Resilience after severe critical illness: a prospective, multicentre, observational study (RESIREA).

BACKGROUND: Critical-illness survivors may experience post-traumatic stress disorder (PTSD) and quality-of-life impairments. Resilience may protect against psychological trauma but has not been adequately studied after critical illness. We assessed resilience and its associations with PTSD and quality of life, and also identified factors associated with greater resilience. METHODS: This prospective, multicentre, study in patients recruited at 41 French ICUs was done in parallel with the NUTRIREA-3 trial in patients given mechanical ventilation and vasoactive amines for shock. Three months to one year after intensive-care-unit admission, survivors completed the Connor-Davidson Resilience Scale (CD-RISC-25), Impact of Event-Revised scale for PTSD symptoms (IES-R), SF-36 quality-of-life scale, Multidimensional Scale of Perceived Social Support (MSPSS), and Brief Illness Perception Questionnaire (B-IPQ). RESULTS: Of the 382 included patients, 203 (53.1%) had normal or high resilience (CD-RISC-25 ≥ 68). Of these resilient patients, 26 (12.8%) had moderate to severe PTSD symptoms (IES-R ≥ 24) vs. 45 (25.4%) patients with low resilience (p = 0.002). Resilient patients had higher SF-36 scores. Factors independently associated with higher CD-RISC-25 scores were higher MSPSS score indicating stronger social support (OR, 1.027; 95%CI 1.008-1.047; p = 0.005) and lower B-IPQ scores indicating a more threatening perception of the illness (OR, 0.973; 95%CI 0.950-0.996; p = 0.02). CONCLUSIONS: Resilient patients had a lower prevalence of PTSD symptoms and higher quality of life scores, compared to patients with low resilience. Higher scores for social support and illness perception were independently associated with greater resilience. Thus, our findings suggest that interventions to strengthen social support and improve illness perception may help to improve resilience. Such interventions should be evaluated in trials with PTSD mitigation and quality-of-life improvement as the target outcomes.

Clinical characteristics and outcomes of immunocompromised critically ill patients with cytomegalovirus end-organ disease: a multicenter retrospective cohort study.

BACKGROUND: Cytomegalovirus (CMV) infection in patients with cellular immune deficiencies is associated with significant morbidity and mortality. However, data on CMV end-organ disease (CMV-EOD) in critically ill, immunocompromised patients are scarce. Our objective here was to describe the clinical characteristics and outcomes of CMV-EOD in this population. METHODS: We conducted a multicenter, international, retrospective, observational study in adults who had CMV-EOD and were admitted to any of 18 intensive care units (ICUs) in France, Israel, and Spain in January 2010-December 2021. Patients with AIDS were excluded. We collected the clinical characteristics and outcomes of each patient. Survivors and non-survivors were compared, and multivariate analysis was performed to identify risk factors for hospital mortality. RESULTS: We studied 185 patients, including 80 (43.2%) with hematologic malignancies, 55 (29.7%) with solid organ transplantation, 31 (16.8%) on immunosuppressants, 16 (8.6%) with solid malignancies, and 3 (1.6%) with primary immunodeficiencies. The most common CMV-EOD was pneumonia (n = 115, [62.2%] including 55 [47.8%] with a respiratory co-pathogen), followed by CMV gastrointestinal disease (n = 64 [34.6%]). More than one organ was involved in 16 (8.8%) patients. Histopathological evidence was obtained for 10/115 (8.7%) patients with pneumonia and 43/64 (67.2%) with GI disease. Other opportunistic infections were diagnosed in 69 (37.3%) patients. Hospital mortality was 61.4% overall and was significantly higher in the group with hematologic malignancies (75% vs. 51%, P = 0.001). Factors independently associated with higher hospital mortality were hematologic malignancy with active graft-versus-host disease (OR 5.02; 95% CI 1.15-27.30), CMV pneumonia (OR 2.57; 95% CI 1.13-6.03), lymphocytes < 0.30 × 109/L at diagnosis of CMV-EOD (OR 2.40; 95% CI 1.05-5.69), worse SOFA score at ICU admission (OR 1.18; 95% CI 1.04-1.35), and older age (OR 1.04; 95% CI 1.01-1.07). CONCLUSIONS: Mortality was high in critically ill, immunocompromised patients with CMV-EOD and varied considerably with the cause of immunodeficiency and organ involved by CMV. Three of the four independent risk factors identified here are also known to be associated with higher mortality in the absence of CMV-EOD. CMV pneumonia was rarely proven by histopathology and was the most severe CMV-EOD.

