The use of enrichment to reduce statistically indeterminate or negative trials in critical care

Efficacy and safety of corticosteroids for the treatment of community-acquired pneumonia: A systematic review and meta-analysis of randomized controlled trials

BACKGROUND

The role of corticosteroids in the treatment of community-acquired pneumonia (CAP) remains uncertain. We conducted an updated meta-analysis to investigate the effectiveness and potential effect modifiers of adjunctive corticosteroids in patients with CAP.

METHODS

The protocol of this meta-analysis was registered with PROSPERO (CRD42022354920). We searched MEDLINE, Embase, the Cochrane Library and trial registers from inception till March 2023 to identify randomized controlled trials (RCTs) investigating corticosteroids in adult patients with CAP. Our primary outcome was the risk of all-cause mortality within 30 days after randomization (if not reported at day 30, we extracted the outcome closest to 30 days). Risk ratios (RR) and mean differences (MDs) were pooled under a random-effects model.

RESULTS

Fifteen RCTs (n = 3252 patients) were included in this review. Corticosteroids reduced the risk of all-cause mortality in CAP patients (RR: 0.69, 95% CI: 0.53-0.89; high certainty). This significant result was restricted to hydrocortisone therapy and patients with severe CAP. Additionally, younger patients demonstrated a greater reduction in mortality. Corticosteroids reduced the incidence of shock and the need for mechanical ventilation (MV), and decreased the length of hospital and ICU stay (moderate certainty).

CONCLUSIONS

Corticosteroids reduce the risk of all-cause mortality, especially in younger patients receiving hydrocortisone, and probably decrease the need for MV, the incidence of shock, and the length of hospital and ICU stay in patients with CAP. Our findings indicate that patients with CAP, especially severe CAP, will benefit from adjunctive corticosteroid therapy.

Extending the 'host response' paradigm from sepsis to cardiogenic shock: evidence, limitations and opportunities

Recent clinical and research efforts in cardiogenic shock (CS) have largely focussed on the restoration of the low cardiac output state that is the conditio sine qua non of the clinical syndrome. This approach has failed to translate into improved outcomes, and mortality has remained static at 30-50%. There is an unmet need to better delineate the pathobiology of CS to understand the observed heterogeneity of presentation and treatment effect and to identify novel therapeutic targets. Despite data in other critical illness syndromes, specifically sepsis, the role of dysregulated inflammation and immunity is hitherto poorly described in CS. High-dimensional molecular profiling, particularly through leukocyte transcriptomics, may afford opportunity to better characterise subgroups of patients with shared mechanisms of immune dysregulation. In this state-of-the-art review, we outline the rationale for considering molecular subtypes of CS. We describe how high-dimensional molecular technologies can be used to identify these subtypes, and whether they share biological features with sepsis and other critical illness states. Finally, we propose how the identification of molecular subtypes of patients may enrich future clinical trial design and identification of novel therapies for CS.

Determinants of Effect of Extracorporeal CO2 Removal in Hypoxemic Respiratory Failure

BACKGROUND: Dead space and respiratory system elastance (Ers) may influence the clinical benefit of a ventilation strategy combining very low tidal volume (VT) with extracorporeal carbon dioxide removal (ECCO2R) in patients with acute hypoxemic respiratory failure. We sought to evaluate whether the effect of ECCO2R on mortality varies according to ventilatory ratio (VR; a composite variable reflective of dead space and shunt) and Ers. METHODS: Secondary analysis of a trial of a strategy combining very low VT and low-flow ECCO2R planned before the availability of trial results. Bayesian logistic regression was used to estimate the posterior probability of effect moderation by VR, Ers, and severity of hypoxemia (ratio of arterial partial pressure of oxygen to fraction of inspired oxygen [PaO2:FiO2]) on 90-day mortality. Credibility of effect moderation was appraised according to the Instrument for Assessing the Credibility of Effect Modification Analyses criteria. RESULTS: A total of 405 patients were available for analysis. The effect of the intervention on mortality varied substantially with VR (posterior probability of interaction, 94%; high credibility). Benefit was more probable than harm in patients with VR 3 or higher. In patients with VR less than 3, the probability of increased mortality with intervention was high (>90%). The effect of the intervention also varied with PaO2:FiO2 (posterior probability of interaction, >99%; low credibility). Benefit was more probable than harm in patients with PaO2:FiO2 110 mm Hg or higher. The effect of the intervention did not vary substantially with Ers (posterior probability of interaction, 68%; low credibility). CONCLUSIONS: VR has a highly credible influence on the effect of a strategy combining very low VT and low-flow ECCO2R on mortality. This intervention may reduce mortality in patients with high VR. (Funded by an Early Career Investigator Award from the Canadian Institutes of Health Research to Dr. Goligher.)

