Beyond 72 Hours: Discharge Barriers and 30-Day Outcomes After Primary PCI in Low-Risk STEMI

Background/Objectives: Guidelines support discharge within 72 h for selected low-risk patients with ST-segment elevation myocardial infarction (STEMI) after primary percutaneous coronary intervention (PCI), yet many stay longer. We characterized the documented reasons for hospitalization beyond 72 h and examined the association between the discharge timing and 30-day clinical outcomes. Methods: In this retrospective single-center cohort of 920 consecutive patients treated with primary PCI (2023–2026), patients with a Zwolle risk score ≤ 3 were grouped by discharge ≤72 h or >72 h, and the principal documented reason for longer stays was abstracted. The thirty-day major adverse cardiovascular events (MACEs) and readmission were compared using propensity score matching, weighting and penalized regression. Results: Of 447 low-risk patients (73.60% men), 164 (36.69%) stayed beyond 72 h. Administrative, diagnostic or social reasons were documented in 130 (79.27%), minor clinical observation in 15 (9.15%), and an in-hospital complication in 19 (11.59%); the first group accounted for 205.45 bed-days beyond 72 h. In 126 matched pairs, a MACE occurred in 3.17% vs. 3.97% (conditional odds ratio [OR] 0.80, 95% confidence interval [CI] 0.16–3.72; p = 1.000) and readmission in 5.56% vs. 7.14% (p = 0.804). The adjusted and sensitivity analyses, including a comparison restricted to patients without in-hospital complications, showed no significant associations, although estimates were imprecise and the readmission estimate changed direction when the peak troponin was excluded from the propensity model. The baseline variables discriminated prolonged stay without complication poorly (optimism-corrected area under the curve [AUC], 0.62); planned staged revascularization showed the strongest association (OR 2.48, 95% CI 1.44–4.28). Conclusions: In Zwolle low-risk patients, documented administrative, diagnostic and social factors accounted for most stays beyond 72 h. No significant difference in 30-day outcomes was detected after matching, although the few events preclude conclusions of equivalence.