ABSTRACT
BACKGROUND: Few studies investigated the outcome of patients admitted to intensive care unit (ICU) for gunshot wounds (GSW). The purpose of this study was to determine the 28-day mortality, and to analyze the impact of variables on the mortality of patients admitted to ICU with GSW in four French University Hospitals level-1 regional trauma centers. METHOD: All medical files of adult patients (above fifteen years old) admitted to four French University Hospitals level-1 regional trauma centers for GSW were retrospectively analyzed from January 1st 2015 to June 30th 2021. The primary aim was to determine 28-day death rate of patients admitted in ICU for GSW. The secondary aim was to describe biological parameters, injuries and management of patients admitted to our ICUs, and to identify the variables associated with the 28-day mortality rate. A multivariate analysis allowed determining independent mortality factors. A Kaplan-Meier analysis compared mortality according to head injury. RESULTS: Among 17,262 patients screened, 173 (1 %) were admitted for GSW and 162 were analyzed. The 28-day mortality rate was 24.7 %. 77.5 % of deaths occurred within the first 48 h after ICU admission, and 87.5 % of deaths within three days of ICU admission. The 28-day death rate of patients with head injury was significantly higher as compared to patients without head injury (p < 0.001). Out of forty deaths, twenty-three (57.5 %) were due to head injury, and nine (22.5 %) were due to bleeding. The mechanisms were assault (45.1 %), suicide (34.6 %), accident (4.9 %) and unidentified (15.4 %). In a multivariate analysis, variables associated with the 28-day death rate were age, pre-hospital Glasgow coma score, and Injury Severity Score. CONCLUSION: GSW represented 1 % of ICU admission. The 28-day mortality rate was 24.7 %. 77.5 % of deaths occurred within the first 48 h due to head injuries and bleeding. Head injuries were associated with significantly higher mortality rate.
ABSTRACT
INTRODUCTION: Speckle tracking technology quantifies lung sliding and detects lung sliding abolition in case of pneumothorax on selected ultrasound loops through the analysis of acoustic markers. OBJECTIVES: We aimed to test the ability of speckle tracking technology to quantify lung sliding using a pleural strain value (PS). METHODS: We performed a prospective study in 30 healthy volunteers in whom we assessed the pleural speckle tracking using ultrasound loops. Seven breathing conditions with and without non-invasive ventilation were tested. Two observers analyzed the ultrasound loops in four lung areas (anterior and posterior, left and right) and compared the obtained PS values. The first endpoint was to determine the feasibility of the PS measurement in different breathing conditions. The secondary endpoints were to assess the intra- and inter-observer's reliability of the measurement to compare PS values between anterior and posterior lung areas and to explore their correlations with the measured tidal volume. RESULTS: We analyzed 1624 ultrasound loops from 29 patients after one volunteer's exclusion. Feasibility of this method was rated at 90.8 [95%CI: 89.6 - 92.4]%. The intra-observer reliability measured through Intraclass Correlation Coefficients was 0.96 [95%CI: 0.91-0.98] and 0.93 [95%CI: 0.86-0.97] depending on the operator. The inter-observer reliability was 0.89 [95%CI: 0.78-0.95]. The PS values were significantly lower in the anterior lung areas compared with the posterior areas in all breathing conditions. A weak positive correlation was found in all the lung areas when a positive end expiratory pressure was applied with r = 0.26 [95%CI: 0.12;0.39]; p < 0.01. CONCLUSION: Speckle tracking lung sliding quantification with PS was applicable in most conditions with an excellent intra- and inter-observer reliability. More studies in patients under invasive mechanical ventilation are needed to explore the correlation between PS values of pleural sliding and tidal volumes. CLINICAL REGISTRATION: NCT05415605.
Subject(s)
Feasibility Studies , Healthy Volunteers , Lung , Ultrasonography , Humans , Male , Adult , Female , Lung/diagnostic imaging , Lung/physiology , Prospective Studies , Ultrasonography/methods , Reproducibility of Results , Pleura/diagnostic imaging , Middle Aged , Tidal Volume , Pneumothorax/diagnostic imaging , Observer Variation , Young Adult , RespirationABSTRACT
Intra-abdominal infections (IAIs) are an important cause of morbidity and mortality in hospital settings worldwide. The cornerstones of IAI management include rapid, accurate diagnostics; timely, adequate source control; appropriate, short-duration antimicrobial therapy administered according to the principles of pharmacokinetics/pharmacodynamics and antimicrobial stewardship; and hemodynamic and organ functional support with intravenous fluid and adjunctive vasopressor agents for critical illness (sepsis/organ dysfunction or septic shock after correction of hypovolemia). In patients with IAIs, a personalized approach is crucial to optimize outcomes and should be based on multiple aspects that require careful clinical assessment. The anatomic extent of infection, the presumed pathogens involved and risk factors for antimicrobial resistance, the origin and extent of the infection, the patient's clinical condition, and the host's immune status should be assessed continuously to optimize the management of patients with complicated IAIs.
