Association between door-to-balloon time ≤60 minutes and short- and long-term outcomes in patients with ST-segment elevation myocardial infarction: a retrospective cohort study

Abstract

Background

Current guidelines diverge on the optimal door-to-balloon time (DTBT) for acute ST-segment elevation myocardial infarction (STEMI), and whether a ≤60 min target benefits all patients equally remains unclear.

Objectives

This study aimed to investigate whether the association between DTBT ≤60 min and short- and long-term prognosis in STEMI patients varies by Global Registry of Acute Coronary Events (GRACE) risk score.

Design

Retrospective cohort study.

Setting

Single high-volume tertiary cardiology centre in China.

Participants

This study initially included 5516 STEMI patients undergoing primary percutaneous coronary intervention (PPCI) treatment. 4513 were included after applying inclusion criteria (age >18 years, presentation within 12 hours of symptom onset, complete medical records) and exclusion criteria (symptom duration ≥12 hours, failure to receive PPCI, incomplete outcome data).

Interventions

Not applicable (observational study).

Primary and secondary outcome measures

The primary outcomes were in-hospital, 1-year and 3-year all-cause mortality. Secondary outcomes included major adverse cardiovascular and cerebrovascular events (MACCE) at 1 and 3 years post-discharge.

Results

Among 4513 STEMI patients, 2433 (54.0%) were high-risk (HR-STEMI) and 2080 (46.0%) low-risk (LR-STEMI). DTBT ≤60 min was achieved in 45.7% of HR-STEMI and 52.0% of LR-STEMI patients. For HR-STEMI patients, DTBT >60 min was associated with significantly higher risks of in-hospital mortality (OR=2.381, 95% CI 1.160 to 4.883, p=0.018), 1-year mortality (HR=1.715, 95% CI 1.194 to 2.464, p=0.003), 1-year MACCE (HR=1.212, 95% CI 1.001 to 1.467, p=0.049), 3-year mortality (HR=1.689, 95% CI 1.267 to 2.253, p<0.001), and 3-year MACCE (HR=1.230, 95% CI 1.042 to 1.453, p=0.014). Among LR-STEMI patients, no significant differences were observed between DTBT groups.

Conclusions

DTBT ≤60 min was significantly associated with better short- and long-term outcomes, particularly in patients with GRACE >140. Sensitivity analysis suggested that the benefit may also extend to patients with GRACE scores between 120 and 140.

Trial registration number

Not applicable (observational study).

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• This is a large retrospective cohort study of 4513 ST-segment elevation myocardial infarction patients undergoing primary percutaneous coronary intervention at a single high-volume centre.

• Patients were stratified a priori by the validated Global Registry of Acute Coronary Events risk score using the guideline-aligned threshold of 140 to evaluate the heterogeneous association of door-to-balloon time.

• Propensity score matching was performed to adjust for a wide range of baseline characteristics, reducing confounding by indication.

• The single-centre design may limit the generalisability of the findings to other populations or healthcare systems.

• As an observational study, residual confounding from unmeasured variables such as total ischaemic time and operator experience cannot be completely excluded.

Introduction

ST-segment elevation myocardial infarction (STEMI) is one of the most critical cardiovascular emergencies, with persistently high mortality and disability rates that impose a heavy burden on global public health systems. Timely and effective reperfusion of the infarct-related artery is fundamental to improving patient outcomes and salvaging ischaemic myocardium. Primary percutaneous coronary intervention (PPCI) is currently the guideline-recommended preferred reperfusion strategy. Against this backdrop, “door-to-balloon time” (DTBT), defined as the interval from the patient’s arrival at the hospital to the moment the PCI device crosses the lesion, has been established as a key quality indicator for STEMI treatment efficiency.^1–3 Numerous quality improvement initiatives—including prehospital electrocardiography, direct catheterisation laboratory activation, and digital health interventions—have been implemented to reduce DTBT.⁴ A recent large multicentre study in Beijing reported a median DTBT of 72 min among STEMI patients undergoing PPCI, indicating that in-hospital delays have been well controlled in this setting.⁵ Current guidelines, however, diverge on the optimal DTBT target. Some international guidelines strongly recommend achieving a DTBT of within 90 min.⁶ The European Society of Cardiology (ESC) and the Chinese Guidelines for PPCI propose pursuing an even shorter target of ≤60 min as the ideal goal to optimise prognostic outcomes.^{7 8} However, recent studies suggested that pursuing extremely short DTBT may not yield equivalent benefits for all patients, indicating a need for more individualised assessment approaches.^{9 10} The Global Registry of Acute Coronary Events (GRACE) risk score is a well-validated tool for mortality risk assessment in acute coronary syndrome (ACS). Nevertheless, current evidence is limited regarding whether the association between DTBT and long-term prognosis varies across GRACE risk strata, and whether a stricter ≤60 min target is equally beneficial for both high- and low-risk patients. Therefore, this retrospective cohort study aimed to investigate whether the association between DTBT ≤60 min and short- and long-term outcomes in STEMI patients varies by GRACE risk score.

