Skip to main content

Volume 14 Supplement 2

Sepsis 2010

Prospective multicenter study of the effect of early fluid resuscitation on trends in IL-6 and TNFα levels in severe sepsis


The prognostic capability of TNFα and IL-6 is limited in septic shock. Previous studies were performed prior to publication of current therapeutic guidelines recommending aggressive early resuscitation. The objective of the present study was to evaluate the impact of early fluid resuscitation on serial TNFα and IL-6 levels and its association with mortality in severe sepsis.


This is a substudy of a previously completed prospective, observational multicenter investigation of patients with severe sepsis. Inclusion criteria were age >17, infection with ≥2 SIRS, hypotension despite fluid challenge, treatment with a standardized quantitative resuscitation protocol, and identification within 3 hours of treatment initiation. Blood samples were obtained at enrollment, 6 hours, and 24 hours. Therapeutic amounts of intravenous crystalloid fluid was defined by ≥5 l and <5 l over 24 hours (initial 2 l fluid challenge over 4 hours followed by 150 ml/hour for 20 hours). Data analysis compared absolute levels of TNFα and IL-6 at each time point between survivors and nonsurvivors. The magnitude and direction of serial cytokine levels was quantified by the percentage difference of each marker for each patient between 0 and 6 hours and 0 and 24 hours. Statistical analysis was performed using the Wilcoxon-rank-sum test or the Student t test.


Forty patients were enrolled; 11 died. Vasopressors were required in 60% of all patients. Absolute values of IL-6 (pg/ml) were higher in nonsurvivors than survivors at enrollment (5,479 vs. 710); 6 hours (4,180 vs. 405), and 24 hours (5,710 vs. 377) (P < 0.05). There was no difference in TNFα values between the two groups (P = NS at 0, 6, 24 hours). Nonsurvivors had a larger percentage (difference) in both TNFα and IL-6 than survivors at 24 hours. See Figure 1. Treatment with ≥5 l intravenous fluid over 24 hours was associated with a 32% decline in IL-6 compared with a 64% increase in IL-6 with <5 l fluid therapy. See Figure 2.

Figure 1

%Δ IL-6 and TNF.

Figure 2

%Δ at 24 hours.


In the context of a quantitative protocol for the treatment of severe sepsis, high-volume fluid resuscitation is associated with a decline in the percentage difference of IL-6. Trends in the percentage difference of both TNFα and IL-6 differentiate survivors from nonsurvivors. Further investigation is needed into the impact fluid resuscitation has on decreasing the inflammatory insult and the use of serial cytokine measurements as a measure of therapeutic effectiveness.


Conducted within the Emergency Medicine Shock Research Network (EMShockNet). RA has no financial disclosures relevant to this study but has received research funding from Hutchinson Technologies. The present study was supported in part by a grant from the Shock Society/Novo Nordisk research grant for Hemorrhagic Shock and Hemostasis to ST. AJ's effort is supported by a grant from the National Institutes of Health/National Institutes of General Medical Sciences K23GM076652. NS is supported in part by grants from the National Institutes of Health L091757 and GM076659.

Author information



Corresponding author

Correspondence to R Arnold.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Arnold, R., Jones, A., Shapiro, N. et al. Prospective multicenter study of the effect of early fluid resuscitation on trends in IL-6 and TNFα levels in severe sepsis. Crit Care 14, P24 (2010).

Download citation


  • Severe Sepsis
  • Percentage Difference
  • Fluid Resuscitation
  • Serial Cytokine
  • Fluid Challenge