Citations

Abstract Rationale Acute respiratory distress syndrome (ARDS) is a common complication after hematopoietic stem cell transplantation (HCT) and is a major contributor to nonrelapse mortality. Objectives To better understand pretransplant risk factors for developing ARDS after HCT. Methods This is a single-center observational study comparing risk factors for ARDS development in 164 patients who went on to develop post-HCT ARDS compared with 492 patients who did not. The patients were matched 1 to 3 on age, sex, type of transplant (allogeneic vs. autologous), and underlying disease. Pertinent risk factors were analyzed separately in multivariable conditional logistic regression after adjustment for a priori variables known to be associated with ARDS development. Results Patients with ARDS were more likely to have a lower pretransplant pulmonary function as measured by forced vital capacity (FVC) (odds ratio [OR], 0.54 [0.42–0.70] per liter increase in FVC; P < 0.001), forced expiratory volume in one second (FEV1) (OR, 0.52 [0.38–0.71] per liter increase in FEV1; P < 0.001) and diffusing capacity (OR, 0.92 [0.88–0.96] per ml/min/mm Hg increase in diffusing capacity; P < 0.001). Several laboratory indices were predictive of subsequent ARDS development including elevated AST (aspartate aminotransferase) (OR, 1.01 [1.00–1.01]; P < 0.008), lower serum albumin (OR, 0.44 [0.30–0.66]; P < 0.001), lower pretransplant hemoglobin (OR, 0.82 [0.73–0.92]; P = 0.001), and lower leukocyte count (OR, 0.88 [0.79–0.99]; P < 0.03). Patients who went on to develop ARDS were more likely to have been hospitalized in the year before the transplant (OR, 1.11 [1.04–1.20]; P = 0.003), and required invasive or noninvasive ventilation during that hospitalization. Lastly, patients with ARDS were significantly more likely to have received carboplatin, thalidomide, methotrexate, and cisplatin than the non-ARDS control subjects. Conclusions Several risk factors for developing ARDS after HCT are identifiable at the time of transplantation, well before the development of critical illness and ARDS. The identification of risk factors long before ARDS develops is relatively unique to the HCT population. Further work is needed to develop usable risk prediction tools in this setting.

Abstract Rationale The periengraftment respiratory distress syndrome (PERDS) is an early important cause of morbidity following autologous hematopoietic cell transplantation (HCT). There are few contemporary data describing PERDS. Objectives To determine prevalence, risk factors, and outcomes of PERDS after autologous HCT. Methods This was a historical cohort study of adults undergoing autologous HCT at Mayo Clinic, Rochester, Minnesota, between 2005 and 2016. PERDS was defined as 1) respiratory failure requiring supplemental oxygen within 5 days on either side of the neutrophil engraftment date, 2) new pulmonary opacities on chest imaging, and 3) exclusion of an infectious or cardiac etiology to explain the clinical presentation. Results Of 3,473 patients undergoing autologous HCT, 167 (4.8%) developed PERDS. Radiographic changes preceded engraftment in 77% of cases. In a multivariable regression model, risk factors for PERDS included female sex (odds ratio [OR], 1.73; P = 0.001), the number of preengraftment platelet transfusions (OR, 1.22; P = 0.002), and more rapid engraftment (OR, 0.72 per day longer; P < 0.001). PERDS cases were more likely to be admitted to the intensive care unit (47.3% vs. 9.5%, P < 0.001) and require intubation (20.4% vs. 1.6%, P < 0.001). In an adjusted 100-day death analysis, those diagnosed with PERDS were more likely to die (hazard ratio, 3.1; 95% confidence interval, 1.5–6.2; P = 0.002). Conclusions PERDS is a common complication of autologous HCT and is associated with increased mortality and healthcare use. Radiographic evidence of pulmonary involvement precedes hematopoietic recovery. A larger number of platelet transfusions and more rapid engraftment appear to increase risk for PERDS.

Objectives: To develop and verify a digital twin model of critically ill patient using the causal artificial intelligence approach to predict the response to specific treatment during the first 24 hours of sepsis. Design: Directed acyclic graphs were used to define explicitly the causal relationship among organ systems and specific treatments used. A hybrid approach of agent-based modeling, discrete-event simulation, and Bayesian network was used to simulate treatment effect across multiple stages and interactions of major organ systems (cardiovascular, neurologic, renal, respiratory, gastrointestinal, inflammatory, and hematology). Organ systems were visualized using relevant clinical markers. The application was iteratively revised and debugged by clinical experts and engineers. Agreement statistics was used to test the performance of the model by comparing the observed patient response versus the expected response (primary and secondary) predicted by digital twin. Setting: Medical ICU of a large quaternary- care academic medical center in the United States. Patients or Subjects: Adult (> 18 year yr old), medical ICU patients were included in the study. Interventions: No additional interventions were made beyond the standard of care for this study. Measurements and Main Results: During the verification phase, model performance was prospectively tested on 145 observations in a convenience sample of 29 patients. Median age was 60 years (54–66 d) with a median Sequential Organ Failure Assessment score of 9.5 (interquartile range, 5.0–14.0). The most common source of sepsis was pneumonia, followed by hepatobiliary. The observations were made during the first 24 hours of the ICU admission with one-step interventions, comparing the output in the digital twin with the real patient response. The agreement between the observed versus and the expected response ranged from fair (kappa coefficient of 0.41) for primary response to good (kappa coefficient of 0.65) for secondary response to the intervention. The most common error detected was coding error in 50 observations (35%), followed by expert rule error in 29 observations (20%) and timing error in seven observations (5%). Conclusions: We confirmed the feasibility of development and prospective testing of causal artificial intelligence model to predict the response to treatment in early stages of critical illness. The availability of qualitative and quantitative data and a relatively short turnaround time makes the ICU an ideal environment for development and testing of digital twin patient models. An accurate digital twin model will allow the effect of an intervention to be tested in a virtual environment prior to use on real patients.