Subject(s)
Intraabdominal Infections , Humans , Intraabdominal Infections/drug therapy , Risk Factors , Anti-Bacterial Agents/therapeutic useABSTRACT
Background: Ultrasound has demonstrated its interest in the analysis of diaphragm function in patients with respiratory failure. The criteria used to diagnose hemidiaphragm paralysis are not well defined. Methods: The aim of this observational retrospective study was to describe the ultrasound findings in 103 patients with diaphragm paralysis, previously diagnosed by conventional methods after various circumstances such as trauma or surgery. The ultrasound study included the recording of excursions of both diaphragmatic domes and the measurement of inspiratory thickening. Results: On paralyzed hemidiaphragm, thickening was less than 20% in all patients during deep inspiration. Thinning was recorded in 53% of cases. In some cases, the recording of the thickening could be difficult. The study of motion during voluntary sniffing reported a paradoxical excursion in all but one patient. During quiet breathing, an absence of movement or a paradoxical displacement was observed. During deep inspiration, a paradoxical motion at the beginning of inspiration followed by a reestablishment of movement in the cranio-caudal direction was seen in 82% of cases. In some patients, there was a lack of movement followed, after an average delay of 0.4 s, by a cranio-caudal excursion. Finally, in 4 patients no displacement was recorded. Evidence of hyperactivity (increased inspiratory thickening and excursion) of contralateral non-paralyzed hemidiaphragm was observed. Conclusion: To accurately detect hemidiaphragm paralysis, it would be interesting to combine the ultrasound study of diaphragm excursion and thickening. The different profiles reported by our study must be known to avoid misinterpretation.
ABSTRACT
BACKGROUND: Current guidelines recommend using antifungals for selected patients with health care-associated intra-abdominal infection (HC-IAI), but this recommendation is based on a weak evidence. This study aimed to assess the association between early empirical use of antifungals and outcomes in intensive care unit (ICU) adult patients requiring re-intervention after abdominal surgery. METHODS: A retrospective, multicentre cohort study with overlap propensity score weighting was conducted in three ICUs located in three medical institutions in France. Patients treated with early empirical antifungals for HC-IAI after abdominal surgery were compared with controls who did not receive such antifungals. The primary endpoint was the death rate at 90 days, and the secondary endpoints were the death rate at 1 year and composite criteria evaluated at 30 days following the HC-IAI diagnosis, including the need for re-intervention, inappropriate antimicrobial therapy and death, whichever occurred first. RESULTS: At 90 days, the death rate was significantly decreased in the patients treated with empirical antifungals compared with the control group (11.4% and 20.7%, respectively, p = 0.02). No differences were reported for the secondary outcomes. CONCLUSION: The use of early empirical antifungal therapy was associated with a decreased death rate at 90 days, with no effect on the death rate at 1 year, the death rate at 30 days, the rate of re-intervention, the need for drainage, and empirical antibiotic and antifungal therapy failure at 30 days.
ABSTRACT
PURPOSE OF REVIEW: The purpose of this review is to investigate the long-term outcomes of cancer patients who experience sepsis or septic shock. RECENT FINDINGS: Sepsis is a frequent cause of ICU admission in cancer patients, accounting for approximately 15% of such cases. Short-term mortality rates among these patients vary widely across studies, but they are consistently found to be slightly higher than those of noncancer patients. However, there is a lack of evidence regarding the long-term outcomes of cancer patients who have experienced sepsis or septic shock. The few available studies have reported relatively high mortality rates, reaching around 80% in a few cohort studies. Although several observational studies have noted a decrease in 1-year mortality rates over time, observational data also suggest that sepsis may increase the risk of cancer in the long run. SUMMARY: As cancer is becoming a chronic disease, there is an urgent need for studies on the quality of life of cancer patients who have experienced sepsis. The relationship between sepsis and cancer extends beyond its impact on the progression of cancer, as sepsis might also contribute to the development of cancer.
