J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S VO L . 1 0 , N O . 2 , 2 0 1 7

ª 2 0 1 7 B Y T H E AM E R I C A N C O L L E G E O F C A R D I O L O G Y F O U N D A T I O N

P U B L I S H E D B Y E L S E V I E R

I S S N 1 9 3 6 - 8 7 9 8 / $ 3 6 . 0 0

h t t p : / / d x . d o i . o r g / 1 0 . 1 0 1 6 / j . j c i n . 2 0 1 6 . 1 0 . 0 1 8

CORONARY

The ReACT TrialRandomized Evaluation of Routine Follow-upCoronary Angiography After Percutaneous CoronaryIntervention Trial

Hiroki Shiomi, MD,a Takeshi Morimoto, MD, PHD,b Shoji Kitaguchi, MD,c Yoshihisa Nakagawa, MD,d

Katsuhisa Ishii, MD,e Yoshisumi Haruna, MD, PHD,c Itaru Takamisawa, MD,f Makoto Motooka, MD,g

Kazuhiro Nakao, MD,h Shintaro Matsuda, MD,i Satoru Mimoto, MD,j Yutaka Aoyama, MD,k Teruki Takeda, MD,l

Koichiro Murata, MD,m Masaharu Akao, MD,n Tsukasa Inada, MD,o Hiroshi Eizawa, MD,i Eiji Hyakuna, MD,p

Kojiro Awano, MD,q Manabu Shirotani, MD,r Yutaka Furukawa, MD,s Kazushige Kadota, MD,t

Katsumi Miyauchi, MD,u Masaru Tanaka, MD,o Yuichi Noguchi, MD,v Sunao Nakamura, MD,j Satoshi Yasuda, MD,h

Shunichi Miyazaki, MD,w Hiroyuki Daida, MD,u Kazuo Kimura, MD,x Yuji Ikari, MD,y Haruo Hirayama, MD, PHD,k

Tetsuya Sumiyoshi, MD,f Takeshi Kimura, MD,a on behalf of the ReACT Investigators

ABSTRACT

Fro

Ep

dio

ba

Ge

OBJECTIVES The purpose of this study was to evaluate long-term clinical impact of routine follow-up coronary

angiography (FUCAG) after percutaneous coronary intervention (PCI) in daily clinical practice in Japan.

BACKGROUND The long-term clinical impact of routine FUCAG after PCI in real-world clinical practice has not been

evaluated adequately.

METHODS In this prospective, multicenter, open-label, randomized trial, patients who underwent successful PCI were

randomly assigned to routine angiographic follow-up (AF) group, in which patients were to receive FUCAG at 8 to 12

months after PCI, or clinical follow-up alone (CF) group. The primary endpoint was defined as a composite of death,

myocardial infarction, stroke, emergency hospitalization for acute coronary syndrome, or hospitalization for heart failure

over a minimum of 1.5 years follow-up.

RESULTS Between May 2010 and July 2014, 700 patients were enrolled in the trial among 22 participating centers

and were randomly assigned to the AF group (n ¼ 349) or the CF group (n ¼ 351). During a median of 4.6 years of

follow-up (interquartile range [IQR]: 3.1 to 5.2 years), the cumulative 5-year incidence of the primary endpoint was

22.4% in the AF group and 24.7% in the CF group (hazard ratio: 0.94; 95% confidence interval: 0.67 to 1.31; p ¼ 0.70).

Any coronary revascularization within the first year was more frequently performed in AF group than in CF group (12.8%

vs. 3.8%; log-rank p < 0.001), although the difference between the 2 groups attenuated over time with a similar cu-

mulative 5-year incidence (19.6% vs. 18.1%; log-rank p ¼ 0.92).

CONCLUSIONS No clinical benefits were observed for routine FUCAG after PCI and early coronary revasculari-

zation rates were increased within routine FUCAG strategy in the current trial. (Randomized Evaluation of Routine

Follow-up Coronary Angiography After Percutaneous Coronary Intervention Trial [ReACT]; NCT01123291)

(J Am Coll Cardiol Intv 2017;10:109–17) © 2017 by the American College of Cardiology Foundation.

m the aDepartment of Cardiovascular Medicine, Graduate School of Medicine, Kyoto University; bDepartment of Clinical

idemiology, Hyogo College of Medicine, Hyogo; cDivision of Cardiology, Hirakata Kohsai Hospital, Hirakata; dDivision of Car-

logy, Tenri Hospital, Nara; eDivision of Cardiology, Kansai Electric Power Hospital, Osaka; fDepartment of Cardiology, Sakaki-

ra Heart Institute, Japan Research Promotion Society for Cardiovascular Diseases, Tokyo; gDivision of Cardiology, Shizuoka

neral Hospital, Shizuoka; hDepartment of Cardiovascular Medicine, National Cerebral and Cardiovascular Center, Osaka;

iDivision o

CardiologymDepartme

tion Kyoto

Shimonose

diology, Ki

pital, Kobe

University

of CardioloyDepartme

Production

ration. All

Manuscrip

ABBR EV I A T I ON S

AND ACRONYMS

ACS = acute coronary

syndrome(s)