The Impact of PEEP on Ventilation Distribution in ARDS.

BACKGROUND: The first aim of this study was to evaluate the capacity of electrical impedance tomography (EIT) to identify the effect of PEEP on regional ventilation distribution and the regional risk of collapse, overdistention, hypoventilation, and pendelluft in mechanically ventilated patients. The second aim was to evaluate the feasibility of EIT for estimating airway opening pressure (AOP). METHODS: The EIT signal was recorded both during baseline cyclic ventilation and slow insufflation for one breath for 9 subjects with moderate-to-severe ARDS. From these data, the AOP and volumes insufflated to lung regions with or without the risk of either collapse, overdistention, hypoventilation, or pendelluft were assessed at 3 PEEP levels (5, 10, and 15 cm H2O). PEEP levels were compared by Friedman analysis of variance and the AOP measured by EIT evaluated using an F-test and the Bland and Altman method. RESULTS: The volume for which there was no specific risk significantly decreased at the highest PEEP from 55 ± 31% tidal volume (VT) at PEEP 5 or 82 ± 18% VT at PEEP 10 to 10 ± 30% VT at PEEP 15 (P = .038 between PEEP 5 vs PEEP 15; P = .01 between PEEP 10 vs PEEP 15). The volume associated with overdistention significantly increased with increasing PEEP, whereas that associated with atelectrauma significantly decreased. Pendelluft significantly decreased with increasing PEEP: VT of 8.9 ± 18.6%, 3.6 ± 7.0%, and 3.2 ± 7.1% for PEEP 5, PEEP 10, and PEEP 15, respectively. The center of ventilation tended to increase in the dependent direction with higher PEEP. The AOPs assessed by EIT and from the pressure-volume curve were in good agreement (bias 0.48 cm H2O). CONCLUSIONS: Our results suggest that EIT could aid clinicians in making personalized and reasoned choices in setting the PEEP for subjects with ARDS.

Exploration of COVID-19 associated bradycardia using heart rate variability analysis in a case-control study of ARDS patients.

BACKGROUND: Bradycardia and dysautonomia observed during SARS-Cov2 infection suggests involvement of the autonomic nervous system (ANS). Limited data exists on ANS dysregulation and its association with outcomes in patients with acute respiratory distress syndrome (ARDS) related to COVID-19 (C-ARDS) or other etiologies (NC-ARDS). OBJECTIVES: We aimed to explore sympathovagal balance, assessed by heart rate variability (HRV), and its clinical prognostic value in C-ARDS compared with NC-ARDS. METHODS: A single-center, prospective case-control study was conducted. Consecutive patients meeting ARDS criteria between 2020 and 2022 were included. HRV was assessed using 1-hour electrographic tracing during a stable, daytime period. RESULTS: Twenty-four patients with C-ARDS and 19 with NC-ARDS were included. Age, sex and ARDS severity were similar between groups. The median heart rate was markedly lower in the C-ARDS group than in the NC-ARDS group (60 [53-72] versus 101 [91-112] bpm, p<.001). Most of HRV parameters were significantly increased in patients with C-ARDS. HRV correlated with heart rate only in patients with C-ARDS. A positive correlation was found between the low-to high-frequency ratio (LF/HF) and length of intensive care unit stay (r = 0.576, p<.001). CONCLUSION: This study confirmed that C-ARDS was associated with marked bradycardia and severe ANS impairment, suggesting a sympathovagal imbalance with vagal overtone. Poor outcomes appeared to be more related to sympathetic rather than parasympathetic hyperactivation.

Joint modeling of monocyte HLA-DR expression trajectories predicts 28-day mortality in severe SARS-CoV-2 patients.

