We made the following changes to the protocol (Martí‐Carvajal 2012).

• 1. The list of authors has changed since the protocol was published. Martí‐Carvajal AJ is not present and Franco VJA, Arevalo‐Rodriguez I and Xu Y were included.

  2. The background has been modified: some text related to the history of the concept of HAI has been deleted; and more recent references have been included. We also provided more details on how the interventions might work.

  3. In the section Types of studies: in the protocol it was mentioned that "we will exclude quasi‐randomized studies and prospective observational studies for evaluating clinical effectiveness. However, we will consider these studies for reports on adverse events". We did not include quasi‐randomized studies and prospective observational studies for reports on adverse events because the methodology to do this was not detailed in the protocol. However, we collected and analysed all information regarding adverse events from included studies.

  4. In the section Types of interventions: in the protocol "frusemide" is listed as an intervention, which is the previous chemical denomination of the loop diuretic. Since the current denomination of this intervention is "furosemide" (Pubchem ‐ Furosemide 2017), we used this denomination throughout the review.

  5. In the section Types of outcome measures: we have included the definition of the outcome 'Complete relief of acute mountain sickness symptoms' by adding the following text: "defined as the complete absence of the acute mountain sickness symptoms by the end of the study".

  6. This outcome was considered as a binary outcome as it was originally stated in the protocol at the section Measures of treatment effect.

  7. In the section Electronic searches: the Chinese database Wanfang (Wanfangdata.com) was included in the search. This decision was taken by the review authors since several studies taking place in the Tibet and other areas of Asia may not appear in CENTRAL, MEDLINE, Embase, LILACS, ISI Web of Science and CINAHL.

  8. In the section Searching other resources: we added the date of search of the World Health Organization International Clinical Trials Registry Platform (ICTRP; search date 24 February 2017) and of the principal investigators (3 March 2017). We added the following text after the dates: "Unpublished trials will be considered in updating of this review".

  9. In the section Measures of treatment effect: the phrases "The unit of analysis will be the patient" and "We will collect and analyse a single measurement for each outcome from each participant" included in the protocol in the section of Measures of treatment effect were moved to the section Unit of analysis issues. We have considered adverse events (stated in the outcomes) as a synonym of safety. We have rewritten this section including the word 'safety' in brackets next to the words 'adverse events' that is the outcome defined for the review.

  10. For analysis of continuous outcomes, we used standardized mean differences instead of mean differences, taking into account that included studies used different scales to measure the improvement of HAI symptoms. This analysis was used to present the findings about reduction in illness severity for acetazolamide versus placebo.

  11. In the sections summary of findings Table for the main comparison and summary of findings Table 2 and GRADE: we have considered adverse events (stated in the outcomes) as a synonym of safety. We have rewritten this section including the word 'safety' in brackets next to the words 'adverse events' that is the outcome defined for the review. We also expanded the description on how we developed the 'Summary of findings' tables, and on how we took them into account to appraise the overall quality of evidence, the magnitude of effect of the interventions examined and the sum of available data on the outcomes we considered.

  12. In Appendix 5: the search strategy in the WHO International Trials Registry Portal has been modified in order to improve sensitivity. Before: advanced search: high‐altitude pulmonary oedema (in the title field). After: altitude Sickness OR Altitude illness OR acute mountain sickness OR High‐altitude oedema OR high‐altitude oedema (in the title field).

  13. Due to scarcity of evidence we were unable to carry out the following methods:

      1. we planned to present the results of continuous outcomes as summary standardized mean difference with 95% CI. Instead, we presented the standardized mean difference for pooled results and the mean difference for individual studies.

      2. exploration of heterogeneity and sensitivity analyses.

      3. assessment of reporting biases.

      4. use of fixed‐effect and random‐effects models.

      5. subgroup analysis