[The preventive effect of four drugs on acute mountain sickness: a Bayesian network meta-analysis].

Objective: To compare and predict the preventive effects of acetazolamide and other drugs on acute mountain sickness(AMS). Methods: Following the retrieval strategy of PRISMA statement of systematic review and meta-analysis, we searched the databases of PubMed, Embase, Cochrane Library, Web of Science, CNKI, Wanfang, etc. from January 1, 1980 to November 30, 2020, and randomized controlled trials (RCT) consistent with drug prevention of AMS were conducted. Using R and other statistical software, Markov chain-Monte Carlo method was carried out for network meta-analysis under Bayesian framework, and node separation method was performed to check the consistency of closed-loop research. Results: Twenty-three literatures (25 studies) were included to compare the preventive effects of 4 drugs on AMS. Bayesian network meta-analysis showed that the incidence of AMS in acetazolamide group (ACE), dexamethasone group (DEX), ginkgo biloba extract group (GBE) and rhodiola group (RHO) was lower than that in placebo group (PLA). In the comparison of drug groups, the incidence of AMS in ACE, DEX and RHO was lower than that in GBE. There was no statistically significant difference in the incidence of AMS among ACE, DEX and RHO groups. Eight of these studies reported the effects of two drugs on pulse oxygen saturation (SpO2) in people entering the target altitude. Bayesian network meta-analysis showed that SpO2 in RHO was higher than that in ACE and PLA, but there was no statistically significant difference in SpO2 between ACE and PLA. The probability ranking of prevention AMS effect grade showed that the rank 5th probability of AMS in ACE, DEX, GBE, RHO and PLA was 45.72%, 48.80%, 0, 5.48% and 0, respectively. The probability ranking of improving the SpO2 level of the target altitude population showed that the probability of the ACE, RHO and PLA ranking 1st in improving the SpO2 effect at the target altitude was 2.27%, 97.66% and 0.07%, respectively; the results of direct comparison were in good agreement with those of Bayesian prediction model indirectly, and there was no statistical difference. Conclusions: Acetazolamide and dexamethasone can effectively prevent AMS, and should be the first choice for related supplementary research in the future. Rhodiola not only improves the SpO2 of people entering high altitude, but also reduces the incidence of AMS, which needs more attention. Ginkgo biloba extract is not as effective as the above three drugs in preventing AMS and should be used depending on clinical situations.目的： 对比和预测乙酰唑胺等药物对急性高山病（AMS）的预防效果。 方法： 遵循系统评价、Meta分析的PRISMA声明的检索策略，检索PubMed、Embase、Cochrane Library、Web of Science、CNKI、万方等数据库1980年1月1日至2020年12月30日符合药物预防AMS的随机对照研究（RCT），分析研究资料。使用R等统计软件在贝叶斯框架下采用马尔科夫链-蒙特卡罗法进行网状Meta分析，另外使用节点分离法对闭环研究进行一致性检验。 结果： 最终纳入23篇文献（25项研究）对比了4种药物对AMS的预防效果，按照药物分组进行贝叶斯网状Meta分析，乙酰唑胺组、地塞米松组、银杏叶提取物组和红景天组中AMS发生率均低于安慰剂组；在不同药物的相互比较中，乙酰唑胺组、地塞米松组和红景天组AMS发生率低于银杏叶提取物组；而乙酰唑胺组、地塞米松组、红景天组之间AMS发生率差异未见统计学意义（P>0.05）。其中8项研究报道了2种药物对进入目标海拔人群脉搏血氧饱和度（SpO2）的影响，贝叶斯网状Meta分析结果显示：在目标海拔，红景天组的SpO2高于乙酰唑胺组和安慰剂组，乙酰唑胺组与安慰剂组的SpO2的差异并无统计学意义。预防AMS效果概率排序显示：乙酰唑胺组、地塞米松组、银杏叶提取物组、红景天组和安慰剂组发生AMS排名第5的概率分别为45.72%、48.80%、0、5.48%和0。提高目标海拔人群SpO2概率排序显示：乙酰唑胺组、红景天组和安慰剂组在目标海拔提高SpO2效果排名第1的概率分别为2.27%、97.66%和0.07%；直接比较结果和贝叶斯预测模型间接比较结果一致性较好，差异无统计学意义（P>0.05）。 结论： 乙酰唑胺和地塞米松可有效预防AMS发生，应作为今后相关补充研究的首选药物。红景天在提升进入高海拔人群的SpO2同时也可降低AMS的发生率，也应受到足够的重视。银杏叶提取物预防AMS效果不如上述3种药物，临床中应视情况使用。.