Delaying reperfusion plus left ventricular unloading reduces infarct size: Sub-analysis of DTU-STEMI pilot study

Introduction: The STEMI-DTU pilot study tested the early safety and practical feasibility of left ventricular (LV) unloading with a trans-valvular pump before reperfusion. In the intent-to-treat cohort, no difference was observed for microvascular obstruction (MVO) or infarct size (IS) normalized to either the area at risk (AAR) at 3-5 days or total LV mass (TLVM) at 3-5 days We now report a per protocol analysis of the STEMI-DTU pilot study. Methods: In STEMI-DTU STUDY 50 adult patients (25 in each arm) with anterior STEMI [sum of precordial ST-segment elevation (Sigma STE) >= 4 mm] requiring primary percutaneous coronary intervention (PCI) were enrolled. Only patients who met all inclusion and exclusion criteria were included in this analysis. Cardiac magnetic resonance (CMR) imaging 3-5 days after PCI quantified IS/AAR and IS/TLVM and MVO. Group differences were assessed using Student's t-tests and linear regression (SAS Version-9.4). Results: Of the 50 patients enrolled, 2 died before CMR imaging. Of the remaining 48 patients those without CMR at 3-5 days (n = 8), without PCI of a culprit left anterior descending artery lesion (n = 2), with OHCA (n = 1) and with Sigma STE < 4 mm (n = 5) were removed from this analysis leaving 32/50 (64 %) patients meeting all inclusion and exclusion criteria (U-IR, n = 15; U-DR, n = 17) as per protocol. Despite longer symptom-to-balloon times in the U-DR arm (228 +/- 80 vs 174 +/- 59 min, p < 0.01), IS/AAR was significantly lower with 30 min of delay to reperfusion in the presence of active LV unloading (47 +/- 16 % vs 60 +/- 15 %, p = 0.02) and remained lower irrespective of the magnitude of precordial Sigma STE =. MVO was not significantly different between groups (1.5 +/- 2.8 % vs 3.5 +/- 4.8 %, p = 0.15). Among patients who received LV unloading within 180 min of symptom onset, IS/AAR was significantly lower in the U-DR group. Conclusion: In this per-protocol analysis of the STEMI-DTU pilot study we observed that LV unloading for 30 min before reperfusion significantly reduced IS/AAR compared to LV unloading and immediate reperfusion, whereas in the ITT cohort no difference was observed between groups. This observation supports the design of the STEMI-DTU pivotal trial and suggests that strict adherence to the study protocol can significantly influence the outcome.