Pharmacoinvasive Approach with Streptokinase in Low to Intermediate Risk ST-Elevation Myocardial Infarction Patients: Insights from the Tamil Nadu-STEMI Initiative

Abstract

For patients presenting with ST-elevation myocardial infarction (STEMI), immediate and timely reperfusion is of paramount importance, and the optimal choice between primary percutaneous coronary intervention (PPCI) and intravenous thrombolytic agents depends upon the timeliness of effective delivery as well as the total ischemic time [1, 2]. Although rapid reperfusion with PPCI is widespread in the USA and Europe [3, 4], limited resources and lack of infrastructure means that patients with STEMI in lowand middle-income countries (LMICs) receive significantly less reperfusion therapy. When they do, these patients are more likely to receive thrombolytic agents over PPCI [5, 6]. For those who receive fibrin-specific thrombolytics, early catheterization and definitive revascularization is the preferred option based on trials supporting clinical equivalence of a pharmacoinvasive approach with PPCI [2, 7]. Conceptually, a pharmacoinvasive approach has particular promise for LMICs; however, use of fibrin-specific thrombolytics such as tenecteplase and alteplase can be cost-prohibitive compared with streptokinase in these communities [6]. We assessed the real-world efficacy of using streptokinase as part of a pharmacoinvasive approach within the Tamil Nadu (TN)-STEMI initiative, a large real-world registry study of reperfusion therapy in south India. Details of the TN-STEMI initiative have been described previously [8]. Briefly, this was a multicenter public–private initiative to improve access to invasive cardiac services for STEMI that connected 35 primary care health clinics and small hospitals (i.e., spokes) with four large PCI centers (i.e., hubs). Consecutive patients with STEMI presenting at these centers were enrolled between 2012 and 2014 in a “preimplementation/post-implementation” quasi-experimental study design. Utilizing a regional ambulance service and an indigenously developed low-cost monitoring and electrocardiogram transmission device, patients at spoke centers were routinely transferred to hub hospitals after initial thrombolysis. A large proportion of patients were living below the poverty line and participated in a statewide governmentsponsored health insurance scheme to cover costs of therapy. Data from all four clusters were combined for the current analysis. Our primary objective was to investigate outcomes among patients undergoing PPCI compared with those undergoing the pharmacoinvasive approach with streptokinase (PI-SK). The primary outcomes were all-cause mortality at 1 and 2 years, and the primary safety endpoint was inhospital major bleeding. We used propensity scores to adjust for baseline differences. Propensity scores for all patients to undergo PI-SK were first estimated using a nonparsimonious multivariable logistic regression model. We then used these scores to generate inverse probability of treatment weights, wherein individuals were weighted by the inverse probability of receiving the treatment that they actually received. Mean weights were further normalized. Multivariable models were further adjusted for age, sex, hypertension, diabetes mellitus, smoking, symptom onset to first medical contact, prior PCI, systolic blood pressure on admission, heart rate * Ajit S. Mullasari [email protected]