Many climbers and trekkers find acetazolamide 500 mg/day to be useful
EDITOR--Dumont et al reach a false, potentially harmful conclusion in their paper on the efficacy and harm of pharmacological prevention of acute mountain sickness.[1] Their claim that acetazolamide 500 mg/day does not work in preventing acute mountain sickness must be challenged. Their analysis is flawed for three reasons.
Firstly, they used only nine of 25 available controlled studies, and analysed only four of 10 trials with acetazolamide 500 mg/day. Consequently, only 143 individuals taking acetazolamide 500 mg/day were included, 120 from one study.[2] In that investigation the rate of ascent in one of the two trials was so slow that acetazolamide made no difference in acute mountain sickness scores. The other trial, with faster rate of ascent, showed acetazolamide to be very effective. Another meta-analysis, of 10 trials of acetazolamide 500 mg/day (306 participants), concluded that it was effective.[3] If Dumont et al included more trials they would find that acetazolamide 500 mg/day is indeed effective.
Secondly, rate of ascent was not adequately controlled. The authors compared acetazolamide 750 mg/day (250 mg three times a day) during particularly rapid ascent with 500 mg/day with much slower ascent. If the 500 mg dose is not tested during an abrupt ascent similar to that by the group taking the 750 mg dose the comparison is not valid.
Thirdly, Dumont et al chose an unnecessarily strict end point for their analysis: the dichotomous presence or absence of acute mountain sickness. Many people with symptoms of acute mountain sickness, while not meeting criteria for acute mountain sickness, do indeed find acetazolamide helpful. In fact, Dumont et al did note that the drug was found to be effective prevention for headache, insomnia, nausea, and dizziness in various studies not included in their analysis.
Rigorous trials comparing dosages of acetazolamide are lacking, but a huge clinical experience cannot be ignored: climbers, trekkers, and tourists find lower doses of acetazolamide very useful, with fewer side effects. Moreover, the clinical impression that acetazolamide works at 250 mg twice daily has been so strong that many clinicians (myself included) have lowered the dose further, to 125 mg twice daily, with no apparent decrease in effectiveness.
Pending further research, I urge doctors to recommend acetazolamide 125-250 mg twice a day (depending on body size) for the prevention of acute mountain sickness. Since renal carbonic anhydrase inhibition is complete with acetazolamide 5 mg/kg/day, even 500 mg/day is more than most people need. Trials directly comparing effectiveness of varied dosages are essential; this meta-analysis is flawed.
Peter Hackett president, International Society of Mountain Medicine 610 Sabeta Drive, Ridgway, CO 81432-9335, USA phack@ruralhealth.org
[1] Dumont L, Mardirosoff C, Tramier M. Efficacy and harm of pharmacological prevention of acute mountain sickness: quantitative systematic review. BMJ 2000;321:267-72. (29 July.)
[2] Hackett PH, Rennie ID, Levine HD. The incidence, importance, and prophylaxis of acute mountain sickness. Lancet 1976;ii:1149-54.
[3] Reid LD, Carter KA, Ellsworth A. Acetazolamide or dexamethasone for prevention of acute mountain sickness: a meta-analysis. Journal of Wilderness Medicine 1994;5:34-48.
Same ascent rates must be used to assess effectiveness of different doses of acetazolamide
EDITOR--Dumont et al's meta-analysis included only controlled studies reporting dichotomous data on acetazolamide for the prevention of acute mountain sickness.[1] Their conclusion that "contrary to widespread belief, 500 mg of acetazolamide does not work" is not supported by the data presented or by many papers excluded from the analysis.
Figure 3 of the paper shows that the relative risk depends on the rate of ascent The effects of different doses of acetazolamide should therefore be compared at the same ascent rate. The ascent rates of four of the five field studies in which 750 mg acetazolamide was used were above 2400 m/day (100 m/h). In contrast, the three studies in which 500 mg acetazolamide was used had ascent rates of [is less than or equal to] 500 m/day (20 m/h). Two datapoints referring to these studies (fig 3) have a high number needed to treat ([is greater than or equal to] 20). One of these can most probably be attributed to a slow ascent rate with a low relative risk and insufficient statistical power. Thus there remains one study that seems to support the main message of the meta-analysis.[2]
[Figure 3 ILLUSTRATION OMITTED]
A careful look at this paper shows, however, that this is not the case. Cerebral symptoms (AMS-C (acute mountain sickness-C) score) and respiratory symptoms (AMS-R score) were assessed by the environmental symptom questionnaire[3] for several days at 3650-4050 m. During the first three days the number needed to treat was always below 4.7 for assessments with both scores (except for AMS-R on day 1, when it was 27.8). Interestingly, Dumont et al consider only this single value for classifying the study. The investigators of the study, however, report an overall incidence of acute mountain sickness assessed by the AMS-C score of 38% for placebo and 16% for acetazolamide, which gives a number needed to treat of 4.5. The corresponding values when the AMS-R score is used are 63% v 31% and 3.0. Thus the paper is falsely classified by Dumont et al.