The recent SarsCov2 pandemic has disrupted healthcare system notably impacting intensive care units (ICU). In severe cases, the immune system is dysregulated, associating signs of hyperinflammation and immunosuppression. In the present work, we investigated, using a joint modeling approach, whether the trajectories of cellular immunological parameters were associated with survival of COVID-19 ICU patients. This study is based on the REA-IMMUNO-COVID cohort including 538 COVID-19 patients admitted to ICU between March 2020 and May 2022. Measurements of monocyte HLA-DR expression (mHLA-DR), counts of neutrophils, of total lymphocytes, and of CD4+ and CD8+ subsets were performed five times during the first month after ICU admission. Univariate joint models combining survival at day 28 (D28), hospital discharge and longitudinal analysis of those biomarkers' kinetics with mixed-effects models were performed prior to the building of a multivariate joint model. We showed that a higher mHLA-DR value was associated with a lower risk of death. Predicted mHLA-DR nadir cutoff value that maximized the Youden index was 5414 Ab/C and led to an AUC = 0.70 confidence interval (95%CI) = [0.65; 0.75] regarding association with D28 mortality while dynamic predictions using mHLA-DR kinetics until D7, D12 and D20 showed AUCs of 0.82 [0.77; 0.87], 0.81 [0.75; 0.87] and 0.84 [0.75; 0.93]. Therefore, the final joint model provided adequate discrimination performances at D28 after collection of biomarker samples until D7, which improved as more samples were collected. After severe COVID-19, decreased mHLA-DR expression is associated with a greater risk of death at D28 independently of usual clinical confounders.

Cefiderocol in Difficult-to-Treat Nf-GNB in ICU Settings.

BACKGROUND: The efficacy and safety of cefiderocol in ICU patients with difficult-to-treat resistance (DTR) non-fermenting Gram-negative bacteria (Nf-GNB) are not as well-established. Consequently, we conducted a cohort study to compare Cefiderocol with the Best Available Therapy (BAT) in ICU patients. METHODS: We included adult patients from 9 different ICUs, including a burn ICU unit, from 2019 to 2023 treated with Cefiderocol for DTR Nf-GNB isolated from the blood or lungs. We matched each patient at a 1:2 ratio based on the same DTR Nf-GBN isolated pathogen, and when possible, within the same type of ICU (burn unit or not). The primary endpoint of the study was the clinical cure at 15 days, with secondary endpoints including clinical cure at 30 days, relapse, and in-ICU mortality. For each outcome, adjusted odds ratios were estimated using bidirectional stepwise regression in a final model, which included 13 preselected confounders. RESULTS: We included 27 patients with cefiderocol, matched with 54 patients receiving the BAT. Four patients were not exactly matched on the type of ICU unit. Characteristics were comparable between groups, mostly male with a Charlson Comorbidity Index of 3 [1-5], and 28% had immunosuppression. Cefiderocol patients were most likely to have higher number of antibiotic lines. The main DTR Nf-GNB identified was Pseudomonas aeruginosa (81.5%), followed by Acinetobater baumanii (14.8%) and Stenotrophomonas maltophilia (3.7%). Pneumonia was the identified infection in 21 (78.8%) patients in the Cefiderocol group and in 51 (94.4%) patients in the BAT group (p = 0.054). Clinical cure at 15 and 30-day and the in-ICU mortality was comparable between groups, however relapse was higher in the cefiderocol group (8-29.6% vs. 4-7.4%;aOR 10.06[1.96;51.53]) CONCLUSION: Cefiderocol did not show an improvement in clinical cure or mortality rates compared to BAT in the treatment of DTR Nf-GNB, but it was associated with a higher relapse rate.

Monitoring monocyte HLA-DR expression and CD4 + T lymphocyte count in dexamethasone-treated severe COVID-19 patients.

