Comparison of Long-Term Clinical Outcomes in Patients Stratified by a Novel Acute Myocardial Infarction Risk Stratification (nARS) System

Abstract

Background: As severity of acute myocardial infarction (AMI) varies widely, several risk stratifications for AMI have been reported. We have introduced a novel AMI risk stratification system linked to a rehabilitation program (novel AMI risk stratification; nARS), which stratified AMI patients into low (L)-, intermediate (I)-, and high (H)-risk groups. The purpose of this retrospective study was to compare the long-term clinical outcomes in patients with AMI among L-, I-, H-risk groups.

Methods and Results: This study included 773 AMI patients, and assigned them into the L-risk group (n=332), the I-risk group (n=164), and the H-risk group (n=277). The primary endpoint was major cardiovascular events (MACE), defined as the composite of all-cause death, readmission for heart failure, non-fatal myocardial infarction, and target vessel revascularization after the discharge of index admission. The median follow-up duration was 686 days. MACE was most frequently observed in the H-risk group (39.4%), followed by the I-risk group (23.2%), and least in the L-risk group (19.9%) (P<0.001). The multivariate Cox hazard analysis revealed that the H-risk was significantly associated with MACE (HR 2.166, 95% CI 1.543–3.041, P<0.001) after controlling for multiple confounding factors.

Conclusions: H-risk according to nARS was significantly associated with long-term adverse events after hospital discharge for patients with AMI. These results support the validity of nARS as a risk marker for long-term outcomes.

In-hospital mortality caused by acute myocardial infarction (AMI) has dramatically decreased owing to multidisciplinary management, including primary percutaneous coronary intervention (PCI).^1–4 Severity of AMI varies widely from preserved cardiac function to severe left ventricular dysfunction.^5,6 Our group has introduced a novel AMI risk stratification system linked to a rehabilitation program (nARS), which stratified AMI patients into low (L)-, intermediate (I)-, and high (H)-risk groups according to our established criteria.⁷ Our group has shown the use of nARS focusing on in-hospital clinical outcomes and complications using an AMI patient cohort.^8,9

The introduction of nARS shortened the length of coronary care unit (CCU) stay and hospital stay.⁸ Nevertheless, the complications requiring CCU care rarely happen in the general wards after the introduction of nARS, which suggests the potential safety of nARS.⁹ Furthermore, in-hospital and subsequent mid-term clinical outcomes were reported, which revealed that H-risk patients had worse clinical outcomes as compared to I-risk or L-risk patients.⁷ However, because the difference in clinical outcomes was mainly caused by the difference in in-hospital outcomes, long-term clinical outcomes after hospital discharge have not been compared among L-, I-, H-risk groups. The purpose of this retrospective study was to compare the long-term clinical outcomes after the discharge from an index admission for patients with AMI among the L-, I-, H-risk groups stratified by nARS.

Methods

Study Design

We reviewed all AMI patients treated at our institution (Saitama Medical Center, Jichi Medical University) between August 2016 and December 2019. The inclusion criterion was patients with AMI. The exclusion criteria were: (1) patients who did not undergo PCI of the culprit lesion for AMI; (2) patients who underwent coronary artery bypass graft surgery (CABG) during hospitalization; (3) patients who died during the index hospitalization; (4) patients who were not followed up after hospital discharge; and (5) same patient with multiple occurrences of AMI during the study period.

The final study population was divided into a L-risk group, a I-risk group, and a H-risk group according to final risk stratified by nARS. The risk stratification was prospectively performed during our daily CCU conference. The L-risk patients needed to satisfy all of the following criteria: (1) primary PCI <12 h from onset of symptoms; (2) final Thrombolysis in Myocardial Infarction (TIMI)-3 flow grade on primary PCI; (3) global ejection fraction (EF) >40%; (4) introduction of angiotensin-converting enzyme inhibitors/angiotensin II receptor blockers (ACEI/ARB) and β-blockers on the day of admission or the day after admission; and (5) no high-risk comorbidities.⁷ In contrast, the H- risk patients needed to have at least one of the following criteria: (1) primary PCI >24 h from onset of symptoms; (2) final TIMI <2 flow grade on primary PCI; (3) global EF <30%; (4) unsuccessful rehabilitation test during hospitalization; (5) right ventricular (RV) infarction that worsens hemodynamic status; (6) >10 mm pericardial effusion on echocardiography; (7) apical aneurysm requiring anticoagulation therapy; (8) presence of a mechanical complication; or (9) requiring intra-aortic balloon pumping (IABP) support >48 h.⁷ Patients who were neither L- nor H-risk were categorized as I-risk.⁷ Patients in the L-risk group had to undergo a 2-min standing test on the day after primary PCI, followed by a 200-m walk test on the day after the successful 2-min standing test, and then a 500-m walk test on the day after the successful 200-m walk test, followed by discharge on the day after a successful 500-m walk test.⁷ Patients in the I-risk group had a 1-day interval between each test, and patients in the H-risk group typically underwent a 2-min standing test on day 3 and had a 2-day interval between each test.⁷ Risk stratification transition (from L-risk to H- risk or from I-risk to H-risk) occurred when the patient could not pass the scheduled rehabilitation test.⁷ The primary endpoint was major cardiovascular events (MACE) after hospital discharge, which was defined as the composite of all-cause death, readmission for heart failure, non-fatal myocardial infarction, and target vessel revascularization (TVR). Information regarding the above clinical outcomes were acquired from hospital records. The day of hospital discharge was defined as the index day (day 1). The study patients were followed until meeting MACE or until the study end date (December 20, 2021).

