Skip to main content

Foundational Trials

APROCCHSS

Image: Shutterstock / ChaNaWiT

Publication

  • Title: Hydrocortisone plus fludrocortisone for adults with septic shock
  • Acronym: APROCCHSS
  • Year: 2018
  • Journal published in: New England Journal of Medicine
  • Citation: Annane D, Renault A, Brun-Buisson C, et al.; CRICS-TRIGGERSEP Network. Hydrocortisone plus fludrocortisone for adults with septic shock. N Engl J Med. 2018;378(9):809-818.

Context & Rationale

  • Background
    Septic shock remains a leading cause of ICU death; corticosteroids had long been used to accelerate shock reversal, but mortality effects were uncertain and trial results conflicted.
    A prior RCT of low-dose hydrocortisone plus fludrocortisone suggested improved survival in septic shock2, whereas a later large trial of hydrocortisone (without routine fludrocortisone) did not show a mortality benefit and raised questions around testing/targeting “relative adrenal insufficiency”3.
    At trial conception, drotrecogin alfa (activated protein C) was also in use for septic shock, with ongoing uncertainty around net benefit (including bleeding risk), motivating a factorial approach in the original APROCCHSS design.
  • Research Question/Hypothesis
    In adults with persistent septic shock, adjunctive low-dose corticosteroid therapy (hydrocortisone plus fludrocortisone) would reduce all-cause 90-day mortality compared with placebo, while also hastening resolution of cardiovascular and organ failure.
  • Why This Matters
    If a low-cost, readily deployable regimen reduces mortality in a high-risk syndrome where absolute mortality is substantial, the population impact is large.
    The trial also directly informs how corticosteroids should be administered (bolus regimen, addition of mineralocorticoid) and helps reconcile divergent results between earlier steroid trials.

Design & Methods

  • Research Question: In adults with septic shock, does hydrocortisone plus fludrocortisone (7 days) reduce 90-day all-cause mortality versus placebo?
  • Study Type: Multicentre, randomised, double-blind, placebo-controlled trial in French ICUs; initially a 2×2 factorial trial including drotrecogin alfa (activated), later continued as a two-group trial focusing on corticosteroids after drotrecogin withdrawal; investigator-initiated/publicly funded; central randomisation with stratification by ICU1.
  • Population:
    • Setting: adult ICU patients with persistent septic shock in 34 centres; enrolment began 2008 and concluded 2015.
    • Key inclusion features: suspected/confirmed infection with shock requiring vasopressors despite fluid resuscitation; randomisation within 24 hours of shock onset; high illness severity (e.g., SAPS II ~56, ~92% mechanically ventilated; ~27–28% receiving renal-replacement therapy at baseline).
    • Key exclusions: vasopressor therapy >24 hours before screening/randomisation; contraindications to trial agents; and other protocol-defined conditions limiting feasibility/safety (including bleeding contraindications during the original factorial design).
  • Intervention:
    • Hydrocortisone hemisuccinate 50 mg IV every 6 hours (200 mg/day) for 7 days.
    • Fludrocortisone 50 micrograms once daily (via oral or nasogastric route) for 7 days.
  • Comparison:
    • Matched IV placebo injections (every 6 hours) plus matched oral/nasogastric placebo tablets (daily) for 7 days.
  • Blinding: Double-blind (patients, clinicians, investigators, and outcome assessors); placebo formulations matched active drugs.
  • Statistics: A total of 1280 patients were planned to detect a 10% absolute reduction in 90-day mortality (from 45% to 35%) with 95% power at a two-sided 5% significance level; primary analysis was intention-to-treat with no imputation for missing data; secondary outcomes were not adjusted for multiplicity1.
  • Follow-Up Period: 180 days (6 months) for mortality and safety follow-up.

Key Results

This trial was not stopped early for efficacy or harm. Recruitment ended when trial-agent expiry dates were reached (1241/1280 planned participants; 97% of target), after earlier protocol disruptions including discontinuation of the activated protein C component.