Methods

Patient population

This retrospective cohort study initially included 5516 STEMI patients who underwent PPCI treatment at the Department of Cardiology, General Hospital of Northern Theatre Command, from March 2016 to March 2022. After screening according to inclusion and exclusion criteria, 4513 patients were ultimately enrolled. Inclusion criteria were: (1) Age >18 years, (2) Presentation within 12 hours of symptom onset and diagnosis of STEMI,¹¹ (3) Admission for PPCI treatment and (4) Complete medical records and post-discharge follow-up data. Exclusion criteria were: (1) Symptoms lasting ≥12 hours, (2) Failure to receive PPCI and (3) Incomplete primary outcome data. The detailed patient selection process is illustrated in the flowchart (online supplemental figure 1). A formal a priori sample size calculation was not performed, as the study aimed to enrol all eligible patients consecutively throughout the study period to minimise selection bias and ensure representativeness of real-world clinical practice. All patients underwent coronary intervention during hospitalisation and received dual antiplatelet therapy (aspirin and P₂Y₁₂ receptor inhibitor) prior to PPCI. Baseline and perioperative clinical data were collected.

Data definitions

We collected clinical data from patients meeting the inclusion and exclusion criteria. Clinical baseline characteristics included age, gender, hypertension, heart failure, GRACE score and arrhythmia status, etc. Laboratory tests included haemoglobin, low-density lipoprotein cholesterol, left ventricular ejection fraction and N-terminal pro-B-type natriuretic peptide (NT-proBNP), etc. Patient’s medication history included aspirin, P₂Y₁₂ receptor inhibitors, tirofiban, etc. Patients were divided into two groups based on GRACE scores at admission: a high risk-STEMI group (GRACE score >140, HR: high risk) and a low risk-STEMI group (GRACE score ≤140, LR: low risk). Within each group, patients were further categorised by DTBT into a short DTBT group (DTBT ≤60 min) and a long DTBT group (DTBT >60 min). Baseline characteristics, laboratory results, medication use, coronary lesions and procedural features were analysed for each group. Follow-up was conducted for 3 years via outpatient visits, telephone calls to examine the association of DTBT duration with long-term outcomes. The 60 min cut-off was selected a priori according to ESC and Chinese guidelines, not by ROC analysis, to directly address international guideline discrepancies.

Endpoints

The primary endpoints of this study were all-cause mortality (in-hospital, 1-year and 3-year). Secondary endpoints included major adverse cardiovascular and cerebrovascular events (MACCE) at 1-year and 3-year post-discharge. MACCE was defined as a composite endpoint of all-cause mortality, recurrent myocardial infarction, stroke and target-vessel revascularisation.

Statistical analysis

Data were analysed using SPSS V.25.0 and R V.4.4.1 software. Propensity score matching (PSM) was performed using the optimal matching algorithm to match HR-STEMI and LR-STEMI patients. PSM adjusted for all baseline variables in online supplemental table 2, including GRACE score, matching patients 1:1 with a calliper of 0.02. Normally distributed continuous data were described as mean±SD, and intergroup comparisons were performed using one-way analysis of variance. For skewed data, median and IQR were used for description, with intergroup comparisons performed using the Wilcoxon signed-rank test. The count data were described as frequency (percentage), with intergroup comparisons conducted using Fisher’s exact test or the χ² test.

Logistic regression was used for in-hospital mortality. Cox proportional hazards regression was used for 1-year and 3-year all-cause mortality and MACCE occurrence. All analyses were performed using the post-PSM cohorts. Variables with a univariate p value ≤0.1 were included in the multivariate models. The variables used were: sex, smoker, hypertension, DM, dyslipidaemia, previous MI, history of CCF, previous PCI, previous CABG, CKD, cardiogenic shock, LM lesion, emergency aspirin use, emergency P2Y12 use, LDL-C, LVEF, GRACE score. The interaction between DTBT and GRACE risk stratum was tested to assess effect modification. A Directed Acyclic Graph outlining the hypothesised causal relationships between DTBT, GRACE score and outcomes, as well as the role of all potential confounders, is provided in online supplemental figure 2.