The Impact of Sample Size Misestimations on the Interpretation of ARDS Trials: Systematic Review and Meta-analysis

BACKGROUND

Indeterminate randomized controlled trials (RCTs) in ARDS may arise from sample size misspecification, leading to abandonment of efficacious therapies.

RESEARCH QUESTIONS

If evidence exists for sample size misspecification in ARDS RCTs, has this led to rejection of potentially beneficial therapies? Does evidence exist for prognostic enrichment in RCTs using mortality as a primary outcome?

STUDY DESIGN AND METHODS

We identified 150 ARDS RCTs commencing recruitment after the 1994 American European Consensus Conference ARDS definition and published before October 31, 2020. We examined predicted-observed sample size, predicted-observed control event rate (CER), predicted-observed average treatment effect (ATE), and the relationship between observed CER and observed ATE for RCTs with mortality and nonmortality primary outcome measures. To quantify the strength of evidence, we used Bayesian-averaged meta-analysis, trial sequential analysis, and Bayes factors.

RESULTS

Only 84 of 150 RCTs (56.0%) reported sample size estimations. In RCTs with mortality as the primary outcome, CER was overestimated in 16 of 28 RCTs (57.1%). To achieve predicted ATE, interventions needed to prevent 40.8% of all deaths, compared with the original prediction of 29.3%. Absolute reduction in mortality ≥ 10% was observed in 5 of 28 RCTs (17.9%), but predicted in 21 of 28 RCTs (75%). For RCTs with mortality as the primary outcome, no association was found between observed CER and observed ATE (pooled OR: β = -0.04; 95% credible interval, -0.18 to 0.09). We identified three interventions that are not currently standard of care with a Bayesian-averaged effect size of > 0.20 and moderate strength of existing evidence: corticosteroids, airway pressure release ventilation, and noninvasive ventilation.

INTERPRETATION

Reporting of sample size estimations was inconsistent in ARDS RCTs, and misspecification of CER and ATE was common. Prognostic enrichment strategies in ARDS RCTs based on all-cause mortality are unlikely to be successful. Bayesian methods can be used to prioritize interventions for future effectiveness RCTs.

Association of specialized enteral nutrition with glycemic control and clinical outcomes in critically ill patients: A meta-analysis of randomized controlled trials

OBJECTIVE

To evaluate the association of glycemic-control formulae (GCF) with measurements of glycemic control and clinical outcomes compared to standard enteral formulae (SF) in critically ill patients.

DATA SOURCES

MEDLINE, EMBASE, Scopus and the Cochrane Central Register of Controlled Trials were searched from inception up to January, 2021.

STUDY SELECTION

RCTs that assessed the effects of GCF relative to SF in adult critically ill patients.

DATA EXTRACTION

Measurements of glycemic control were the primary outcomes. Secondary outcomes included insulin requirements, mechanical ventilation (MV), length of intensive care unit (ICU) stay and mortality. Two authors independently extracted data and assessed risk of bias using the Cochrane's RoB 2 tool and the GRADE approach was used to assess the quality of evidence.

DATA SYNTHESIS

Ten studies (12 reports, 685 patients) were included. The use of GCFs was associated with lower blood glucose (WMD, -16.06 mg/dL; 95% CI -23.48 to -8.63; I² = 47%) and lower daily administered insulin (WMD, -7.20 IU; 95% CI -13.92 to -0.48; I² = 53%). Glycemic variability, measured by the coefficient of variation, was also associated with the use of GCFs (WMD, -6.84%; 95% CI, -13.57 to -0.11; I² = 95%). In contrast, analyses for length of ICU stay (WMD, -0.12, 95% CI -1.77 to 1.52; I² = 0%), duration of MV (WMD, -0.34 days; 95% CI, -1.72 to 1.04; I² = 0%) and mortality (RR, 1.13; 95% CI 0.82 to 1.56; I² = 0%) were not statistically significant. Quality of evidence ranged from low to very low, and only one study was judged as at low risk of bias.

CONCLUSIONS

In this meta-analysis, GCFs were significantly associated with lower insulin requirements and improved glycemic control. Although results for clinical outcomes were not statistically significant, there is insufficient evidence to confirm or exclude important differences due to serious imprecision in the effect estimates and overall low quality of evidence. The effects of GCFs on clinical outcomes require confirmation in larger randomized trials.

