Comment on: Calleja et al. Insulin Resistance Is Associated With a Poor Response to Intravenous Thrombolysis in Acute Ischemic Stroke. Diabetes Care 2011;34:2413–2417

Abstract

In a recent article, Calleja et al. (1) reported that insulin resistance (IR), as reflected in homeostasis model assessment of IR (HOMA-IR), is a major determinant of adverse outcomes in stroke treated with thrombolysis. Although the data are consistent with the conclusion, it is extremely important that adequate attention is given to plasma glucose concentrations, since they are known to affect outcomes even when they appear to be modestly elevated. Though the authors provide data on glucose concentrations at the time of presentation, they do not provide data on glucose and insulin concentrations prior to the injection of the thrombolytic agent, which they used for the calculation of HOMA-IR, probably collected the next morning. This is important since the readers would want to know whether the glucose or the insulin concentrations were responsible for the elevation of HOMA-IR. It has now been shown that not only the glucose concentrations at admission but also those after admission are related to clinical outcomes. In a study involving 700 patients with ischemic stroke treated with intra-arterial intervention, glucose concentration after 48 h of admission and the change in glucose after admission in addition to the glucose at admission also determined the clinical outcomes (2). An admission glucose concentration of 112 mg/dL was associated with 20% mortality, a glucose concentration above 142 mg/dL was associated with 45% mortality, and patients with glucose concentrations in between had 29% mortality. An increase of 30 mg/dL in the highest tertile leads to an increase in mortality to 69%, whereas a decrease in the lowest tertile by 30 mg/dL led to a reduction in mortality to 8%. Thus, an increasing glucose concentration, which is known to be related to adverse outcomes, will also increase HOMA-IR. In order to establish an independent effect of HOMA-IR, therefore, one would need to carry out a multivariable analysis with values obtained at that time so that a contribution by the glucose concentrations can be excluded. It is relevant that the outcomes of acute myocardial infarction are also determined by not only the glucose concentrations at admission but also those at 48 h after admission (3). The dangerous threshold concentration of glucose in these studies is also around 140 mg/dL, demonstrated by the Clinical Trial of Metabolic Modulation in Acute Myocardial Infarction Treatment Evaluation–Estudios Cardiologicos Latinoamerica (CREATE-ECLA) study (n = 10,000) and the pioneering studies by Kosiborod et al. (4). Clearly, there are pathogenic mechanisms that are common to acute ischemic syndromes in the brain and the heart, which are triggered by elevated glucose concentrations. It is important to realize that these elevations are modest and are yet indicative of seriously adverse outcomes.