Handling of missing data

Missing data were limited to LVEF (5.7%) and NT-proBNP (4.7%). Complete case analysis was used for the primary models. Multiple imputation was conducted as a sensitivity analysis, which yielded consistent results, confirming robustness (online supplemental table 1).

Availability of data and materials

The datasets analysed in the present study are available from the corresponding author on reasonable request.

Patient and public involvement

None.

Results

Clinical baseline data

From March 2016 to March 2022, 4513 STEMI patients undergoing PCI at the Department of Cardiology, General Hospital of Northern Theatre Command were included in the analysis. Among them, 2433 (54.0%) were classified as HR-STEMI and 2080 (46.0%) were classified as LR-STEMI. Among HR-STEMI patients, 1113 (45.7%) had DTBT ≤60 min. Among LR-STEMI patients, 1083 (52.0%) had DTBT ≤60 min. The median DTBT time for LR-STEMI patients was 60 (48, 76) min. The median time in the DTBT ≤60 min group was 49 (41, 55) min, while the median time in the DTBT >60 min group was 77 (68, 90) min. For HR-STEMI patients, the median DTBT duration was 63 (50, 81) min; the median duration in the DTBT ≤60 min group was 50 (43, 55) min, and in the DTBT >60 min group, it was 79 (70, 98) min. The distribution of GRACE scores in the two risk strata is shown in online supplemental figure 3. Baseline, medication, laboratory and procedural characteristics as well as clinical outcomes were analysed for each group (online supplemental tables 2–5). After PSM, the matched cohort consisted of 1107 patients in the HR-STEMI group and 990 patients in each cohort for the LR-STEMI group. Patients with DTBT ≤60 min in the LR-STEMI group had a significantly higher prevalence of hyperlipidaemia (33.0% vs 28.2%, p=0.019). The remaining baseline characteristics demonstrated no discernible differences between the comparison groups. (p>0.05) (table 1).

Table 1. Baseline characteristics of the Post-PSM cohort for HR-STEMI and LR-STEMI depending on DTBT.

Characteristics	HR-STEMI (n=2214)		P value	LR-STEMI (n=1980)		P value
	DTBT ≤60 min (n=1107)	DTBT >60 min (n=1107)		DTBT ≤60 min (n=990)	DTBT >60 min (n=990)
Female, %	27.0	27.7	0.703	13.0	12.1	0.542
Age, median (IQR)	66 (61–73)	66 (60–73)	0.607	52 (45–58)	52 (46–58)	0.696
Heart rate, median (IQR)	83 (72–93)	84 (73–95)	0.162	80 (71–90)	80 (71–90)	0.607
SBP, median (IQR)	119 (106–135)	117 (105–133)	0.118	127 (114–143)	129 (117–144)	0.184
Smoker, %	60.1	59.3	0.729	72.9	72.2	0.724
Drinker, %	29.5	30.8	0.517	37.7	35.9	0.401
Hypertension,%	52.3	52.9	0.766	49.7	51.0	0.559
DM, %	25.7	26.6	0.629	24.5	24.7	0.917
Dyslipidaemia,%	24.9	26.4	0.436	33.0	28.2	0.019
Stroke, %	16.8	17.7	0.574	10.5	10.1	0.767
History of CCF, %	2.7	2.6	0.895	1.1	0.8	0.489
Previous MI, %	7.3	8.1	0.474	5.6	6.2	0.566
PVD, %	0.5	0.5	0.762	0.7	0.5	0.563
Previous-PCI, %	7.4	8.7	0.274	5.9	6.2	0.777
Previous-CABG, %	0.2	0.3	1.0	0.3	0.2	1.0
Cancer, %	1.1	1.4	0.561	0.7	0.4	0.364
CKD, %	0.8	0.8	1.0	0.7	0.6	0.781
Arrhythmia, %	4.6	4.9	0.764	1.6	1.5	0.856
Cardiogenic shock, %	5.3	5.9	0.579	0.9	0.7	0.616
OHCA, %	0.2	0.2	1.0	0	0	–

CABG, coronary artery bypass graft surgery; CCF, congestive heart failure; CKD, chronic kidney disease; DM, diabetes mellitus; DTBT, door-to-balloon time; HR-STEMI, high-risk STEMI; IQR, interquartile range; LR-STEMI, low-risk STEMI; MI, myocardial infarction; OHCA, out-of-hospital cardiac arrest; PCI, percutaneous coronary intervention; PVD, peripheral vascular disease; SBP, Systolic Blood Pressure; STEMI, ST-segment elevation myocardial infarction.

