Glutamate, as the major excitatory neurotransmitter in the brain, promotes neuronal depolarization [
29]. Extracellular glutamate levels are directly correlated to levels of neuronal hyperexcitability and seizure intensity in animal models of epilepsy [
30,
31]. Here, we evaluated the combined concentration of the glutamate and glutamine as these metabolites are not differentiable at 3.0 T
1H-MRS. In healthy volunteers, the majority of the combined concentration consists of glutamate (~ 80%) [
32] and 13%–22% of the glutamate concentration in the healthy brain is present in the extracellular space [
29]. Most likely, the glutamate concentrations measured by
1H-MRS largely reflect the extracellular glutamate levels. In support of this, a
1H-MRS study reported lower glutamate levels in amyotrophic lateral sclerosis patients treated with riluzole, a drug that increases glutamate uptake in central nervous system (CNS) neurons, compared to riluzole-naive amyotrophic lateral sclerosis patients and healthy controls [
16]. The transient sildenafil-induced increase of glutamate in the brainstem in the present study thus likely reflects increased extracellular glutamate levels and possibly increased neuronal excitability. In support, sildenafil is able to cross the blood-brain barrier [
23,
33] and some individuals report CNS side effects, such as dizziness and confusion [
33‐
36]. Thus, sildenafil may be able to directly affect the neurons in deep brain structures such as the brainstem. In contrast, a functional MRI (fMRI) study of the visual cortex suggested that oral sildenafil intake did not change the neuronal activation threshold either at 1 or 2 h after administration [
2]. The plasma t
max of oral 100 mg sildenafil is about 1 h with a close to 4 h half-life in the fasting state [
36]. Here, we detected an increased glutamate level at scan 1 (40–70 min after sildenafil), around the time of t
max, but not at scan 2 (140–170 min after sildenafil). Possibly, plasma concentrations of sildenafil above a certain level are needed to alter the glutamate levels. Another possibility is that the transient changes may be attributed to adaptation of sildenafil’s effect at scan 2.
We detected no difference in the glutamate levels between groups of participants developing headache vs. no headache. It should be noted that the participants were healthy with no family history of migraine developing merely a mild to moderate non-migraine headache after the drug administration. Therefore, we speculate that a “healthy” trigeminonociceptive system would not be sufficiently activated to produce detectable changes in the glutamate level. This may also explain the lack of changes in the glutamate levels after CGRP as well as in the thalamus. Given the transient glutamate changes and lack of correlation to headache status, it is likely that the observed changes are related to the pharmacological effects of the drug rather than the headache per se.
The lactate concentration was decreased in the brainstem at scan 1 after sildenafil compared to the corresponding placebo change. This observation is very interesting since brain lactate levels under normal conditions increase during neuronal activation [
37]. Therefore, we would expect the brainstem lactate levels to increase following sildenafil administration, along with the observed increase in glutamate. A possible explanation could be that the lactate decrease reflects a neuronal energy consumption via conversion to pyruvate [
38]. The lactate concentration finding in the present study should be interpreted with caution due to the relatively large standard deviations. Also of note, the lactate concentration is very low in the healthy brain (below 1.0 mmol/L) [
39]. This contributes to the risk of lactate signal loss in the spectrum due to chemical shift displacement or J-modulations deviations during the MRS measurements, which are known issues [
39].