The Paper
Sammour YM, Khan SU, Hong H, et al. Institutional Variability in Processes of Care and Outcomes Among Patients With STEMI in the US. JAMA Cardiol. Published online June 11, 2025. doi:10.1001/jamacardio.2025.1411. PMID: 40498491
In STEMI, “time is myocardium”—but are we actually achieving timely PCI in the United States? Current ACC/AHA guidelines recommend:
- ≤90 min First Medical Contact (FMC)-to-device time for primary presentations to PCI centers
- ≤120 min for interhospital transfers
This study explored how well hospitals met guideline-recommended FMC-to-device time targets and whether delays actually translate into worse in-hospital outcomes, especially mortality. It’s one of the largest and most recent analyses of STEMI care performance.
Study Design
Clinical Question
Are delays in PCI—measured by failure to meet FMC-to-device time goals—associated with higher in-hospital mortality and worse outcomes in STEMI patients?
Study Design
- Type: Retrospective cross-sectional
- Data Source: AHA Get With The Guidelines–CAD Registry (2020–2022)
- Population: 73,826 patients from 503 US hospitals
- Inclusion: STEMI/STEMI equivalents undergoing primary PCI split into two cohorts
- Primary (walked into PCI-capable hospital)
- Transfers (initial hospital → PCI hospital)
- Exclusion: Patients treated with fibrinolysis, missing data, or who didn’t receive PCI
- Exposures: First medical contact (FMC) to device time with target times ≤ 90 min and ≤ 120 min for primary and transfer sites respectively
- Primary Outcomes: In-hospital mortality, length of stay
Hospitals Stratified by Performance
Hospital Performance was grouped into tertiles (high, intermediate, and low) by how often they met:
- ≤90 min FMC-to-device time for primary presentations
- ≤120 min FMC-to-device time for transfer patients
Outcomes Measured
- Primary: In-hospital mortality and length of stay (LOS)
- Secondary:
- Cardiogenic shock
- Heart failure
- Cardiac arrest
- Stroke
- Major bleeding
Statistical Methods
- Hierarchical logistic regression models: These allow adjusting for both patient-level factors (age, sex, vitals, comorbidities) and hospital-level differences (size, teaching status, region). Covariates included:
- Patient factors: age, sex, race, insurance status, BMI, comorbidities (HTN, DM, CAD, CKD, PAD, CVA, LVEF), vitals, arrest/shock/heart failure on arrival, and labs (creatinine)
- Site factors: hospital size, teaching status, urban/rural location, regional distribution
- Adjusted Odds Ratio (aOR): Measures how much more likely an outcome is if a patient missed the PCI time target, controlling for other factors (see above)
- Example: aOR 2.21 for mortality = 2.2x greater odds of death if PCI was delayed.
- Confidence Interval (CI): Tells us the range where the “true” value probably lies. If the CI does not cross 1.0 for an odds ratio, it's statistically significant.
Results
Primary Outcome
Outcome | Primary Presentations | Transfers |
% Meeting Target Time | 59.5% (≤90 min) | 50.3% (≤120 min) |
In-Hospital Mortality Risk (missed target FMC to device time) | aOR 2.21 (95% CI 2.02–2.42) | aOR 2.44 (95% CI 1.90–3.12) |
LOS (missed target FMC to device time) | aOR 1.35 (95% CI 1.30–1.41) | aOR 1.35 (95% CI 1.28–1.42) |
Explanation: Only about half of STEMI patients met guideline-recommended PCI time targets, and those who missed them had more than double the odds of in-hospital death and significantly longer hospital stays.
Primary Outcome by Hospital Performance (High vs. Low)
Group | % Met PCI Target Time | Median FMC-to-Device |
Primary, High Performance | 72.2% | 76 min |
Primary, Low Performance | 46.0% | 93 min |
Transfer, High Performance | 72.3% | 100 min |
Transfer, Low Performance | 21.9% | 155 min |
Outcome | Primary Presentations | Transfers |
Mortality Risk (low vs high performing hospitals) | aOR 1.16 (1.00–1.34) | aOR 1.09 (0.96–1.25)
❌ NS (not statistically significant) |
LOS (low vs high performers) | aOR 1.21 (1.15–1.27) | aOR 1.57 (1.48–1.67) |
Explanation: High-performing hospitals achieved PCI time goals in over 70% of cases with substantially shorter FMC-to-device times, yet showed only a modest or non-significant reduction in mortality compared to low-performing sites. However, patients at low-performing hospitals had significantly longer hospital stays, especially among transfers.
