CLOVERS
PubMed • Full text • PDF • ClinicalTrials.gov
Clinical Question
In patients with sepsis induced hypotension who have already received 1-3L of IVF, is a strategy that prioritizes early use of vasopressors superior to one that prioritizes fluid boluses.
Bottom Line
There is no mortality difference between a strategy that prioritizes early vasopressors (restrictive group) compared to maximizing IV fluids (liberal group) in patients with sepsis induced hypotension who have already received 1-3L of IVF resuscitation.
Major Points
Several observational trials showed potential benefits of restrictive approaches to fluids in septic shock compared to large volumes of fluids. A recent clinical trial (CLASSIC) compared restrictive versus liberal fluid protocols among patients admitted to the ICU which did not show a difference in mortality. CLASSIC used dynamic techniques to identify fluid responsiveness to guide further fluid resuscitation. CLOVERS adds to this literature by randomizing patients earlier into presentation (most patients randomized in ER). CLOVERS also collected data on safety of administration of peripheral vasopressors.
Study was halted by the DSMB at the second interim analysis for futility in the primary outcome.
Guidelines
As of February 2023, no guidelines have been published that reflect the results of this trial.
Design
- Multicenter, unblinded, randomized, controlled trial
- N=1,563
- Restrictive (n=782)
- Liberal (n=781)
- Analysis: Intention-to-treat
- Primary outcome: death from any cause before discharge home at 90 days
Population
Inclusion Criteria
- Adults (>18 years old)
- Sepsis (suspected or confirmed infection)
- Sepsis induced hypotension - systolic BP <100 after > 1L IVF
Exclusion Criteria
- >4 hours since meeting criteria before randomization
- >24 hours since presentation to the hospital
- Receipt of >3L IVF prior to randomization (including administration by EMS)
- Presence of fluid overload
- Severe volume depletion from non-sepsis causes
Baseline Characteristics
- Mean age: 59 years
- Female sex: 47%
- Race
- White - 70%
- Black - 15%
- Asian - 3.5%
- Other - 1%
- Not reported - 10%
- Hispanic or Latino ethnic group: 14.5%
- Coexisting conditions
- Diabetes - 29%
- Chronic CHF - 11.5%
- ESKD on HD - 4.7%
- Mean SOFA score: 3.4
- Mean systolic BP: 93.5
- Median time to randomization: 61 min
- Location at randomization
- ER - 92%
- ICU - 7%
- Other - 1%
- Median volume of fluid given prior to randomization: 2050 mL
Interventions
- Trial went for 24 hours post randomization
- Randomized 1:1 to restrictive or liberal fluid groups
- Restrictive group prioritized vasopressor administration
- Halt all maintenance and bolus fluids
- Could administer boluses up to 2L (including pre-randomization fluids)
- If MAP <65 or SBP <90 - start vasopressors (norepinephrine) to achieve MAP of 65, could add 2nd vasopressor if needed
- Could give 'rescue fluids' if one if the following was met:
- severe hypotension (systolic < 70, MAP <55)
- refractory hypotension (systolic <90, MAP < 65 with norepinephrine at dose of 20 ug/min or equivalent dose of other pressor)
- lactate level >4 mmol/L and increasing after 2 hrs of therapy
- sinus HR > 130 for >15 min
- echocardiographic or hemodynamic evidence of extreme hypovolemia
- 'rescue fluids' could be given at any time by clinical team at their discretion
- Liberal group
- Halt all maintenance fluids
- Give 2L bolus at randomization (could stop after 1L if clinician determined patient is volume replete)
- If the patient has any of the following, give 500cc bolus
- MAP <65 or SBP <90
- Lactate >4mmol/L and increasing
- Decreased UOP (<30cc/hr)
- Sinus HR >110 BPM
- receipt of vasopressors to maintain BP
- In the following situations, 'rescue vasopressors' could be used
- severe hypotension (systolic < 70, MAP <55)
- lactate level >4 mmol/L and increasing after 2 hrs of therapy
- clinical manifestations of fluid overload (also halt fluids)
- >5L total IVF administered
- 'rescue vasopressors' could be given at any time by clinical team at their discretion
- Restrictive group prioritized vasopressor administration
Outcomes
Comparisons are restrictive group vs. liberal group.