AMI = acute myocardial

infarction

DES = drug-eluting stent(s)

FUCAG = follow-up coronary

angiography

HF = heart failure

IQR = interquartile range

MI = myocardial infarction

PCI = percutaneous coronary

intervention

TLR = target lesion

revascularization

Shiomi et al. J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 1 0 , N O . 2 , 2 0 1 7

Routine Follow-up Coronary Angiography After PCI J A N U A R Y 2 3 , 2 0 1 7 : 1 0 9 – 1 7

110

I n several previous studies, routinefollow-up coronary angiography (FUCAG)after percutaneous coronary interven-

tion (PCI) increased the rate of coronaryrevascularization, but did not improve clin-ical outcomes (1–4). Based on these study re-sults, the current clinical guidelines in theUnited States have already disregardedroutine FUCAG, even after PCI for left maincoronary artery disease, whereas the currentclinical guidelines in Europe regarded routineFUCAG after high-risk PCI as Class IIb (5,6).However, previous studies in the drug-eluting stents (DES) era were conducted inthe context of pivotal randomized trials ofDES and there have been no randomized clin-

ical trial evaluating long-term clinical impact ofroutine FUCAG after PCI in the real-world clinicalpractice, including in high-risk patients for cardiovas-cular events risk such as complex coronary artery dis-ease and acute myocardial infarction (AMI)presentation (3,7,8). The current randomized clinicaltrial, therefore, was conducted to evaluate the long-term clinical impact of routine FUCAG after PCI inreal-world clinical practice in Japan, where routineFUCAG after PCI is still performed commonly as usualcare (4,9,10).

SEE PAGE 118

METHODS

STUDY DESIGN AND PATIENT SELECTION. TheReACT (Randomized Evaluation of Routine Follow-upCoronary Angiography After Percutaneous CoronaryIntervention Trial) is a prospective, multicenter,open-label, randomized trial comparing the routineangiographic follow-up (AF) strategy with the clinicalfollow-up (CF) strategy in daily clinical practice inJapan. In this all-comer design trial, patients whounderwent successful PCI without planned staged PCIwere enrolled from 22 participating centers (List A in

f Cardiology, Nishikobe Medical Center, Kobe; jDepartment of Card

, Nagoya Second Red Cross Hospital, Nagoya; lDivision of C

nt of Cardiology, Shizuoka City Shizuoka Hospital, Shizuoka; nDe

Medical Center, Kyoto; oCardiovascular Center Osaka Red Cross H

ki General Hospital, Yamaguchi; qDepartment of Cardiology, Kita

ndai University Nara Hospital, Nara; sDepartment of Cardiovascul

; tDivision of Cardiology, Kurashiki Central Hospital, Kurashiki;

, Graduate School of Medicine, Tokyo; vDepartment of Cardiology, T

gy, Kindai University, Osaka; xDivision of Cardiology, Yokohama

nt of Cardiology, Tokai University, Kanagawa, Japan. Supported b

Development (Kyoto, Japan). Dr. Morimoto received honoraria fo

other authors have reported that they have no relationships releva

t received October 16, 2016; accepted October 20, 2016.

the Online Appendix) without any exclusion criteria.The study protocol was approved by the institutionalreview board at each participating center. Writteninformed consent was obtained from all the studypatients. The trial was registered with NCT01123291.

STUDYPROCEDURES. Patients were randomly allocatedin a 1:1 ratio to the routine AF group or CF group.Randomization was performed before hospitaldischarge after the index PCI and stratified by centersandbare-metal stent use. In theAFgroup, patientswerescheduled to undergo routine FUCAG at 8 to 12 monthsafter the index PCI, whereas in the CF group, patientswere scheduled to receive CF without routine FUCAG.During follow-up, any physiological stress tests such astreadmill exercise test or stress nuclear study wereallowed to be performed, but coronary computed to-mography angiography was not allowed in eithergroup. Clinically indicated coronary angiographicstudies, such as those for acute coronary syndrome(ACS), for recurrence of angina, and/or for objectiveevidence of myocardial ischemia, were allowed basedon the decision by the attending physicians.

Follow-up data were collected by the clinicalresearch coordinators belonging to the participatingcenters, or to the academic research organization(Research Institute for Production Development,Kyoto, Japan). Follow-up assessments were per-formed by means of hospital visit or telephone con-tact with the patient and/or the referring physician at1 year and final follow-up. Data collection for the finalfollow-up was started at February 1, 2016, which was1.5 years after the last patient was enrolled.

