publications

2025

1. Predicting benefit from adjuvant therapy with corticosteroids in community-acquired pneumonia: a data-driven analysis of randomised trials

   Jim M Smit, Philip A Van Der Zee, Sara C M Stoof, Michel E Van Genderen, and 32 more authors

   The Lancet Respiratory Medicine, 2025

   Abs DOI

   Summary Background Despite several randomised controlled trials (RCTs) on the use of adjuvant treatment with corticosteroids in patients with community-acquired pneumonia (CAP), the effect of this intervention on mortality remains controversial. We aimed to evaluate heterogeneity of treatment effect (HTE) of adjuvant treatment with corticosteroids on 30-day mortality in patients with CAP. Methods In this individual patient data meta-analysis, we included RCTs published before July 1, 2024, comparing adjuvant treatment with corticosteroids versus placebo in patients hospitalised with CAP. The primary endpoint was 30-day all-cause mortality, collected across all trials, and analyses followed the intention-to-treat principle. We analysed HTE using risk and effect modelling. For risk modelling, patients were classified as having less severe or severe CAP based on the pneumonia severity index (PSI), comparing PSI class I–III versus class IV–V. For effect modelling, we trained a corticosteroid-effect model on six trials and externally validated it using data from two trials, received after model preregistration. This model classified patients into two groups: no predicted benefit and predicted benefit from adjuvant treatment with corticosteroids. The literature search was registered on PROSPERO, CRD42022380746. Findings We included eight RCTs with 3224 patients. Across all eight trials, 246 (7\textperiodcentered6%) patients died within 30 days (106 [6\textperiodcentered6%] of 1618 in the corticosteroid group vs 140 [8\textperiodcentered7%] of 1606 in the placebo group; odds ratio [OR] 0\textperiodcentered72 [95% CI 0\textperiodcentered56–0\textperiodcentered94], p=0\textperiodcentered017). The corticosteroid-effect model, which selected C-reactive protein (CRP), showed significant HTE during external validation in the two most recent trials. In these trials, 154 (11\textperiodcentered4%) of 1355 patients died within 30 days (88 [13\textperiodcentered1%] of 671 in the placebo group vs 66 [9\textperiodcentered6%] of 684 in the corticosteroid group; OR 0\textperiodcentered71 [95% CI 0\textperiodcentered50–0\textperiodcentered99], p=0\textperiodcentered044). Among patients predicted to have no benefit (CRP ≤204 mg/L, n=725), no significant effect was observed (OR 0\textperiodcentered98 [95% CI 0\textperiodcentered63–1\textperiodcentered50]), whereas for those with predicted benefit (CRP >204 mg/L, n=630), 39 (13\textperiodcentered0%) of 301 patients died in the placebo group compared with 20 (6\textperiodcentered1%) of 329 in the corticosteroid group (0\textperiodcentered43 [0\textperiodcentered25–0\textperiodcentered76], pinteraction=0\textperiodcentered026). No significant HTE was found between less severe CAP (PSI class I–III, n=229) and severe CAP (PSI class IV–V, n=1126). Corticosteroid therapy significantly increased hyperglycaemia risk (44 [12\textperiodcentered8%] of 344 in the placebo group vs 84 [24\textperiodcentered8%] of 339 in the corticosteroid group; OR 2\textperiodcentered50 [95% CI 1\textperiodcentered63–3\textperiodcentered83], p<0\textperiodcentered0001) and hospital re-admission risk (30 [3\textperiodcentered7%] of 814 in the placebo group vs 57 [7\textperiodcentered0%] of 819 in the corticosteroid group; 1\textperiodcentered95 [1\textperiodcentered24–3\textperiodcentered07], p=0\textperiodcentered0038). Interpretation Overall, adjuvant therapy with corticosteroids significantly reduces 30-day mortality in patients hospitalised with CAP. The treatment effect varied significantly among subgroups based on CRP concentrations, with a substantial mortality reduction observed only in patients with high baseline CRP. Funding None.

2. The Heterogeneous Effect of High PEEP strategies on Survival in Acute Respiratory Distress Syndrome: preliminary results of a data-driven analysis of randomized trials

   Jim M Smit, Jesse H Krijthe, Jasper Bommel, Demet S Sulemanji, and 24 more authors