ABSTRACT
BACKGROUND: Hospital-acquired bloodstream infections are common in the intensive care unit (ICU) and have a high mortality rate. Patients with cirrhosis are especially susceptible to infections, yet there is a knowledge gap in the epidemiological distinctions in hospital-acquired bloodstream infections between cirrhotic and non-cirrhotic patients in the ICU. It has been suggested that cirrhotic patients, present a trend towards more gram-positive infections, and especially enterococcal infections. This study aims to describe epidemiological differences in hospital-acquired bloodstream infections between cirrhotic and non-cirrhotic patients hospitalized in the ICU regarding infection sources, microorganisms and mortality. METHODS: Using prospective Eurobact-2 international cohort study data, we compared hospital-acquired bloodstream infections sources and microorganisms in cirrhotic and non-cirrhotic patients. The association between Enterococcus faecium and cirrhosis was studied using a multivariable mixed logistic regression. The association between cirrhosis and mortality was assessed by a multivariable frailty Cox model. RESULTS: Among the 1059 hospital-acquired bloodstream infections patients included from 101 centers, 160 had cirrhosis. Hospital-acquired bloodstream infection source in cirrhotic patients was primarily abdominal (35.6%), while it was pulmonary (18.9%) for non-cirrhotic (p < 0.01). Gram-positive hospital-acquired bloodstream infections accounted for 42.3% in cirrhotic patients compared to 33.2% in non-cirrhotic patients (p = 0.02). Hospital-acquired bloodstream infections in cirrhotic patients were most frequently caused by Klebsiella spp (16.5%), coagulase-negative Staphylococci (13.7%) and E. faecium (11.5%). E. faecium bacteremia was more frequent in cirrhotic patients (11.5% versus 4.5%, p < 0.01). After adjusting for possible confounding factors, cirrhosis was associated with higher E. faecium hospital-acquired bloodstream infections risk (Odds ratio 2.5, 95% CI 1.3-4.5, p < 0.01). Cirrhotic patients had increased mortality compared to non-cirrhotic patients (Hazard Ratio 1.3, 95% CI 1.01-1.7, p = 0.045). CONCLUSIONS: Critically ill cirrhotic patients with hospital-acquired bloodstream infections exhibit distinct epidemiology, with more Gram-positive infections and particularly Enterococcus faecium.
ABSTRACT
Background: new-onset atrial fibrillation remains a common complication in critical care settings, often necessitating treatment when the correction of triggers is insufficient to restore hemodynamics. The treatment strategy includes electric cardioversion in cases of hemodynamic instability and either rhythm control or rate control in the absence of instability. Landiolol, an ultrashort beta-blocker, effectively controls heart rate with the potential to regulate rhythm. Objectives This review aims to compare the efficacy of landiolol in controlling heart rate and converting to sinus rhythm in the critical care setting. Methods: We conducted a comprehensive review of the published literature from 2000 to 2022 describing the use of landiolol to treat atrial fibrillation in critical care settings, excluding both cardiac surgery and medical cardiac care settings. The primary outcome assessed was sinus conversion following landiolol treatment. Results: Our analysis identified 17 publications detailing the use of landiolol for the treatment of 324 critical care patients. While the quality of the data was generally low, primarily comprising non-comparative studies, landiolol consistently demonstrated similar efficacy in controlling heart rate and facilitating conversion to sinus rhythm in both non-surgical (75.7%) and surgical (70.1%) settings. The incidence of hypotension associated with landiolol use was 13%. Conclusions: The use of landiolol in critical care patients with new-onset atrial fibrillation exhibited comparable efficacy and tolerance in both non-surgical and surgical settings. Despite these promising results, further validation through randomized controlled trials is necessary.
ABSTRACT
Norepinephrine is the first-line vasopressor used in acute cardio-circulatory failure. At the bedside, its dose is critical as it serves to determine the severity of patients, to compare the patients between different studies, and to start specific interventions. Recently, several investigators underlined differences in the expression of norepinephrine doses according to countries and manufacturers. For various reasons, all norepinephrine products are processed to a salt formulation thereby generating a stable and soluble conjugated acid in a slightly acidic solution. Depending on the salt used, the total weight will differ, but the weight of pure norepinephrine base is the same. This results in at-risk situations with large variations in terms of practices, evaluation and treatment. In this technical note, we summarized the evidence and provided a few suggestions to improve the practices at different levels.
