As an important antioxidant, vitamin C deficiency has frequently been observed in critically ill patients [1]. The treatment with early high-dose intravenous vitamin C has shown beneficial effects on patients with sepsis, septic shock, and myocardial ischemia in both preclinical and clinical studies [2]. Recently, 2 original articles caught my attention. In December 2019, a nationwide cohort study involving 2713 patients reported in Critical Care demonstrated that high-dose vitamin C therapy was associated with reduced mortality in patients with severe burns [3]. The pleiotropic effects of vitamin C may result from its protection against oxidative stress-mediated cell damage and organ dysfunction [1]. However, the CITRIS-ALI trial, reported earlier by Fowler and colleagues in JAMA, October 2019, showed an inconsistent result. In this study, 167 adults with sepsis and acute respiratory distress syndrome were randomized to receive either high-dose vitamin C or placebo for 96 h, and the primary outcomes including modified Sequential Organ Failure Assessment (mSOFA) scores were not statistically different between the groups [4], despite the expected protective effects of vitamin C against multiple organ failure, inflammation, and endothelial injury as shown in their phase I safety trial [5]. After scrutinizing the results in the CITRIS-ALI trial, we proposed that a survivorship bias may have contributed to the conflicting result. As we noted from their results (Fig. 3, page 1268), there were sudden increases in the overall mortality probability from day 0 to day 7 in the placebo group. Thus, we performed further analyses according to Fig. 3 and discovered that the overall mortalities in the placebo group had more intensive increases on day 2 (10.8 vs. 1.2%; P = 0.021), day 3 (14.5 vs. 2.4%; P = 0.005), and day 4 (19.3 vs. 3.6%; P = 0.001) than those in the vitamin C group (shown in Table 1). Those sudden increases in mortality in the placebo group may result from the deteriorated conditions of certain patients, leaving the survived patients with less severe conditions for statistical analyses at the primary end points. Therefore, when the mSOFA at 96 h and the levels of C-reactive protein and thrombomodulin at 168 h were evaluated, the survivorship bias rendered the differences between the vitamin C group and placebo group uncomparable, causing decreased reliability of the results. Nevertheless, a positive effect of vitamin C still gleams behind the results presented, as a misunderstood ally.

中文翻译：

---

维生素C：一个被误解的盟友？

作为一种重要的抗氧化剂，维生素C缺乏症经常在危重患者中观察到[1]。在临床前和临床研究中，早期大剂量静脉注射维生素 C 对脓毒症、感染性休克和心肌缺血患者均显示出有益效果[2]。最近，2篇原创文章引起了我的注意。2019 年 12 月，一项涉及 2713 名重症监护患者的全国队列研究表明，高剂量维生素 C 治疗与降低严重烧伤患者的死亡率有关 [3]。维生素 C 的多效作用可能源于其对氧化应激介导的细胞损伤和器官功能障碍的保护作用 [1]。然而，Fowler 及其同事于 2019 年 10 月在 JAMA 上早些时候报道的 CITRIS-ALI 试验显示出不一致的结果。在这项研究中，167 名患有败血症和急性呼吸窘迫综合征的成人随机接受高剂量维生素 C 或安慰剂治疗 96 小时，主要结局包括改良的序贯器官衰竭评估 (mSOFA) 评分在各组之间没有统计学差异 [4]，尽管维生素 C 对多器官衰竭、炎症和内皮损伤具有预期的保护作用，如他们的 I 期安全试验所示 [5]。在仔细审查 CITRIS-ALI 试验的结果后，我们提出生存偏差可能导致了相互矛盾的结果。正如我们从他们的结果中注意到的那样（图 3，第 1268 页），安慰剂组从第 0 天到第 7 天的总体死亡率概率突然增加。因此，我们根据图 3 进行了进一步的分析。3 并发现安慰剂组的总体死亡率在第 2 天（10.8 对 1.2%；P = 0.021）、第 3 天（14.5 对 2.4%；P = 0.005）和第 4 天（19.3 对 1.2%；P = 0.005）和第 4 . 3.6%; P = 0.001) 比维生素 C 组（见表 1）。安慰剂组死亡率的突然增加可能是由于某些患者的病情恶化导致存活的患者病情较轻，以便在主要终点进行统计分析。因此，当评估 96 小时的 mSOFA 和 168 小时的 C 反应蛋白和血栓调节蛋白水平时，生存偏差使得维生素 C 组和安慰剂组之间的差异无法比较，导致结果的可靠性降低。尽管如此，维生素 C 的积极作用仍然在所呈现的结果背后闪闪发光，