Lower or higher oxygenation targets for acute Hypoxaemic respiratory failure: Protocol for an individual patient data meta-analysis.

BACKGROUND: Supplemental oxygen therapy is central to the treatment of acute hypoxaemic respiratory failure, a condition which remains a major driver for morbidity and mortality in intensive care. Despite several large randomised clinical trials comparing a higher versus a lower oxygenation target for these patients, significant differences in study design impede analysis of aggregate data and final clinical recommendations. METHODS: This paper presents the protocol for conducting an individual patient data meta-analysis where full individual patient data according to the intention-to-treat principle will be pooled from the HOT-ICU and HOT-COVID trials in a one-step procedure. The two trials are near-identical in design. We plan to use a hierarchical general linear mixed model that accounts for data clustering at a trial and site level. The primary outcome will be 90-day all-cause mortality while the secondary outcome will be days alive without life-support at 90 days. Further, we outline 14 clinically relevant predefined subgroups which we will analyse for heterogeneity in the intervention effects and interactions, and we present a plan for assessing the credibility of the subgroup analyses. CONCLUSION: The presented individual patient data meta-analysis will synthesise individual level patient data from two of the largest randomised clinical trials on targeted oxygen therapy in intensive care. The results will provide a re-analysis of the intervention effects on the pooled intention-to-treat populations and facilitate subgroup analyses with an increased power to detect clinically important effect modifications.

Developing a Bayesian hierarchical model for a prospective individual patient data meta-analysis with continuous monitoring.

BACKGROUND: Numerous clinical trials have been initiated to find effective treatments for COVID-19. These trials have often been initiated in regions where the pandemic has already peaked. Consequently, achieving full enrollment in a single trial might require additional COVID-19 surges in the same location over several years. This has inspired us to pool individual patient data (IPD) from ongoing, paused, prematurely-terminated, or completed randomized controlled trials (RCTs) in real-time, to find an effective treatment as quickly as possible in light of the pandemic crisis. However, pooling across trials introduces enormous uncertainties in study design (e.g., the number of RCTs and sample sizes might be unknown in advance). We sought to develop a versatile treatment efficacy assessment model that accounts for these uncertainties while allowing for continuous monitoring throughout the study using Bayesian monitoring techniques. METHODS: We provide a detailed look at the challenges and solutions for model development, describing the process that used extensive simulations to enable us to finalize the analysis plan. This includes establishing prior distribution assumptions, assessing and improving model convergence under different study composition scenarios, and assessing whether we can extend the model to accommodate multi-site RCTs and evaluate heterogeneous treatment effects. In addition, we recognized that we would need to assess our model for goodness-of-fit, so we explored an approach that used posterior predictive checking. Lastly, given the urgency of the research in the context of evolving pandemic, we were committed to frequent monitoring of the data to assess efficacy, and we set Bayesian monitoring rules calibrated for type 1 error rate and power. RESULTS: The primary outcome is an 11-point ordinal scale. We present the operating characteristics of the proposed cumulative proportional odds model for estimating treatment effectiveness. The model can estimate the treatment's effect under enormous uncertainties in study design. We investigate to what degree the proportional odds assumption has to be violated to render the model inaccurate. We demonstrate the flexibility of a Bayesian monitoring approach by performing frequent interim analyses without increasing the probability of erroneous conclusions. CONCLUSION: This paper describes a translatable framework using simulation to support the design of prospective IPD meta-analyses.

Characteristics and Outcomes of Patients With Frailty Admitted to ICU With Coronavirus Disease 2019: An Individual Patient Data Meta-Analysis.

Frailty is often used in clinical decision-making for patients with coronavirus disease 2019, yet studies have found a variable influence of frailty on outcomes in those admitted to the ICU. In this individual patient data meta-analysis, we evaluated the characteristics and outcomes across the range of frailty in patients admitted to ICU with coronavirus disease 2019. DATA SOURCES: We contacted the corresponding authors of 16 eligible studies published between December 1, 2019, and February 28, 2021, reporting on patients with confirmed coronavirus disease 2019 admitted to ICU with a documented Clinical Frailty Scale. STUDY SELECTION: Individual patient data were obtained from seven studies with documented Clinical Frailty Scale were included. We classified patients as nonfrail (Clinical Frailty Scale = 1-4) or frail (Clinical Frailty Scale = 5-8). DATA EXTRACTION: We collected patient demographics, Clinical Frailty Scale score, ICU organ supports, and clinically relevant outcomes (ICU and hospital mortality, ICU and hospital length of stays, and discharge destination). The primary outcome was hospital mortality. DATA SYNTHESIS: Of the 2,001 patients admitted to ICU, 388 (19.4%) were frail. Increasing age and Sequential Organ Failure Assessment score, Clinical Frailty Scale score greater than or equal to 4, use of mechanical ventilation, vasopressors, renal replacement therapy, and hyperlactatemia were risk factors for death in a multivariable analysis. Hospital mortality was higher in patients with frailty (65.2% vs 41.8%; p < 0.001), with adjusted mortality increasing with a rising Clinical Frailty Scale score beyond 3. Younger and nonfrail patients were more likely to receive mechanical ventilation. Patients with frailty spent less time on mechanical ventilation (median days [interquartile range], 9 [5-16] vs 11 d [6-18 d]; p = 0.012) and accounted for only 12.3% of total ICU bed days. CONCLUSIONS: Patients with frailty with coronavirus disease 2019 were commonly admitted to ICU and had greater hospital mortality but spent relatively fewer days in ICU when compared with nonfrail patients. Patients with frailty receiving mechanical ventilation were at greater risk of death than patients without frailty.