Intravenous Heroin Induces Rapid Brain Hypoxia and Hyperglycemia that Precede Brain Metabolic Response

Abstract

Heroin use and overdose have increased in recent years as people transition from abusing prescription opiates to using the cheaper street drug. Despite a long history of research, many physiological effects of heroin and their underlying mechanisms remain unknown. Here, we used high-speed amperometry to examine the effects of intravenous heroin on oxygen and glucose levels in the nucleus accumbens (NAc) in freely-moving rats. Heroin within the dose range of human drug use and rat self-administration (100-200 μg/kg) induced a rapid, strong, but transient drop in NAc oxygen that was followed by a slower and more prolonged rise in glucose. Using oxygen recordings in the subcutaneous space, a densely-vascularized site with no metabolic activity, we confirmed that heroin-induced brain hypoxia results from decreased blood oxygen, presumably due to drug-induced respiratory depression. Respiratory depression and the associated rise in CO₂ levels appear to drive tonic increases in NAc glucose via local vasodilation. Heroin-induced changes in oxygen and glucose were rapid and preceded the slow and prolonged increase in brain temperature and were independent of enhanced intra-brain heat production, an index of metabolic activation. A very high heroin dose (3.2 mg/kg), corresponding to doses used by experienced drug users in overdose conditions, caused strong and prolonged brain hypoxia and hyperglycemia coupled with robust initial hypothermia that preceded an extended hyperthermic response. Our data suggest heroin-induced respiratory depression as a trigger for brain hypoxia, which leads to hyperglycemia, both of which appear independent of subsequent changes in brain temperature and metabolic neural activity.

Keywords: brain temperature; electrochemistry; opioids; rats; respiratory depression; vascular tone.

Graphical abstract

Figure 1.

Relative changes in NAc oxygen (A) and glucose (B) induced by intravenous heroin at 100 and 200 μg/kg doses. Circles represent 1-min average (±SEM) changes calibrated in μM. Bars represent percentage changes with respect to baseline concentrations. n = number of averaged responses. Filled symbols show values significantly different from preinjection baseline. Vertical hatched lines show the onset of heroin injection.

Figure 2.

Rapid changes in NAc oxygen (A) and glucose (B) induced by intravenous heroin at 100 and 200 μg/kg doses. Circles represent 4-s mean (±SEM) changes calibrated in μM. n = number of averaged responses. Filled symbols show values significantly different from preinjection baseline. Vertical hatched lines show the onset and offset of heroin injection.

Figure 3.

Correlative relationships between slow (A, 1-min bin) and rapid (B, 4-s bin) changes in oxygen and glucose induced by iv heroin at 100 and 200 μg/kg doses. Initial value immediately preceding heroin injection (0 for both oxygen and glucose) is marked as baseline. The period assessed for correlation is shown as the period between filled black circles. A line of best fit is shown with the correlation coefficient (r). Data are shown as percentage change from baseline. For details, see Results.

Figure 4.

Mean (±SEM) changes in oxygen in the subcutaneous space (A) induced by heroin at 100 and 200 μg/kg doses. Changes in glucose concentration in the subcutaneous space (B) induced by iv injection of 100 μg/kg heroin and 10% glucose solution. Graphs of heroin injections also show data obtained in the NAc for comparison. n = number of averaged responses. Filled symbols indicate values significantly different from the preinjection baseline. Vertical hatched lines show the onset of heroin injection.

Figure 5.

Mean (±SEM) changes in several temperature parameters (A, temperatures recorded from the NAc, temporal muscle, and skin; B, NAc-muscle and skin-muscle temperature differentials) and locomotion (C) induced by heroin at 100 and 200 μg/kg doses. Percentage changes in NAc oxygen and glucose are superimposed on each graph. Filled symbols for temperature parameters show values significantly different from preinjection baseline. Data for oxygen and glucose are shown without standard errors for clarity.

Figure 6.

Original examples of changes in oxygen (A) and glucose (B) currents (1-s bins) following intravenous injections of heroin at a large dose (3.2 mg/kg). Each electrochemical record also shows changes induced by heroin at 100 μg/kg dose. C shows changes in NAc, muscle, and skin temperature induced by heroin at the same large dose. Small black circles on C show mean NAc temperature response induced by heroin at 100 μg/kg dose.