To the Editor:
We read with interest the findings of Gertsch et al about the efficacy of ibuprofen in chemoprophylaxis of acute mountain sickness (AMS). 1 Although pleased to see confirmation of our recently published study, 2 we found their alternative conclusion that “ibuprofen merely masks the [headache] of AMS,” implying its effects are analgesic rather than preventing AMS, questionable in light of our findings. Their conclusion seems conveniently expedient as the primary investigator of the aforementioned paper ignored conflicting results stemming from our previously published study, which he coauthored.
We previously found that ibuprofen was robust in the prevention of AMS versus placebo (26% decrease in incidence; number needed to treat, 4), but by AMS symptom subgroup high altitude headache (HAH) incidence was not significantly different between the 2 arms (66% vs 76%; odds ratio, 0.6; 95% confidence interval, 0.2 to 1.6). We did not power the study to examine HAH specifically, so a type II error may not have allowed us to detect a change. However, as HAH has almost twice the prevalence as AMS,1,3 it leads to the reasonable conclusion that significance in HAH prevention would have been detected if present. As Gertsch et al did not include a subgroup analysis of AMS symptoms, they were unable to verify whether ibuprofen had a class effect on AMS symptoms (as our research showed) resulting in a decrease in incidence of disease, or was only “masking” symptoms of headache as promoted. We are disappointed that our published results to the contrary were ignored, and such strong claims propagated without statistical substantiation.
If one concurs that a nonsteroidal anti-inflammatory drug prevents the constellation of symptoms that comprise AMS, then this drug may be viewed in a similar light to another anti-inflammatory, dexamethasone, which experts suggest as an excellent alternative prophylactic agent if used with care. 4 Any time science finds a promising new result, we agree it is prudent to view it with caution until there are confirmatory studies that have further clarified the risks and benefits. Our concern in implementing ibuprofen into the high altitude pharmacopeia mirrors that of using dexamethasone for chemoprophylaxis in the big mountains of Asia or South America, where there is potential for “rebound” of AMS symptoms if abrupt cessation occurs, physiologically stranding someone up high. However, at our studied altitude of 3800 m, ibuprofen has the potential to provide comfort and subsequently increased enjoyment in the mountains of the western United States and Europe where the AMS incidence ranges from 25% to 40%. Dismissing potential applicability of ibuprofen in the appropriate setting demeans the very real benefit this drug may provide, with minimal risk.
Finally, we applaud Gertsch et al for running an intent-to-treat analysis, although their large number of study participants was matched by the equally impressive 38% attrition rate after initial enrollment and randomization. Their reported positive findings included the 183 fully compliant and 49 noncompliant participants (21% of follow-up completion who took nonstudy medications or missed 2 or more doses). However, they erroneously omitted from their calculations the additional 62 people lost to follow-up (21% of enrolled and randomized). A rigorous “intent-to-treat” strategy includes “all patients, regardless of…the treatment actually received, and subsequent withdrawal or deviation from the protocol”; if this is not done correctly, clinical effectiveness may be overestimated. 5 After reading the paper and following the statistical leaps of faith required to reach their conclusion, while having our “smaller” study's comprehensive results summarily dismissed, it left us asking the question, “is bigger really better?”