   medRxiv, Jan 2025

   Abs DOI

   Background: Mixed trial results suggest that some ventilated patients with acute respiratory distress syndrome (ARDS) benefit from high PEEP while others may be harmed, indicating heterogeneity of treatment effect (HTE). This study applies data-driven predictive approaches to uncover HTE and re-examines previously hypothesized HTE. This manuscript serves as a pre-registration of planned external validation of our trained models. Methods: We identified eight randomized trials, and obtained individual patient data (IPD) from three of them (ALVEOLI, LOVS, EXPRESS), as our train cohort. We used effect modelling to predict individualized treatment effects (predicted 28-day mortality risk difference between PEEP strategies) across patient subgroups stratified by observed tertiles (<=8 cmH2O, 9-11 cmH2O, >=12 cmH2O). Candidate effect modelling methods included meta-learners and technique-specific methods. Optimal methods were selected through 'leave-one-trial-out' cross-validation, evaluating the methods' performances in each PEEP tertile using AUC-benefit. We trained final models using the best performing methods implemented with or without forward selection (which yielded sufficient AUC-benefit), and additional final models by selecting the variables that yielded consistency in the forward selections performed in the cross validation, if any. We further evaluated earlier hypothesized HTE comparing (1) patients with baseline PaO2/FiO2 <=200 versus >200 mmHg, and (2) patients with hypoinflammatory versus hyperinflammatory subphenotypes. Preliminary findings: In the lower PEEP tertile (<=8 cmH2O), an X-learner implemented without, and an S-learner implemented with forward selection (both with flexible base learners), yielded the highest AUC benefits and were used to train final models. In the high PEEP tertile (>=12 cmH2O), only the causal forest implemented with forward selection yielded an AUC benefit exceeding zero. Respiratory-system compliance (CRS) was consistently selected in the forward selections of cross validation, and was used to train an extra final causal forest model, with predicted effects shifting from harm to benefit for CRS 26.5 mL/cmH2O or higher. Higher PEEP benefited patients with baseline PaO2/FiO2 <=200 mmHg (OR 0.80, 95% CI 0.66-0.98), incurred harm among those with PaO2;/FiO2 >200 mmHg (OR 1.74, 95% CI 1.02-2.98; interaction P=0.01). This HTE was strongest when PaO2/FiO2 was measured at low PEEP (<=8 cmH2O), reduced at mid-level PEEP (9-11 cmH2O), and negligible at high PEEP (>=12 cmH2O). A second-order interaction showed significant heterogeneity of HTE (ie, second-order heterogeneity) across PEEP tertiles (P=0.03). Preliminary Conclusions: Our preliminary findings indicated that baseline CRS >=26.5 mL/cmH2O predicts benefit, while CRS <26.5 mL/cmH2O predicts harm from high PEEP when CRS is measured at high baseline PEEP (>=12 cmH2O). Similarly, baseline PaO2/FiO2; <=200 mmHg predicts benefit, while PaO2/FiO2 >200 mmHg predicts harm from high PEEP when PaO2/FiO2 is measured at a low baseline PEEP (<=8 cmH2O). Using data from the LOVS trial, we investigated HTE for high PEEP between hypo- and hyperinflammatory subphenotypes but found none, despite significant HTE observed earlier in the ALVEOLI trial.Competing Interest StatementA.H. Jonkman declares research funding paid to the institution by Pulmotech B.V., for validation of new esophageal pressure sensor. D. Talmor declares support from the National Institutes of Health and lecture honoraria from Mindray. J. Villar was funded by Instituto de Salud Carlos III, Madrid, Sapin (CB06/06/1088, PI19/00141, AC-21_2/00039), ERAPerMed(JTC_2021), ERAPerMed(JTC_2021), The European Regional Development Funds, Fundacion Canaria Intituto de Investigacipn Sanitaria de Canarias, and Asociacion Cientifica Pulmon y Ventilacion Mecanica. C.S. Calfee declares grants from NIH, Roche Genentech, and Quantum Leap Healthcare Collaborative to her institution, consulting fees from Vasomune, Gen1e Life Sciences, NGM Bio, Cellenkos, Calcimedica, Arrowhead, EnliTISA, Novartis, and Merck, being speaker at a symposium on ESICM guidelines supported by Fisher-Paykel, a Patent on metagenomic sequencing for sepsis diagnosis (co-recipient) issued to Regents of University of California and Chan Zuckerberg BioHub, and being council member of the International Sepsis Forum (unpaid). Funding StatementThis study did not receive any funding.Author DeclarationsI confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained.YesThe details of the IRB/oversight body that provided approval or exemption for the research described are given below:This study included individual patient data from three randomized controlled trials. The trial by Mercat et al. enrolled patinets at 37 intensive care units in France, and their study protocol was approved for all centers by the ethics committee of the Angers University Hospital (Comite Consultatif de Protection des Personnes dans la Recherche Biomedicale), according to French law. The trial by Brower et al. enrolled patients at 23 hospitals of the National HeartLung, and Blood Institute (NHLBI) ARDS Clinical Trials Network, and was approved by the institutional review board of each hospital. The trial by Meade et al. enrolled patients in 30 hospitals in Canada, Australia, and Saudi Arabia, and the research ethics board of each hospital approved the trial.I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals.YesI understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance).Yes I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable.YesAll data produced in the present study are available upon reasonable request to the authors

3. Analyzing PaO2/FiO2?: mind the interaction with PEEP!

   J M Smit, J H Krijthe, J Van Bommel, M E Van Genderen, and 2 more authors

   Intensive Care Medicine, Jan 2025

   DOI

2023

1. The future of artificial intelligence in intensive care: moving from predictive to actionable AI.

   Jim M Smit, Jesse H Krijthe, and Jasper Bommel

   Intensive care medicine, Sep 2023

   DOI
2. Causal inference using observational intensive care unit data: a scoping review and recommendations for future practice

