to ascertain the exact rate of very late stent thrombosis because the data was not included in the NHIRD.(24, 25) In accordance with those studies,

AC

our results suggested a possible late catch-up phenomenon in the AMI patients with CS who received DES compared with those who received BMS. Further studies of these implications is warranted to elucidate this issue.

5. Study Limitations This study had several limitations including its retrospective nature. In addition, the claims database did not include personal information on tobacco use, body mass index, family history of CV disease, or lipid profile. However, we were able to include a wide range of 16

ACCEPTED MANUSCRIPT variables related to outcomes to show that our two study groups were well matched. Because

T

we collected follow-up data from NHIRD, it was not possible for us to ascertain the

IP

occurrence of stent thrombosis or target vessel revascularization through ICD-9 coding.

SC R

Finally, selection bias affecting physician decision-making regarding choice of stent type may

NU

have affected our findings, although we did adjust for multiple baseline differences.

MA

6. Conclusions

DES implantation significantly reduced the risk of major CV events, especially coronary

TE

D

revascularization and death, without increasing the risk of major bleeding or hemorrhagic

CE P

stroke compared with BMS implantation among AMI patients with CS undergoing percutaneous coronary interventions. Our results suggest that DES appears safe and more

AC

efficacious than BMS in patients with CS complicating AMI.

Acknowledgments We would like to thank Hsing-Fen Lin for his statistical assistance.

References 17

ACCEPTED MANUSCRIPT 1.

Reynolds HR, Hochman JS. Cardiogenic shock: current concepts and improving

outcomes. Circulation. 2008;117(5):686-97. Babaev A, Frederick PD, Pasta DJ, Every N, Sichrovsky T, Hochman JS. Trends in

T

2.

IP

management and outcomes of patients with acute myocardial infarction complicated by

3.

SC R

cardiogenic shock. Jama. 2005;294(4):448-54.

Goldberg RJ, Spencer FA, Gore JM, Lessard D, Yarzebski J. Thirty-year trends (1975 to

2005) in the magnitude of, management of, and hospital death rates associated with

NU

cardiogenic shock in patients with acute myocardial infarction: a population-based

4.

MA

perspective. Circulation. 2009;119(9):1211-9.

Sleeper LA, Reynolds HR, White HD, Webb JG, Dzavik V, Hochman JS. A severity scoring

system for risk assessment of patients with cardiogenic shock: a report from the SHOCK Trial

5.

TE

D

and Registry. American heart journal. 2010;160(3):443-50. Hochman JS, Sleeper LA, Webb JG, Sanborn TA, White HD, Talley JD, et al. Early

CE P

revascularization in acute myocardial infarction complicated by cardiogenic shock. SHOCK Investigators. Should We Emergently Revascularize Occluded Coronaries for Cardiogenic

6.

AC

Shock. The New England journal of medicine. 1999;341(9):625-34. Hochman JS, Sleeper LA, Webb JG, Dzavik V, Buller CE, Aylward P, et al. Early

revascularization and long-term survival in cardiogenic shock complicating acute myocardial infarction. Jama. 2006;295(21):2511-5. 7.

Bangalore S, Gupta N, Guo Y, Lala A, Balsam L, Roswell RO, et al. Outcomes with invasive

vs conservative management of cardiogenic shock complicating acute myocardial infarction. The American journal of medicine. 2015;128(6):601-8. 8.

Champion S, Gauzere BA, Lefor Y. Drug-eluting stents should not be used in ST-elevated

myocardial infarction with cardiogenic shock. Archives of internal medicine. 2012;172(20):1613-4. 9.

Jaguszewski M, Ghadri JR, Seifert B, Hiestand T, Herrera P, Gaemperli O, et al. 18

ACCEPTED MANUSCRIPT Drug-eluting stents vs. bare metal stents in patients with cardiogenic shock: a comparison by propensity score analysis. Journal of cardiovascular medicine (Hagerstown, Md).

T

2015;16(3):220-9.

IP

10. Hannan EL, Racz M, Walford G, Holmes DR, Jones RH, Sharma S, et al. Drug-eluting

SC R

versus bare-metal stents in the treatment of patients with ST-segment elevation myocardial infarction. JACC Cardiovascular interventions. 2008;1(2):129-35.

11. Mauri L, Silbaugh TS, Garg P, Wolf RE, Zelevinsky K, Lovett A, et al. Drug-eluting or

NU

bare-metal stents for acute myocardial infarction. The New England journal of medicine.

MA

2008;359(13):1330-42.

