Reply to “Reassessing partial efficacy or not responding to acute migraine treatment”

We appreciate the opportunity to respond to the letter by Yang et al., ¹ which commented on our paper published in the journal. ²

Yang et al. ¹ provide thoughtful comments on our paper and raise important issues that we are happy to address.

Regarding the first observation, we acknowledge that our practice recommendations allow for some overlap between “partial efficacy” and “not responding”. While more precise definitions of “effective treatment”, “partial efficacy” and “non-response” would be valuable for clinical trials and future studies, this publication was designed to provide practical guidance for real-world situations in different healthcare and geographic settings. In this context, the distinction between ‘partial efficacy’ and ‘not responding’ can reasonably blur because our primary goal was to provide practical solutions for optimizing care outcome, rather than to establish strict criteria for treatment failure. Along this line of reasoning, when considering that triptans are generally more effective than non-steroidal anti-inflammatory drugs (NSAIDs) ² and widely available, ³ we adopted the more strict pain-freedom criterion (Q1) to avoid delaying the switch from NSAIDs to triptans and to discourage trial-and-error approaches where patients are exposed to multiple analgesics, some of which with limited evidence of efficacy in migraine treatment, or, even worse, to opioids alone or in combinations with analgesics.

While we welcome and commend critical reasoning and new field-tested proposals around the optimal outcome measures, we feel that trying to calculate a specific percentage of pain reduction from the 4-point nominal scale is methodologically incorrect. Furthermore, the definition of pain relief does not include the 100% improvement because the response is defined by a two-level change (from moderate to none or from severe to mild). Of course, some overlap between pain relief and pain freedom may still occur if a subject takes an acute drug when their pain is moderate and becomes pain-free at two hours. While this overlap may pose challenges in clinical trials, we believe it is acceptable in real-world settings.

The possibility to reliably adopt percentages to more precisely define the response to acute migraine treatment is intriguing. In this regard, the continuous visual analogue scale may be more suitable than the ordinal Numerical Rating Scale, although there are statistical controversies on this issue. ⁴ Specifically targeted studies are necessary to verify the agreement between the percentage pain reductions calculated from numeric rating scores of migraine pain intensity and other validated clinical outcomes.

The second aspect highlighted by Yang et al. ¹ concerns the lack of specific recommendations for patients showing only a partial response to NSAIDs. Adding the option to use the combination aspirin–acetaminophen–caffeine to IHS practice recommendations is suggested. While the literature on the efficacy of the combination aspirin–acetaminophen–caffeine versus placebo is convincing, the comparative evidence versus other NSAIDs or sumatriptan used individually is either weak or non-existent. ⁵ Therefore, such an additional suggestion would not rely on a strong scientific basis. Furthermore, the combination aspirin-acetaminophen-caffeine is not mentioned in the World Health Organization essential list of drugs (https://www.who.int/publications/i/item/WHO-MHP-HPS-EML-2023.02) and we lack data on its availability around the world. Therefore, while we acknowledge that no explicit guidance was provided for this group of individuals, the broader recommendations, and particularly those around triptan use, remain applicable to them. Additionally, we emphasize that the nature of the recommendations document itself dictated a focus on addressing a defined set of practical questions, meaning that not all aspects of acute treatment could be covered in this initial publication.