Laboratory findings and medication

Regarding laboratory findings and medication (table 2) during hospitalisation, within the HR-STEMI group, patients with DTBT ≤60 min received a loading dose of aspirin (85.9% vs 81.6%, p=0.006) and a loading dose of P2Y12 receptor antagonists (85.0% vs 80.8%, p=0.008) more frequently than patients with DTBT >60 min. Conversely, tirofiban use (13.6% vs 17.8%, p=0.007) during hospitalisation was significantly lower in the DTBT ≤60 min group. In the LR-STEMI group, patients with DTBT ≤60 min had significantly higher rates of loading dose of aspirin use (86.8% vs 83.3%, p=0.032), loading dose of P2Y12 receptor antagonist use (83.2% vs 78.5%, p=0.007) compared with patients with DTBT >60 min. Regarding laboratory parameters, total cholesterol (5.0 (4.3, 5.8) mmol/L vs 4.8 (4.2, 5.6) mmol/L, p=0.001), low-density lipoprotein cholesterol (3.0 (2.5, 3.6) mmol/L vs 2.9 (2.4, 3.5) mmol/L, p=0.007) and LVEF (52.5% (47%, 57%) vs 53% (48%, 58%), p=0.001) showed statistically significant differences between DTBT ≤60 min and DTBT >60 min.

Table 2. Medication laboratory and procedural characteristics of the Post-PSM cohort for HR-STEMI and LR-STEMI depending on DTBT.

Characteristics	HR-STEMI (n=2214)		P value	LR-STEMI (n=1980)		P value
	DTBT ≤60 min (n=1107)	DTBT >60 min (n=1107)		DTBT ≤60 min (n=990)	DTBT >60 min (n=990)
Emergency medication, %
Aspirin	85.9	81.6	0.006	86.8	83.3	0.032
P2Y12 Inhibitor	85.0	80.8	0.008	83.2	78.5	0.007
In-hospital medication, %
Aspirin	99.4	98.7	0.125	99.5	99.1	0.283
P2Y12 Inhibitor	99.2	98.5	0.115	99.3	99.0	0.465
Tirofiban	13.6	17.8	0.007	16.2	15.6	0.712
Total cholesterol, median (IQR)	4.7 (4.1–5.5)	4.8 (4.0–5.6)	0.811	5 (4.3–5.8)	4.8 (4.2–5.6)	0.001
Triglycerides, median (IQR)	1.2 (0.9–1.7)	1.3 (0.9–1.7)	0.147	1.5 (1.0–2.2)	1.5 (1.0–2.2)	0.887
LDL-C, median (IQR)	2.8 (2.3–3.4)	2.8 (2.3–3.4)	0.428	3.0 (2.5–3.6)	2.9 (2.4–3.5)	0.007
Platelet, median (IQR)	215 (182–250)	218 (184–249)	0.707	232 (200–266)	233.5 (196–268)	0.817
WBC, median (IQR)	10 (8–12)	11 (8–13)	0.092	11.3 (9.4–13.3)	11.1 (9.1–13.1)	0.221
Haemoglobin, median (IQR)	139 (130–150)	139 (129–149)	0.183	147.5 (139–157)	147.5 (139–156.3)	0.476
Serum creatinine, median (IQR)	71 (59–83)	72 (60–83)	0.126	67.5 (57.8–77.0)	68.4 (59.1–77.9)	0.160
CK-MB, median (IQR)	39 (16–111)	36 (17–101)	0.837	44 (17–110)	39 (17–96)	0.270
Troponin, median (IQR)	4 (2–7)	5 (2–7)	0.629	3.4 (1.6–5.4)	3.6 (1.5–5.4)	0.838
Creatine kinase, median (IQR)	347 (129–1097)	355 (124–1015)	0.996	475.5 (145–1224.8)	405.5 (141–1076.5)	0.251
NT-proBNP, median (IQR)	358 (101–1260)	428 (124–1260)	0.126	205.1 (54.0–743.9)	222 (64.9–705.8)	0.717
LVEF, median (IQR)	50 (45–57)	50 (45–57)	0.423	52.5 (47–57)	53 (48–58)	0.001
Target vessels, %
LM	0.9	1.3	0.412	0.6	0.7	0.781
LAD	46.4	42.8	0.087	48.4	49.3	0.686
LCX	8.4	7.4	0.386	10.7	9.6	0.413
RCA	41.4	43.8	0.246	35.9	35.5	0.851
Pre-PCI TIMI flow 0/1, %	79.9	82.6	0.103	72.2	73.2	0.614
Post-PCI TIMI flow 3, %	95.6	92.6	0.003	95.7	93.9	0.085
Thrombus aspiration, %	9.2	14.2	<0.001	10.3	11.7	0.315
Stents number, median (IQR)	1.0 (1.0–1.0)	1.0 (1.0–1.0)	0.003	1.0 (1.0–1.0)	1.0 (1.0–1.0)	0.819
Stents length, median (IQR)	29.0 (18.0–36.0)	28.0 (18.0–36.0)	0.052	28.0 (18.0–36.0)	24.0 (18.0–36.0)	0.103
Stents diameter, median (IQR)	3.0 (2.5–3.0)	3.0 (2.5–3.0)	0.010	3.0 (2.8–3.5)	3.0 (2.5–3.5)	0.185
Temporary pacing, %	2.3	3.4	0.128	0.7	1.5	0.086
IABP, %	4.0	6.0	0.031	1.2	1.6	0.446