Outcome Hydrocortisone + Fludrocortisone Placebo Effect p value / 95% CI Notes
All-cause mortality at day 90 (primary) 264/614 (43.0%) 308/627 (49.1%) RR 0.88 95% CI 0.78 to 0.99; P=0.03 Primary comparison after continuation as two-group corticosteroid trial.
All-cause mortality at day 28 207/614 (33.7%) 244/627 (38.9%) RR 0.87 95% CI 0.75 to 1.01; P=0.06 Numerical reduction; not statistically significant at 5%.
All-cause mortality at ICU discharge 217/613 (35.4%) 257/627 (41.0%) RR 0.86 95% CI 0.75 to 0.99; P=0.04 Objective endpoint; denominators reflect available discharge status.
All-cause mortality at hospital discharge 239/613 (39.0%) 284/627 (45.3%) RR 0.86 95% CI 0.76 to 0.98; P=0.02 Consistent direction with primary outcome.
All-cause mortality at day 180 285/611 (46.6%) 328/625 (52.5%) RR 0.89 95% CI 0.79 to 0.99; P=0.04 Small attrition by day 180 (denominators < randomised N).
Decision to withhold/withdraw active treatment by day 90 119/604 (19.7%) 157/620 (25.3%) RR 0.78 95% CI 0.63 to 0.97; P=0.03 Potentially both a mediator and an outcome influenced by clinical trajectory.
Vasopressor-free days to day 28 17.8 ± 10.4 15.0 ± 11.1 Not reported P<0.001 Mean ± SD; indicates faster cardiovascular recovery.
Ventilator-free days to day 28 11.5 ± 11.0 10.9 ± 11.1 Not reported P=0.07 Mean ± SD; borderline.
Organ-failure–free days to day 28 14.8 ± 11.5 13.6 ± 11.9 Not reported P=0.003 Composite recovery metric over 28 days.
Hyperglycaemia (safety), by day 28 547/614 (89.1%) 515/626 (82.3%) RR 1.07 95% CI 1.03 to 1.12; P=0.002 Also more days with blood glucose ≥150 mg/dL: 4.3 ± 4.1 vs 3.4 ± 3.8; P<0.001.
Gastroduodenal bleeding (safety), by day 28 39/614 (6.4%) 45/626 (7.2%) RR 0.88 95% CI 0.58 to 1.34; P=0.56 No signal for excess GI bleeding with steroids in this cohort.
Superinfection (safety), by day 180 191/614 (31.1%) 178/626 (28.4%) RR 1.09 95% CI 0.92 to 1.30; P=0.30 Definition per trial report; interpret alongside antimicrobial stewardship context.
Neurologic sequelae (MDRS >1), by day 180 153/614 (24.9%) 130/626 (20.8%) RR 1.20 95% CI 0.98 to 1.47; P=0.08 Non-significant trend; ascertainment depends on survival and follow-up.
  • Hydrocortisone plus fludrocortisone reduced 90-day mortality (43.0% vs 49.1%; RR 0.88; 95% CI 0.78 to 0.99; P=0.03) and was directionally consistent across multiple mortality timepoints.
  • The regimen accelerated shock recovery (vasopressor-free days to day 28: 17.8 ± 10.4 vs 15.0 ± 11.1; P<0.001) and increased organ-failure–free days (14.8 ± 11.5 vs 13.6 ± 11.9; P=0.003).
  • Expected metabolic harms were observed (hyperglycaemia 89.1% vs 82.3%; RR 1.07; 95% CI 1.03 to 1.12; P=0.002), without clear increases in gastroduodenal bleeding or superinfection in reported endpoints.