PRIMARY AND SECONDARY ENDPOINTS. Theprimary endpoint was defined as a composite ofdeath, myocardial infarction (MI), stroke, emergencyhospitalization for ACS, or hospitalization for heartfailure (HF) during the minimum of 1.5 years CF afterthe index PCI.

The secondary endpoints included all-causedeath, MI, stroke, emergency hospitalization for ACS,

iology, New Tokyo Hospital, Tokyo; kDepartment of

ardiology, Koto Memorial Hospital, Higashioumi;

partment of Cardiology, National Hospital Organiza-

ospital, Osaka; pDepartment of Cardiology, Saiseikai

harima Medical Center, Hyogo; rDepartment of Car-

ar Medicine, Kobe City Medical Center General Hos-uDepartment of Cardiovascular Medicine, Juntendo

sukubaMedical Center Hospital, Tsukuba; wDivision

City University Medical Center, Yokohama; and the

y an educational grant from the Research Institute for

r education consulting from Boston Scientific Corpo-

nt to the contents of this paper to disclose.

J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 1 0 , N O . 2 , 2 0 1 7 Shiomi et al.J A N U A R Y 2 3 , 2 0 1 7 : 1 0 9 – 1 7 Routine Follow-up Coronary Angiography After PCI

111

hospitalization for HF, definite stent thrombosis, ma-jor bleeding, target lesion revascularization (TLR),clinically driven TLR, any coronary revascularization,and clinically driven coronary revascularization.

Stent thrombosis and MI were defined according tothe Academic Research Consortium definitions (11).Stroke during follow-up was defined as ischemic orhemorrhagic stroke requiring hospitalization withsymptoms lasting >24 hrs. ACS was diagnosed ac-cording to clinical symptoms, electrocardiographicchanges compatible with acute myocardial ische-mia, and elevation of cardiac biomarkers. AMI(ST-segment elevation AMI and non–ST-segmentelevation AMI) and unstable angina were distin-guished according to the presence or absence of car-diac biomarker elevation. Unstable angina wasadjudicated only in the presence of an angiographic-ally evident culprit lesion. Hospitalization for HF wasdefined as hospitalization due to worsening HFrequiring intravenous drug therapy. Major bleedingwas defined as moderate or severe bleeding accordingto the GUSTO (Global Utilization of Streptokinase andTissue Plasminogen Activator for Occluded CoronaryArteries) classification (12). TLR was defined as eitherPCI or coronary artery bypass grafting due to reste-nosis or thrombosis of the target lesion that includedthe proximal and distal edge segments as well as theostium of the side branches. Only those lesionstreated at the time of the index PCI procedure wereregarded as target lesions. Any coronary revasculari-zation was defined as either PCI or coronary arterybypass grafting for any reasons. A coronary revascu-larization was considered clinically indicated if one ofthe following occurred: 1) a positive history of recur-rent angina pectoris; 2) objective signs of ischemia atrest (changes in the electrocardiograph) or duringexercise test (or the equivalent); and 3) abnormalresults of any invasive functional diagnostic test(e.g., fractional flow reserve).

Adjudication of endpoint events by an indepen-dent clinical event committee (List B in theOnline Appendix) was conducted in a blindedfashion regarding the assigned study groups. Clinicaloutcomes are analyzed according to the intention-to-treat principle.

STATISTICAL ANALYSES. Categorical variables wereexpressed as number (%), and were compared with thechi-square test or Fisher’s exact test. Continuous var-iables were expressed as mean � SD or medianwith interquartile range (IQR). Continuous variableswere compared using the Student t test or Wilcoxonrank-sum test based on their distributions. The cu-mulative incidence of a clinical event was assessed by

the Kaplan-Meier method and compared by the log-rank test. We developed the Cox proportional hazardmodel incorporating the random effect of center totake the differences in management between facilitiesinto consideration. The effect of routine follow-upangiography for the primary endpoint was expressedby hazard ratio with its 95% confidence interval. As asubgroup analysis, the treatment effect of routine AFstrategy relative to CF strategy was evaluated inseveral clinically relevant subgroups, including thosepatients with diabetes mellitus, restenotic lesion, leftmain coronary artery disease, chronic total occlusionlesion, bifurcation lesion, multivessel disease, totalstent length $40 mm, and “post hoc” high-risk groupdefined as having $1 high-risk features, such as leftmain coronary artery disease, bifurcation lesion,multivessel disease, and total stent length $40 mm.

The trial was originally designed to enroll 3,300patients to ensure a power of 80% to detect a 15%relative reduction of the primary endpoint rate at3 years in the AF group as compared with that in the CFgroup, in which the estimated primary endpoint eventrate was 25% at 3 years based on the data from the j-Cypher registry (13). The enrollment of study patientswas started at May 2010. In June 2014, however, theprotocol was amended to have a target enrollment of700 patients with an estimated median follow-upduration of 5 years, because of slow enrollment, andresulting in longer follow-up interval. The sample sizecalculation were based on an estimated primary eventrate of 47.7% in CF group, with a power of 80% todetect a relative reduction of 25% in AF group ascompared with CF group for the primary endpoint,assuming 25% crossover and loss to follow-up. Interimanalysis was not performed during the study period.