This study was approved by the institutional review board of the Saitama Medical Center, Jichi Medical University (S21-106), and written informed consent was waived because of the retrospective study design. The data collection and storage were performed anonymously, according to the Japan Ministry of Health, Labour and Welfare guidelines. All procedures were performed in accordance with the Declaration of Helsinki.

Definitions

AMI was defined according to the universal definition.¹⁰ Diagnostic ST elevation was defined as new ST elevation at the J point in at least 2 contiguous leads of 2 mm (0.2 mV), and the AMI patients with ST elevation were diagnosed as ST-elevation myocardial infarction (STEMI).¹¹ The definitions of hypertension, diabetes mellitus, and dyslipidemia are described elsewhere.^12,13 We used the laboratory data at admission. Because we could not measure some laboratory data such as HbA1c or low-density lipoprotein (LDL) cholesterol levels during out-of-hours (at night or on holidays), we substituted the earliest HbA1c or LDL cholesterol levels with laboratory data at admission.¹⁴ Left ventricular EF (LVEF) was measured by transthoracic echocardiography during the index hospitalization. LVEF was calculated through either modified Simpson’s method, Teichholz method, or eyeball estimation. A Teichholz method was adopted only when a modified Simpson’s method was not available, and an eyeball estimation was adopted only when both the modified Simpson’s method and Teichholz method were not available.¹⁵ We also calculated the estimated glomerular filtration rate (eGFR) using serum creatinine (Cr), age, weight, and gender according to the following formulas: eGFR = 194 × Cr^−1.094 × age^−0.287 (male), or eGFR = 194 × Cr^−1.094 × age^−0.287 × 0.739 (female).¹⁶ The initial TIMI flow grade and final TIMI flow grade were recorded from coronary angiography.¹³

Statistical Analysis

Data are presented as a percentage for categorical variables, a mean±standard deviation (SD) for normally distributed continuous variables, and median (quartile 1–quartile 3) for non-parametric variables. Categorical variables were presented as numbers (percentage) and were compared using a Chi-squared test. The Shapiro-Wilk test was performed to determine whether continuous variables were normally distributed. Normally distributed continuous variables were compared using a one-way ANOVA. Otherwise, continuous variables were compared by using the Kruskal-Wallis test. Event-free survival curves were constructed using the Kaplan-Meier method, and statistical differences between curves were assessed by using the log-rank test. We also performed a multivariate Cox hazard analysis to investigate the association between nARS stratification and MACE after controlling confounding factors. In this model, the variables were clinically selected from those that were not related to the nARS criteria to avoid multicollinearity. Moreover, the variables with ≥10 missing values were not included in the model. Hazard ratios and the 95% confidence intervals (CI) were calculated. A P value <0.05 was considered statistically significant. All analyses were performed using statistical software, SPSS ver. 25 for Windows (SPSS, Chicago, IL, USA).

Results

From August 2016 to December 2019, a total of 1,001 AMI patients were admitted to our medical center. After excluding 228 patients who met exclusion criteria, the final study population consisted of 737 AMI patients, which was divided into the L-risk group (n=332), I-risk group (n=164) and H-risk group (n=277) (Figure 1).

Figure 1.

Study flowchart. AMI, acute myocardial infarction; CABG, coronary artery bypass grafting surgery; H, high; I, intermediate; L, low; PCI, percutaneous coronary intervention.

Table 1 shows the comparison of patient’s characteristics between the 3 groups. Patients were oldest in the H-risk group, followed by the I-risk group, and youngest in the L-risk group. The prevalence of shock at admission was highest in the H-risk group, followed by the I-risk group, and lowest in the L-risk group. Estimated GFR was highest in the L-risk group, followed by the I-risk group, and lowest in the H-risk group. Peak creatinine kinase (CK) and peak CK-myocardial band were significantly higher in the H-risk group than in the I-risk and L-risk groups. B-type natriuretic peptide (BNP) at admission was highest in the H-risk group, followed by the I-risk group, and lowest in the I-risk group. LVEF was highest in the L-risk group, followed by the I-risk group, and lowest in the H-risk group. The lengths of CCU stay and hospital stay were shortest in the L-risk group, followed by the I-risk group, and longest in the H-risk group.