Subject(s)
Norepinephrine , Vasoconstrictor Agents , Humans , Norepinephrine/administration & dosage , Norepinephrine/therapeutic use , Vasoconstrictor Agents/administration & dosage , Vasoconstrictor Agents/therapeutic use , FranceABSTRACT
BACKGROUND: There have been conflicting reports regarding the case-fatality outcomes associated with sepsis and septic shock in patients with severe mental illness (SMI). METHODS: We searched Medline®, Web of Science® and the Cochrane Library® databases (from inception to 4-July-2023) for papers reporting outcomes associated with sepsis and septic shock in adult with (cases) vs. without SMI (controls). The main study outcome was the unadjusted case-fatality rate at hospital discharge, or 30 days if unavailable. Secondary outcomes included the rates of adjusted case-fatality at hospital discharge. RESULTS: A total of six studies were included in the systematic review, of which four provided data for meta-analysis involving 2,124,072 patients. Compared to controls, patients with SMI were younger and more frequently women. Unadjusted analyses showed that SMI patients had a lower case-fatality rate associated with sepsis and septic shock than their non-SMI counterparts (OR 0.61, 95% CI [0.58-0.65], PI 95% CI [0.49-0.77], I2 = 91%). Meta-regression and subgroup analyses showed that the denominator of the study population (i.e. septic shock or sepsis) was associated with the outcome with an R2 of 59.7%. CONCLUSION: In conclusion, our study reveals a survival advantage of SMI patients over their non-SMI counterparts. Further research is needed to fully elucidate the mechanisms involved and to develop targeted interventions that can improve the prognosis of both SMI and non-SMI patients facing sepsis.
Subject(s)
Sepsis , Thrombocytopenia , Humans , Thrombocytopenia/etiology , Thrombocytopenia/therapy , Sepsis/complicationsABSTRACT
PURPOSE: The primary objective of this study was to evaluate the associations between centre/country-based factors and two important process and outcome indicators in patients with hospital-acquired bloodstream infections (HABSI). METHODS: We used data on HABSI from the prospective EUROBACT-2 study to evaluate the associations between centre/country factors on a process or an outcome indicator: adequacy of antimicrobial therapy within the first 24 h or 28-day mortality, respectively. Mixed logistical models with clustering by centre identified factors associated with both indicators. RESULTS: Two thousand two hundred nine patients from two hundred one intensive care units (ICUs) were included in forty-seven countries. Overall, 51% (n = 1128) of patients received an adequate antimicrobial therapy and the 28-day mortality was 38% (n = 839). The availability of therapeutic drug monitoring (TDM) for aminoglycosides everyday [odds ratio (OR) 1.48, 95% confidence interval (CI) 1.03-2.14] or within a few hours (OR 1.79, 95% CI 1.34-2.38), surveillance cultures for multidrug-resistant organism carriage performed weekly (OR 1.45, 95% CI 1.09-1.93), and increasing Human Development Index (HDI) values were associated with adequate antimicrobial therapy. The presence of intermediate care beds (OR 0.63, 95% CI 0.47-0.84), TDM for aminoglycoside available everyday (OR 0.66, 95% CI 0.44-1.00) or within a few hours (OR 0.51, 95% CI 0.37-0.70), 24/7 consultation of clinical pharmacists (OR 0.67, 95% CI 0.47-0.95), percentage of vancomycin-resistant enterococci (VRE) between 10% and 25% in the ICU (OR 1.67, 95% CI 1.00-2.80), and decreasing HDI values were associated with 28-day mortality. CONCLUSION: Centre/country factors should be targeted for future interventions to improve management strategies and outcome of HABSI in ICU patients.
