Neurophysiological correlates of clinical improvement after greater occipital nerve (GON) block in chronic migraine: relevance for chronic migraine pathophysiology

CM is a complication of episodic migraine affecting 2% of the general population [1] and most commonly encountered in tertiary headache clinics [2]. It is frequently associated with Medication Overuse Headache (MOH), a secondary headache, which could be interpreted as a comorbidity as well as a marker of poor control of the migraine itself [3, 4]. In the migraine pathology spectrum, chronic migraine (CM) represents the most severe form and its disabling effects are not only related to the physical health of patients but also reverberate on social and economic aspects. The available pharmacological treatments of CM are quite limited. Only a few drug treatments like topiramate [5] and onabotulinumtoxin A [6, 7], have an efficacy proven by randomized controlled trials. Most drugs commonly used in the treatment of CM have been investigated only by small open studies (for a review, see [8]). Management of CM patients is further complicated by poor therapeutic adherence due to the insufficient efficacy of drug therapies and their side effects [9].

Since no predictors of treatment response are available to date, preventive medications are selected by trial and error, taking into account the patients’ clinical profile, drug side effects and the patients’ comorbidities more so than by their efficacy differences. Each drug is commonly recommended for 4–6 months before assessing (in)efficacy and CM patients try on average 4 therapeutic lines with severe patient discomfort and frustration of the doctor. Half of patients interrupt treatment due to lack of efficacy or side effects and only about 9% achieve clinical improvement [10].

Greater occipital nerve blocks with a local anesthetic alone [11, 12] or with a steroid-local anaesthetic combination [13, 14] were found useful for CM with varying effect sizes and durations, but their mechanism of action remains elusive.

Repetition of attacks and failure of preventive therapies produce a longer exposure to headaches that might promote central sensitization and maintain migraine chronicity and possibly refractoriness [8, 15].

Understanding neurophysiological mechanisms of chronification might allow identifying biomarkers able to predict transition and remission from chronic to episodic migraine (EM).

Possible candidates could be markers of brain excitability and responsivity, such as habituation and amplitude-stimulus function of cortical evoked responses (measured by IDAP) [16, 17].

Detailed description of the lack of habituation in migraineurs has been widely reported elsewhere [17, 18]. In EM between attacks, during repeated and unmodified sensory stimulation, cortical evoked responses are characterised by a reduced initial amplitude (lower preactivation level) and a progressive amplitude increase (i.e. hyperresponsivity) instead of a physiological habituation, regardless of the sensory modality, visual, auditory, somatosensory, laser, including cognitive potentials like contigent negative variation or P3 [19]. All evoked potentials, besides laser evoked potentials, normalize during migraine attacks [20]. Interestingly, in CM the electrophysiological pattern is similar to that of ictal EM recordings, suggesting genuine hyperexcitability and supporting the idea that CM is a sort of “never-ending attack” [21, 22]. MOH patients have an intermediate pattern between CM and EM with an increased initial response but a lack of habituation, a pattern that is found in the pre-ictal EM phase [23].

When CM patients improve and revert to EM, they acquire again the interictal electrophysiological profile consisting of a low initial response and a deficient habituation [24].

Intensity Dependence of Auditory evoked Potentials (IDAP) is chiefly modulated by activity in central brainstem-cortical serotonergic projections [25]. In interictal EM IDAP is increased reflecting low serotonergic tone, and correlates with deficient habituation [26]. So far, no IDAP data are available in CM. It is thus not known if IDAP is correlated or not with habituation as in EM nor if it changes when patients revert from CM to EM.

In this study, we aimed to evaluate changes in IDAP before and 1 week after a greater occipital nerve block (GON-B) with a steroid-local anaesthetic combination, as well as clinical changes 4 weeks after the block in CM/MOH patients without preventive therapy. Besides confirming a possible therapeutic effect, our aim was to search for a correlation between such an effect and baseline IDAP or its GON-B induced early change in order to identify a biomarker able to predict the treatment effect.

Due to the low rate of response and the need to treat patients for a long time before having any clinical feedback about the efficacy of the treatment itself, having an early predictive tool to identify if a specific treatment is effective or not for a given patient is valuable information from a medical, social and economic point of view. In this study we tested if a greater occipital nerve block (GON-B) could be such a tool by assessing its long-term clinical efficacy and its short effects on brain physiology.

We found that a single GON-B produced after 1 month a clinical benefit with a 34% reduction in headache days and a return to episodic migraine in 50% of the cases. In addition, we showed that this beneficial effect was associated with a decrease of intensity dependence of auditory evoked potentials (IDAP) 1 week after the block in clinical responders, suggestive of an increase in central serotonergic tone.