CK-MB, Creatine Kinase Isoenzymes; DTBT, door-to-balloon time; HR-STEMI, high-risk STEMI; IABP, intra-aortic balloon pump; IQR, interquartile range; LAD, Left Anterior Descending Artery; LCX, Left Circumflex Artery; LDL-C, low density lipoprotein cholesterol; LM, left main coronary artery; LR-STEMI, low risk STEMI; LR-STEMI, low risk STEMI; LVEF, left ventricular ejection fraction; NT-proBNP, N-terminal pro-B-type natriuretic peptide; PCI, percutaneous coronary intervention; P₂Y₁₂, clopidogrel or ticagrelor; RCA, Right Coronary Artery; STEMI, ST-segment elevation myocardial infarction; TIMI, thrombolysis in myocardial infarction; WBC, white blood cell.

Procedural characteristics

For HR-STEMI patients, DTBT ≤60 min versus DTBT >60 min showed significant differences in the number of implanted stents (1.0 (1.0, 1.0) vs 1.0 (1.0, 1.0), p=0.003), stent diameter (3.0 (2.5–3.0) mm vs 3.0 (2.5–3.0) mm, p=0.010), post-procedure TIMI flow 3 (95.6% vs 92.6%, p=0.003), thrombus aspiration (9.2% vs 14.2%, p<0.001) and IABP implantation (4% vs 6%, p<0.031). For LR-STEMI patients, no statistically significant differences were observed in procedure-related variables between the DTBT ≤60 min and DTBT >60 min groups (p>0.05) (table 2).

Clinical outcomes

Based on the PSM-adjusted cohort analysis, for HR-STEMI patients, comparisons between DTBT ≤60 min versus DTBT >60 min showed significant differences in in-hospital mortality (1.1% vs 2.5%, p=0.011), 1-year mortality (4.5% vs 7.2%, p=0.007), 1-year MACCE incidence (17.6% vs 21.1%, p=0.036), 3-year mortality (6.9% vs 11.6%, p<0.001) and 3-year MACCE incidence (22.8% vs 29.1%, p=0.001). In patients with LR-STEMI, there were no significant differences in the above indexes between those with a DTBT of >60 min and those with a DTBT of ≤60 min (table 3).

Table 3. Clinical outcomes of the Post-PSM cohort for HR-STEMI and LR-STEMI depending on DTBT.

Characteristics	HR-STEMI (n=2214)		P value	LR-STEMI (n=1980)		P v alue
	DTBT ≤60 min (n=1107)	DTBT >60 min (n=1107)		DTBT ≤60 min (n=990)	DTBT >60 min (n=990)
In-hospital mortality, %(n)	1.1 (12)	2.5 (28)	0.011	0.2 (2)	0.7 (7)	0.179
One-year outcomes, %(n)
MACCE	17.6 (195)	21.1 (234)	0.036	11.9 (118)	11.8 (117)	0.945
Mortality	4.5 (50)	7.2 (80)	0.007	1.3 (13)	1.9 (19)	0.285
MI	0.2 (2)	0.6 (7)	0.179	0.6 (6)	0.1 (1)	0.124
TVR	12.3 (136)	12.5 (138)	0.897	9.3 (92)	9.4 (93)	0.938
Stroke	0.6 (7)	0.8 (9)	0.616	0.7 (7)	0.4 (4)	0.364
Three-year outcomes, %(n)
MACCE	22.8 (252)	29.1 (322)	0.001	16.9 (167)	16.9 (167)	1.0
Mortality	6.9 (76)	11.6 (128)	<0.001	2.5 (25)	3.3 (33)	0.286
MI	0.7 (8)	0.8 (9)	0.808	0.7 (7)	0.7 (7)	1.0
TVR	14.4 (159)	15.4 (170)	0.511	12.1 (120)	11.8 (117)	0.835
Stroke	0.8 (9)	1.4 (15)	0.218	1.5 (15)	1.0 (10)	0.314