12. Di Lorenzo E, De Luca G, Sauro R, Varricchio A, Capasso M, Lanzillo T, et al. The PASEO (PaclitAxel or Sirolimus-Eluting Stent Versus Bare Metal Stent in Primary Angioplasty)

TE

D

Randomized Trial. JACC Cardiovascular interventions. 2009;2(6):515-23. 13. Stone GW, Lansky AJ, Pocock SJ, Gersh BJ, Dangas G, Wong SC, et al. Paclitaxel-eluting

CE P

stents versus bare-metal stents in acute myocardial infarction. The New England journal of medicine. 2009;360(19):1946-59.

AC

14. Chen DY, Mao CT, Tsai ML, Hsieh MJ, Lin YS, Cherng WJ, et al. Clinical Outcomes of Drug-Eluting Stents vs. Bare-Metal Stents in Acute Myocardial Infarction Patients Under Dialysis- A Nationwide Cohort Study. Circulation journal : official journal of the Japanese Circulation Society. 2016;80(2):363-70. 15. Chen DY, Wang SH, Mao CT, Tsai ML, Lin YS, Su FC, et al. Sitagliptin After Ischemic Stroke in Type 2 Diabetic Patients: A Nationwide Cohort Study. Medicine. 2015;94(28):e1128. 16. Chen DY, Wang SH, Mao CT, Tsai ML, Lin YS, Chou CC, et al. Sitagliptin and cardiovascular outcomes in diabetic patients with chronic kidney disease and acute myocardial infarction: A nationwide cohort study. International journal of cardiology. 2015;181:200-6. 17. Cheng CL, Lee CH, Chen PS, Li YH, Lin SJ, Yang YH. Validation of acute myocardial infarction cases in the national health insurance research database in taiwan. Journal of 19

ACCEPTED MANUSCRIPT epidemiology / Japan Epidemiological Association. 2014;24(6):500-7. 18. Cheng CL, Kao YH, Lin SJ, Lee CH, Lai ML. Validation of the National Health Insurance

T

Research Database with ischemic stroke cases in Taiwan. Pharmacoepidemiology and drug

IP

safety. 2011;20(3):236-42.

SC R

19. Chang CH, Lee YC, Tsai CT, Chang SN, Chung YH, Lin MS, et al. Continuation of statin therapy and a decreased risk of atrial fibrillation/flutter in patients with and without chronic kidney disease. Atherosclerosis. 2014;232(1):224-30.

NU

20. Thiele H, Zeymer U, Neumann FJ, Ferenc M, Olbrich HG, Hausleiter J, et al. Intraaortic

of medicine. 2012;367(14):1287-96.

MA

balloon support for myocardial infarction with cardiogenic shock. The New England journal

21. Tsai TT, Messenger JC, Brennan JM, Patel UD, Dai D, Piana RN, et al. Safety and efficacy

TE

D

of drug-eluting stents in older patients with chronic kidney disease: a report from the linked CathPCI Registry-CMS claims database. Journal of the American College of Cardiology.

CE P

2011;58(18):1859-69.

22. Kirtane AJ, Gupta A, Iyengar S, Moses JW, Leon MB, Applegate R, et al. Safety and

AC

efficacy of drug-eluting and bare metal stents: comprehensive meta-analysis of randomized trials and observational studies. Circulation. 2009;119(25):3198-206. 23. Tu JV, Bowen J, Chiu M, Ko DT, Austin PC, He Y, et al. Effectiveness and safety of drug-eluting stents in Ontario. The New England journal of medicine. 2007;357(14):1393-402. 24. Kastrati A, Mehilli J, Pache J, Kaiser C, Valgimigli M, Kelbaek H, et al. Analysis of 14 trials comparing sirolimus-eluting stents with bare-metal stents. The New England journal of medicine. 2007;356(10):1030-9. 25. Park KW, Kim CH, Lee HY, Kang HJ, Koo BK, Oh BH, et al. Does "late catch-up" exist in drug-eluting stents: insights from a serial quantitative coronary angiography analysis of sirolimus versus paclitaxel-eluting stents. American heart journal. 2010;159(3):446-53.e3. 20

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

21

ACCEPTED MANUSCRIPT Figure legends

T

Fig. 1. Enrollment schema of the study patients. Patients with AMI complicating cardiogenic

IP

shock were identified as the study cohort after relevant exclusions. Abbreviations: AMI, acute

SC R

myocardial infarction; DES, drug-eluting stent; BMS, bare metal stent.

NU

Fig. 2. Cumulative incidence of the primary composite endpoint of (A) overall follow-up and

MA

(B) stratified by 1-year follow-up. The primary end point was a composite of death,

TE

D

myocardial infarction, stroke and coronary revascularization.

CE P

Fig.3. Cumulative incidence of (A) revascularization, (B) all-cause mortality, (C) myocardial

AC

infarction, and (D) stroke stratified by 1-year follow-up.

Fig. 4. Subgroup analysis for primary composite endpoint. The primary end point was a composite of myocardial infarction, coronary revascularization, stroke, and death.