Secondary Outcomes: Missed FMC to Device Target
Outcome | Primary Presentation | Transfers |
Cardiogenic Shock | aOR 1.62 (95% CI 1.51–1.74) | aOR 1.86 (95% CI 1.56–2.22) |
New/Worsening HF | aOR 1.41 (95% CI 1.33–1.49) | aOR 1.52 (95% CI 1.29–1.78) |
Cardiac Arrest | aOR 1.58 (95% CI 1.47–1.70) | aOR 1.81 (95% CI 1.50–2.19) |
Major Bleeding | aOR 1.03 (95% CI 0.94–1.13) ❌ NS | aOR 1.13 (95% CI 0.95–1.34) ❌ NS |
Stroke | aOR 1.02 (95% CI 0.87–1.21) ❌ NS | aOR 1.10 (95% CI 0.83–1.46) ❌ NS |
Explanation: Missing PCI time goals was associated with significantly higher odds of cardiogenic shock, heart failure, and cardiac arrest—but not with major bleeding or stroke—highlighting that delays primarily increase cardiopulmonary complications, not procedural ones.
Author Conclusions
“Failure to meet guideline-recommended FMC-to-device time targets was associated with higher in-hospital mortality and longer length of stay. However, inter-hospital differences in performance had minimal effect on mortality, suggesting that patient-level factors may dominate hospital-level differences.”
Critical Appraisal
RAMMBO Assessment (FIRST10EM)
BEEM Prognostic Appraisal (Best Evidence in Emergency Medicine)
Strengths
- Huge sample size
- Detailed patient-level variables (shock, cardiac arrest, LVEF, vitals, labs) allow robust statistical adjustment for confounders
- Separately analyzes primary vs transfer STEMI
- Real-world hospital performance stratification
- In-hospital mortality and LOS are hard endpoints that minimize reporting bias.
- Mortality and LOS are highly relevant outcomes
Limitations
- Observational study → can’t prove causation
- Does not include patients treated with fibrinolytics
- No data on symptom onset-to-device time
- Some delays may reflect sicker patients, even with adjustment
- Outcomes beyond hospital discharge (e.g., 30-60 day mortality, rehospitalization, heart failure progression) were not measured.
- Excludes fibrinolysis cohort which limits applicability to centers or patients where lytics are used, especially in rural or resource-limited settings.
Bottom Line
In STEMI care, faster PCI saves lives—yet nearly half of patients in this national cohort missed guideline-recommended door-to-balloon time targets. These delays were independently associated with increased in-hospital mortality, cardiogenic shock, heart failure, and cardiac arrest. The findings reinforce the clinical value of timely PCI and highlight the need to streamline STEMI systems of care, particularly for transfer patients. However, as an observational study limited to in-hospital outcomes, the analysis may either underestimate or overstate the true impact of PCI delays on long-term prognosis. Additionally, the study did not include patients treated with fibrinolytics—a strategy that may offer life-saving benefit when timely PCI is not achievable—limiting its applicability to resource-constrained settings where lytics are often the only option.
References
- Rao SV, Tamis-Holland JE, Nallamothu BK, et al. 2025 ACC/AHA/ACEP/NAEMSP/SCAI Guideline for the Management of Patients With Acute Coronary Syndromes: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. J Am Coll Cardiol. 2025;S0735-1097(24)06538-2. doi:10.1016/j.jacc.2024.05.001.
- Sammour YM, Khan SU, Hong H, et al. Institutional Variability in Processes of Care and Outcomes Among Patients With STEMI in the US. JAMA Cardiol. Published online June 11, 2025. doi:10.1001/jamacardio.2025.1411. PMID: 40498491