Emergency Medicine · Medical Literature
Vasopressors or Fluids in Early Septic Shock
Abstract
New England Journal of Medicine, Ahead of Print.
Clinical relevance
In adult patients presenting to the emergency department with early septic shock, does a strategy of restricted intravenous fluid volume with early vasopressor initiation result in more days alive and out of the hospital at 90 days compared with a strategy of greater fluid volume with later vasopressor use? __SNAPSHOT__ {"studyDesign":"Investigator-initiated, multicenter, open-label, parallel-group randomized controlled trial (ARISE FLUIDS; ClinicalTrials.gov NCT04569942)","setting":"51 emergency departments across Australia, New Zealand, and the Republic of Ireland; 2021–2025","sampleSize":"n=963 intention-to-treat population (vasopressor group: n=481; fluids group: n=482); 1,000 patients randomized, 37 excluded post-randomization due to absence of informed consent","population":"Adults presenting to the ED with clinically suspected infection, SBP 2.0 mmol/L, and IV antimicrobials initiated. Median age 69 years; 60.6% male. Exclusions included receipt of >2,000 mL IV fluids before enrollment, >6 hours since ED presentation, care limitations, immediate need for surgery, and clinician-determined unsuitability.","intervention":"Restricted fluid volume with early vasopressor therapy: IV resuscitation fluids stopped at randomization, vasopressors commenced immediately; 250 mL boluses permitted only for specified signs of persistent hypoperfusion. Intervention applied for at least 6 hours and up to 24 hours.","comparator":"Greater fluid volume with later vasopressor therapy: up to 1,000 mL initially, then 500 mL boluses for persistent hypotension or hypoperfusion; 30 mL/kg within 3 hours recommended unless contraindicated. Vasopressors used later per standard practice.","primaryOutcome":"Days alive and out of the hospital from randomization to day 90 (DAOH-90); patients dying on or before day 90 assigned zero days.","followUp":"90 days for primary outcome; secondary outcomes included mortality at 28 and 90 days, organ support-free days to day 28, and quality of life at 6 and 12 months (12-month disability data not yet reported at time of publication).","fundingSource":"Government (Australian National Health and Medical Research Council Medical Research Future Fund, APP1200084; New Zealand Health Research Council, 22/409). No industry funding reported."}
Key findings
The primary outcome did not differ between groups. The median days alive and out of the hospital at day 90 was 76 days (IQR 55–83) in the vasopressor group and 76 days (IQR 55–82) in the fluids group (difference 0.0 days; 95% CI −2.7 to 2.7; P=1.00). Treatment separation was achieved rapidly: patients in the vasopressor group received a median of 1,108 mL less IV fluid in the first 24 hours than the fluids group (95% CI −1,395 to −850 mL), and vasopressor use was 18.9 percentage points higher in the vasopressor group (95% CI 13.3 to 24.5). Mortality at 28 days was 12.9% (vasopressor) vs. 10.0% (fluids), and at 90 days was 16.4% (vasopressor) vs. 14.4% (fluids); these differences were described as non-significant (exact p-values not reported in the abstract). Pulmonary edema occurred in 0.6% of the vasopressor group vs. 5.0% of the fluids group (P<0.001), representing an approximately 8-fold relative reduction (calculated), though this outcome was assessed by unblinded clinicians and must be interpreted cautiously. Despite the pulmonary edema signal, rates of mechanical ventilation were equal at approximately 15% in both groups. ICU admission was more frequent in the vasopressor group (approximately 77% vs. 68%). Other adverse events occurred at similar rates between groups. NNT for the primary outcome: not applicable (null result). NNH for pulmonary edema favoring vasopressor strategy: approximately 23 (calculated from 5.0% vs. 0.6% absolute risk difference of 4.4%).