Table 1. Comparison of Patient Clinical Characteristics for the L-Risk, I-Risk, and H-Risk Groups

	All (n=773)	L-risk group (n=332)	I-risk group (n=164)	H-risk group (n=277)	P value
Age, years	69.7 (62.0–79.0)	68.2 (59.25–77.0)	70.4 (64.0–80.0)	71.1 (63.5–80.0)	0.006
Male sex, n (%)	580 (75.0)	258 (77.7)	121 (73.8)	201 (72.6)	0.318
Body mass index, kg/m2	24.0 (21.5–26.1)	24.3 (21.9–26.3)	23.9 (21.6–15.9)	23.2 (21.1–26.1)	0.068
Hypertension, n (%)	631/768 (82.2)	275 (83.3)	137 (84.6)	219 (79.3)	0.302
Diabetes mellitus, n (%)	345/768 (44.9)	140 (42.4)	68 (41.5)	137 (50.0)	0.108
Dyslipidemia, n (%)	446/760 (58.7)	206 (63.2)	100 (62.1)	140 (51.3)	0.008
Current smoker, n (%)	239/756 (31.6)	107 (32.7)	51 (32.3)	81 (29.9)	0.842
Hemodialysis, n (%)	68 (4.2)	28 (8.4)	18 (11.0)	22 (7.9)	0.538
History of previous MI, n (%)	89 (11.5)	45 (13.6)	18 (11.0)	26 (9.4)	0.283
History of previous CABG, n (%)	24 (3.1)	12 (3.6)	5 (3.0)	7 (2.5)	0.743
History of previous PCI, n (%)	148 (19.1)	71 (21.4)	36 (22.0)	41 (14.8)	0.066
STEMI, n (%)	410 (53.0)	147 (44.3)	104 (63.4)	159 (57.4)	<0.001
Killip classification, n (%)					<0.001
1 or 2	629 (81.4)	313 (94.3)	128 (78.0)	188 (67.9)
3	80 (10.3)	10 (3.1)	26 (15.9)	44 (15.9)
4	64 (8.6)	9 (2.7)	10 (6.1)	45 (16.2)
Cardiac arrest out of hospital, n (%)	24 (3.1)	2 (0.6)	4 (2.4)	18 (6.5)	<0.001
Shock at admission, n (%)	61 (7.9)	7 (2.1)	10 (6.1)	44 (15.9)	<0.001
SBP at admission, mmHg	143.8 (123.0–164.0)	149.2 (130.2–168.0)	148.3 (125.3–167.8)	134.6 (115.5–158.5)	<0.001
DBP at admission, mmHg	83.0 (71.0–95.0)	84.4 (73.0–96.0)	85.8 (74.0–98.8)	79.5 (67.0–94.0)	0.005
Heart rate at admission, beats/min	82.1 (67.0–95.0)	77.2 (66.0–88.0)	84.8 (69.3–96.8)	86.4 (70.0–102.0)	<0.001
Hemoglobin levels, g/dL	13.2 (11.8–14.6)	13.6 (12.5–14.9)	13.1 (11.8–14.5)	13.1 (11.3–14.3)	<0.001
Serum creatinine, mg/dL	2.1 (0.7–1.1)	1.4 (0.7–1.0)	1.7 (0.6–1.2)	3.0 (0.8–1.6)	<0.001
eGFR, mL/min/1.73 m2	62.3 (44.2–80.7)	69.0 (55.9–84.5)	62.4 (40.8–83.0)	54.2 (32.1–73.4)	<0.001
Peak CK, mg/dL	1,459.3 (169.0–1,977.5)	1,017.6 (121.3–1,465.8)	1,583.3 (221.3–2,210.5)	1,915.0 (274.0–2,561.0)	<0.001
Peak CK-MB, mg/dL	133.4 (15.0–192.5)	98.8 (6.0–136.3)	160.8 (16.0–232.2)	158.7 (15.5–216.5)	<0.001
BNP at admission, pg/mL	500.0 (37.1–561.8) n=737	237.4 (26.2–158.9) n=313	510.3 (36.5–624.5) n=158	664.3 (86.9–986.1) n=266	<0.001
LVEF, %	52.1 (41.8–62.6)	58.4 (52.1–66.1)	51.9 (41.1–60.0)	44.5 (32.9–56.2)	<0.001
Medical therapy at hospital discharge, n (%)
Aspirin	770 (99.6)	332 (100)	164 (100)	274 (98.9)	0.055
Thienopyridine	755 (97.7)	328 (98.8)	162 (98.8)	265 (95.7)	0.030
Statin	770 (99.6)	332 (100)	164 (100)	274 (98.9)	0.055
ACE inhibitors or ARB	763 (98.7)	329 (99.1)	164 (100)	270 (97.5)	0.054
β-blockers	763 (98.7)	328 (98.8)	162 (98.8)	273 (98.8)	1.000
Calcium channel blocker	135 (17.5)	54 (16.3)	41 (25.0)	40 (14.4)	0.017
Diuretics	234 (30.3)	44 (13.3)	57 (34.8)	133 (48.0)	<0.001
Oral antidiabetic	188 (24.3)	75 (22.6)	38 (23.2)	75 (27.1)	0.409
Insulin	40 (5.2)	10 (3.0)	6 (3.7)	24 (8.7)	0.006
DOAC	51 (6.6)	9 (2.7)	14 (8.5)	28 (10.1)	<0.001
Warfarin	19 (2.5)	2 (0.6)	1 (0.6)	16 (5.8)	<0.001
Length of CCU stay, days	2.5 (2.0–3.0)	1.6 (0.0–2.0)	2.3 (2.0–3.0)	3.7 (2.0–5.0)	<0.001
Length of hospital stay, days	9.1 (5.0–15.0)	5.9 (4.0–6.0)	8.3 (7.0–8.0)	13.4 (9.0–14.0)	<0.001

Data are expressed as median and interquartile range, the mean±standard deviation or number (percentage). Normally distributed continuous variables were compared by using the one-way ANOVA. Otherwise, continuous variables were compared by using the Kruskal-Wallis test. A Chi-squared test was used for categorical variables. ACE inhibitors, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blockers; BNP, B-type natriuretic peptide; CABG, coronary artery bypass grafting surgery; CCU, cardiac care unit; CK, creatine kinase; CK-MB, creatine kinase MB; DBP, diastolic blood pressure; DOAC, direct oral anticoagulant; eGFR, estimated glomerular filtration rate; H, high; L, low; I, intermediate; LVEF, left ventricular ejection fraction; MI, myocardial infarction; PCI, percutaneous coronary intervention; SBP, systolic blood pressure; STEMI, ST-elevated myocardial infarction.