   J M Smit, J H Krijthe, W M R Kant, J A Labrecque, and 5 more authors

   npj Digital Medicine, Sep 2023

   Abs DOI

   This scoping review focuses on the essential role of models for causal inference in shaping actionable artificial intelligence (AI) designed to aid clinicians in decision-making. The objective was to identify and evaluate the reporting quality of studies introducing models for causal inference in intensive care units (ICUs), and to provide recommendations to improve the future landscape of research practices in this domain. To achieve this, we searched various databases including Embase, MEDLINE ALL, Web of Science Core Collection, Google Scholar, medRxiv, bioRxiv, arXiv, and the ACM Digital Library. Studies involving models for causal inference addressing time-varying treatments in the adult ICU were reviewed. Data extraction encompassed the study settings and methodologies applied. Furthermore, we assessed reporting quality of target trial components (i.e., eligibility criteria, treatment strategies, follow-up period, outcome, and analysis plan) and main causal assumptions (i.e., conditional exchangeability, positivity, and consistency). Among the 2184 titles screened, 79 studies met the inclusion criteria. The methodologies used were G methods (61%) and reinforcement learning methods (39%). Studies considered both static (51%) and dynamic treatment regimes (49%). Only 30 (38%) of the studies reported all five target trial components, and only seven (9%) studies mentioned all three causal assumptions. To achieve actionable AI in the ICU, we advocate careful consideration of the causal question of interest, describing this research question as a target trial emulation, usage of appropriate causal inference methods, and acknowledgement (and examination of potential violations of) the causal assumptions.

2022

1. Developing, implementing and governing artificial intelligence in medicine: a step-by-step approach to prevent an artificial intelligence winter.

   Davy Sande, Michel E Van Genderen, Jim M Smit, Joost Huiskens, and 6 more authors

   BMJ health & care informatics, Feb 2022

   Abs DOI

   OBJECTIVE: Although the role of artificial intelligence (AI) in medicine is increasingly studied, most patients do not benefit because the majority of AI models remain in the testing and prototyping environment. The development and implementation trajectory of clinical AI models are complex and a structured overview is missing. We therefore propose a step-by-step overview to enhance clinicians’ understanding and to promote quality of medical AI research. METHODS: We summarised key elements (such as current guidelines, challenges, regulatory documents and good practices) that are needed to develop and safely implement AI in medicine. CONCLUSION: This overview complements other frameworks in a way that it is accessible to stakeholders without prior AI knowledge and as such provides a step-by-step approach incorporating all the key elements and current guidelines that are essential for implementation, and can thereby help to move AI from bytes to bedside.

2. Development and validation of an early warning model for hospitalized COVID-19 patients: a multi-center retrospective cohort study

   Jim M Smit, Jesse H Krijthe, Andrei N Tintu, Henrik Endeman, and 20 more authors

   Intensive Care Medicine Experimental, Feb 2022

   Abs DOI

   Timely identification of deteriorating COVID-19 patients is needed to guide changes in clinical management and admission to intensive care units (ICUs). There is significant concern that widely used Early warning scores (EWSs) underestimate illness severity in COVID-19 patients and therefore, we developed an early warning model specifically for COVID-19 patients.

3. Dynamic prediction of mortality in COVID-19 patients in the intensive care unit: A retrospective multi-center cohort study

   J M Smit, J H Krijthe, H Endeman, A N Tintu, and 43 more authors

   Intelligence-Based Medicine, Feb 2022

   Abs DOI

   Background The COVID-19 pandemic continues to overwhelm intensive care units (ICUs) worldwide, and improved prediction of mortality among COVID-19 patients could assist decision making in the ICU setting. In this work, we report on the development and validation of a dynamic mortality model specifically for critically ill COVID-19 patients and discuss its potential utility in the ICU. Methods We collected electronic medical record (EMR) data from 3222 ICU admissions with a COVID-19 infection from 25 different ICUs in the Netherlands. We extracted daily observations of each patient and fitted both a linear (logistic regression) and non-linear (random forest) model to predict mortality within 24 h from the moment of prediction. Isotonic regression was used to re-calibrate the predictions of the fitted models. We evaluated the models in a leave-one-ICU-out (LOIO) cross-validation procedure. Results The logistic regression and random forest model yielded an area under the receiver operating characteristic curve of 0.87 [0.85; 0.88] and 0.86 [0.84; 0.88], respectively. The recalibrated model predictions showed a calibration intercept of −0.04 [−0.12; 0.04] and slope of 0.90 [0.85; 0.95] for logistic regression model and a calibration intercept of −0.19 [−0.27; −0.10] and slope of 0.89 [0.84; 0.94] for the random forest model. Discussion We presented a model for dynamic mortality prediction, specifically for critically ill COVID-19 patients, which predicts near-term mortality rather than in-ICU mortality. The potential clinical utility of dynamic mortality models such as benchmarking, improving resource allocation and informing family members, as well as the development of models with more causal structure, should be topics for future research.

2021

1. Demystifying machine learning for mortality prediction

   J. M. Smit, M. E. Genderen, M. J. T. Reinders, D. A. M. P. J. Gommers, and 2 more authors

   Critical Care, Feb 2021

   DOI