Heroin Contaminated with Fentanyl Dramatically Enhances Brain Hypoxia and Induces Brain Hypothermia

Abstract

While opioid abuse is an established medical and public health issue, the increased availability of highly potent synthetic opioids, such as fentanyl, has given rise to acute health complications, including a comatose state and death during drug overdose. Since respiratory depression that leads to acute hypoxia is the most dangerous complication of opioid drug use, we examined the effects of intravenous heroin and heroin contaminated with 10% fentanyl on oxygen levels in the nucleus accumbens (NAc) monitored using high-speed amperometry in freely moving rats. Additionally, we examined the effects of heroin, fentanyl, and their mixture on locomotion and temperatures in the NAc, temporal muscle, and skin. Both fentanyl and heroin at human-relevant doses (400 and 40 μg/kg, respectively) induced rapid, strong and transient decreases in NAc oxygen, indicative of brain hypoxia. When the heroin-fentanyl mixture was injected, the NAc hypoxic response was greatly potentiated in its duration, suggesting sustained hypoxia. In contrast to modest, monophasic brain temperature increases caused by heroin alone, the heroin-fentanyl mixture induced a biphasic temperature response, with a prominent postinjection decrease resulting from peripheral vasodilation. This hypothermic effect, albeit much smaller and more transient, was typical of fentanyl injected alone. Our findings indicate that accidental use of fentanyl instead of heroin, or even a relatively minor contamination of "street heroin" with fentanyl, poses great danger for acute health complications, including a comatose state and death.

Keywords: brain temperature; metabolic activation; opioid; oxygen; vasoconstriction; vasodilation.

Graphical abstract

Figure 1.

Changes in NAc oxygen levels (normalized as percentage vs baseline = 100%) induced by intravenous injections of fentanyl (40 µg/kg; A), heroin (400 µg/kg; B), and their mixture (40 µg/kg fentanyl + 360 µg/kg heroin; B) in awake, freely moving rats. Changes in NAc oxygen were significant for each of the three treatments. Values significantly different from baseline (p < 0.05; horizontal hatched lines) are shown as filled symbols. Moment of drug injection (time = 0 min) is shown as vertical hatched line. C, Mean values of three parameters of oxygen response [peak amplitude, s; latency to peak, s; and area under the curve for the total oxygen decrease (AUC)]. Asterisks denote significant between-group differences (p < 0.05). Symbols in peak amplitude graph show individual values of oxygen decreases in percentage of baseline.

Figure 2.

Changes in temperature in the NAc, temporal muscle, and skin (top row), NAc-muscle and skin-muscle temperature differentials (middle row), and locomotor activity (bottom row) induced by intravenous injections of fentanyl (40 µg/kg; A), heroin (400 µg/kg; B), and their mixture (40 µg/kg fentanyl + 360 µg/kg heroin; C) in awake, freely moving rats. Values significantly different from baseline (p < 0.05; horizontal hatched lines) are shown as filled symbols. Moment of drug injection (time = 0 min) is shown as vertical hatched line.

Figure 3.

High-resolution analyses of changes in temperature parameters (A, NAc temperature; B, NAc-muscle differential; C, skin-muscle differential) and NAc oxygen levels (D) for the first 20 min following our treatment (fentanyl, 40 µg/kg; heroin, 400 µg/kg; heroin-fentanyl mixture, 40 µg/kg fentanyl + 360 µg/kg heroin). Each point in all graphs represents mean values for a 10-s bin. Filled symbols in A–C show values significantly different from baseline (p < 0.01).

Figure 4.

Correlative relationships between drug-induced changes in NAc oxygen and NAc temperature (A) and between changes in NAc oxygen and NAc-muscle temperature differentials (B). Each graph shows changes in mean values of parameters from the baseline (red point) and span the duration of the temperature responses. Blue symbols show values for the duration of the oxygen response. All graphs show coefficients of correlation calculated for the duration of the oxygen response. Regression lines are shown in red for cases having a significant correlation (**p < 0.01; ***p < 0.001).