Table 2 shows the comparison of angiographic and procedural findings between the 3 groups. The prevalence of left main trunk stenosis and chronic total occlusion (CTO) in non-culprit arteries was highest in the H-risk group, followed by the I-risk group, and lowest in the L-risk group. The prevalence of triple-vessel disease was highest in the H-risk group, followed by the I-risk group, and lowest in the L-risk group. Intra-aortic balloon pump support (IABP) and percutaneous cardiopulmonary support device (PCPS) were most frequently used in the H-risk group, followed by the I-risk group, and least in the L-risk group.

Table 2. Comparison of Angiographic Lesions and Procedural Characteristics for the L-Risk, I-Risk, and H-Risk Groups

	All (n=773)	L-risk group (n=332)	I-risk group (n=164)	H-risk group (n=277)	P value
Infarct-related artery					0.729
Left main trunk-left anterior descending artery, n (%)	385 (49.8)	163 (49.1)	86 (52.4)	136 (49.1)
Left circumflex artery, n (%)	117 (15.1)	50 (15.1)	20 (12.2)	47 (17.0)
Right coronary artery, n (%)	254 (32.9)	110 (33.1)	57 (34.8)	87 (31.4)
Graft of CABG, n (%)	7 (0.9)	3 (0.9)	1 (0.6)	3 (1.1)
Not determined, n (%)	10 (1.3)	6 (1.8)	0 (0)	4 (1.4)
Left main trunk stenosis >50%, n (%)	80 (10.3)	26 (7.8)	14 (8.5)	40 (14.4)	0.024
AMI with non-culprit CTO, n (%)	96 (12.4)	29 (8.7)	22 (13.4)	45 (16.2)	0.017
Number of narrowed coronary arteries					<0.001
1, n (%)	335 (43.3)	174 (52.4)	58 (35.4)	103 (37.2)
2, n (%)	261 (33.8)	86 (25.9)	77 (47.0)	98 (35.4)
3, n (%)	177 (22.9)	72 (21.7)	29 (17.7)	76 (27.4)
Initial TIMI flow grade					0.006
0, n (%)	278 (36.0)	95 (28.6)	60 (36.6)	123 (44.4)
1, n (%)	52 (6.7)	22 (6.6)	12 (7.3)	18 (6.5)
2, n (%)	121 (15.7)	63 (19.0)	25 (15.2)	33 (11.9)
3, n (%)	322 (41.7)	152 (45.8)	67 (40.9)	103 (37.2)
Final TIMI flow grade					<0.001
0, n (%)	5 (0.6)	0 (0)	0 (0)	5 (1.8)
1, n (%)	6 (0.8)	0 (0)	0 (0)	6 (2.2)
2, n (%)	21 (2.7)	2 (0.6)	2 (1.2)	17 (6.1)
3, n (%)	741 (95.9)	330 (99.4)	160 (98.8)	249 (89.9)
Final PCI procedure					<0.001
Drug coated balloon, n (%)	36 (4.7)	22 (6.6)	6 (3.7)	8 (2.9)
Bare metal stent, n (%)	10 (1.3)	3 (0.9)	3 (1.8)	4 (1.4)
Drug eluting stent, n (%)	670 (86.7)	298 (89.8)	145 (88.4)	227 (81.9)
POBA and/or Aspiration, n (%)	57 (7.4)	9 (2.7)	10 (6.1)	38 (13.7)
Intra-aortic balloon pump support, n (%)	51 (6.6)	3 (0.9)	10 (6.1)	38 (13.8)	<0.001
Percutaneous cardiopulmonary support device, n (%)	16 (2.1)	0 (0)	2 (1.2)	14 (5.1)	<0.001
Access site					0.002
Radial, n (%)	576 (74.5)	270 (81.3)	110 (67.1)	196 (70.8)
Brachial, n (%)	13 (1.7)	5 (1.5)	2 (1.2)	6 (2.2)
Femoral, n (%)	184 (23.8)	57 (17.2)	52 (31.7)	75 (27.1)
Catheter size					0.006
6Fr, n (%)	522 (67.5)	242 (72.9)	110 (67.1)	170 (61.4)
7Fr, n (%)	246 (31.8)	89 (26.8)	51 (31.1)	106 (38.3)
8Fr, n (%)	5 (0.6)	1 (0.3)	3 (1.8)	1 (0.4)

Data are expressed as the mean±standard deviation or number (percentage). A Chi-squared test was used for categorical variables. CTO, chronic total occlusion; TIMI, Thrombolysis in Myocardial Infarction; POBA, plain old balloon angioplasty. Other abbreviations as in Table 1.

