现在的位置: 首页研究点评, 进展交流>正文
[Chest读者来信]:治疗全身性感染的魔弹抑或偶然发现的夸大?
2018年06月23日 研究点评, 进展交流 暂无评论

Correspondence

The Magic Bullet in Sepsis or the Inflation of Chance Findings?

To the Editor:

In the June 2017 issue of CHEST, Marik et al1 presented a single-center retrospective before and after study assessing the effects of treating patients with severe sepsis/septic shock with IV vitamin C, hydrocortisone, and thiamine. Forty-seven patients were treated with IV vitamin C, hydrocortisone, and thiamine within 24 hours of ICU admission during a 7-month period (treatment group) and compared with 47 patients admitted to the same ICU over the preceding 7 months (control group). The authors reported a hospital mortality of 8.5% in the treatment group compared with 40.4% in the control group (P < .001; adjusted odds of mortality, 0.13; 95% CI, 0.04-0.48). They concluded that vitamin C, corticosteroids, and thiamine may reduce organ dysfunction and mortality in patients with severe sepsis/septic shock.1

Although we commend the authors for exploring strategies to improve outcomes in this vulnerable population,2 we are concerned that spurious findings, biased results, and overstated conclusions are presented. First, there is no high-quality evidence that any of the three interventions individually improves survival in patients with sepsis.2 Second, an absolute risk reduction of > 30% (relative risk reduction of 87%) in mortality is biologically implausible. Third, the nonrandomized retrospective design, and lack of blinding and a predefined protocol and statistical analysis plan substantially increase the risk of bias.3Fourth, the single-center design increases the risk of inflated estimates.4 Fifth, hospital mortality is considered an inadequate outcome measure, as it is affected by discharge criteria. Finally, with a study population of 94 patients, with 47 in each group implying selection bias, the results are almost certainly affected by a type I error.

The substantial methodological flaws of the study bring into question its external validity and veracity. Our confidence in the estimates is very low due to concerns about risk of bias and imprecision, implying large uncertainty about the results. This is at variance with the authors’ interpretation of the results and conclusions.

Many apparently promising ICU interventions have proved neutral or even harmful following detailed assessment and confirmation by high-quality trials. In a review of critical care trials, eight of 15 ICU interventions used in daily clinical practice over the years proved to increase mortality.5

In conclusion, we highly recommend meticulous assessment of all patient-important benefits and harms of vitamin C, hydrocortisone, and thiamine prior to adopting this unproven strategy in clinical ICU practice.

References

P.E. Marik, V. Khangoora, R. Rivera, M.H. Hooper, J. CatravasHydrocortisone, vitamin C and thiamine for the treatment of severe sepsis and septic shock: a retrospective before-after study
Chest, 151 (6) (2017), pp. 1229-1238
A. Rhodes, L.E. Evans, W. Alhazzani, et al.Surviving Sepsis Campaign: international guidelines for management of sepsis and septic shock: 2016
Intensive Care Med, 43 (3) (2017), pp. 304-377
J. Savovic, H.E. Jones, D.G. Altman, et al.Influence of reported study design characteristics on intervention effect estimates from randomized, controlled trials
Ann Intern Med, 157 (6) (2012), pp. 429-438
A. Bafeta, A. Dechartres, L. Trinquart, A. Yavchitz, I. Boutron, P. RavaudImpact of single centre status on estimates of intervention effects in trials with continuous outcomes: meta-epidemiological study
BMJ, 344 (2012), p. e813
G. Landoni, M. Comis, M. Conte, et al.Mortality in multicenter critical care trials: an analysis of interventions with a significant effect
Crit Care Med, 43 (8) (2015), pp. 1559-1568

给我留言

您必须 [ 登录 ] 才能发表留言!

×
腾讯微博