DTBT, door-to-balloon time; HR-STEMI, high-risk STEMI; LR-STEMI, low-risk STEMI; MACCE, major adverse cardiovascular and cerebrovascular events; MI, myocardial infarction; STEMI, ST-segment elevation myocardial infarction; TVR, target-vessel revascularization.

The unadjusted and adjusted ORs for in-hospital mortality are presented in online supplemental table 4. The unadjusted and adjusted HRs for 1-year and 3-year MACCE and mortality, along with person-years and mortality rates, are presented in online supplemental table 5 and figure 6, respectively.

Multivariate logistic regression analysis revealed that DTBT >60 min was associated with a higher risk of in-hospital mortality compared with DTBT ≤60 min (OR=2.381, 95% CI 1.160 to 4.883, p=0.018). Multivariate Cox proportional hazards regression analysis revealed that DTBT >60 min was associated with an increased risk of 1-year post-discharge mortality (HR=1.715, 95% CI 1.194 to 2.464, p=0.003) and an increased risk of MACCE occurrence within 1-year post-discharge (HR=1.212, 95% CI 1.001 to 1.467, p=0.049) compared with DTBT ≤60 min. Furthermore, patients with DTBT >60 min had a higher risk of mortality within 3-year post-discharge (HR=1.689, 95% CI 1.267 to 2.253, p<0.001) and an increased risk of MACCE within 3-year post-discharge (HR=1.230, 95% CI 1.042 to 1.453, p=0.014) (online supplemental table 6).

After multivariable adjustment, a significant interaction was observed between DTBT and GRACE risk stratum for all endpoints (p for interaction <0.05), indicating that the association of DTBT with prognosis was modified by baseline risk. Based on the PSM cohort, figures 1–4 show the survival curves for all-cause mortality and MACCE incidence in the HR- and LR-STEMI groups according to DTBT.

Figure 1. One-year survival curves for HR-STEMI stratified by DTBT. One-year Kaplan-Meier curves for mortality and MACCE in HR-STEMI patients stratified by door-to-balloon time (DTBT ≤60 min vs >60 min). The log-rank test revealed a statistically significant difference for mortality (p=0.003) and a borderline significant difference for MACCE (p=0.05), indicating that shorter DTBT (≤60 min) was associated with lower incidence of both endpoints. DTBT, door-to-balloon time; HR-STEMI, high-risk STEMI; MACCE, major adverse cardiovascular and cerebrovascular events; STEMI, ST-segment elevation myocardial infarction.

Figure 4. Three-year survival curves for LR-STEMI stratified by DTBT. Three-year Kaplan-Meier curves for mortality and MACCE in LR-STEMI patients stratified by door-to-balloon time (DTBT ≤60 min vs >60 min). The log-rank test showed no statistically significant difference between the two groups for mortality (p=0.098) or MACCE (p=0.595). DTBT, door-to-balloon time; LR-STEMI, low-risk STEMI; MACCE, major adverse cardiovascular and cerebrovascular events; STEMI, ST-segment elevation myocardial infarction.

Figure 2. Three-year survival curves for HR-STEMI stratified by DTBT. Three-year Kaplan-Meier curves for mortality and MACCE in HR-STEMI patients stratified by door-to-balloon time (DTBT ≤60 min vs >60 min). The log-rank test revealed statistically significant differences between the two groups, indicating that shorter DTBT (≤60 min) was associated with lower incidences of both mortality (p<0.001) and MACCE (p=0.014). DTBT, door-to-balloon time; HR-STEMI, high-risk STEMI; MACCE, major adverse cardiovascular and cerebrovascular events; STEMI, ST-segment elevation myocardial infarction.