Studies reporting symptom scores rather than presence or absence of acute mountain sickness show that 500 mg acetazolamide daily significantly reduces symptoms. A meta-analysis of all controlled trials until 1993 lists five studies (122 subjects altogether) that found a significant decrease in symptom scores with 500 mg acetazolamide daily, while in two studies (32 subjects) the decrease was not significant.[4]
We conclude that the recommendation of 500 mg acetazolamide daily for the prevention of acute mountain sickness during rapid ascent is supported by scientific evidence. The question whether 500 mg or 750 mg or another daily dose is more effective should be investigated under equal ascent rates.
Peter Bartsch head of division peter.bartsch@med.uni-heidelberg.de
Michael Schneider research associate Division of Sports Medicine, Department of Internal Medicine, University Hospital, D-69115 Heidelberg, Germany
[1] Dumont L, Mardirosoff C, Tramer MR. Efficacy and harm of pharmacological prevention of acute mountain sickness: quantitative systematic review. BMJ 2000;321:267-72. (29 July.)
[2] Zell SC, Goodman PH. Acetazolamide and dexamethasone in the prevention of acute mountain sickness. West J Med 1988;148:541-5.
[3] Sampson JB, Cymerman A, Burse RL, Maher JT, Rock PB. Procedures for the measurement of acute mountain sickness. Aviat Space Environ Med 1983;54:1063-73.
[4] Ried LD, Carter KA, Ellsworth A. Acetazolamide or dexamethasone for prevention of acute mountain sickness: a meta-analysis. Journal of Wilderness Medicine 1994;5:34-48.
Authors' reply
EDITOR--Some of these comments on our systematic review (and others sent as rapid responses to bmj.com[1]) are based more on assumptions than on evidence. Statements that acetazolamide 250 mg is often sufficient to prevent acute mountain sickness are not helpful. Mate de Coca has been recommended for centuries by the Incas, and even magnesium,[2] pervincamine,[3] and methylene blue[4] may be beneficial. In the absence of valid randomised controlled trials, however, an observation stays an observation.
We would like to clarify some of the issues raised.
Firstly, it seems odd to us that some people make a fuss about acetazolamide. Acetazolamide is an old and cheap drug with a low adverse effect profile. There was evidence that 750 mg was effective in the prevention of acute mountain sickness, and there was a lack of evidence for 500 mg with the same end point.
Secondly, the increased efficacy of acetazolamide 750 mg is plausible. Only a few people would argue that a 50% increase in the dose of a drug may not lead to increased efficacy.
Thirdly, our main efficacy end point was complete prevention of acute mountain sickness. This is different from improving symptoms. We concentrated on this high hurdle of efficacy to avoid both observational bias and unnecessary heterogeneity of the data. As with all systematic reviews, the advantage of such rigorous analyses is that readers may get the papers and redo the analysis using their own end points. To reanalyse the 500 mg data using a different end point as suggested would change the pooled number needed to treat for prevention of acute mountain sickness from 7.1 to 6.6. This difference is unlikely to be of clinical relevance.
Finally, the suspicion has been raised that our analysis was based on a biased selection of studies and that we did not include all 10 trials of acetazolamide 500 mg, as in a previously published meta-analysis.[5] We did not include 10 trials for three reasons. Firstly, there are only nine; secondly, four were not randomised; and, thirdly, one did not report any dichotomous data on efficacy or harm. We could have analysed continuous data. The clinical relevance of weighted mean differences and P values, however, is not obvious. What, for instance, does an effect size of -0.61 (95% confidence interval -0.29 to -0.93) indicate? For rational decision making we need to know how well something works, and not only that it works.
We agree that the pivotal trial should randomise subjects at similar rates of ascent to different doses of acetazolamide.
L Dumont staff anaesthesiologist lionel.dumont@hcuge.ch
M R Tramer staff anaesthesiologist Division of Anaesthesiology, Geneva University Hospital, CH-1211 Geneva 14, Switzerland
C Mardirosoff staff anaesthesiologist Departement d'Anesthesie-Reanimation, Hopital Sud-Leman-Valserine, 74164 Saint-Julien en Genevois, France
[1] Electronic responses. Efficacy and harm of pharmacological prevention of acute mountain sickness: quantitative systematic review.bmj.com 2000;321 www.bmj.com/cgi/ content/full/321/7256/267#responses; accessed 7 Dec 2000.
[2] Dumont L, Mardirosoff C, Soto-Debef G, Tassonyi E. Magnesium and acute mountain sickness. Aviat Space Environ Med 1999;70:625.
[3] Adenis L. Expedition noel 1974 au Kilimandjaro (5885 m). Essais de la pervincamine forte retard. J Sci Med Lille 1976;94:125-6.
[4] Brooks MM. Methylene blue, an antidote to altitude sickness. Aviat Med 1948;Aug:298-9.
[5] Ried LD, Carter KA, Ellsworth A. Acetazolamide or dexamethasone for prevention of acute mountain sickness: a meta-analysis. Journal of Wilderness Medicine 1994;5:34-48.
COPYRIGHT 2001 British Medical Association
COPYRIGHT 2001 Gale Group