22

ACCEPTED MANUSCRIPT Appendix. ICD-9-CM code used for diagnosis in the current study Source

Code

Old myocardial infarction

ICD-9 CM

412.xx

Acute myocardial infarction

ICD-9 CM

410.xx

Any cerebral vascular accident

ICD-9 CM

430.xx–437.xx

Hemorrhage stroke

ICD-9 CM

430.xx–432.xx

Ischemic stroke

ICD-9 CM

433.xx–435.xx

IP

SC R

NU

MA

Unspecified stroke

ICD-9 CM

436.xx–437.xx

ICD-9 CM

440.0, 440.2x, 440.3x, 440.4, 440.9,

CE P

TE

D

Peripheral arterial disease

443.9, 444.2, 444.22, 444.8, 444.81, 445.0, 445.02, 250.7x, 707.1x ICD-9 CM

401.xx–405.xx

ICD-9 CM

272.xx

Diabetes mellitus

ICD-9 CM

250.xx

Coronary artery disease

ICD-9 CM

413.xx,414.0x

Heart failure

ICD-9 CM

428.xx

Chronic kidney disease

ICD-9 CM

585.xx

Atrial fibrillation

ICD-9 CM

427.31

Gout

ICD-9 CM

274.xx

Dyslipidemia

AC

Hypertension

T

Variable

23

ACCEPTED MANUSCRIPT Source

Code

Chronic obstructive pulmonary disease

ICD-9 CM

490.xx–496.xx

Malignancy

ICD-9 CM

140.xx–208.xx

GI bleeding

ICD-9 CM

531.xx–534.xx, 578.xx

Major bleeding

ICD-9 CM

MA

gastrointestinal tract hemarthrosis

719.1x

TE

626.2, 626.6, 626.8, 627.0, 627.1 786.3

AC

epistaxis

578.x

599.7

CE P

hematuria

430.xx–432.xx

423.0

D

hemopericardium

hemoptysis

IP

SC R

NU

intra-cerebral

vaginal bleeding

T

Variable

784.7

hemorrhage not otherwise specified

459.0

24

Fig. 1

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

25

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Fig. 2

26

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

Fig. 3

27

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

Fig. 4

28

ACCEPTED MANUSCRIPT Table 1. Baseline characteristics of the patients

T

After PSM matching BMS

SC R

DES (n=1,017)

IP

Characteristics

Age (years)

68.8±12.8

NU

Gender

731 (71.9)

MA

Male Female

P

(n=2,034)

68.5±13.5

0.609 0.713

1,449 (71.2)

286 (28.1)

585 (28.8)

448 (44.1)

899 (44.2)

0.938

122 (12.0)

240 (11.8)

0.874

153 (15.0)

308 (15.1)

0.943

73 (7.2)

147 (7.2)

0.961

23 (2.3)

41 (2.0)

0.655

Hypertension

554 (54.5)

1,136 (55.9)

0.471

Dyslipidemia

682 (67.1)

1,349 (66.3)

0.684

Previous known CAD

188 (18.5)

378 (18.6)

0.947

Heart failure

122 (12.0)

239 (11.8)

0.843

83 (8.2)

189 (9.3)

0.301

Prior stroke

CE P

Prior myocardial infarction

TE

D

Diabetes mellitus

AC

Known peripheral arterial disease Prior CABG

Comorbidities

Chronic kidney disease 29

ACCEPTED MANUSCRIPT After PSM matching Characteristics

BMS

T

DES (n=1,017)

P

52 (5.1)

104 (5.1)

1.000

Atrial fibrillation

72 (7.1)

140 (6.9)

0.840

Gout

63 (6.2)

130 (6.4)

0.833

110 (10.8)

216 (10.6)

0.868

69 (6.8)

124 (6.1)

0.462

NU

SC R

Dialysis

IP

(n=2,034)

MA

Chronic obstructive pulmonary disease

TE

D

Malignancy

1

AC

2 3

0.718

CE P

No. of stented diseased vessels

728 (71.6)

1,463 (71.9)

256 (25.2)

495 (24.3)

33 (3.2)

76 (3.7)

No. of stents implanted per patient

0.967

1

680 (66.9)

1,354 (66.6)

2

235 (23.1)

466 (22.9)

3

75 (7.4)

154 (7.6)

4 or more

27 (2.7)

60 (2.9)

203 (20.0)

403 (19.8)

Aspiration catheter used 30

0.923

ACCEPTED MANUSCRIPT After PSM matching Characteristics

BMS

T

DES (n=1,017)

P

454 (44.6)

912 (44.8)

0.918

Intubation

358 (35.2)

687 (33.8)

0.434

ECMO use

44 (4.3)

86 (4.2)