Figure 3. One-year survival curves for LR-STEMI stratified by DTBT. One-year Kaplan-Meier curves for mortality and MACCE in LR-STEMI patients stratified by door-to-balloon time (DTBT ≤60 min vs >60 min). The log-rank test showed no statistically significant difference between the two groups for mortality (p=0.363) or MACCE (p=0.970). DTBT, door-to-balloon time; LR-STEMI, low-risk STEMI; MACCE, major adverse cardiovascular and cerebrovascular events; STEMI, ST-segment elevation myocardial infarction.

In a sensitivity analysis using alternative GRACE cut-offs (120, 130 and 140), the association between DTBT >60 min and outcomes was consistently observed in patients above each cut-off. Notably, patients with GRACE scores between 120 and 140 also showed benefit (online supplemental table 7).

Discussion

Timely reperfusion is the cornerstone of STEMI management, and DTBT has been established as a key quality indicator. However, whether a stricter ≤60 min target is equally beneficial across different risk strata remains unclear.^{12 13} Through this longitudinal retrospective cohort analysis, the survival advantage associated with shortening DTBT to ≤60 min was not uniformly distributed across all STEMI patients but was almost exclusively concentrated in the HR-STEMI group with GRACE >140. HR-STEMI patients who achieved a DTBT ≤60 min had significantly better outcomes, including lower in-hospital mortality and reduced risks of mortality and MACCE at 1 and 3 years. However, no significant additional improvement was observed in LR-STEMI patients when DTBT was reduced to ≤60 min.

A large observational study previously reported a significant association between DTBT reduction and mortality decrease, with DTBT shortening from 90 to 60 min reducing in-hospital mortality by 0.8%.¹⁴ A large Korean registry study further found that in-hospital mortality was 3.6 times higher in patients with DTBT ≥90 min compared with those with DTBT <60 min, identifying 59 min as the optimal cut-off point via ROC curve analysis.¹⁵ Moreover, a contemporary patient-level analysis by Park et al¹⁶ showed that each 30 min reduction in DTBT was continuously associated with a lower risk of 1 year all-cause mortality, showing an absolute risk reduction of 2.4% specifically when DTBT was reduced from 90 to 60 min. However, research by Menees et al¹⁷ indicated that despite significant nationwide improvements in DTBT across the United States, corresponding reductions in in-hospital mortality were not observed. Zahidi et al¹⁸ also reported no significant difference in outcomes between patients with DTBT ≤60 min and those with DTBT >60 min after adjusting for baseline characteristics. This conflicting evidence suggests that focusing solely on a universal target for DTBT may overlook inherent differences among distinct patient populations. Additionally, there is a paucity of data regarding the long-term prognostic implications of DTBT, as existing literature has predominantly centred on short-term mortality. Therefore, this study incorporated the GRACE score to stratify STEMI patients, separately validating the short-term and long-term benefits of reducing DTBT to 60 min.

The observed survival advantage of achieving DTBT ≤60 min in HR-STEMI patients may be related to two interrelated mechanisms: securing effective immediate reperfusion and mitigating delay-associated procedural risks. First, HR-STEMI patients present with more severe conditions. These patients typically exhibit larger infarct sizes¹⁹ (as indicated by higher Creatine Kinase Isoenzymes peaks). Additionally, this group includes a higher proportion of elderly patients who have poorer physiological reserves and reduced tolerance to ischaemia.^20–24 These factors result in an extremely narrow therapeutic window, where myocardial necrosis can occur with any delay in treatment. Second, our data suggest that timely reperfusion is associated with significantly better procedural outcomes. In HR-STEMI patients, DTBT ≤60 min was significantly associated with achieving Post-procedure TIMI flow 3, indicating superior microvascular reperfusion. Furthermore, these patients required fewer stents and less thrombus aspiration, suggesting that timely intervention may be associated with reduced progression of thrombus burden and increased lesion complexity. Combined with findings from the IABP-SHOCK II study,²⁵ significantly higher IABP usage among patients with longer DTBT further indicates more critical baseline or intraoperative conditions in these individuals. This inherent haemodynamic instability inherently demands stricter adherence to rapid reperfusion principles, reinforcing why achieving the ≤60 min target is particularly crucial for HR-STEMI patients.

Another key finding of this study is that shortening DTBT from >60 min to ≤60 min was not significantly associated with improved outcomes in patients with LR-STEMI. This observation aligns with the current research shift in focus from DTBT to total ischaemic time. A recent large-scale study suggests that DTBT itself may no longer be an independent predictor of long-term mortality, whereas total ischaemic time holds decisive value.²⁶ This perspective may help interpret the findings in LR-STEMI patients in our study: their smaller infarct size, stable haemodynamics and fewer comorbidities collectively form a solid foundation for ischaemia tolerance. Therefore, LR-STEMI patients in this study may fall within a time window characterised by relatively short total ischaemic time and favourable prognosis, meaning variations in DTBT within the 60 to 90 min range have minimal impact on final outcomes. Conversely, HR-STEMI patients often experience significantly prolonged total ischaemic time due to complex conditions and prehospital delays, and any delay in DTBT may be associated with a substantially increased myocardial necrosis risk.