All statistical analyses were performed using JMP8.0 (SAS Institute Inc., Cary, North Carolina) and SAS9.4 (SAS Institute, Inc.) software. All reported p valueswere 2-sided and p < 0.05 were regarded as statisti-cally significant.

RESULTS

STUDY POPULATION. Between May 2010 and July2014, a total of 700 patients were enrolled in the trialamong 22 participating centers and were randomlyassigned to AF group (n ¼ 349) or CF group (n ¼ 351)(Figure 1). The 2 study groups were balancedwith regard to clinical, angiographic, and proce-dural characteristics (Table 1). The study populationreflected the real-world clinical practice in Japan,including large proportions of patients with advancedage, diabetes mellitus, prior PCI, and multivessel dis-ease, and a significant proportions of patients with AMI

FIGURE 1 Study Flow Chart

ACS ¼ acute coronary syndrome(s); AF ¼ angiographic follow-up; CF ¼ clinical follow-up; FUCAG ¼ follow-up coronary angiography;

PCI ¼ percutaneous coronary intervention.

Shiomi et al. J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 1 0 , N O . 2 , 2 0 1 7

Routine Follow-up Coronary Angiography After PCI J A N U A R Y 2 3 , 2 0 1 7 : 1 0 9 – 1 7

112

presentation, target of bifurcation lesion, and target ofchronic total occlusion. Regarding the PCI procedure,DES was used in 85% of patients, in whom second-generation DES was used in 89% of patients (Table 1).

During the first year, FUCAG, including those due toclinical reasons, were actually performed in 298 pa-tients (85.4%) in the AF group and in 42 patients(12.0%) in the CF group. Median time to FUCAG inpatients receiving FUCAG within 1 year after index PCIwas 287 days (IQR: 253 to 322 days) in the AF group(n ¼ 298 of 349) and 235 days (IQR: 174 to 299 days) inthe CF group (n¼42 of 351). In the AF group, 21 patients(7%) underwent coronary angiography for clinicalreasons. In the CF group, the reasons for coronaryangiography within the first year included 6 patients(14%) for ACS, 25 patients (60%) for recurrence ofangina, 6 patients (14%) for other clinical reasons, and5 patients (12%) without any clinical reason (protocolviolation) (Figure 1). Noninvasive physiological stresstests such as treadmill exercise test and stress nuclearstudy were more often performed in the CF group thanthe AF group within the first year after PCI (33.6% and25.2%; p ¼ 0.01), and during the entire follow-upperiod (52.7% and 40.1%; p < 0.001).

CLINICAL OUTCOMES. Median follow-up durationafter the index PCI was 4.6 (IQR: 3.1 to 5.2) years in theentire study population (AF group 4.5 [IQR: 3.1 to 5.2]years, and CF group: 4.6 [IQR: 3.1 to 5.2] years;

p¼0.87). The CF rate was 98.6% at 1 year and 95.5% at 3years (277 eligible patients) in the AF group, and 99.4%at 1 year and 96.2% at 3 years (279 eligible patients) inthe CF group. The cumulative 5-year incidence of theprimary endpoint was 22.4% in the AF group and 24.7%in the CF group (hazard ratio: 0.94, 95% confidenceinterval: 0.67 to 1.31; p ¼ 0.70) (Table 2, Figure 2).

The cumulative 5-year incidences of the individualcomponents of the primary endpoint such as all-causedeath, MI, stroke, emergency hospitalization for ACS,and hospitalization for HF were also not significantlydifferent between the AF and CF groups (Table 2). Thecumulative 5-year incidence of major bleeding wasalso not different between the 2 groups (Table 2).

TLR within the first year after the index PCI wasperformed more frequently in AF group than in CFgroup (7.0% vs. 1.7%; log-rank p < 0.001) (Figure 3A,Online Figure 1A). However, the cumulative 5-yearincidence of TLR in the AF group was not signifi-cantly different with that in the CF group (10.4% vs.8.5%; log-rank p ¼ 0.12) (Table 2, Figure 3A). Any cor-onary revascularization within the first year after theindex PCI was also more frequently performed in theAF group than in the CF group (12.8% vs. 3.8%; log-rank p < 0.001) (Figure 3B). However, any coronaryrevascularization beyond the first year after the indexPCI was performed more frequently in the CF groupthan in the AF group, and the difference in anycoronary revascularization between the 2 groups

TABLE 1 Baseline Clinical, Angiographic, and Procedural Characteristics, and Medications