0.899

8.5±10.3

8.4±9.4

0.705

1.8±2.6

1.8±2.6

0.718

10.2±24.6

9.9±24.5

0.761

4.7±8.9

4.7±8.8

0.841

964 (94.8)

1,932 (95.0)

0.816

1,000 (98.3)

2,000 (98.3)

1.000

Dual antiplatelet

957 (94.1)

1,927 (94.7)

0.464

B-blocker

629 (61.8)

1,239 (60.9)

0.618

ACEI/ARB

712 (70.0)

1,390 (68.3)

0.347

Statin

613 (60.3)

1,209 (59.4)

0.657

PPI

273 (26.8)

532 (26.2)

0.684

NU

SC R

IABP use

IP

(n=2,034)

MA

Duration in intensive care unit (days) Dosage of inotropic medication

TE

CE P

Norepinephrine (mg)

D

Dopamine (mg×103)

Epinephrine (mg)

Aspirin

AC

Medication at discharge

Clopidogrel

31

ACCEPTED MANUSCRIPT After PSM matching Characteristics

BMS

T

DES (n=1,017)

P

294 (28.9)

SC R

Calcium channel blocker

IP

(n=2,034)

Index admission duration (days)

16.4±18.8

568 (27.9)

0.570

16.0±17.9

0.639

AC

CE P

TE

D

MA

NU

Abbreviations: ACEI , angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; BMS, bare-metal stent; CABG, coronary artery bypass graft; CAD, coronary artery disease; DES, drug-eluting stent; ECMO, extracorporeal membrane oxygenation; IABP, intra-aortic balloon pump; PPI, proton-pump inhibitor.

32

ACCEPTED MANUSCRIPT Table 2. Primary outcomes over various follow-up periods

T

Number of events (%) BMS

DES vs. BMS

Primary composite endpoint§

(n = 1,017)

(n = 2,034)

3-month follow-up

344 (33.8)

872 (42.9)

0.74 (0.65–0.83)

<0.001

6-month follow-up

423 (41.6)

1,064 (52.3)

0.73 (0.65–0.81)

<0.001

1-year follow-up

485 (47.7)

1,219 (59.9)

0.71 (0.64–0.79)

<0.001

At the last follow-up

571 (56.1)

0.74 (0.67–0.81)

<0.001

MA

NU

SC R

IP

DES

1,347 (66.2)

HR (95% CI) †

D

Abbreviations: BMS, bare-metal stent; DES, drug-eluting stent; HR, hazard ratio; CI,

AC

CE P

TE

confidence interval. † Adjusted by propensity score. § Any one of the following: myocardial infarction, any cerebral vascular accident, revascularization, and all-cause mortality.

33

P

ACCEPTED MANUSCRIPT Table 3. Secondary outcomes at the last follow-up

T

Number of events (%) BMS

(n = 1,017)

(n = 2,034)

HR (95% CI)†

P

Myocardial infarction

69 (6.8)

140 (6.9)

0.89 (0.66–1.18)

0.409

Revascularization

228 (22.4)

493 (24.2)

0.78 (0.67–0.91)

0.002

PCI

208 (20.5)

445 (21.9)

0.79 (0.67–0.94)

0.006

CABG

31 (3.0)

65 (3.2)

0.87 (0.57–1.33)

0.520

346 (34.0)

909 (44.7)

0.70 (0.62–0.79)

<0.001

88 (8.7)

142 (7.0)

1.12 (0.86–1.46)

0.413

43 (4.2)

73 (3.6)

1.03 (0.71–1.50)

0.888

Ischemic stroke

36 (3.5)

62 (3.0)

1.02 (0.67–1.53)

0.937

Hemorrhagic stroke

7 (0.7)

9 (0.4)

1.34 (0.50–3.61)

0.559

Unspecified stroke

1 (0.1)

9 (0.4)

0.19 (0.02–1.52)

0.118

Major bleeding

44 (4.3)

97 (4.8)

0.80 (0.56–1.14)

0.213

GI bleeding

87 (8.6)

156 (7.7)

0.99 (0.76–1.28)

0.920

CE P

Death from any cause

TE

D

MA

NU

CV outcome

Heart failure admission

SC R

Outcome

DES vs. BMS

IP

DES

Any CVA

AC

Neurologic outcome

Safety outcome

34

ACCEPTED MANUSCRIPT Major or GI bleeding

98 (9.6)

202 (9.9)

0.85 (0.67–1.08)

0.189

AC

CE P

TE

D

MA

NU

SC R

IP

T

Abbreviations: BMS, bare-metal stent; CABG, coronary artery bypass graft; DES, drug-eluting stent; CV, cardiovascular; PCI, percutaneous coronary intervention; CVA = cerebral vascular accident; GI, gastrointestinal; HR, hazard ratio; CI, confidence interval. † Adjusted by propensity score

35