Adjunctive treatments for the management of septic shock - a narrative review of the current evidence

Septic shock is a leading cause of death and morbidity worldwide. The cornerstones of management include prompt identification of sepsis, early initiation of antibiotic therapy, adequate fluid resuscitation and organ support. Over the past two decades, there have been considerable improvements in our understanding of the pathophysiology of sepsis and the host response, including regulation of inflammation, endothelial disruption and impaired immunity. This has offered opportunities for innovative adjunctive treatments such as vitamin C, corticosteroids and beta-blockers. Some of these approaches have shown promising results in early phase trials in humans, while others, such as corticosteroids, have been tested in large, international, multicentre randomised controlled trials. Contemporary guidelines make a weak recommendation for the use of corticosteroids to reduce mortality in sepsis and septic shock. Vitamin C, despite showing initial promise in observational studies, has so far not been shown to be clinically effective in randomised trials. Beta-blocker therapy may have beneficial cardiac and non-cardiac effects in septic shock, but there is currently insufficient evidence to recommend their use for this condition. The results of ongoing randomised trials are awaited. Crucial to reducing heterogeneity in the trials of new sepsis treatments will be the concept of enrichment, which refers to the purposive selection of patients with clinical and biological characteristics that are likely to be responsive to the intervention being tested.

Population enrichment for critical care trials: phenotypes and differential outcomes

PURPOSE OF REVIEW

Sepsis and acute respiratory distress syndrome (ARDS) are two heterogenous acute illnesses where numerous RCTs have indeterminate results. We present a narrative review on the recent developments in enriching patient populations for future sepsis and ARDS trials.

RECENT FINDINGS

Many researchers are actively pursuing enrichment strategies to reduce heterogeneity to increase the sensitivity of future trials. Enrichment refers to the use of measurable patient characteristics, known before randomisation, to refine trial populations. Biomarkers could increase the diagnostic certainty of sepsis, whereas chest radiology training to enhance reliability of interpretation and stabilisation period of mechanical ventilation have been considered to increase the diagnostic certainty of ARDS. Clinical and biomarker data analyses identifies four to six sepsis clinical phenotypes and two ARDS clinical phenotypes. Similarly, leukocyte gene expression data identifies two to four sepsis molecular phenotypes. Use of a test-dose identifies ARDS subpopulations who are likely to benefit from higher PEEP. Early-phase trials report how a biomarker that is altered by the intervention, such as lymphocyte count for recombinant interleukin-7 therapy and higher check point inhibitor expression for anti-check point treatments in sepsis, could identify a higher treatment effect population for future trials.

SUMMARY

Enrichment reduces heterogeneity and will enhance the sensitivity of future trials. However, enrichment, even when it identifies more homogenous populations, may not be efficient to deploy in trials or clinical practice.

Plasma sTNFR1 and IL8 for prognostic enrichment in sepsis trials: a prospective cohort study

Background

Enrichment strategies improve therapeutic targeting and trial efficiency, but enrichment factors for sepsis trials are lacking. We determined whether concentrations of soluble tumor necrosis factor receptor-1 (sTNFR1), interleukin-8 (IL8), and angiopoietin-2 (Ang2) could identify sepsis patients at higher mortality risk and serve as prognostic enrichment factors.

Methods

In a multicenter prospective cohort study of 400 critically ill septic patients, we derived and validated thresholds for each marker and expressed prognostic enrichment using risk differences (RD) of 30-day mortality as predictive values. We then used decision curve analysis to simulate the prognostic enrichment of each marker and compare different prognostic enrichment strategies.

Measurements and main results

An admission sTNFR1 concentration > 8861 pg/ml identified patients with increased mortality in both the derivation (RD 21.6%) and validation (RD 17.8%) populations. Among immunocompetent patients, an IL8 concentration > 94 pg/ml identified patients with increased mortality in both the derivation (RD 17.7%) and validation (RD 27.0%) populations. An Ang2 level > 9761 pg/ml identified patients at 21.3% and 12.3% increased risk of mortality in the derivation and validation populations, respectively. Using sTNFR1 or IL8 to select high-risk patients improved clinical trial power and efficiency compared to selecting patients with septic shock. Ang2 did not outperform septic shock as an enrichment factor.

Conclusions

Thresholds for sTNFR1 and IL8 consistently identified sepsis patients with higher mortality risk and may have utility for prognostic enrichment in sepsis trials.

Host-Response Subphenotypes Offer Prognostic Enrichment in Patients With or at Risk for Acute Respiratory Distress Syndrome

OBJECTIVES

Classification of patients with acute respiratory distress syndrome into hyper- and hypoinflammatory subphenotypes using plasma biomarkers may facilitate more effective targeted therapy. We examined whether established subphenotypes are present not only in patients with acute respiratory distress syndrome but also in patients at risk for acute respiratory distress syndrome (ARFA) and then assessed the prognostic information of baseline subphenotyping on the evolution of host-response biomarkers and clinical outcomes.