Internal Validity

  • Randomisation and allocation: Central randomisation with permuted blocks (size 8) and stratification by ICU; allocation concealment maintained through identical placebo preparations and blinded dispensing.
  • Dropout/exclusions: Primary endpoint ascertainment was robust; longer-term follow-up had small attrition (day 180 denominators 611/614 and 625/627), unlikely to overturn effect estimates but relevant to disability/neurologic outcomes.
  • Performance/detection bias: Double blinding reduces co-intervention and ascertainment bias; primary endpoint (mortality) is objective; free-day outcomes depend on survival and discharge practices but were prespecified.
  • Protocol adherence and separation: Clear biological separation is evident via metabolic effects (days with blood glucose ≥150 mg/dL: 4.3 ± 4.1 vs 3.4 ± 3.8; P<0.001; hyperglycaemia 89.1% vs 82.3%; P=0.002) and physiological response (vasopressor-free days 17.8 ± 10.4 vs 15.0 ± 11.1; P<0.001).
  • Baseline comparability: Groups were clinically similar at baseline (e.g., mechanical ventilation at inclusion 92.3% vs 91.3%; renal-replacement therapy 27.0% vs 28.1%; norepinephrine-equivalent dose ~1.02 ± 1.61 vs 1.14 ± 1.66 μg/kg/min; SAPS II ~56 ± 19).
  • Heterogeneity: Multicentre ICU delivery introduces practice heterogeneity; blinding and objective endpoint selection mitigate bias, but discharge-related outcomes may vary by centre-level practice.
  • Timing: Randomisation within 24 hours of shock onset supports a biologically plausible window for modifying vasoplegia and inflammatory response; delayed initiation would be expected to attenuate benefit.
  • Dose: Hydrocortisone 200 mg/day is consistent with “low-dose” stress replacement regimens; addition of 50 micrograms/day fludrocortisone targets mineralocorticoid pathways potentially relevant to catecholamine responsiveness.
  • Outcome assessment: Mortality endpoints were prespecified; secondary outcomes were numerous and not adjusted for multiplicity, necessitating caution in interpreting borderline findings (e.g., ventilator-free days P=0.07).
  • Statistical rigour: Trial achieved 97% of target sample size, preserving near-planned power; primary analysis was ITT, but trial evolution (factorial discontinuation; suspensions) creates interpretive complexity without directly invalidating randomised comparisons within the final two-group phase.

Conclusion on Internal Validity: Overall, internal validity appears strong for the primary mortality endpoint given concealed randomisation, blinding, and objective outcome assessment, with moderate caution warranted for secondary/discharge-dependent outcomes and for interpretive complexity introduced by trial interruptions and design evolution.

External Validity

  • Population representativeness: Enrolled patients were severely ill (high vasopressor dose; ~92% ventilated), aligning with typical ICU septic shock populations in high-income settings, but may over-represent more severe shock compared with mixed ED/ward cohorts.
  • Healthcare system context: Conducted in French ICUs with structured sepsis management and access to vasopressors, ventilation, and renal replacement; transferability to resource-limited settings depends on availability of close glucose monitoring, vasopressor titration, and fludrocortisone formulations.
  • Intervention feasibility: Hydrocortisone bolus dosing is widely feasible; adding daily fludrocortisone requires enteral access and supply, and is not a universal component of sepsis protocols.
  • Applicability across subgroups: Subsequent analyses suggest potential effect heterogeneity by infection phenotype (e.g., community-acquired pneumonia-related septic shock)11, supporting cautious extrapolation beyond the dominant phenotypes enrolled.

Conclusion on External Validity: Generalisability is good to adult ICU septic shock in comparable high-resource systems, but is less certain in lower-acuity sepsis populations, settings without routine enteral drug delivery/monitoring capacity, and phenotypes where treatment effect may differ.

Strengths & Limitations

  • Strengths: Large, multicentre, double-blind randomised design; clinically meaningful primary endpoint (90-day mortality); clear and practical dosing regimen; comprehensive follow-up to 180 days; objective adjudication for key outcomes.
  • Limitations: Complex trial history (initial factorial design, discontinuation of one intervention component, enrolment interruptions); near-but-not-complete achievement of planned sample size; extensive secondary outcomes without multiplicity adjustment; discharge-related endpoints influenced by centre practice; limited direct evidence within the trial to isolate the incremental contribution of fludrocortisone versus hydrocortisone alone.

Interpretation & Why It Matters

  • Practice implication
    In severe septic shock with ongoing vasopressor dependence, a 7-day regimen of hydrocortisone (200 mg/day in divided doses) plus fludrocortisone (50 micrograms/day) provides evidence of a modest but clinically relevant mortality reduction and faster haemodynamic recovery.
  • Mechanistic inference
    The combination targets both glucocorticoid and mineralocorticoid pathways; compared with hydrocortisone-alone strategies, APROCCHSS supports the hypothesis that mineralocorticoid augmentation (or regimen differences such as bolus dosing) may influence outcomes.
  • Decision-making balance
    Benefits must be weighed against predictable metabolic harms (notably hyperglycaemia), with careful glucose monitoring and avoidance of high-dose/short-duration steroid regimens.