To our knowledge, this study is the first to examine the correlation of serotonergic tone, indexed by IDAP, with CM and in particular its variation in association with clinical improvement. We found that patients with CM had a low baseline serotonergic tone, which increased after a clinically successful GON-B. This occurred especially in those patients who switched from a chronic form of migraine to an episodic one. Furthermore, the degree of IDAP flattening after 1 week was directly correlated to the improvement of migraine in terms of reduction of headache days at 1 month post-GON-B. This result may thus be of some utility for clinicians who plan to perform a GON-B and like to have an early feedback of the long-term efficacy of GON-B.

The change in IDAP is not unexpected because of its strong correlation with central serotonin pathways. Migraine in general is considered since a long time to be a low serotonin disorder and in CM in particular serotonin levels are thought to be even lower on the basis of clinical (e.g. high comorbidity with depression) and biochemical data [31, 32]. Contrary to the IDAP abnormality in CM, we also found a normal pattern of VEP habituation in CM with or without MOH, which is consistent with previous findings and with the hypothesis that CM is a sort of a “never ending attack” [21, 33].

VEP habituation returns to the normal non-habituation pattern of episodic migraines, when patients with CM are successfully treated [24]. In line with these results, we found that habituation decreased for the PR-VEP N1P2 component with clinical improvement so that at T1 it differed from that of HV.

In previous studies of episodic migraineurs abnormal IDAP and VEP habituation were thought to be linked interictally because decreased habituation of auditory potentials may lead to a steep IDAP slope and low central serotonin activities could explain both abnormalities [26] .

A possible explanation for the dissociation between IDAP and VEP habituation changes in our study could be that in chronic migraine the apparently normal habituation pattern is due to changes in brain plasticity more than to a simple change in serotonergic tone [34, 35]. Chronic migraine is characterized by persistent hyperexcitability of sensory cortices [22], suggesting that the repetition of attacks induces plastic alterations of the excitatory-inhibitory balance that is fundamental for tuning synaptic plasticity and circuitry [36]. In this concept, inhibitory responses tend to decrease more than excitatory ones following stimulus repetition [37] and are also negatively modulated by neurotransmitters, such as acetylcholine, noradrenalin and oxytocin. The resulting disinhibition might thus promote hyperesponsivity [36].

In the presence of reduced inhibitory activity, a higher cortical preactivation level might be achieved and supported by a low serotonin activity. For this reason, the pattern of a normal habituation found in CM may actually be a “pseudo-normal habituation”, related to LTD-mediated inhibitory responses more that just a “ceiling effect” as hypothesized for episodic migraine [38] and lately validated by several consistent experiments (see [17] for a review). LTD-mediated inhibition responses were found more commonly in CM and in EM with high attack frequency than in episodic migraineurs with a low frequency of attacks [39].

In patients reversing from a chronic to an episodic pattern of migraine, we observed an early flattening of the IDAP response after the GON-B, indicating that clinical improvement is likely to be associated with an increase in serotonergic firing.

Although the precise mechanism of action of GON-B is unknown, it is thought to modulate brain excitability acting on input gate at the brainstem level. Although cervical stimulation has been shown to increase directly brain serotonin [40, 41], such evidence is lacking for GON-B.

The IDAP slope flattening, and thus probably central serotonin activity, positively correlated with the reduction of total headache days after a month at group level, reinforcing a possible role of serotonin in remission from CM. Both at cortical and thalamic levels, serotonin is able to modulate the excitatory/inhibitory balance, both directly and by modifying the action of other neurotransmitters like dopamine [42‐44]. Serotonin-related metaplasticity, characterized by a shift from inhibitory to excitatory activity, was already found in the hippocampus [45]. Taken together, these results corroborate the idea that serotonin may be a crucial actor in the plastic brain changes that accompany migraine chronification and reversal. In our study we provide indirect evidence for such a role, which needs to be confirmed by more direct assessments of central serotonergic neurotransmission.

A single greater occipital nerve block showed a good clinical response in chronic migraine patients with or without acute medication overuse. Clinical improvement, and in particular the reversal to an episodic migraine pattern was significantly associated with a decrease of intensity dependence of auditory evoked potentials (IDAP) one week after the block. As IDAP is thought to reflect central serotonergic activity and it is in addition increased at baseline in chronic migraineurs, it is likely that serotonin plays a crucial role in migraine chronification and associated plastic brain changes.