DESIGN

Prospective, observational cohort study.

SETTING

Medical ICU at a tertiary academic medical center.

PATIENTS

Mechanically ventilated patients with acute respiratory distress syndrome or ARFA.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

We performed longitudinal measurements of 10 plasma biomarkers of host injury and inflammation. We applied unsupervised latent class analysis methods utilizing baseline clinical and biomarker variables and demonstrated that two-class models (hyper- vs hypoinflammatory subphenotypes) offered improved fit compared with one-class models in both patients with acute respiratory distress syndrome and ARFA. Baseline assignment to the hyperinflammatory subphenotype (39/104 [38%] acute respiratory distress syndrome and 30/108 [28%] ARFA patients) was associated with higher severity of illness by Sequential Organ Failure Assessment scores and incidence of acute kidney injury in patients with acute respiratory distress syndrome, as well as higher 30-day mortality and longer duration of mechanical ventilation in ARFA patients (p < 0.0001). Hyperinflammatory patients exhibited persistent elevation of biomarkers of innate immunity for up to 2 weeks postintubation.

CONCLUSIONS

Our results suggest that two distinct subphenotypes are present not only in patients with established acute respiratory distress syndrome but also in patients at risk for its development. Hyperinflammatory classification at baseline is associated with higher severity of illness, worse clinical outcomes, and trajectories of persistently elevated biomarkers of host injury and inflammation during acute critical illness compared with hypoinflammatory patients. Our findings provide strong rationale for examining treatment effect modifications by subphenotypes in randomized clinical trials to inform precision therapeutic approaches in critical care.

Heterogeneity of treatment effect by baseline risk of mortality in critically ill patients: re-analysis of three recent sepsis and ARDS randomised controlled trials

Background

Randomised controlled trials (RCTs) enrolling patients with sepsis or acute respiratory distress syndrome (ARDS) generate heterogeneous trial populations. Non-random variation in the treatment effect of an intervention due to differences in the baseline risk of death between patients in a population represents one form of heterogeneity of treatment effect (HTE). We assessed whether HTE in two sepsis and one ARDS RCTs could explain indeterminate trial results and inform future trial design.

Methods

We assessed HTE for vasopressin, hydrocortisone and levosimendan in sepsis and simvastatin in ARDS patients, on 28-day mortality, using the total Acute Physiology And Chronic Health Evaluation II (APACHE II) score as the baseline risk measurement, comparing above (high) and below (low) the median score. Secondary risk measures were the acute physiology component of APACHE II and predicted risk of mortality using the APACHE II score. HTE was quantified both in additive (difference in risk difference (RD)) and multiplicative (ratio of relative risks (RR)) scales using estimated treatment differences from a logistic regression model with treatment risk as the interaction term.

Results

The ratio of the odds of death in the highest APACHE II quartile was 4.9 to 7.4 times compared to the lowest quartile, across the three trials. We did not observe HTE for vasopressin, hydrocortisone and levosimendan in the two sepsis trials. In the HARP-2 trial, simvastatin reduced mortality in the low APACHE II group and increased mortality in the high APACHE II group (difference in RD = 0.34 (0.12, 0.55) (p = 0.02); ratio of RR 3.57 (1.77, 7.17) (p < 0.001). The HTE patterns were inconsistent across the secondary risk measures. The sensitivity analyses of HTE effects for vasopressin, hydrocortisone and levosimendan were consistent with the main analyses and attenuated for simvastatin.

Conclusions

We assessed HTE in three recent ICU RCTs, using multivariable baseline risk of death models. There was considerable within-trial variation in the baseline risk of death. We observed potential HTE for simvastatin in ARDS, but no evidence of HTE for vasopressin, hydrocortisone or levosimendan in the two sepsis trials. Our findings could be explained either by true lack of HTE (no benefit of vasopressin, hydrocortisone or levosimendan vs comparator for any patient subgroups) or by lack of power to detect HTE. Our results require validation using similar trial databases.

Electronic supplementary material

The online version of this article (10.1186/s13054-019-2446-1) contains supplementary material, which is available to authorized users.

Early PREdiction of sepsis using leukocyte surface biomarkers: the ExPRES-sepsis cohort study

PURPOSE

Reliable biomarkers for predicting subsequent sepsis among patients with suspected acute infection are lacking. In patients presenting to emergency departments (EDs) with suspected acute infection, we aimed to evaluate the reliability and discriminant ability of 47 leukocyte biomarkers as predictors of sepsis (Sequential Organ Failure Assessment score ≥ 2 at 24 h and/or 72 h following ED presentation).