Controversies & Subsequent Evidence

  • Trial evolution as a validity stress-test: The protocol’s transition from a factorial design to a two-group corticosteroid trial, plus recruitment suspensions and near-target (not complete) sample size, complicate causal storytelling but do not negate the randomised comparison within the final corticosteroid phase1.
  • Mortality discordance across landmark trials: APROCCHSS demonstrated a statistically significant 90-day mortality reduction with hydrocortisone + fludrocortisone, while ADRENAL (hydrocortisone infusion without routine fludrocortisone) did not show a mortality benefit, although it improved shock resolution and ICU processes4. Earlier CORTICUS found no mortality benefit and revived concerns about patient selection, timing, and adrenal testing paradigms3.
  • Role of fludrocortisone (mineralocorticoid) remains an evidence gap: APROCCHSS cannot disentangle whether benefit stems from hydrocortisone dosing strategy, fludrocortisone addition, or both; no large head-to-head RCT has definitively compared hydrocortisone + fludrocortisone versus hydrocortisone alone using mortality as a primary endpoint.
  • Contemporary synthesis of the totality of evidence: A patient-level meta-analysis of IV hydrocortisone trials (excluding the routine mineralocorticoid component) found no mortality reduction, while showing a modest increase in vasopressor-free days (mean difference 1.24 days; 95% CI 0.74 to 1.73) and specific adverse effects such as hypernatraemia and muscle weakness9.
  • Comparative effectiveness across corticosteroid regimens: A Bayesian network meta-analysis comparing placebo/usual care, hydrocortisone alone, and hydrocortisone + fludrocortisone suggested lower mortality with hydrocortisone + fludrocortisone versus placebo (RR 0.85; 95% CrI 0.72 to 0.99) and a non-definitive difference versus hydrocortisone alone (RR 0.88; 95% CrI 0.74 to 1.03), highlighting ongoing uncertainty about incremental benefit over hydrocortisone monotherapy10.
  • Phenotype-dependent effects: In a targeted analysis of APROCCHSS participants with community-acquired pneumonia-related septic shock, hydrocortisone + fludrocortisone was associated with lower 90-day mortality (OR 0.60; 95% CI 0.43 to 0.83), raising the possibility of clinically meaningful heterogeneity by infection source/phenotype11.
  • Observational evidence (confounding risk): A retrospective target-trial emulation in routine practice compared hydrocortisone + fludrocortisone against hydrocortisone alone and reported lower in-hospital mortality with combination therapy, supporting plausibility but remaining vulnerable to residual confounding and treatment-selection bias12.
  • Editorial and correspondence: The paired 2018 NEJM steroid trials prompted editorial emphasis on careful interpretation of “borderline” mortality findings, co-interventions, and regimen differences5, alongside formal correspondence debating whether fludrocortisone explains divergent trial outcomes and whether hydrocortisone trial designs were sufficiently comparable6.
  • Guideline positioning: Surviving Sepsis Campaign 2021 conditionally recommends corticosteroids for septic shock with ongoing vasopressor requirement (commonly hydrocortisone 200 mg/day), without mandating mineralocorticoid co-therapy7. A 2024 focused update similarly issued a conditional recommendation for corticosteroids in septic shock and recommended against high-dose/short-duration regimens8.
  • Living evidence: The most recent Cochrane update concludes that corticosteroids probably reduce short- and longer-term mortality in sepsis overall, while uncertainties persist for optimal dosing strategy and adverse effects, which frames APROCCHSS as part of a broader, evolving evidence base rather than a standalone “final answer”13.

Summary

  • APROCCHSS tested a pragmatic, blinded 7-day regimen of hydrocortisone (50 mg IV q6h) plus fludrocortisone (50 micrograms daily) versus placebo in severe ICU septic shock.
  • The regimen reduced 90-day all-cause mortality (43.0% vs 49.1%; RR 0.88; 95% CI 0.78 to 0.99; P=0.03) with consistent direction across ICU, hospital, and 180-day mortality.
  • Physiological recovery improved: more vasopressor-free days (17.8 ± 10.4 vs 15.0 ± 11.1; P<0.001) and more organ-failure–free days (14.8 ± 11.5 vs 13.6 ± 11.9; P=0.003).
  • Expected harms were mainly metabolic (hyperglycaemia increased), while major bleeding and superinfection signals were not clearly increased in reported endpoints.
  • Interpretation is strengthened by blinding and objective endpoints but complicated by trial evolution and the absence of a definitive head-to-head mortality trial versus hydrocortisone alone.