BACKGROUND: A 10-day dexamethasone regimen has emerged as the internationally adopted standard-of-care for severe COVID-19 patients. However, the immune response triggered by SARS-CoV-2 infection remains a complex and dynamic phenomenon, leading to various immune profiles and trajectories. The immune status of severe COVID-19 patients following complete dexamethasone treatment has yet to be thoroughly documented. RESULTS: To analyze monocyte HLA-DR expression (mHLA-DR) and CD4 + T lymphocyte count (CD4) in critically ill COVID-19 patients after a dexamethasone course and evaluate their association with 28-day ICU mortality, adult COVID-19 patients (n = 176) with an ICU length of stay of at least 10 days and under dexamethasone treatment were included. Associations between each biomarker value (or in combination) measured at day 10 after ICU admission and 28-day mortality in ICU were evaluated. At day 10, the majority of patients presented decreased values of both parameters. A significant association between low mHLA-DR and 28-day mortality was observed. This association remained significant in a multivariate analysis including age, comorbidities or pre-existing immunosuppression (adjusted Hazard ratio (aHR) = 2.86 [1.30-6.32], p = 0.009). Similar results were obtained with decreased CD4 + T cell count (aHR = 2.10 [1.09-4.04], p = 0.027). When combining these biomarkers, patients with both decreased mHLA-DR and low CD4 presented with an independent and significant elevated risk of 28-day mortality (i.e., 60%, aHR = 4.83 (1.72-13.57), p = 0.001). CONCLUSIONS: By using standardized immunomonitoring tools available in clinical practice, it is possible to identify a subgroup of patients at high risk of mortality at the end of a 10-day dexamethasone treatment. This emphasizes the significance of integrating immune monitoring into the surveillance of intensive care patients in order to guide further immumodulation approaches.

Comparison of mortality and outcomes of four respiratory viruses in the intensive care unit: a multicenter retrospective study.

This retrospective study aimed to compare the mortality and burden of respiratory syncytial virus (RSV group), SARS-CoV-2 (COVID-19 group), non-H1N1 (Seasonal influenza group) and H1N1 influenza (H1N1 group) in adult patients admitted to intensive care unit (ICU) with respiratory failure. A total of 807 patients were included. Mortality was compared between the four following groups: RSV, COVID-19, seasonal influenza, and H1N1 groups. Patients in the RSV group had significantly more comorbidities than the other patients. At admission, patients in the COVID-19 group were significantly less severe than the others according to the simplified acute physiology score-2 (SAPS-II) and sepsis-related organ failure assessment (SOFA) scores. Using competing risk regression, COVID-19 (sHR = 1.61; 95% CI 1.10; 2.36) and H1N1 (sHR = 1.87; 95% CI 1.20; 2.93) were associated with a statistically significant higher mortality while seasonal influenza was not (sHR = 0.93; 95% CI 0.65; 1.31), when compared to RSV. Despite occurring in more severe patients, RSV and seasonal influenza group appear to be associated with a more favorable outcome than COVID-19 and H1N1 groups.

Clinical description and outcome of overall varicella-zoster virus-related organ dysfunctions admitted in intensive care units: the VAZOREA cohort study.

BACKGROUND: Due to aging population and increasing part of immunocompromised patients, a raise in life-threatening organ damage related to VZV can be expected. Two retrospective studies were already conducted on VZV in ICU but focused on specific organ injury. Patients with high-risk of VZV disease still must be identified. The objective of this study was to report the clinical features and outcome of all life-threatening VZV manifestations requiring intensive care unit (ICU) admission. This retrospective cohort study was conducted in 26 French ICUs and included all adult patients with any life-threatening VZV-related event requiring ICU admission or occurring in ICU between 2010 and 2019. RESULTS: One-hundred nineteen patients were included with a median SOFA score of 6. One hundred eight patients (90.8%) were admitted in ICU for VZV disease, leaving 11 (9.2%) with VZV disease occurring in ICU. Sixty-one patients (51.3%) were immunocompromised. Encephalitis was the most prominent organ involvement (55.5%), followed by pneumonia (44.5%) and hepatitis (9.2%). Fifty-four patients (45.4%) received norepinephrine, 72 (60.5% of the total cohort) needed invasive mechanical ventilation, and 31 (26.3%) received renal-replacement therapy. In-hospital mortality was 36.1% and was significantly associated with three independent risk factors by multivariable logistic regression: immunosuppression, VZV disease occurring in ICU and alcohol abuse. Hierarchical clustering on principal components revealed five phenotypically distinct clusters of patients: VZV-related pneumonia, mild encephalitis, severe encephalitis in solid organ transplant recipients, encephalitis in other immunocompromised hosts and VZV disease occurring in ICU. In-hospital mortality was highly different across phenotypes, ranging from zero to 75% (p < 0.001). CONCLUSION: Overall, severe VZV manifestations are associated with high mortality in the ICU, which appears to be driven by immunosuppression status rather than any specific organ involvement. Deciphering the clinical phenotypes may help clinicians identify high-risk patients and assess prognosis.

