EBQ:Lactate clearance vs central venous oxygen saturation

Complete Journal Club Article

Jones AE.. "Lactate clearance vs central venous oxygen saturation as goals of early sepsis therapy: a randomized clinical trial". JAMA. 2010. 303(8):739-746.
PubMed Full text PDF

Clinical Question

Is lactate clearance as good as SVO2 as a measure of oxygen delivery to tissues in patients presenting with severe sepsis and septic shock?

Conclusion

For patients with septic shock who were treated by normalizing CVP and MAP; attempt to normalize lactate clearance as opposed to normalize SVO2 showed no significant difference in in-hospital mortality.

Major Points

• Lactate clearance-directed therapy was non-inferior to ScvO2-directed therapy for in-hospital mortality in severe sepsis and septic shock
• Lactate clearance of >=10% over 6 hours was used as the resuscitation target in the intervention group
• Both groups received early antibiotics, fluid resuscitation, and vasopressors per an EGDT-based protocol
• The lactate group had fewer central venous catheter hours, which may reduce catheter-related complications
• Results suggested that lactate clearance could serve as a simpler, more accessible alternative to continuous ScvO2 monitoring

Study Design

• Prospective, randomized, non-inferiority trial
• Single center: Beth Israel Deaconess Medical Center, Boston
• N = 300 patients with severe sepsis or septic shock
• Non-inferiority margin: 10% absolute difference in in-hospital mortality
• Study period: January 2007 - January 2009

Population

Inclusion Criteria

• Severe sepsis or septic shock meeting SIRS criteria with suspected or confirmed infection
• Lactate >=4 mmol/L or SBP <90 mmHg after fluid bolus

Exclusion Criteria

• Age <18 years
• Need for immediate surgery
• Anticipated survival <24 hours
• Contraindication to central venous catheter placement

Inclusion Criteria

• >17 years old AND
• confirmed or presumed infection meeting criteria for severe sepsis or septic shock:
  • 2 or more SIRS criteria AND
  • SBP <90 after 20 mL/kg bolus or blood lactate at least 36 mg/dL

Exclusion Criteria

• Pregnancy
• Primary diagnosis other than sepsis
• Likely surgery required within 6 hours of diagnosis
• Contraindication to chest or neck CVC
• Cardiopulmonary resuscitation
• Transfer from an institution with sepsis protocol already underway
• Advance directive restricting study protocol

Interventions

• Randomized into 1 of 2 resuscitation groups
• CVP was managed first in both groups to achieve a CVP of at least 8
  • Isotonic boluses given
• SBP was managed second to maintain a MAP of at least 65
  • Fluid resuscitation followed by vasopressors (dopamine and norepi)
• The groups differed in the third physiologic parameter that was targeted: SVO2 vs. lactate clearance
  • SVO2 of 70%
  • lactate clearance of 10%
  • If hematocrit <30 and either target not achieved, PRBC transfusion to achieve hematocrit of at least 30
  • If hematocrit was at least 30 and either target not achieved, then dopamine titrated to achieve effect

Outcome

*Intention to treat analysis

Primary Outcomes

• Absolute in-hospital mortality rate
  • SVO2: 23%
  • Lactate: 17%
  • Did not reach -10% threshold

Secondary Outcomes

• ICU length of stay
  • SVO2: 8.46%
  • Lactate: 7.39%
  • p-value = 0.75
• Hospital length of stay
  • SVO2: 10.89%
  • Lactate: 11.68%
  • p-value = 0.60
• Ventilator-free days
  • SVO2: 10.39%
  • Lactate: 11.09%
  • p-value = 0.67
• New onset of multiple organ failure
  • SVO2: 25%
  • Lactate: 22%
  • p-value = 0.68

Criticisms

• Single-center study limits generalizability
• Non-inferiority design with a wide 10% margin may not detect clinically meaningful differences
• Both groups still required central venous access, so the practical advantage of the lactate strategy was limited
• A single lactate measurement at 6 hours may miss important trends in the intervening period
• The study was underpowered to detect superiority of either approach

Funding

• National Institutes of Health (NIH)