Subject(s)
Critical Illness , Cross Infection , Intensive Care Units , Humans , Cross Infection/mortality , Cross Infection/drug therapy , Male , Female , Critical Illness/mortality , Intensive Care Units/statistics & numerical data , Intensive Care Units/organization & administration , Middle Aged , Prospective Studies , Aged , Anti-Bacterial Agents/therapeutic use , Bacteremia/mortality , Bacteremia/drug therapy , Europe/epidemiology , Drug Monitoring/methods , Drug Monitoring/statistics & numerical data , Outcome and Process Assessment, Health CareABSTRACT
BACKGROUND: Posttraumatic stress disorder (PTSD) after a stay in the intensive care unit (ICU) can affect one in five ICU survivors. At the beginning of the coronavirus disease 2019 (COVID-19) pandemic, admission to the ICU for COVID-19 was stressful due to the severity of this disease. This study assessed whether admission to the ICU for COVID-19 was associated with a higher prevalence of PTSD compared with other causes of ICU admission after adjustment for pre-ICU psychological factors. METHODS: This prospective observational comparative cohort study included 31 ICUs. Eligible patients were adult ICU survivors hospitalized during the first wave of COVID-19 pandemic in France, regardless of the reason for admission. The prevalence of presumptive diagnosis of PTSD at 6 months was assessed using the PTSD Checklist for DSM-5 (PCL-5). Sociodemographics, clinical data, history of childhood trauma (Childhood Trauma Questionnaire [CTQ]), and exposure to potentially traumatic events (Life Events Checklist for DSM-5 [LEC-5]) were assessed. RESULTS: Of the 778 ICU survivors included during the first wave of COVID-19 pandemic in France, 417 and 361 were assigned to the COVID-19 and non-COVID-19 cohorts, respectively. Fourteen (4.9%) and 11 (4.9%), respectively, presented with presumptive diagnosis of PTSD at 6 months (p = 0.976). After adjusting for age, sex, severity score at admission, use of invasive mechanical ventilation, ICU duration, CTQ and LEC-5, COVID-19 status was not associated with presumptive diagnosis of PTSD using the PCL-5. Only female sex was associated with presumptive diagnosis of PTSD. However, COVID-19 patients reported significantly more intrusion and avoidance symptoms than non-COVID patients (39% vs. 29%, p = 0.015 and 27% vs. 19%, p = 0.030), respectively. The median PCL-5 score was higher in the COVID-19 than non-COVID-19 cohort (9 [3, 20] vs. 4 [2, 16], p = 0.034). CONCLUSION: Admission to the ICU for COVID-19 was not associated with a higher prevalence of PTSD compared with admission for another cause during the first wave of the COVID-19 pandemic in France. However, intrusion and avoidance symptoms were more frequent in COVID-19 patients than in non-COVID-19 patients. TRIAL REGISTRATION: Clinicaltrials.gov Identifier NCT03991611, registered on June 19, 2019.
Subject(s)
COVID-19 , Psychological Tests , Self Report , Stress Disorders, Post-Traumatic , Adult , Female , Humans , Cohort Studies , COVID-19/epidemiology , COVID-19/complications , Intensive Care Units , Pandemics , Stress Disorders, Post-Traumatic/psychology , Survivors , MaleABSTRACT
PURPOSE: The effect of renal replacement therapy (RRT) in comatose patients with acute kidney injury (AKI) remains unclear. We compared two RRT initiation strategies on the probability of awakening in comatose patients with severe AKI. METHODS: We conducted a post hoc analysis of a trial comparing two delayed RRT initiation strategies in patients with severe AKI. Patients were monitored until they had oliguria for more than 72 h and/or blood urea nitrogen higher than 112 mg/dL and then randomized to a delayed strategy (RRT initiated after randomization) or a more-delayed one (RRT initiated if complication occurred or when blood urea nitrogen exceeded 140 mg/dL). We included only comatose patients (Richmond Agitation-Sedation scale [RASS] < - 3), irrespective of sedation, at randomization. A multi-state model was built, defining five mutually exclusive states: death, coma (RASS < - 3), incomplete awakening (RASS [- 3; - 2]), awakening (RASS [- 1; + 1] two consecutive days), and agitation (RASS > + 1). Primary outcome was the transition from coma to awakening during 28 days after randomization. RESULTS: A total of 168 comatose patients (90 delayed and 78 more-delayed) underwent randomization. The transition intensity from coma to awakening was lower in the more-delayed group (hazard ratio [HR] = 0.36 [0.17-0.78]; p = 0.010). Time spent awake was 10.11 days [8.11-12.15] and 7.63 days [5.57-9.64] in the delayed and the more-delayed groups, respectively. Two sensitivity analyses were performed based on sedation status and sedation practices across centers, yielding comparable results. CONCLUSION: In comatose patients with severe AKI, a more-delayed RRT initiation strategy resulted in a lower chance of transitioning from coma to awakening.
Subject(s)
Acute Kidney Injury , Coma , Humans , Acute Kidney Injury/etiology , Coma/etiology , Coma/therapy , Proportional Hazards Models , Renal Replacement Therapy/methods , Multicenter Studies as Topic , Randomized Controlled Trials as TopicABSTRACT
Patients with septic shock who experience refractory hypotension despite adequate fluid resuscitation and high-dose noradrenaline have high mortality rates. To improve outcomes, evidence-based guidelines recommend starting a second vasopressor, such as vasopressin, if noradrenaline doses exceed 0.5 µg/kg/min. Recently, promising results have been observed in treating refractory hypotension with angiotensin II, which has been shown to increase mean arterial pressure and has been associated with improved outcomes. This narrative review aims to provide an overview of the pathophysiology of the renin-angiotensin system and the role of endogenous angiotensin II in vasodilatory shock with a focus on how angiotensin II treatment impacts clinical outcomes and on identifying the population that may benefit most from its use.