Figure 2 shows the Kaplan-Meier curves for MACE between the 3 groups. The median follow-up duration was 686 days (Q1: 215 days to Q3: 1,040 days). The MACE free survival rate was significantly lower in the H-risk group as compared with other groups (P<0.001 for overall). Table 3 shows the comparison of clinical outcomes between the 3 groups. MACE were most frequently observed in the H-risk group (39.4%), followed by the I-risk group (23.2%) and least in the L-risk group (19.9%) (P<0.001).

Figure 2.

Kaplan-Meier curves for MACE-free survival between the low (L)-risk group, intermediate (I)-risk group and high (H)-risk group. A log-rank test was used. MACE, major cardiovascular events.

Table 3. Clinical Outcomes of the L-Risk, I-Risk, and H-Risk Groups

Variables	All (n=773)	L-risk group (n=332)	I-risk group (n=164)	H-risk group (n=277)	P value
MACE, n (%)	213 (27.6)	66 (19.9)	38 (23.2)	109 (39.4)	<0.001
All cause death, n (%)	70 (9.1)	27 (8.1)	9 (5.5)	34 (12.3)	0.045
Heart failure, n (%)	93 (12.0)	10 (3.0)	22 (13.4)	61 (22.0)	<0.001
Non-fatal AMI, n (%)	57 (7.4)	17 (5.1)	16 (9.8)	24 (8.7)	0.094
TVR, n (%)	80 (10.3)	26 (7.8)	18 (11.0)	36 (13.0)	0.101

Data are expressed as median, inter-quartile range and number (percentage). Continuous variables were compared by using the Kruskal-Wallis test. A Chi-squared test was used for categorical variables. AMI, acute myocardial infarction; MACE, major cardiovascular evets; TVR, target vessel revascularization. Other abbreviations as in Table 1.

The multivariate Cox hazard analysis is shown in Table 4. H-risk was significantly associated with MACE (HR 2.166, 95% CI 1.543–3.041, P<0.001) after controlling for multiple confounding factors including age, sex, history of previous PCI, history of previous CABG, diabetes mellitus, shock at admission, peak CK, thienopyridine at discharge, calcium channel blocker at discharge, diuretics at discharge, oral antidiabetic at discharge, insulin at discharge, direct oral anticoagulant (DOAC) at discharge, infarct-related artery, access site, and catheter size.

Table 4. Multivariance Cox Regression Analysis Prediction of MACE

Composite endpoints	HR	95% CI	P value
MACE
L-risk	Ref.
Unadjusted I-risk	1.307	0.876–1.949	0.189
Adjusted I-risk	1.157	0.763–1.756	0.493
Unadjusted H-risk	2.352	1.732–3.195	<0.001
Adjusted H-risk	2.166	1.543–3.041	<0.001
Component endpoints	HR	95% CI	P value
All cause death
L-risk	Ref.
Unadjusted I-risk	0.750	0.352–1.596	0.455
Adjusted I-risk	0.609	0.271–1.366	0.229
Unadjusted H-risk	1.692	1.021–2.807	0.041
Adjusted H-risk	1.629	0.920–2.874	0.094
Heart failure
L-risk	Ref.
Unadjusted I-risk	4.871	2.306–10.290	<0.001
Adjusted I-risk	3.857	1.762–8.442	<0.001
Unadjusted H-risk	8.353	4.277–16.311	<0.001
Adjusted H-risk	6.594	3.305–13.560	<0.001
Non-fatal AMI
L-risk	Ref.
Unadjusted I-risk	2.054	1.037–4.070	0.039
Adjusted I-risk	1.861	0.896–3.864	0.096
Unadjusted H-risk	1.868	1.003–3.479	0.049
Adjusted H-risk	1.993	1.001–3.969	0.050
TVR
L-risk	Ref.
Unadjusted I-risk	1.49	0.816–2.720	0.194
Adjusted I-risk	1.406	0.744–2.658	0.295
Unadjusted H-risk	1.837	1.109–3.045	0.018
Adjusted H-risk	2.041	1.154–3.609	0.014

In the adjusted model, L-risk group was adjusted for age, sex, history of previous PCI, history of previous CABG, diabetes mellitus, shock at admission, peak CK, thienopyridine at discharge, calcium channel blocker at discharge, diuretics at discharge, oral antidiabetic at discharge, insulin at discharge, DOAC at discharge, infarct-related artery, access site and catheter size. CI, confidence interval; HR, hazard ratio. Other abbreviations as in Tables 1,3.

Discussion

We included 773 AMI patients and divided those into the L-risk group (n=332), the I-risk group (n=164), and the H-risk group (n=277) according to the nARS. We followed up those patients after hospital discharge with a median duration of 686 days. MACE were more frequently observed in the H-risk group than in other groups. The multivariate Cox hazard analysis revealed that H-risk was significantly associated with MACE after controlling for multiple confounding factors. Although nARS was introduced to facilitate in-hospital rehabilitation and to reduce the length of hospital stay for AMI patients, our results suggest that nARS would be useful to stratify high-risk patients with AMI after hospital discharge.

