Predictive factors of effectiveness of occipital nerve stimulation for chronic cluster headache: data from for the French ONS registry

Effectiveness of Occipital Nerve Stimulation (ONS) in refractory chronic cluster headache (rCCH) is supported by series of cases and a unique controlled trial, and its risk/benefit ratio has been questioned. Our objective was to identify predictive factors of ONS effectiveness in rCCH patients, to optimize this risk/benefit ratio and better select eligible patients. We analyzed 125 patients (43 women, mean age 46,4) included prospectively in the “French ONS registry”, suffering from rCCH, treated by ONS for more than one year, and with data concerning putative preoperative predictive factors of effectiveness, including demographic, CCH characteristics and severity, treatment use and co-morbidities. We studied factors associated with ONS response, defined as a ≥ 50% reduction of weekly attack frequency (WAF) between baseline and one year follow-up. Factors predictor of good response to ONS were high preoperative WAF (p = 0.0003), high number of days with attack treatment use (p = 0.0006) and absence of epilepsy (p = 0.019). The best cut-off to predict ONS effectiveness was a WAF of 14 attacks/w. ONS response’s rates were 77.0% and 37.3% in patients with WAF ≥ 14/w and WAF < 14/w, respectively. However, in both groups, quality of life was significantly improved after ONS (p < 0.001). Preoperative WAF was the main predictive factors of ONS response in rCCH, with a cut-off of 14 attacks/w, which may help to select patients and increase the ONS risk/benefit ratio. However, this threshold should be used with caution, as patients with WAF < 14/w also benefited from ONS. The study has been registered (clinicaltrials.gov identifier NCT01842763).