ABSTRACT
BACKGROUND: Glucocorticoids probably improve outcomes in patients hospitalised for community acquired pneumonia (CAP). In this a priori planned exploratory subgroup analysis of the phase 3 randomised controlled Activated Protein C and Corticosteroids for Human Septic Shock (APROCCHSS) trial, we aimed to investigate responses to hydrocortisone plus fludrocortisone between CAP and non-CAP related septic shock. METHODS: APROCCHSS was a randomised controlled trial that investigated the effects of hydrocortisone plus fludrocortisone, drotrecogin-alfa (activated), or both on mortality in septic shock in a two-by-two factorial design; after drotrecogin-alfa was withdrawn on October 2011, from the market, the trial continued on two parallel groups. It was conducted in 34 centres in France. In this subgroup study, patients with CAP were a preselected subgroup for an exploratory secondary analysis of the APROCCHSS trial of hydrocortisone plus fludrocortisone in septic shock. Adults with septic shock were randomised 1:1 to receive, in a double-blind manner, a 7-day treatment with daily administration of intravenous hydrocortisone 50 mg bolus every 6h and a tablet of 50 µg of fludrocortisone via the nasogastric tube, or their placebos. The primary outcome was 90-day all-cause mortality. Secondary outcomes included all-cause mortality at intensive care unit (ICU) and hospital discharge, 28-day and 180-day mortality, the number of days alive and free of vasopressors, mechanical ventilation, or organ failure, and ICU and hospital free-days to 90-days. Analysis was done in the intention-to-treat population. The trial was registered at ClinicalTrials.gov (NCT00625209). FINDINGS: Of 1241 patients included in the APROCCHSS trial, CAP could not be ruled in or out in 31 patients, 562 had a diagnosis of CAP (279 in the placebo group and 283 in the corticosteroid group), and 648 patients did not have CAP (329 in the placebo group and 319 in the corticosteroid group). In patients with CAP, there were 109 (39%) deaths of 283 patients at day 90 with hydrocortisone plus fludrocortisone and 143 (51%) of 279 patients receiving placebo (odds ratio [OR] 0·60, 95% CI 0·43-0·83). In patients without CAP, there were 148 (46%) deaths of 319 patients at day 90 in the hydrocortisone and fludrocortisone group and 157 (48%) of 329 patients in the placebo group (OR 0·95, 95% CI 0·70-1·29). There was significant heterogeneity in corticosteroid effects on 90-day mortality across subgroups with CAP and without CAP (p=0·046 for both multiplicative and additive interaction tests; moderate credibility). Of 1241 patients included in the APROCCHSS trial, 648 (52%) had ARDS (328 in the placebo group and 320 in the corticosteroid group). There were 155 (48%) deaths of 320 patients at day 90 in the corticosteroid group and 186 (57%) of 328 patients in the placebo group. The OR for death at day 90 was 0·72 (95% CI 0·53-0·98) in patients with ARDS and 0·85 (0·61-1·20) in patients without ARDS (p=0·45 for multiplicative interaction and p=0·42 for additive interaction). The OR for observing at least one serious adverse event (corticosteroid group vs placebo) within 180 days post randomisation was 0·64 (95% CI 0·46-0·89) in the CAP subgroup and 1·02 (0·75-1·39) in the non-CAP subgroup (p=0·044 for multiplicative interaction and p=0·042 for additive interaction). INTERPRETATION: In a pre-specified subgroup analysis of the APROCCHSS trial of patients with CAP and septic shock, hydrocortisone plus fludrocortisone reduced mortality as compared with placebo. Although a large proportion of patients with CAP also met criteria for ARDS, the subgroup analysis was underpowered to fully discriminate between ARDS and CAP modifying effects on mortality reduction with corticosteroids. There was no evidence of a significant treatment effect of corticosteroids in the non-CAP subgroup. FUNDING: Programme Hospitalier de Recherche Clinique of the French Ministry of Health, by Programme d'Investissements d'Avenir, France 2030, and IAHU-ANR-0004.