METHODS

In a multi-centre cohort study in four EDs and intensive care units (ICUs), we standardised flow-cytometric leukocyte biomarker measurement and compared patients with suspected acute infection (cohort-1) with two comparator cohorts: ICU patients with established sepsis (cohort-2), and ED patients without infection or systemic inflammation but requiring hospitalization (cohort-3).

RESULTS

Between January 2014 and February 2016, we recruited 272, 59 and 75 patients to cohorts 1, 2, and 3, respectively. Of 47 leukocyte biomarkers, 14 were non-reliable, and 17 did not discriminate between the three cohorts. Discriminant analyses for predicting sepsis within cohort-1 were undertaken for eight neutrophil (cluster of differentiation antigens (CD) CD15; CD24; CD35; CD64; CD312; CD11b; CD274; CD279), seven monocyte (CD35; CD64; CD312; CD11b; HLA-DR; CD274; CD279) and a CD8 T-lymphocyte biomarker (CD279). Individually, only higher neutrophil CD279 [OR 1.78 (95% CI 1.23-2.57); P = 0.002], higher monocyte CD279 [1.32 (1.03-1.70); P = 0.03], and lower monocyte HLA-DR [0.73 (0.55-0.97); P = 0.03] expression were associated with subsequent sepsis. With logistic regression the optimum biomarker combination was increased neutrophil CD24 and neutrophil CD279, and reduced monocyte HLA-DR expression, but no combination had clinically relevant predictive validity.

CONCLUSIONS

From a large panel of leukocyte biomarkers, immunosuppression biomarkers were associated with subsequent sepsis in ED patients with suspected acute infection.

CLINICAL TRIAL REGISTRATION

NCT02188992.

Lymphocyte subset expression and serum concentrations of PD-1/PD-L1 in sepsis - pilot study

Background

Sepsis remains a major cause of mortality in critical care, for which specific treatments are lacking. The dysregulated response to infection seen in sepsis includes features of lymphocyte dysfunction and exhaustion, suggesting that immune-stimulatory therapy may improve outcomes in certain patient groups. Monoclonal antibodies targeting checkpoint molecules, such as programmed-death 1 protein (PD-1) and its ligand PD-L1, have shown success in stimulating the immune response in patients with cancer, and are being considered for future sepsis trials. The aims of this pilot study were to compare lymphocyte subset expression of PD-1 and its ligands between patients with sepsis and controls; to characterize serum levels of PD-1 and PD-L1 in patients with sepsis and controls, and determine if serum concentrations correlated with cell surface expression.

Methods

Expression levels of PD-1, PD-L1 and PD-L2 on four lymphocyte subsets (CD27 + CD19+ B cells, CD27-CD19+ B cells, CD27 + CD4+ T cells and CD27-CD4+ T cells) were compared between 22 patients with sepsis (including 11 survivors and 11 non-survivors) and 11 healthy controls using flow cytometry. Levels of soluble PD-1 and PD-L1 were also compared using commercially available ELISA kits.

Results

Expression of PD-1 and PD-L1 was higher on all lymphocyte subsets in patients with sepsis compared to controls (p < 0.05). PD-L2 expression on CD27+ B cells was also higher in patients with sepsis (p = 0.0317). There was differential expression of PD-1 by CD27 status, with expression being higher in the B and T cell subsets associated with memory status (CD27+ and CD27-, respectively; p < 0.001). Higher PD-1 and PD-L1 expression was not associated with mortality or with a higher risk of nosocomial infection. There were no differences in levels of soluble PD-1 or PD-L1 between patients with sepsis and controls.

Conclusions

Higher expression of PD-1 by memory subpopulations of B cells and CD4+ T cells, with normal soluble PD-1 and PD-L1 in patients with sepsis, are novel findings. This information may be useful to enrich sepsis populations for trials of PD-1/PD-L1 blockade.

Electronic supplementary material

The online version of this article (10.1186/s13054-018-2020-2) contains supplementary material, which is available to authorized users.

How could we enhance translation of sepsis immunology to inform immunomodulation trials in sepsis?

Sepsis results in complex alterations to the immune system. Our understanding of how these alterations in immune responses could help characterize extreme immune phenotypes, identify biomarkers with the ability to stratify patients for therapeutic interventions, surrogates in the causal pathway of clinical end-points, and treatable traits are still rudimentary. A methodologically rigorous, consensus-based approach should enrich sepsis immune subpopulations to increase the probability of successful trials.