Greater occipital nerve blockade in cluster headache: effectiveness and safety of a combined injection protocol

Excerpt

Cluster headache (CH) is the most common disorder among trigeminal autonomic cephalalgia (TACs). It is characterized by attacks of harrowing unilateral pain associated with cranial autonomic symptoms and a restless and agitated behavior, occurring in bouts or clusters. Optimal management includes acute, prophylactic, and transitional treatments. Acute therapy is aimed to interrupt a single attack, while prophylaxes are intended to reduce intensity and frequency of pain as well as to shorten cluster duration. However, preventive dosage titration can cause a delayed pain relief, overuse of triptans, and poor quality of life. In this scenario, transitional treatments represent a middle ground between acute and preventive treatments that may allow a sustained analgesic effect and acute treatment intake reduction during the process of preventive titration. Oral corticosteroids are the most widely used transitional therapy; nonetheless, long-term systemic administration is burdened with potentially serious adverse events. Greater occipital nerve blockade (GONB) has proved to be an effective and safe transitional treatment option, thanks to its targeted action and absence of systemic adverse events (level of evidence A according to American Headache Society) [1]. Plenty of studies have evaluated GONB effectiveness and safety [2]; however, a wide range of different GONB protocols are described, and it is uncertain which leads to better results. This interventional, open-label study aimed to evaluate the effectiveness and safety of the GONB protocol with methylprednisolone and lidocaine in CH. We included all consecutive patients aged more than 18 years old referred to the Bologna Headache Centre between January 2020 and February 2021, for Episodic Cluster Headache (ECH) with active cluster or chronic cluster headache (CCH) who were clinically judged to potentially benefit from a GONB by a headache specialist and without contraindication to steroid and lidocaine treatment. Patients underwent GONB with slow-release methylprednisolone 80 mg and lidocaine 40 mg injected in the suboccipital area ipsilateral to pain at a point lying on the medial third of a line drawn between the inion and mastoid process ipsilateral to CH attacks. Patients were then followed up 30 days after the procedure; the number and intensity of attacks were registered through a self-filled headache diary. The primary outcome was the complete absence of CH attacks at 1 month from the injection and the secondary outcome was the reduction of at least 50% of daily attacks; Beck depression inventory scale and Zung anxiety scale were completed at baseline and after 1 month. A total of 32 patients were recruited (9 females, 23 males): 23 with ECH and nine with CCH. Among ECH patients, the mean cluster duration was 1.6 months (range 0.5–6 months), and the mean temporal distance at treatment time from active cluster beginning was 17.5 days (range 3–50 days). Forty-eight percent of patients had a concomitant prophylaxis, represented mostly by Verapamil (93.75%). Thirty days after the injection, ten patients (31%) were attack-free, reaching the primary outcome, while a total amount of 19 patients (59%) showed a reduction of at least 50% of daily attacks reaching the secondary outcome. Furthermore, in non-attack-free responders, the daily frequency of attacks decreased from a median of 2 (range: 0.7–8) to 0.46 (range: 0.2–1.5) (p < 0.05); pain intensity, evaluated through the number rating scale, decreased from a median of 8 (range: 2–10) to 6.3 (range: 3–9) (p < 0.05) (Table 1). Two patients (6%) experienced a daily attack reduction of less than 50% and eleven patients (34%) were considered non-responders, since they needed further therapies. We found no statistically significant difference in responsiveness rate for both primary and secondary outcomes in chronic patients compared to episodic ones. After 30 days, mean scores for the Beck depression inventory scale and Zung anxiety scale among responders dropped by 8.61% and 8.22%, respectively. No serious adverse events were reported: four patients (13.3%) had mild local adverse events such as transient neck stiffness and local pain. This study has several limitations. First, we cannot evaluate and measure a possible placebo effect related to the injection itself, due to the lack of a controlled group, although a previous study reported a virtual absence of placebo effect for a single injection [3]. Second, patients’ selection was based on neurologists’ clinical judgement and patients’ willingness to participate; therefore, we cannot exclude a selection bias. In conclusion, GONB is already part of clinical practice for the management of cluster headaches; nevertheless, there is still no agreement on the ideal administration protocol. Our study suggests that GONB with slow-release methylprednisolone 80 mg and lidocaine 40 mg, in a single administration, represents a safe and effective transitional therapy: in our sample, 1 month after GONB, 31% of patients were attack-free and 59% showed a reduction of at least 50% of daily attacks. Considering the complex methodology of designing randomized clinical trials in an unpredictable and rare condition as cluster headaches, studies like ours are warranted to expand evidence and optimize clinical practice. …