In the field of risk classification for AMI, the Global Registry of Acute Coronary Events (GRACE) score and the TIMI risk score are well-known risk scores.^17,18 However, since these risk scores were invented before the development of primary PCI, these risk scores do not reflect the results of primary PCI. In other words, the GRACE score of patients who underwent primary PCI may be the same as that of patients who did not undergo primary PCI. Furthermore, the GRACE score of patients who acquired TIMI flow grade 3 by primary PCI may be the same as that of patients who could not acquire TIMI flow grade 3. Because these classical risk scores do not reflect the results of primary PCI, new risk scores or classifications that reflect the results of primary PCI would be necessary to stratify AMI patients more efficiently. Therefore, we introduced nARS to provide AMI patients with an in-hospital rehabilitation program according to reasonable risk stratification.

As the mortality caused by AMI would be highest in the acute phase, followed by the subacute phase, and lowest in the chronic phase, patients with the most severe AMI would die during the index admission.^19–21 Therefore, the risk factors for in-hospital death may not necessarily correspond to the risk factors for death or cardiovascular events after hospital discharge. Aissaoui et al found that AMI patients who experienced cardiogenic shock (CS) had a worse 1-year prognosis than those patients who did not experience CS in a landmark analysis based on the 30-days post-discharge.²² Wada et al also reported that AMI patients who experienced CS had worse long-term clinical outcomes than those who did not experience CS in a landmark analysis based on the 30-days post-discharge.²³ These studies suggest that CS is a strong risk factor for short- and long-term clinical outcomes. Wright et al showed that post-discharge mortality was higher in AMI patients with renal dysfunction, even for those with mild renal dysfunction.²⁴ Primary PCI >12 h from onset of symptoms was also reported as a risk factor for long-term clinical outcomes.²⁵

We should address why the pre-discharge H-risk was associated with long-term MACE. First, the H-risk group included patients with more serious disease than the L-risk or I-risk groups, because patients in the H-risk group had at least 1 high-risk feature such as primary PCI >24 h from onset of symptoms, final TIMI flow grade ≤2, or low LVEF (≤30%).^7,8 Furthermore, among 4 components of MACE, readmission for heart failure was the most significantly associated with the H-risk group. As AMI patients with low LVEF would suffer from readmission for heart failure,^26–28 a feature of low LVEF in the H-risk group might be a reason for increased readmission for heart failure. Second, patients in the L- or I-risk groups who could not pass the rehabilitation tests, including standing for 2 min, a 200-m walk test, and a 500-m walk test, were reclassified into the H-risk group before discharge. Previous studies showed that AMI patients with slow gait speed in the 200-m walk test had a poor long-term outcomes.²⁹ Moreover, our group reported that readmission for heart failure was more frequently observed in patients with AMI who failed the 500-m walk test.³⁰ Furthermore, unsuccessful 200-m or 500-m rehabilitation tests might reflect frailty. Frail AMI patients would have long-term MACE, because frailty was closely associated with future cardiovascular events.^31–33

Clinical implications of the present study should be noted. First, the present study shed light on nARS as a risk marker for long-term outcomes in patients with AMI. Although the use of nARS was reported previously, the emphasis was on the acute phase of AMI.^7–9 The present study suggests the use of nARS as a risk marker for long-term outcomes as well as in-hospital outcomes in patients with AMI. Second, nARS may be useful in selecting a high-risk group for readmission, and these patients require close follow up in outpatient clinics. Because medical resources are limited in an unprecedented aging society, it would be difficult to plan for the close follow up for all AMI patients. Close follow up for patients with H-risk may prevent readmission for heart failure,^34,35 which would save medical resources.

Several limitations warrant mention. As this study was a single-center, retrospective study, there was a potential selection bias. Although we have proposed nARS in previous studies,^7–9 the validity of nARS has not been assessed in other institutions yet. As this study assessed clinical events after hospital discharge, the severest patients who died during the index hospitalization were not included. Although H-risk was significantly associated with long-term MACE, I-risk was only associated with long-term readmission for heart failure. In view of the long-term risk marker, I-risk was insufficient as a risk marker. Finally, the structure of nARS is not simple as compared to other long-term risk markers such as low LVEF or Killip class 4. We originally proposed nARS to facilitate in-hospital rehabilitation including the 200-m walk test or the 500-m walk test, and to reduce the length of hospitalization safely. Therefore, some of the criteria for H-risk, such as pericardial effusion in the acute phase, would not be associated with long-term MACE. In this study, our goal was not to prove nARS as the best marker for long-term outcomes, but to add a new value as long-term risk marker to nARS.

Conclusions

Pre-discharge H-risk according to nARS was significantly associated with long-term adverse events after hospital discharge in patients with AMI. These results support the validity of nARS as a risk marker for long-term outcomes.

Acknowledgments

The authors acknowledge all staff in the catheter laboratory, cardiology units, and emergency and critical care units in Saitama Medical Center, Jichi Medical University, for their technical support in this study.

Disclosures

Dr. Sakakura has received speaking honoraria from Abbott Vascular, Boston Scientific, Medtronic Cardiovascular, Terumo, OrbusNeich, Japan Lifeline, Kaneka, and NIPRO. Dr. Jinnouchi has received speaking honoraria from Abbott Vascular. Prof. Fujita has served as a consultant for Mehergen Group Holdings, Inc.

