CLASSIC Trial – Liberal vs Restrictive Fluid strategy in Septic Shock – Less is More?

The CLASSIC trial 

Blog written by Dr Jose Chacko 

Setting and design

The CLASSIC trial was a multicentric randomized clinical trial carried out in 31 ICUs in Denmark, Norway, Sweden, Switzerland, Italy, the Czech Republic, the United Kingdom, and Belgium. The study was conducted over a 3-year period between November 27, 2018, and November 16, 2021.

Inclusion criteria

• Patients with septic shock
• Suspected or confirmed infection
• Plasma lactate level of ≥2 mmol/l
• On a vasopressor or inotrope infusion
• Minimum 1 liter of intravenous fluids in the 24 hours before the screening
• Shock for ≥12 hours

Excluded

• No consent
• Life-threatening hemorrhage
• Had acute burns with >10% involvement
• Pregnant

Randomized in 1:1 ratio, to receive restrictive intravenous vs. standard intravenous fluid therapy. Randomization was in permuted blocks of 6 or 8, and stratified according to trial site and the presence of metastatic or hematologic malignancy.

Intervention

Restrictive

Intravenous fluid bolus of 250 to 500 ml of isotonic crystalloid (normal saline or buffered solutions) if any of the following criteria were met

• Serum lactate ≥4 mmol/l
• MAP <50 mm Hg, with vasopressor or inotrope infusion
• Mottling above the knee
• Urine output <0.1 ml/kg in the first 2 hours after randomization
• Intravenous fluids to replace fluid losses (gastrointestinal or drain losses)
• Correction of dehydration or electrolyte loss if not possible enterally
• Provide a total daily fluid intake of 1 liter, including fluids associated with administration of medication and nutrition (if enteral nutrition not possible)

Standard therapy

• Intravenous fluid if hemodynamic improvement (2016 SS
• Replace losses
• Correction of dehydration or electrolyte abnormalities
• Maintenance fluid in ICUs with a protocol that had a recommendation
• No upper limit

Common to both groups

• Recommendations regarding type of fluid (isotonic crystalloids for circulatory impairment)
• Co-interventions including antibiotic therapy, source control, norepinephrine as vasopressor, and renal replacement therapy
• Assigned type of care until discharge from ICU or for a maximum of 90 days
• Rest of the management based on clinician judgment

Sample size

• 90-day mortality as primary endpoint
• Assumed 45% mortality at baseline
• 7% difference in 90-day mortality, 15% relative risk reduction
• 80% power, two-sided alpha of 0.05
• Sample size of 1554

 Baseline characteristics

The median total volume of IV fluids in the ICU excluding medication fluid and nutrition: 1798 ml vs. 3811 ml

Median cumulative volume including all fluids: 10,433 ml vs. 12,747 ml

Subgroup analysis

No difference in the 90-day mortality in pre-defined subgroups: With our without respiratory support, AKI, lactate 4.0 mmol/l or less, body weight 76 kg or more, IV fluids received prior to randomization 30 ml/kg or more

Limitations

• 2 vs. 3.0 litres already administered in the 24 hours prior to randomization
• The difference in the volume of fluid administered between groups may not have been large enough to observe a difference in outcome (median total volume of IV fluids in the ICU excluding medication fluid and nutrition: 1798 ml vs. 3811 ml; median cumulative volume including all fluids: 10,433 ml vs. 12,747 ml). Patients in the control group may have received less than the usual volumes of fluid as observed in previous studies
• The difference between groups by day 5 was approximately 2L (8.8L vs. 10.8L); by day 5, the difference was approximately 800mls
• Had the practice already changed to a restrictive strategy before the commencement of this trial?
• No information on respiratory endpoints: P/F ratio, duration of ventilator support, ventilator-free days
• Unblinded
• Large volumes of fluid were administered outside the limits of the protocol

Summary – 

A restrictive fluid strategy is safe to use in patients with septic shock.

One should avoid unnecessary use of iV fluids in critically ill patients.