AF Group (N ¼ 349) CF Group (N ¼ 351)

Clinical characteristics

Age, yrs 68.9 � 10.0 68.2 � 9.1

Male 260 (75) 291 (83)

Body mass index, kg/m2 24.3 � 3.4 24.2 � 3.2

Hypertension 256 (73) 279 (79)

Diabetes mellitus 144 (41) 169 (48)

Dyslipidemia 268 (77) 280 (80)

eGFR, ml/min/1.73 m2 65.9 � 21.7 66.1 � 21.7

eGFR <30 ml/min/1.73 m2, not on hemodialysis 6 (1.8) 5 (1.5)

Hemodialysis 13 (3.7) 12 (3.4)

Current smoker 62 (18) 65 (19)

Prior myocardial infarction 60 (17) 65 (19)

Prior percutaneous coronary intervention 105 (30) 121 (34)

Prior coronary artery bypass grafting 9 (2.6) 12 (3.4)

Prior stroke 25 (7.2) 36 (10)

Past history of heart failure 18 (5.2) 23 (6.6)

Atrial fibrillation 19 (5.4) 28 (8.0)

Stable coronary artery disease 222 (64) 222 (63)

Unstable angina 56 (16) 62 (18)

Acute myocardial infarction 71 (20) 67 (19)

Peripheral artery disease 43 (12) 41 (12)

Malignancy 57 (16) 38 (11)

Angiographic and procedural characteristics

Multivessel disease 145 (42) 153 (44)

Target vessel location

LMCA 15 (4.3) 13 (3.7)

LAD 194 (56) 196 (56)

LCx 98 (28) 87 (25)

RCA 123 (35) 126 (36)

Bypass graft 3 (0.9) 3 (0.9)

Target of STEMI culprit lesion 62 (18) 51 (15)

Target of bifurcation lesion 120 (34) 107 (30)

Target of chronic total occlusion 21 (6.0) 15 (4.3)

Target of restenosis lesion 25 (7.2) 25 (7.1)

Number of treated lesions per patient 1.29 � 0.57 1.28 � 0.55

Number of stents used (per patient) 1.54 � 0.94 1.46 � 0.81

Total stent length, mm (per patient) 32.9 � 23.9 31.5 � 20.8

Drug-eluting stents use 298 (85) 299 (85)

First-generation drug-eluting stents 33 (11) 34 (11)

Second-generation drug-eluting stents 265 (89) 265 (89)

Bare-metal stents use 53 (15) 52 (15)

Medications

Aspirin 348 (99.7) 345 (98)

Thienopyridine 345 (99) 347 (99)

Cilostazole 8 (2.3) 12 (3.4)

Statins 288 (83) 290 (83)

ACE-I/ARB 207 (59) 213 (61)

Beta blockers 124 (36) 158 (45)

Calcium-channel blocker 133 (38) 144 (41)

Nitrates 76 (22) 78 (22)

Warfarin 19 (5.4) 23 (6.6)

Proton pump inhibitor 184 (53) 167 (48)

H2 blocker 46 (13) 53 (15)

Values are mean � SD or n (%).

ACE-I ¼ angiotensin-converting enzyme inhibitor; AF ¼ angiographic follow-up; ARB ¼ angiotensin receptorblockers; CF ¼ clinical follow-up; eGFR ¼ estimated glomerular filtration rates; H2 blocker ¼ histamine type-2receptor blockers; LAD ¼ left anterior descending coronary artery; LCx ¼ left circumflex coronary artery; LMCA ¼left main coronary artery; RCA ¼ right coronary artery; STEMI ¼ ST-segment elevation myocardial infarction.

J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 1 0 , N O . 2 , 2 0 1 7 Shiomi et al.J A N U A R Y 2 3 , 2 0 1 7 : 1 0 9 – 1 7 Routine Follow-up Coronary Angiography After PCI

113

attenuated over time with similar cumulative 5-yearincidence (19.6% vs. 18.1%; log-rank p ¼ 0.92)(Table 2, Figure 3B, Online Figure 1B).

Regarding the subgroup analyses, there was nointeraction between the subgroup factors and the effectof AF relative to CF on the primary endpoint (Figure 4).

DISCUSSION

The main findings of the current trial were as follows:1) routine FUCAG after PCI did not provide any clin-ical benefits as compared with CF alone; and 2) anincreased 1-year rate of repeat coronary revasculari-zation with routine AF attenuated with long-termfollow-up.

