Cervicogenic headache and onabotulinumtoxinA: Where do we stand?

Abstract

Recent clinical trials have shown benefit for onabotulinumtoxinA (OBTA) in chronic migraine, and it certainly will be questioned whether OBTA might be beneficial in other headache types as well. Cervicogenic headache is one of the more controversial of these other headache types. In this issue of Cephalalgia, Linde and colleagues report on the results of a randomized placebo controlled trial on the use of OBTA in the treatment of cervicogenic headache (1). Their trial was negative, with OBTA showing treatment results no better than placebo. These results are of course disappointing, as cervicogenic headache is often a difficult headache syndrome to treat, but does this study provide a definitive answer to the question of whether OBTA can be beneficial in cervicogenic headache? In our interpretation of the results of this trial, at least two levels of consideration are important. The first has to do with the limitations of the trial itself. Although the authors point out that the cross-over design of their trial gives it considerable power, the size of the trial is still relatively small (1). Only 13 patients were randomized to OBTA and 15 to placebo at study onset. Furthermore, subject drop-outs during the trial were significant. Five of the 13 patients randomized initially to OBTA dropped out before the end of the study. Those who follow the medical literature in this area will recall that much larger trials were negative when OBTA was first tested for efficacy in patients with chronic daily headache. The recent PREEMPT trials that finally supported efficacy for OBTA in chronic migraine randomized a total of 688 patients to OBTA treatment and 696 to placebo (2,3). In addition, the study by Linde et al. (1) can be criticized in that subjects could be crossed over to the other treatment arm in as little as 8 weeks after initiation of treatment. What this means is that patients who received OBTA in the initial part of the study could be crossed over to placebo after 8 weeks, even though it is generally recognized that the effects of OBTA can last as long as three months. Therefore, in theory, for those patients who received placebo during the second arm of this cross-over study, the first four weeks of placebo treatment could be contaminated by the lingering effects of the initial OBTA injection. The authors indicate that, statistically, the treatment sequence did not affect their results, but given the size of the study, this might still be questioned. The second level of consideration lies in the very nature of cervicogenic headache itself. There are many potential sources of pain in the neck (4) and these might respond differently to attempted treatment with OBTA. The authors undertook the trial on the premise that the soft tissues of the neck might be a significant source of pain in cervicogenic headache and that the headache itself is a result of referred pain due to the convergence of nociceptive inputs from the head and neck onto the same second-order sensory neurons in the trigeminocervical complex (5,6). However, if cervicogenic headache is a heterogeneous condition with different sources of pain in different patients, it might be difficult to show a treatment effect with any treatment that is not specifically tailored to the underlying pathology. The most proven pain generator in cervicogenic headache is the facet joint (7,8). In the trial by Linde et al. (1), the OBTA was injected in a standardized fashion and in the same way in all patients. Injections were apparently given into only six injection sites. Even in patients whose pain may have been ofmuscular origin, because injectionswere not individualized based on individual patient soft tissue assessment; for many patients the injection paradigm used by Linde et al. may have been suboptimal. Standardized injection paradigms may be effective in relatively diffuse conditions such as chronic migraine, but the situation may be very different in conditions where the source of the pain is focal. On the other hand, if a potential therapeutic effect of OBTA in cervicogenic headache depends on a diffuse reduction of neural inputs into the trigeminocervical complex, then a more widespread