IRB Information

This study was approved by the Saitama Medical Center, Jichi Medical University (S21-106).

References

• 1.   Rothwell PM, Coull AJ, Silver LE, Fairhead JF, Giles MF, Lovelock CE, et al. Population-based study of event-rate, incidence, case fatality, and mortality for all acute vascular events in all arterial territories (Oxford Vascular Study). Lancet 2005; 366: 1773–1783.
• 2.   Nakayama N, Kimura K, Endo T, Fukui K, Himeno H, Iwasawa Y, et al. Current status of emergency care for ST-elevation myocardial infarction in an urban setting in Japan. Circ J 2009; 73: 484–489.
• 3.   Daida H, Miyauchi K, Ogawa H, Yokoi H, Matsumoto M, Kitakaze M, et al. Management and two-year long-term clinical outcome of acute coronary syndrome in Japan: Prevention of atherothrombotic incidents following ischemic coronary attack (PACIFIC) registry. Circ J 2013; 77: 934–943.
• 4.   Sabaté M, Brugaletta S, Cequier A, Iñiguez A, Serra A, Hernádez-Antolín R, et al. The EXAMINATION trial (Everolimus-Eluting Stents Versus Bare-Metal Stents in ST-Segment Elevation Myocardial Infarction): 2-year results from a multicenter randomized controlled trial. JACC Cardiovasc Interv 2014; 7: 64–71.
• 5.   Jiang Y, Hu S, Cao M, Li X, Zhou J, Ding B, et al. Evaluation of acute myocardial infarction patients with mid-range ejection fraction after emergency percutaneous coronary intervention. Postgrad Med J 2019; 95: 355–360.
• 6.   Jiang Y, Boris AF, Zhu Y, Gan H, Hu X, Xue Y, et al. Incidence, clinical characteristics and short-term prognosis in patients with cardiogenic shock and various left ventricular ejection fractions after acute myocardial infarction. Am J Cardiol 2022; 167: 20–26.
• 7.   Yamamoto K, Sakakura K, Akashi N, Watanabe Y, Noguchi M, Taniguchi Y, et al. Clinical outcomes after acute myocardial infarction according to a novel stratification system linked to a rehabilitation program. J Cardiol 2018; 72: 227–233.
• 8.   Yamamoto K, Sakakura K, Akashi N, Watanabe Y, Noguchi M, Seguchi M, et al. Novel Acute Myocardial Infarction Risk Stratification (nARS) system reduces the length of hospitalization for acute myocardial infarction. Circ J 2019; 83: 1039–1046.
• 9.   Sawano S, Sakakura K, Yamamoto K, Taniguchi Y, Tsukui T, Seguchi M, et al. Further validation of a Novel Acute Myocardial Infarction Risk Stratification (nARS) system for patients with acute myocardial infarction. Int Heart J 2020; 61: 463–469.
• 10.   Sawano M, Yamaji K, Kohsaka S, Inohara T, Numasawa Y, Ando H, et al. Contemporary use and trends in percutaneous coronary intervention in Japan: An outline of the J-PCI registry. Cardiovasc Interv Ther 2020; 35: 218–226.
• 11.   Shibata K, Sakakura K, Taniguchi Y, Yamamoto K, Tsukui T, Seguchi M, et al. Comparison of clinical outcomes of acute myocardial infarction between prasugrel and clopidogrel. Int Heart J 2021; 62: 479–486.
• 12.   Yamamoto K, Sakakura K, Hamamoto K, Hasegawa H, Tsukui T, Seguchi M, et al. Determinants of Greater Peak radiation skin dose in percutaneous coronary intervention for chronic total occlusion. J Cardiol 2020; 76: 217–223.
• 13.   Murakami T, Sakakura K, Jinnouchi H, Taniguchi Y, Tsukui T, Watanabe Y, et al. Complications related to veno-arterial extracorporeal membrane oxygenation in patients with acute myocardial infarction: VA-ECMO complications in AMI. J Cardiol 2022; 79: 170–178.
• 14.   Otani A, Sakakura K, Yamamoto K, Taniguchi Y, Ito M, Ibe T, et al. Comparison of mid-term clinical outcomes after acute myocardial infarction in diabetic men between living alone and living together. Heart Vessels 2019; 34: 1288–1296.
• 15.   Ban S, Sakakura K, Jinnouchi H, Taniguchi Y, Tsukui T, Watanabe Y, et al. Association of asymptomatic low ankle-brachial index with long-term clinical outcomes in patients after acute myocardial infarction. J Atheroscler Thromb, doi:10.5551/jat.62998.
• 16.   Matsuo S, Imai E, Horio M, Yasuda Y, Tomita K, Nitta K, et al. Revised equations for estimated GFR from serum creatinine in Japan. Am J Kidney Dis 2009; 53: 982–992.
• 17.   Granger CB, Goldberg RJ, Dabbous O, Pieper KS, Eagle KA, Cannon CP, et al. Predictors of hospital mortality in the global registry of acute coronary events. Arch Intern Med 2003; 163: 2345–2353.
• 18.   Morrow DA, Antman EM, Charlesworth A, Cairns R, Murphy SA, de Lemos JA, et al. TIMI risk score for ST-elevation myocardial infarction: A convenient, bedside, clinical score for risk assessment at presentation: An intravenous nPA for treatment of infarcting myocardium early II trial substudy. Circulation 2000; 102: 2031–2037.