Renal replacement therapy initiation strategies in comatose patients with severe acute kidney injury: a secondary analysis of a multicenter randomized controlled trial.

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.

Acute liver failure after out-of-hospital cardiac arrest: An observational study.

RATIONALE: Apart from hypoxic hepatitis (HH), the hepatic consequences of out-of-hospital cardiac arrest (OHCA) have been little studied. This cohort study aimed to investigate the characteristics of liver dysfunction resulting from OHCA and its association with outcomes. METHODS: Among the conventional static liver function tests used to define acute liver failure (ALF), we determined which one correlated more closely with the reference indocyanine green (ICG) clearance test in a series of OHCA patients from the CYRUS trial (NCT01595958). Subsequently, we assessed whether ALF, in addition to HH (i.e., acute liver injury), was an independent risk factor for death in a large cohort of OHCA patients admitted to two intensive care units between 2007 and 2017. RESULTS: ICG clearance, available for 22 patients, was impaired in 17 (77.3%) cases. Prothrombin time (PT) ratio was the only static liver function test that correlated significantly (r = -0.66, p < 0.01) with ICG clearance and was therefore used to define ALF, with the usual cutoff of < 50%. Of the 418 patients included in the analysis (sex ratio: 1.4; median age: 64 [53-75] years; non-shockable rhythm: 73%), 67 (16.0%) presented with ALF, and 61 (14.6%) had HH at admission. On day 28, 337 (80.6%) patients died. Following multivariate analysis, ALF at admission, OHCA occurring at home, absence of bystander, non-cardiac cause of OHCA, low-flow duration ≥ 20 min, and SOFA score excluding liver subscore at admission were independently associated with day 28 mortality. CONCLUSIONS: ALF occurred frequently after OHCA and, unlike HH, was independently associated with day 28 mortality.

Assessing respiratory epidemic potential in French hospitals through collection of close contact data (April-June 2020).

The transmission risk of SARS-CoV-2 within hospitals can exceed that in the general community because of more frequent close proximity interactions (CPIs). However, epidemic risk across wards is still poorly described. We measured CPIs directly using wearable sensors given to all present in a clinical ward over a 36-h period, across 15 wards in three hospitals in April-June 2020. Data were collected from 2114 participants and combined with a simple transmission model describing the arrival of a single index case to the ward to estimate the risk of an outbreak. Estimated epidemic risk ranged four-fold, from 0.12 secondary infections per day in an adult emergency to 0.49 per day in general paediatrics. The risk presented by an index case in a patient varied 20-fold across wards. Using simulation, we assessed the potential impact on outbreak risk of targeting the most connected individuals for prevention. We found that targeting those with the highest cumulative contact hours was most impactful (20% reduction for 5% of the population targeted), and on average resources were better spent targeting patients. This study reveals patterns of interactions between individuals in hospital during a pandemic and opens new routes for research into airborne nosocomial risk.

Resilience and Mental-Health Symptoms in ICU Healthcare Professionals Facing Repeated COVID-19 Waves.

Rationale: Psychological resilience (the ability to thrive in adversity) may protect against mental-health symptoms in healthcare professionals during coronavirus disease (COVID-19) waves. Objectives: To identify determinants of resilience in ICU staff members. Methods: In this cross-sectional survey in 21 French ICUs, staff members completed the 10-item Connor-Davidson Resilience Scale, Hospital Anxiety and Depression Scale, and Impact of Event Scale-Revised (for post-traumatic stress disorder [PTSD]). Factors independently associated with resilience were identified. Measurements and Main Results: The response rate was 73.1% (950 of 1,300). The median 10-item Connor-Davidson Resilience Scale score was 29 (interquartile range, 25-32). Symptoms of anxiety, depression, and PTSD were present in 61%, 39%, and 36% of staff members, respectively. Distress associated with the COVID-19 infodemic was correlated with symptoms of depression and PTSD. More resilient respondents less often had symptoms of anxiety, depression, and PTSD. Greater resilience was independently associated with male sex, having provided intensive care during the early waves, having managed more than 50 patients with COVID-19, and, compared with earlier waves, working longer hours, having greater motivation, and more often involving families in end-of-life decisions. Independent risk factors for lower resilience were having managed more than 10 patients who died of COVID-19, having felt frightened or isolated, and greater distress from the COVID-19 infodemic. Conclusions: This study identifies modifiable determinants of resilience among ICU staff members. Longitudinal studies are needed to determine whether prior resilience decreases the risk of mental ill health during subsequent challenges. Hospital and ICU managers, for whom preserving mental well-being among staff members is a key duty, should pay careful attention to resilience.