Further Reading

Other Trials

Systematic Review & Meta Analysis

Observational Studies

Guidelines

Notes

  • APROCCHSS provides the strongest randomised evidence to date for hydrocortisone plus fludrocortisone versus placebo in severe ICU septic shock, but definitive isolation of fludrocortisone’s incremental effect over hydrocortisone alone still relies on indirect evidence and observational comparisons.

Overall Takeaway

APROCCHSS is a landmark septic shock trial because it demonstrated a statistically significant reduction in 90-day mortality with a practical 7-day regimen of hydrocortisone plus fludrocortisone, alongside faster haemodynamic recovery. Its findings strongly influenced how clinicians interpret steroid use in septic shock, while also catalysing debate about whether mineralocorticoid co-therapy (rather than hydrocortisone alone) is the key determinant of any mortality benefit and where treatment effects may vary by phenotype.

Overall Summary

  • In severe ICU septic shock, hydrocortisone (200 mg/day in divided doses) plus fludrocortisone (50 micrograms/day) for 7 days reduced 90-day mortality and increased vasopressor-free days.
  • Benefits are counterbalanced primarily by metabolic harms (hyperglycaemia), without a clear bleeding or superinfection signal in reported endpoints.
  • Key remaining uncertainty: incremental benefit of fludrocortisone over hydrocortisone alone.

Bibliography

  • 1.Annane D, Brun-Buisson C, Cariou A, et al.; APROCCHSS Investigators for the TRIGGERSEP Network. Design and conduct of the activated protein C and corticosteroids for human septic shock (APROCCHSS) trial. Ann Intensive Care. 2016;6(1):43. DOI
  • 2.Annane D, Sébille V, Charpentier C, et al. Effect of treatment with low doses of hydrocortisone and fludrocortisone on mortality in patients with septic shock. JAMA. 2002;288(7):862-871. DOI
  • 3.Sprung CL, Annane D, Keh D, et al. Hydrocortisone therapy for patients with septic shock. N Engl J Med. 2008;358(2):111-124. DOI
  • 4.Venkatesh B, Finfer S, Cohen J, et al.; ADRENAL Trial Investigators and the Australian–New Zealand Intensive Care Society Clinical Trials Group. Adjunctive glucocorticoid therapy in patients with septic shock. N Engl J Med. 2018;378(9):797-808. DOI
  • 5.Suffredini AF. A role for hydrocortisone therapy in septic shock? N Engl J Med. 2018;378(9):860-861. DOI
  • 6.Gunst J, Van den Berghe G. Glucocorticoids with or without fludrocortisone in septic shock. N Engl J Med. 2018;379(9):894. DOI
  • 7.Evans L, Rhodes A, Alhazzani W, et al. Surviving Sepsis Campaign: International guidelines for management of sepsis and septic shock 2021. Intensive Care Med. 2021;47(11):1181-1247. DOI
  • 8.Chaudhuri D, Nei AM, Rochwerg B, et al. 2024 Focused Update: Guidelines on Use of Corticosteroids in Sepsis, Acute Respiratory Distress Syndrome, and Community-Acquired Pneumonia. Crit Care Med. 2024;52(5):e219-e233. DOI
  • 9.Pirracchio R, Annane D, Waschka AK, et al. Patient-level meta-analysis of low-dose hydrocortisone in adults with septic shock. NEJM Evid. 2023;2(6):EVIDoa2300034. DOI
  • 10.Teja B, Bosch NA, Walkey AJ, et al. Comparative effectiveness and safety of fludrocortisone plus hydrocortisone, hydrocortisone alone, and placebo/usual care in adults with septic shock: a Bayesian network meta-analysis. Am J Respir Crit Care Med. 2024;209(11):1362-1370. DOI
  • 11.Heming N, Renault A, Megarbane B, et al. Hydrocortisone plus fludrocortisone for adults with community-acquired pneumonia-related septic shock: a subgroup analysis of the APROCCHSS phase 3 randomised clinical trial. Lancet Respir Med. 2024;12(5):366-374. DOI
  • 12.Bosch NA, Teja B, Law AC, et al. Comparative effectiveness of fludrocortisone and hydrocortisone vs hydrocortisone alone among patients with septic shock. JAMA Intern Med. 2023;183(5):451-459. DOI
  • 13.Annane D, Bellissant E, Bollaert P-E, et al. Corticosteroids for treating sepsis in children and adults. Cochrane Database Syst Rev. 2025;6:CD002243. DOI