Previous randomized trials in the balloon angio-plasty or bare-metal stent era and nonrandomizedstudies in the DES era consistently reported thatroutine FUCAG increased repeat coronary revasculari-zation, but did not reduce major adverse cardiacevents, although the rate of MI was slightly lower in AFthan in CF alone in the substudies of the BAAS (BalloonAngioplasty and Anticoagulation Study) and theTAXUS-IV trial (1–3,7,8). However, the impact ofroutine FUCAG for high-risk patients in real-worldpractice has not been evaluated fully, because all theprevious studies included patients with relatively low-risk profile in terms of comorbidity and lesioncomplexity. Cassese et al. (14) reported that the pres-ence of restenosis at FUCAG after PCI was predictive of4-year mortality in their cohort of 10,004 patients withroutine FUCAG. However, as the authors correctlystated in their article, the understanding of a potentialrole for routine follow-up angiography was beyond thescope of their study. According to the findings of thecurrent trial, which included high-risk patients in realclinical practice, routine FUCAG did not provide anyclinical benefit, including preventive effect of MI.Routine FUCAG after PCI is still commonly performedas usual care in Japan without evidence of clinicalbenefit (4,9,10). Considering the invasive nature ofcoronary angiography and increasedmedical expenses,routine FUCAG after PCI would not be considered usualclinical practice, unless patients have recurrent symp-tom or objective evidence of ischemia. In contrast,there was no excess of adverse clinical events with aroutine AF strategy, except for the increased rate of 1-year repeat coronary revascularization. Therefore, thescheduled AFwould still be acceptable in a first-in-mancoronary device trials, or as the mechanistic substudyin the pivotal coronary device trials.

“Oculostenotic reflex” phenomenon, which meanscoronary revascularization for angiographic stenosiswithout objective evidence of ischemia, was reported

TABLE 2 Long-Term (5-Yr) Clinical Outcomes

AF Group (n ¼ 349),Number of Patients With $1 Event

(Cumulative 5-Year Incidence)

CF Group (n ¼ 351),Number of Patients With $1 Event

(Cumulative 5-Year Incidence) HR (95% Cl) p Value

Primary endpoint

Death/MI/stroke/ACS/HF 65 (22.4) 70 (24.7) 0.94 (0.67–1.31) 0.70

Secondary endpoint

All-cause death 30 (10.8) 37 (12.5) 0.81 (0.50–1.31) 0.39

MI 6 (2.7) 9 (3.6) 0.66 (0.24–1.86) 0.43

Stroke 11 (3.5) 12 (4.2) 0.92 (0.41–2.08) 0.84

Emergency hospitalization for ACS 24 (9.0) 16 (6.3) 1.50 (0.79–2.82) 0.21

Hospitalization for heart failure 14 (4.3) 15 (6.0) 0.91 (0.44–1.88) 0.79

Definite stent thrombosis 0 (0) 2 (0.7) – 0.16

Major bleeding 9 (3.3) 14 (5.3) 0.63 (0.27–1.45) 0.28

Target lesion revascularization 34 (10.4) 23 (8.5) 1.51 (0.89–2.57) 0.12

Clinically driven 19 (5.9) 21 (7.9) 0.89 (0.48–1.66) 0.72

Any coronary revascularization 63 (19.6) 49 (18.1) 1.36 (0.93–1.97) 0.11

Clinically driven 47 (15.1) 46 (16.8) 1.02 (0.68–1.53) 0.94

Values are n (%). Number of patients with $1 event was evaluated during the entire follow-up period; the cumulative incidence was estimated at 5 years.

ACS ¼ acute coronary syndrome(s); CI ¼ confidence interval; HF ¼ heart failure; HR ¼ hazard ratio; MI ¼ myocardial infarction; other abbreviations as in Table 1.

Shiomi et al. J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 1 0 , N O . 2 , 2 0 1 7

Routine Follow-up Coronary Angiography After PCI J A N U A R Y 2 3 , 2 0 1 7 : 1 0 9 – 1 7

114

as a negative aspect of routine FUCAG, which resultedin approximately 2-fold higher rate of repeat coronaryrevascularization as compared with CF alone in severalprevious studies (1–3). In a substudy of the SPIRIT III(Clinical Evaluation of the Xience V Everolimus Eluting

FIGURE 2 Cumulative Incidence of the Primary Endpoint

MI ¼ myocardial infarction; other abbreviations as in Figure 1.

Coronary Stent System in the Treatment of Patientswith de novo Native Coronary Artery Lesions) trialevaluating newer generation DES; however, the cu-mulative incidence of repeat coronary revasculariza-tion was not significantly different between routine

FIGURE 3 Cumulative Incidence of Target Lesion Revascularization and Any Coronary Revascularization

Abbreviations as in Figure 1.

J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 1 0 , N O . 2 , 2 0 1 7 Shiomi et al.J A N U A R Y 2 3 , 2 0 1 7 : 1 0 9 – 1 7 Routine Follow-up Coronary Angiography After PCI

115

FIGURE 4 Subgroup Analyses

Subgroup analyses for the effect of AF relative to CF on the primary endpoint. The “post hoc” high-risk subgroup was defined as having at

least 1 high-risk feature such as LMCA disease, bifurcation lesion, multivessel disease, and total stent length $40 mm. CI ¼ confidence

interval; CTO ¼ chronic total occlusion; LMCA ¼ left main coronary artery; other abbreviations as in Figures 1 and 2.