Multicenter Retrospective Study of Invasive Fusariosis in Intensive Care Units, France.

Invasive fusariosis can be life-threatening, especially in immunocompromised patients who require intensive care unit (ICU) admission. We conducted a multicenter retrospective study to describe clinical and biologic characteristics, patient outcomes, and factors associated with death and response to antifungal therapy. We identified 55 patients with invasive fusariosis from 16 ICUs in France during 2002----2020. The mortality rate was high (56%). Fusariosis-related pneumonia occurred in 76% of patients, often leading to acute respiratory failure. Factors associated with death included elevated sequential organ failure assessment score at ICU admission or history of allogeneic hematopoietic stem cell transplantation or hematologic malignancies. Neither voriconazole treatment nor disseminated fusariosis were strongly associated with response to therapy. Invasive fusariosis can lead to multiorgan failure and is associated with high mortality rates in ICUs. Clinicians should closely monitor ICU patients with a history of hematologic malignancies or stem cell transplantation because of higher risk for death.

Single-drug versus combination antimicrobial therapy in critically ill patients with hospital-acquired pneumonia and ventilator-associated pneumonia due to Gram-negative pathogens: a multicenter retrospective cohort study.

KEY MESSAGES: In this study including 391 critically ill patients with nosocomial pneumonia due to Gram-negative pathogens, combination therapy was not associated with a reduced hazard of death at Day 28 or a greater likelihood of clinical cure at Day 14. No over-risk of AKI was observed in patients receiving combination therapy. BACKGROUND: The benefits and harms of combination antimicrobial therapy remain controversial in critically ill patients with hospital-acquired pneumonia (HAP), ventilated HAP (vHAP) or ventilator-associated pneumonia (VAP) involving Gram-negative bacteria. METHODS: We included all patients in the prospective multicenter OutcomeRea database with a first HAP, vHAP or VAP due to a single Gram-negative bacterium and treated with initial adequate single-drug or combination therapy. The primary endpoint was Day-28 all-cause mortality. Secondary endpoints were clinical cure rate at Day 14 and a composite outcome of death or treatment-emergent acute kidney injury (AKI) at Day 7. The average effects of combination therapy on the study endpoints were investigated through inverse probability of treatment-weighted regression and multivariable regression models. Subgroups analyses were performed according to the resistance phenotype of the causative pathogens (multidrug-resistant or not), the pivotal (carbapenems or others) and companion (aminoglycosides/polymyxins or others) drug classes, the duration of combination therapy (< 3 or ≥ 3 days), the SOFA score value at pneumonia onset (< 7 or ≥ 7 points), and in patients with pneumonia due to non-fermenting Gram-negative bacteria, pneumonia-related bloodstream infection, or septic shock. RESULTS: Among the 391 included patients, 151 (38.6%) received single-drug therapy and 240 (61.4%) received combination therapy. VAP (overall, 67.3%), vHAP (16.4%) and HAP (16.4%) were equally distributed in the two groups. All-cause mortality rates at Day 28 (overall, 31.2%), clinical cure rate at Day 14 (43.7%) and the rate of death or AKI at Day 7 (41.2%) did not significantly differ between the groups. In inverse probability of treatment-weighted analyses, combination therapy was not independently associated with the likelihood of all-cause death at Day 28 (adjusted odd ratio [aOR], 1.14; 95% confidence interval [CI] 0.73-1.77; P = 0.56), clinical cure at Day 14 (aOR, 0.79; 95% CI 0.53-1.20; P = 0.27) or death or AKI at Day 7 (aOR, 1.07; 95% CI 0.71-1.63; P = 0.73). Multivariable regression models and subgroup analyses provided similar results. CONCLUSIONS: Initial combination therapy exerts no independent impact on Day-28 mortality, clinical cure rate at Day 14, and the hazard of death or AKI at Day 7 in critically ill patients with mono-bacterial HAP, vHAP or VAP due to Gram-negative bacteria.