Abstract Aims This network meta-analysis of randomized controlled trials (RCTs) evaluates the comparative safety and efficacy of intra-coronary (IC) pharmacological and procedural treatments - on top of balloon angioplasty and stent placement - on clinical outcomes and surrogate endpoints of coronary microvascular obstruction (CMVO) in patients with ST-elevation myocardial infarction (STEMI). Methods and Results Two electronic databases were searched for eligible studies. Primary efficacy endpoints included all-cause mortality, non-fatal myocardial infarction (MI), and heart failure (HF) hospitalization. Primary safety endpoints included peri-procedural arrhythmias including atrioventricular blocks (AVBs) and ventricular fibrillation/sustained ventricular tachycardia (VF/SVT), any bleeding, and stroke. Secondary efficacy endpoints included the occurrence of post-procedural thrombolysis in myocardial infarction (TIMI) flow grade 0-2 and ST-segment resolution. A total of 64 RCTs involving 27,243 patients were included. In mixed comparisons, no treatment significantly reduced the incidence of primary efficacy endpoints compared to conventional primary PCI during a mean follow-up of 8 months. Several treatments significantly reduced the occurrence of post-PCI TIMI 0-2 flow grade adenosine: 0.40 (odds ratio), (95% Confidence Interval 0.24-0.68); verapamil: 0.22 (0.07-0.69); tirofiban:0.43 (0.27-0.71); manual thrombus aspiration (TA): 0.61 (0.45-0.82); fibrinolytic + manual TA: 0.24 (0.12-0.48); tirofiban + manual TA: 0.32 (0.14-0.75), compared to conventional primary PCI. IC administration of tirofiban increased the risk of any bleeding incidence rate ratio: 1.65 (1.11-2.45), while IC adenosine increased the risk of peri-procedural AVBs OR: 2.80 (1.14-6.84).Nicorandil reduced the incidence of peri-procedural VF/SVT OR: 0.31 (0.12-0.81). Conclusions Adjunctive IC treatments during primary PCI do not influence hard clinical outcomes compared to conventional therapy within a mean 8-month follow-up, although several of them lead to an improvement in surrogate endpoints of CMVO.
Laborante et al. (Sun,) studied this question.