Practice implications
IMPORTANT CAVEAT: This summary is based on the published abstract and publicly available data only; the full manuscript was not available for review. Findings should be interpreted with awareness that additional analyses, subgroup data, and safety detail in the full text may materially affect interpretation. What the study established: Among ED patients with early septic shock who had already received an initial fluid bolus of approximately 1,000 mL, a strategy of restricting further fluids and initiating vasopressors early is feasible, achieves meaningful treatment separation, and does not result in worse days alive and out of the hospital at 90 days compared with a more fluid-liberal approach. The early vasopressor strategy was associated with a statistically significant reduction in clinician-reported pulmonary edema, though this is a secondary outcome subject to detection bias given the open-label design. What the study did not establish: The trial does not demonstrate superiority of the vasopressor-first strategy for the primary patient-centered outcome. It does not establish an improvement in 90-day mortality, ventilator-free days, or functional outcomes. The open-label design precludes definitive interpretation of the pulmonary edema signal, which was not mirrored in a reduction in mechanical ventilation rates. Longer-term disability outcomes (12-month WHODAS) were not yet reported. Applicability: Findings apply to ED patients who meet Sepsis-3 criteria for septic shock, have received an initial 1,000 mL fluid bolus, have lactate >2 mmol/L, and have already commenced antimicrobials — and who present within 6 hours of ED arrival. Patients who had received >2,000 mL before enrollment, those requiring immediate surgery, and those with care limitations were excluded. The trial was conducted in high-resource Australasian/Irish emergency systems, which may limit generalizability to under-resourced or non-ED settings. The median age of 69 years and predominance of respiratory and urinary infection sources should be considered when applying findings to different epidemiological contexts. Clinical importance: The result is concordant with prior trials (CLOVERS, CLASSIC) showing no significant outcome difference between restrictive and liberal fluid approaches in septic shock. Taken together, these trials challenge the historical mandate of 30 mL/kg fluid as a universal standard and provide clinicians with evidence that early vasopressor initiation — with reduced fluid administration — is safe and does not harm outcomes in monitored settings. The result supports individualizing resuscitation rather than adopting either strategy as universally superior. __LIMITATIONS__ 1. Open-label design: unblinded clinicians assessed the pulmonary edema outcome, introducing high risk of detection bias; the lack of a corresponding reduction in mechanical ventilation rates suggests clinician-reported pulmonary edema overestimates the true treatment effect on respiratory harm. 2. Exclusion of patients receiving >2,000 mL before enrollment and those presenting >6 hours after ED arrival likely selects for earlier-presenting, potentially less severely resuscitated patients, limiting applicability to those already fluid-loaded at arrival. 3. The trial was powered to detect a 7-day difference in DAOH-90; smaller but potentially clinically meaningful differences in either direction cannot be excluded, though the point estimate of 0.0 days makes a large undetected effect very unlikely. 4. Consent was waived for 37 patients, introducing a minor risk of selection and consent bias. 5. Three patients in the fluids group were lost to follow-up for the primary outcome. 6. The 12-month disability outcome (WHODAS 2.0) was pre-specified but not yet reported, leaving open whether long-term functional recovery differs between strategies. 7. The trial was conducted predominantly in Australasian ED systems with rapid access and monitoring infrastructure; generalizability to resource-limited or non-ED settings is uncertain. 8. The open-label design may have influenced downstream clinical decisions beyond the 24-hour intervention window, potentially confounding secondary outcomes. 9. Nominal mortality differences (12.9% vs. 10% at 28 days; 16.4% vs. 14.4% at 90 days) favored the fluids group in point estimates; while non-significant, these trends warrant continued attention in meta-analyses. __BOTTOM_LINE_CATEGORY__ potentially_practice_informing __BOTTOM_LINE__ In ED patients with early septic shock after an initial ~1,000 mL fluid bolus, a strategy of restricting further fluids and starting vasopressors early produces equivalent 90-day days-alive-and-out-of-hospital outcomes compared with a greater-fluid, later-vasopressor approach (median 76 days in both groups; difference 0.0 days, 95% CI −2.7 to 2.7; P=1.00). This null result, converging with CLOVERS and CLASSIC, provides clinicians with evidence to safely abandon the universal 30 mL/kg fluid mandate and to individualize resuscitation based on patient physiology, fluid tolerance, and monitoring capability. Neither strategy is proven superior; early vasopressor use is safe but does not improve patient-centered outcomes when applied routinely to all comers.