Areas of cerebral blood flow changes on arterial spin labelling with the use of symmetric template during nitroglycerin triggered cluster headache attacks

Cluster headache is a rare, strictly unilateral, severe episodic primary headache disorder. Due to the unpredictable and episodic nature of the attacks, nitroglycerin has been used to trigger attacks for research purposes to further our understanding of cluster headache pathophysiology.

We aimed to identify regions of significant cerebral blood flow (CBF) changes during nitroglycerin triggered cluster headache attacks, using MRI with arterial spin labelling (ASL).

Thirty-three subjects aged 18-60 years with episodic and chronic cluster headache were recruited and attended an open clinical screening visit without scanning to receive an intravenous nitroglycerin infusion (0.5 μg/kg/min over 20 min). Those for whom nitroglycerin successfully triggered a cluster headache attack, were invited to attend two subsequent scanning visits. They received either single-blinded intravenous nitroglycerin (0.5 μg/kg/min) or an equivalent volume of single-blinded intravenous 0.9% sodium chloride over a 20-minute infusion. Whole-brain CBF maps were acquired using a 3 Tesla MRI scanner pre-infusion and post-infusion. As cluster headache is a rare condition and purely unilateral disorder, an analysis strategy to ensure all the image data corresponded to symptomatology in the same hemisphere, without losing coherence across the group, was adopted. This consisted of spatially normalising all CBF maps to a standard symmetric reference template before flipping the images about the anterior-posterior axis for those CBF maps of subjects who experienced their headache in the right hemisphere. This procedure has been employed in previous studies and generated a group data set with expected features on the left hemisphere only.

Twenty-two subjects successfully responded to the nitroglycerin infusion and experienced triggered cluster headache attacks. A total of 20 subjects completed the placebo scanning visit, 20 completed the nitroglycerin scanning visit, and 18 subjects had completed both the nitroglycerin and placebo scanning visits. In a whole-brain analysis, we identified regions of significantly elevated CBF in the medial frontal gyrus, superior frontal gyrus, inferior frontal gyrus and cingulate gyrus, ipsilateral to attack side, in CBF maps acquired during cluster headache attack; compared with data from the placebo session. We also identified significantly reduced CBF in the precuneus, cuneus, superior parietal lobe and occipital lobe contralateral to the attack side. Of particular interest to this field of investigation, both the hypothalamus and ipsilateral ventral pons showed higher CBF in a separate region of interest analysis.

Our data demonstrate that severe cluster headache leads to significant increases in regional cerebral perfusion, likely to reflect changes in neuronal activity in several regions of the brain, including the hypothalamus and the ventral pons. These data contribute to our understanding of cluster headache pathophysiology; and suggest that non-invasive ASL technology may be valuable in future mechanistic studies of this debilitating condition.