Primary Outcomes
- Death before discharge home by day 90
- 14% vs. 14.9% (Difference -0.9, 95% CI -4.4-2.6; P=0.61)
Secondary Outcomes
- # days free from organ support at 28 days
- 24 vs. 23.6 (Difference 0.3; 95% CI -0.5-1.2)
- # days free from organ ventilator use at 28 days
- 23.4 vs. 22.8 (Difference 0.6; 95% CI -0.5-1.2)
- # days free from renal-replacement therapy at 28 days
- 24.1 vs. 23.9 (Difference 0.2; 95% CI -0.4-1.6)
- # days free from vasopressor use at 28 days
- 22 vs. 21.6 (Difference 0.4; 95% CI -0.8-1.2)
- # days free out of ICU from day 1 to day 28
- 22.8 vs. 22.7 (Difference 0.1; 95% CI -0.8-1.0)
- # days free out of hospital by day 28
- 16.2 vs. 15.4 (Difference 0.8; 95% CI -0.3-1.7)
- New intubation and mechanical ventilation by 28 days - # (%)
- 77 (11%) vs. 87 (12.7%) (Difference ; -1.7 95% CI -5.1-1.6)
- Initiation of renal-replacement therapy by 28 days - # (%)
- 24 (3.3%) vs. 24 (3.3%) (Difference 0; 95% CI -1.8-1.8)
- KDIGO score on day 3
- 0.35 vs 0.34 (Difference 0; 95% CI -0.1-0.1)
- Change in SOFA score from baseline to 72 hr
- -0.7 vs -0.8 (Difference 0.1; 95% CI -0.3-0.4)
- Death from any cause at /any/ location - # (%)
- 172 (22.4%) vs. 169 (21.9%) (Difference ; 0.5 95% CI -3.6-4.7)
- ARDS onset between day 1 and day 7 - # (%)
- 59 (7.6%) vs. 67 (8.6%) (Difference ; -1.0 95% CI -3.7-1.7)
- severe adverse events - #
- 21 vs. 19 (Difference ; 2.0 95% CI -10-14)
Additional Outcomes
- Median volume of fluids given - mL (IQR)
- First 6 hrs: 500 (130-1907) vs 2300 (2000-3000)
- Over 24 hrs: 1267 (555-2279)vs 3400 (2500-4495)
- Vasopressors in first 24 hrs - no/total (%)
- 460/780 (59%) vs 290/779 (37%)
- Time to first vasopressor use (hrs)
- 1.8 +/- 3.4 vs 3.2 +/- 4.7
- Duration of vasopressor use in first 24 hours (hrs)
- 9.6 +/- 10 vs 5.4 +/- 8.6
Subgroup Analysis
Multiple subgroups were compared for the primary outcome including: age (greater or less than 65), Sex, Race, Hispanic/Latino ethnicity, location at randomization, chronic CHF, ESKD, baseline SBP <90, baseline HTN, Total SOFA score (0-1, 2, 3-5, 6-16), and primary source of infection (pneumonia vs other). For all subgroups the 95% CI crossed 1 indicating no significant difference. The only subgroup of note was ESKD where the primary outcome was met in 27.3% vs 47.5% (95% CI -41.9-1.5).
Adverse Events
- Number of serious adverse events
- 21 vs 19
Criticisms
- Stopped early meaning that some subgroubs (ie ESKD) could have reached significance with more patients.
- Clinicians did not use dynamic measures of fluid responsiveness to guide fluid administration.
- There was no standard of care arm in this trial but rather a comparison of two strict protocols.
- Trial protocol only went for first 24 hours.
Funding
NIH/NHLBI