To better understand which patient groups may have a survival advantage associated with a shorter DTBT, the use of a validated risk stratification tool is fundamental. This study employed the GRACE risk score to stratify STEMI patients. The GRACE score was selected as it is currently the only risk assessment tool validated in large-scale prospective cohorts, predictive of long-term prognosis and applicable to the entire spectrum of ACSs, including both STEMI and NSTEMI.^{27 28} Its use in this study provided a validated framework suggesting that the survival advantage of DTBT ≤60 min is concentrated in HR-STEMI patients.

A sensitivity analysis using alternative GRACE cut-offs (120, 130 and 140) suggested that the benefit of DTBT ≤60 min may extend to patients with GRACE scores between 120 and 140 (online supplemental table 7). This finding indicates that the association between shorter DTBT and better outcomes is not strictly limited to the guideline-defined high-risk group, and that a subset of patients traditionally classified as low-risk (GRACE 120–140) may also benefit from achieving a ≤60 min target.

It is also important to acknowledge that our findings reflect the outcomes of STEMI patients treated with conventional PPCI, with IABP used as needed for haemodynamic support, but without routine use of ECMO or Impella. In our centre, IABP is the preferred mechanical circulatory support (MCS) device for STEMI patients when indicated, whereas ECMO and Impella are not routinely employed for these patients. However, an increasing number of HR-STEMI patients may require advanced MCS with ECMO or Impella. These patients not only present with complex conditions but also require large-bore arterial access for device implantation, which carries substantial risks of vascular and bleeding complications that can significantly impact recovery and survival. A comprehensive review by Sinning et al highlighted these risks in Impella-assisted high-risk PCI.²⁹ As such, our findings may be most directly applicable to settings where PPCI is performed with IABP but without routine use of ECMO or Impella, and further studies are needed to determine whether the observed association between DTBT and outcomes in HR-STEMI patients is consistent in MCS-supported populations employing different device strategies.

Limitations

This study has several limitations. First, the single-centre, retrospective design limits patient diversity and the generalisability of results, requiring further validation through multicentre, prospective studies. Second, despite using PSM, observational studies cannot eliminate all confounding factors. Unmeasured variables such as total ischaemic time, infarct location, operator experience, medication adherence, socioeconomic status and education level may be associated with outcomes. Third, multivessel coronary artery disease was not specifically adjusted for in the primary analysis. Although our centre’s protocol prioritises culprit-vessel PCI during the acute phase, with staged revascularisation for non-culprit lesions performed electively, the presence of multivessel disease may reflect greater overall atherosclerotic burden and potentially influence long-term outcomes. This should be considered as a potential source of residual confounding. Fourth, no patients in our cohort received MCS (ECMO or Impella) during PCI, as these devices are not routinely used for STEMI patients in our centre. Therefore, our findings may not be generalisable to centres where such support is commonly used. Fifth, the GRACE score was measured only at admission and did not capture dynamic changes during hospitalisation or after treatment, which may have prognostic implications. Sixth, whether the study conclusions fully apply to hospitals lacking direct PCI capabilities and requiring remote transport requires further validation. As with all observational studies, causality cannot be inferred, and our findings should be considered hypothesis-generating.

Conclusions

Our findings suggest that the association of DTBT with clinical outcomes is not uniform but is critically contingent on the patient’s baseline risk stratification. For HR-STEMI patients, DTBT >60 min was independently associated with poorer outcomes. Compared with patients with DTBT ≤60 min, those with DTBT >60 min had significantly higher in-hospital mortality and higher risks of mortality and MACCE at 1 and 3 years post-discharge. However, no such associations were observed in LR-STEMI patients. A sensitivity analysis using alternative cut-offs further suggested that the benefit of DTBT ≤60 min may extend to patients with GRACE scores between 120 and 140. Consequently, the current uniform DTBT target may not optimally meet the clinical needs of patients across different risk strata. If validated in future prospective studies, a risk-stratified approach prioritising DTBT ≤60 min for HR-STEMI patients could have implications for clinical practice and resource allocation.

Data availability statement

Data are available upon reasonable request.