Shiomi et al. J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 1 0 , N O . 2 , 2 0 1 7

Routine Follow-up Coronary Angiography After PCI J A N U A R Y 2 3 , 2 0 1 7 : 1 0 9 – 1 7

116

FUCAG and CF alone in 3-year CF (12.4% vs. 11.3%; log-rank p ¼ 0.45) (8). Consistent with the results of theSPIRIT III trial, long-term risks for TLR and any coro-nary revascularization were not significantly differentbetween AF and CF groups in the current trial. Despitethe 3-fold higher 1-year rate of TLR and any coronaryrevascularization in AF group, this large differencegradually attenuated with long-term follow-up in thecurrent trial. The annual 1.7% rate of late TLR beyond 1year after PCI in the CF group, which resulted in theattenuation of the difference in TLR between the 2groups, was consistent with the annual 2.0% to 2.2%rate of late TLR beyond 1 year reported in first-generation and newer generation DES studies (15,16).In contrast, the relatively lower 0.9% annual rate of lateTLR in AF group might suggest that many of the lesionswith late TLR in CF group actually had early restenosiswithin 1 year. Late TLR is one of the unsolved issues incontemporary PCI using DES. It would be a clinicallyrelevant question as to whether the fully bioresorbablecoronary scaffold could overcome the late adverseevents related to the target-lesion after completeresorption of the scaffold (17). The other possible reasonfor the attenuation of the between group difference inthe coronary revascularization rate during long-termfollow-up was the higher rate of coronary

revascularization for new lesions or progression ofnontarget lesions in the CF group. The lesions treated atthe time of FUCAG might undergo clinically drivenrevascularization with long-term follow-up, even if notdetected by FUCAG within 1 year.STUDY LIMITATIONS. First, the current trial was un-derpowered to detect modest differences in the pri-mary endpoint due to the reduced final sample sizeand the actual event rate lower than anticipated,although the size of the present study was similar toprevious studies. Therefore, the current trial resultmight be “inconclusive” rather than “negative,”warranting future, larger scale studies. Furthermore,we were unable to address the role of routine AF inthe high-risk subgroups, such as left main or multi-vessel coronary artery disease. Future dedicatedstudies are warranted to evaluate the role of routineAF in these high-risk subsets of patients. Second,slow patient enrollment might indicate patient se-lection bias that would potentially influence thestudy results. Finally, because patient demographics,practice patterns including the indication of coronaryrevascularization, and clinical outcomes in Japan maybe different from those outside Japan, generalizingthe present study results to populations outsideJapan should be done with caution.

PERSPECTIVES

WHAT IS KNOWN? Routine FUCAG after PCI could not

improve clinical outcomes but increased the rate of coronary

revascularization due to “oculostenotic reflex” in the previous

studies. However, there have been no randomized clinical trials

evaluating the clinical impact of routine FUCAG after PCI in real-

world clinical practice, including high-risk patients for cardio-

vascular events risk.

WHAT IS NEW? In this trial, which included a large proportion

of high-risk patients for cardiovascular events risk in daily clinical

practice in Japan, no clinical benefits were observed for routine

FUCAG after PCI and early revascularization rates were increased

within this approach.

WHAT IS NEXT? Future, larger scale trials (especially in high-

risk patients) are warranted to address definitively the role of

routine FUCAG after PCI in high-risk subsets of patients, such as

those with left main or multivessel coronary artery disease.

J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 1 0 , N O . 2 , 2 0 1 7 Shiomi et al.J A N U A R Y 2 3 , 2 0 1 7 : 1 0 9 – 1 7 Routine Follow-up Coronary Angiography After PCI

117

CONCLUSIONS

No clinical benefits were observed for routine FUCAGafter PCI and early revascularization rates wereincreased within this approach in the current trial.Thus, routine FUCAG cannot be recommended as aclinical strategy. However, the present study wasunderpowered to detect modest benefits (or harm) ofroutine FUCAG, and larger scale trials (especially inhigh-risk patients) are warranted to definitivelyaddress this issue.

ACKNOWLEDGMENTS The authors thank the mem-bers of the cardiac catheterization laboratories ofthe participating centers and the clinical researchcoordinators.

REPRINT REQUESTS AND CORRESPONDENCE: Dr.Takeshi Kimura, Department of Cardiovascular Med-icine, Kyoto University Graduate School of Medicine,54 Shogoin Kawahara-cho, Sakyo-ku, Kyoto 606-8507, Japan. E-mail: taketaka@kuhp.kyoto-u.ac.jp.