• 19.   Ito R, Yamashita J, Chikamori T, Kondo S, Mitsuhashi Y, Iwata H, et al. Clinical differences of recent myocardial infarction compared with acute myocardial infarction: Insights from the Tokyo CCU Network Multicenter Registry. Circ J 2020; 84: 1511–1518.
• 20.   Ventura M, Belleudi V, Sciattella P, Di Domenicantonio R, Di Martino M, Agabiti N, et al. High quality process of care increases one-year survival after acute myocardial infarction (AMI): A cohort study in Italy. PLoS One 2019; 14: e0212398. doi:10.1371/journal.pone.0212398.
• 21.   Miyachi H, Takagi A, Miyauchi K, Yamasaki M, Tanaka H, Yoshikawa M, et al. Current characteristics and management of ST elevation and non-ST elevation myocardial infarction in the Tokyo metropolitan area: From the Tokyo CCU network registered cohort. Heart Vessels 2016; 31: 1740–1751.
• 22.   Aissaoui N, Puymirat E, Simon T, Bonnefoy-Cudraz E, Angoulvant D, Schiele F, et al. Long-term outcome in early survivors of cardiogenic shock at the acute stage of myocardial infarction: A landmark analysis from the French registry of Acute ST-elevation and non-ST-elevation Myocardial Infarction (FAST-MI) Registry. Crit Care 2014; 18: 516.
• 23.   Wada H, Ogita M, Suwa S, Nakao K, Ozaki Y, Kimura K, et al. Long-term clinical impact of cardiogenic shock and heart failure on admission for acute myocardial infarction. Int Heart J 2021; 62: 520–527.
• 24.   Wright RS, Reeder GS, Herzog CA, Albright RC, Williams BA, Dvorak DL, et al. Acute myocardial infarction and renal dysfunction: A high-risk combination. Ann Intern Med 2002; 137: 563–570.
• 25.   Nepper-Christensen L, Lønborg J, Høfsten DE, Sadjadieh G, Schoos MM, Pedersen F, et al. Clinical outcome following late reperfusion with percutaneous coronary intervention in patients with ST-segment elevation myocardial infarction. Eur Heart J Acute Cardiovasc Care, doi:10.1177/2048872619886312.
• 26.   Møller JE, Hillis GS, Oh JK, Reeder GS, Gersh BJ, Pellikka PA. Wall motion score index and ejection fraction for risk stratification after acute myocardial infarction. Am Heart J 2006; 151: 419–425.
• 27.   Ersbøll M, Valeur N, Mogensen UM, Andersen MJ, Møller JE, Velazquez EJ, et al. Prediction of all-cause mortality and heart failure admissions from global left ventricular longitudinal strain in patients with acute myocardial infarction and preserved left ventricular ejection fraction. J Am Coll Cardiol 2013; 61: 2365–2373.
• 28.   Serrao GW, Lansky AJ, Mehran R, Stone GW. Predictors of left ventricular ejection fraction improvement after primary stenting in ST-segment elevation myocardial infarction (from the Harmonizing Outcomes With Revascularization and Stents in Acute Myocardial Infarction Trial). Am J Cardiol 2018; 121: 678–683.
• 29.   Matsuzawa Y, Konishi M, Akiyama E, Suzuki H, Nakayama N, Kiyokuni M, et al. Association between gait speed as a measure of frailty and risk of cardiovascular events after myocardial infarction. J Am Coll Cardiol 2013; 61: 1964–1972.
• 30.   Ohashi J, Sakakura K, Sasaki W, Taniguchi Y, Kani K, Yamamoto K, et al. Usefulness of 500-m walk electrocardiogram test on clinical outcomes in patients with ST-segment elevation myocardial infarction. Heart Vessels 2021; 36: 48–57.
• 31.   Newman AB, Gottdiener JS, McBurnie MA, Hirsch CH, Kop WJ, Tracy R, et al. Associations of subclinical cardiovascular disease with frailty. J Gerontol A Biol Sci Med Sci 2001; 56: M158–M166.
• 32.   Singh M, Rihal CS, Lennon RJ, Spertus JA, Nair KS, Roger VL. Influence of frailty and health status on outcomes in patients with coronary disease undergoing percutaneous revascularization. Circ Cardiovasc Qual Outcomes 2011; 4: 496–502.
• 33.   Sanchis J, Bonanad C, Ruiz V, Fernández J, García-Blas S, Mainar L, et al. Frailty and other geriatric conditions for risk stratification of older patients with acute coronary syndrome. Am Heart J 2014; 168: 784–791.
• 34.   Matsukawa R, Masuda S, Matsuura H, Nakashima H, Ikuta H, Okabe K, et al. Early follow-up at outpatient care after discharge improves long-term heart failure readmission rate and prognosis. ESC Heart Fail 2021; 8: 3002–3013.
• 35.   Tung YC, Chang GM, Chang HY, Yu TH. Relationship between early physician follow-up and 30-day readmission after acute myocardial infarction and heart failure. PLoS One 2017; 12: e0170061. doi:10.1371/journal.pone.0170061.