Service of Internal Medicine, Lausanne University Hospital, Lausanne, Switzerland.
Service of Endocrinology, Diabetes, and Metabolism, Lausanne University Hospital, Lausanne, Switzerland.
JAMA. 2017 Nov 14;318(18):1810-1819. doi: 10.1001/jama.2017.16192.
Acute mountain sickness (AMS) affects more than 25% of individuals ascending to 3500 m (11 500 ft) and more than 50% of those above 6000 m (19 700 ft). AMS may progress from nonspecific symptoms to life-threatening high-altitude cerebral edema in less than 1% of patients. It is not clear how to best diagnose AMS.
To systematically review studies assessing the accuracy of AMS diagnostic instruments, including the visual analog scale (VAS) score, which quantifies the overall feeling of sickness at altitude (VAS[O]; various thresholds), Acute Mountain Sickness-Cerebral score (AMS-C; ≥0.7 indicates AMS), and the clinical functional score (CFS; ≥2 indicates AMS) compared with the Lake Louise Questionnaire Score (LLQS; score of ≥5).
Searches of MEDLINE and EMBASE from inception to May 2017 identified 1245 publications of which 91 were suitable for prevalence analysis (66 944 participants) and 14 compared at least 2 instruments (1858 participants) using a score of 5 or greater on the LLQS as a reference standard. To determine the prevalence of AMS for establishing the pretest probability of AMS, a random-effects meta-regression was performed based on the reported prevalence of AMS as a function of altitude.
AMS prevalence, likelihood ratios (LRs), sensitivity, and specificity of screening instruments.
The final analysis included 91 articles (comprising 66 944 study participants). Altitude predicted AMS and accounted for 28% of heterogeneity between studies. For each 1000-m (3300-ft) increase in altitude above 2500 m (8200 ft), AMS prevalence increased 13% (95% CI, 9.5%-17%). Testing characteristics were similar for VAS(O), AMS-C, and CFS vs a score of 5 or greater on the LLQS (positive LRs: range, 3.2-8.2; P = .22 for comparisons; specificity range, 67%-92%; negative LRs: range, 0.30-0.36; P = .50 for comparisons; sensitivity range, 67%-82%). The CFS asks a single question: "overall if you had any symptoms, how did they affect your activity (ordinal scale 0-3)?" For CFS, moderate to severe reduction in daily activities had a positive LR of 3.2 (95% CI, 1.4-7.2) and specificity of 67% (95% CI, 37%-97%); no reduction to mild reduction in activities had a negative LR of 0.30 (95% CI, 0.22-0.39) and sensitivity of 82% (95% CI, 77%-87%).
The prevalence of acute mountain sickness increases with higher altitudes. The visual analog scale for the overall feeling of sickness at altitude, Acute Mountain Sickness-Cerebral, and clinical functional score perform similarly to the Lake Louise Questionnaire Score using a score of 5 or greater as a reference standard. In clinical and travel settings, the clinical functional score is the simplest instrument to use. Clinicians evaluating high-altitude travelers who report moderate to severe limitations in activities of daily living (clinical functional score ≥2) should use the Lake Louise Questionnaire Score to assess the severity of acute mountain sickness.
急性高山病（AMS）影响超过 25%的上升至 3500 米（11500 英尺）的个体，以及超过 50%的上升至 6000 米（19700 英尺）的个体。在不到 1%的患者中，AMS 可能从非特异性症状进展为危及生命的高海拔脑水肿。目前尚不清楚如何最好地诊断 AMS。
系统回顾评估 AMS 诊断工具准确性的研究，包括视觉模拟量表（VAS）评分，该评分量化了在高海拔时的整体不适感觉（VAS[O]；各种阈值）、急性高山病-大脑评分（AMS-C；≥0.7 表示有 AMS）和临床功能评分（CFS；≥2 表示有 AMS）与 Lake Louise 问卷评分（LLQS；≥5 表示有 AMS）的比较。
从 MEDLINE 和 EMBASE 中检索了从开始到 2017 年 5 月的 1245 篇文献，其中 91 篇适合进行患病率分析（66944 名参与者），14 篇比较了至少 2 种工具（1858 名参与者），使用 LLQS 评分为 5 或更高作为参考标准。为了确定 AMS 的患病率以确定 AMS 的术前概率，根据 AMS 的报告患病率作为海拔的函数进行了随机效应荟萃回归。
AMS 的患病率、筛选工具的似然比（LR）、敏感性和特异性。
最终分析包括 91 篇文章（共 66944 名研究参与者）。海拔预测了 AMS，占研究间异质性的 28%。每升高 1000 米（3300-英尺），海拔超过 2500 米（8200 英尺），AMS 的患病率增加 13%（95%置信区间，9.5%-17%）。VAS[O]、AMS-C 和 CFS 与 LLQS 评分为 5 或更高的检测特征相似（阳性 LR：范围，3.2-8.2；P=.22 用于比较；特异性范围，67%-92%；阴性 LR：范围，0.30-0.36；P=.50 用于比较；敏感性范围，67%-82%）。CFS 只问一个问题：“总的来说，如果您有任何症状，它们对您的活动有什么影响（等级量表 0-3）？”对于 CFS，中度至重度减少日常活动的阳性 LR 为 3.2（95%置信区间，1.4-7.2）和特异性为 67%（95%置信区间，37%-97%）；没有减少到轻度减少活动的阴性 LR 为 0.30（95%置信区间，0.22-0.39）和敏感性为 82%（95%置信区间，77%-87%）。
急性高山病的患病率随着海拔的升高而增加。用于整体高海拔不适感觉的视觉模拟量表、急性高山病-大脑和临床功能评分与使用 LLQS 评分为 5 或更高作为参考标准的 Lake Louise 问卷评分表现相似。在临床和旅行环境中，临床功能评分是最容易使用的工具。对于报告日常生活活动（临床功能评分≥2）有中度至重度限制的高海拔旅行者，临床医生应使用 Lake Louise 问卷评分来评估急性高山病的严重程度。