RE F E RENCE S

1. Ruygrok PN, Melkert R, Morel MA, et al. Doesangiography six months after coronary interven-tion influence management and outcome? Bene-stent II Investigators. J Am Coll Cardiol 1999;34:1507–11.

2. tenBerg JM,Kelder JC, SuttorpMJ,Verheugt FW,Thijs Plokker HW. Influence of planned six-monthfollow-up angiography on late outcome afterpercutaneous coronary intervention: a randomizedstudy. J Am Coll Cardiol 2001;38:1061–9.

3. Pinto DS, Stone GW, Ellis SG, et al. Impact ofroutine angiographic follow-up on the clinical ben-efits of paclitaxel-eluting stents: results from theTAXUS-IV trial. J Am Coll Cardiol 2006;48:32–6.

4. Shimada K, Kasanuki H, Hagiwara N, Ogawa H,Yamaguchi N. Routine coronary angiographicfollow-up and subsequent revascularization inpatients with acute myocardial infarction. HeartVessels 2008;23:383–9.

5. Levine GN, Bates ER, Blankenship JC, et al. 2011ACCF/AHA/SCAI Guideline for Percutaneous Cor-onary Intervention: a report of the American Col-lege of Cardiology Foundation/American HeartAssociation Task Force on Practice Guidelines andthe Society for Cardiovascular Angiography andInterventions. J Am Coll Cardiol 2011;58:e44–122.

6. Authors/Task Force m, Windecker S, Kolh P,et al. 2014 ESC/EACTS Guidelines on myocardialrevascularization: The Task Force on MyocardialRevascularization of the European Society of Car-diology (ESC) and the European Association forCardio-Thoracic Surgery (EACTS)Developed withthe special contribution of the European Associa-tion of Percutaneous Cardiovascular Interventions(EAPCI). Eur Heart J 2014;35:2541–619.

7. Uchida T, Popma J, Stone GW, et al. The clinicalimpact of routine angiographic follow-up in ran-domized trials of drug-eluting stents: a criticalassessment of “oculostenotic” reintervention inpatients with intermediate lesions. J Am CollCardiol Intv 2010;3:403–11.

8. Lansky AJ, Brar SS, Yaqub M, et al. Impact ofroutine angiographic follow-up after percutaneouscoronary intervention with drug-eluting stents inthe SPIRIT III randomized trial at three years. Am JCardiol 2012;110:21–9.

9. Toyofuku M, Kimura T, Morimoto T, et al.Comparison of 5-year outcomes in patients withand without unprotected left main coronary arterydisease after treatment with sirolimus-elutingstents: insights from the j-Cypher registry. J AmColl Cardiol Intv 2013;6:654–63.

10. Tazaki J, Shiomi H, Morimoto T, et al. Three-year outcome after percutaneous coronary inter-vention and coronary artery bypass grafting inpatients with triple-vessel coronary artery disease:observations from the CREDO-Kyoto PCI/CABGregistry cohort-2. EuroIntervention 2013;9:437–45.

11. Serruys PW, Unger F, Sousa JE, et al. Com-parison of coronary-artery bypass surgery andstenting for the treatment of multivessel disease.N Engl J Med 2001;344:1117–24.

12. An international randomized trial comparingfour thrombolytic strategies for acute myocardialinfarction. The GUSTO investigators. N Engl J Med1993;329:673–82.

13. Kimura T, Morimoto T, Nakagawa Y, et al. Verylate stent thrombosis and late target lesion

revascularization after sirolimus-eluting stent im-plantation: five-year outcome of the j-CypherRegistry. Circulation 2012;125:584–91.

14. Cassese S, Byrne RA, Schulz S, et al. Prognosticrole of restenosis in 10 004 patients undergoingroutine control angiography after coronary stent-ing. Eur Heart J 2015;36:94–9.

15. Natsuaki M, Morimoto T, Furukawa Y, et al. Lateadverse events after implantation of sirolimus-eluting stent and bare-metal stent: long-term (5-7years) follow-up of the coronary revascularizationdemonstrating outcome study-Kyoto RegistryCohort-2. Circ Cardiovasc Interv 2014;7:168–79.

16. Brener SJ, Kereiakes DJ, Simonton CA, et al.Everolimus-eluting stents in patients undergoingpercutaneous coronary intervention: final 3-yearresults of the Clinical Evaluation of the XIENCE VEverolimus Eluting Coronary Stent System in theTreatment of Subjects With de Novo Native Cor-onary Artery Lesions trial. Am Heart J 2013;166:1035–42.

17. Stone GW, Gao R, Kimura T, et al. 1-year out-comes with the Absorb bioresorbable scaffold inpatients with coronary artery disease: a patient-level, pooled meta-analysis. Lancet 2016;387:1277–89.

KEY WORDS angiographic follow-up,percutaneous coronary intervention,prognosis, stent(s)

APPENDIX For supplemental materials andfigures, please see the online version of thisarticle.