AbstractClinical decision-making is based on knowledge, expertise, and authority, with clinicians approving almost every intervention—the starting point for delivery of “All the right care, but only the right care,” an unachieved healthcare quality improvement goal. Unaided clinicians suffer from human cognitive limitations and biases when decisions are based only on their training, expertise, and experience. Electronic health records (EHRs) could improve healthcare with robust decision-support tools that reduce unwarranted variation of clinician decisions and actions. Current EHRs, focused on results review, documentation, and accounting, are awkward, time-consuming, and contribute to clinician stress and burnout. Decision-support tools could reduce clinician burden and enable replicable clinician decisions and actions that personalize patient care. Most current clinical decision-support tools or aids lack detail and neither reduce burden nor enable replicable actions. Clinicians must provide subjective interpretation and missing logic, thus introducing personal biases and mindless, unwarranted, variation from evidence-based practice. Replicability occurs when different clinicians, with the same patient information and context, come to the same decision and action. We propose a feasible subset of therapeutic decision-support tools based on credible clinical outcome evidence: computer protocols leading to replicable clinician actions (eActions). eActions enable different clinicians to make consistent decisions and actions when faced with the same patient input data. eActions embrace good everyday decision-making informed by evidence, experience, EHR data, and individual patient status. eActions can reduce unwarranted variation, increase quality of clinical care and research, reduce EHR noise, and could enable a learning healthcare system.

BACKGROUND: Liberal oxygenation during mechanical ventilation is harmful in critically ill patients and in certain subsets of patients, including those with stroke, acute myocardial infarction, and cardiac arrest. Surveillance through electronic medical records improves safety of mechanical ventilation in the ICU. To date, this practice has not been used for oxygen titration ( F IO 2 ) in adults. We hypothesize that a surveillance system based on the electronic medical record to alert respiratory therapists to titrate F IO 2 is feasible, safe, and efficacious. METHODS: In this pilot study, mechanically ventilated subjects were randomized to respiratory therapist-driven F IO 2 titration after an electronic alert versus standard of care (ie, titration based on physician order). An automated surveillance system utilizing a hyperoxemia-detection algorithm generated an electronic alert to a respiratory therapist’s pager. Hyperoxemia was defined as F IO 2 > 0.5 and S pO 2 > 95% for > 30 min. No other aspects of treatment were changed. We assessed feasibility, safety, and preliminary efficacy. Primary outcome was duration of hyperoxemia during mechanical ventilation. An unsafe outcome was identified as hypoxemia ( S pO 2 < 88%) within 1 h after titration per alert. Feasibility was assessed by a survey of respiratory therapists. RESULTS: Of 226 randomized subjects, 31 were excluded (eg, programming errors of the electronic alerts, no consent, physician discretion). We included 195 subjects, of whom 86 were in the intervention arm. Alert accuracy was 78%, and respiratory therapists responded to 64% of the alerts. During mechanical ventilation, exposure to hyperoxemia significantly decreased in the intervention group (median 13.5 h [interquartile range 6.2–29.4] vs 18.8 h [interquartile range 9.6–37.4]). No episodes of significant hypoxemia were registered. Most respiratory therapists agreed that the alert was helpful in reducing excessive oxygen exposure. CONCLUSIONS: Use of an electronic surveillance system to titrate F IO 2 was safe and feasible and showed preliminary efficacy in reducing hyperoxemia. Our study serves to justify larger randomized controlled trials for F IO 2 titration.

Objectives: Use of observational data to inform the response and care of patients during a pandemic faces unique challenges. Design: The Society of Critical Care Medicine Discovery Viral Infection and Respiratory Illness Universal Study COVID 2019 Registry Core data and research methodology team convened over virtual meetings throughout March to June 2020 to determine best practice goals for development of a pandemic disease registry to support rapid data collection and analysis. Setting: International, multi-center registry of hospitalized patients. Patients: None. Interventions: None. Measurements and Main Results: Large-scale observational data collection requires: 1) quality assurance and harmonization across many sites; 2) a transparent process for selecting from among many potential research questions; 3) the use of best practices in design of descriptive, predictive, and inferential studies; (4) innovative approaches to characterize random error in the setting of constantly updated data; (5) rapid peer-review and reporting; and (6) transitions from a focus on discovery to implementation. Herein, we describe the guiding principles to best practices and suggestions for innovations to study design and reporting within the coronavirus disease 2019 Viral Infection and Respiratory Illness Universal Study pandemic registry. Conclusions: Society of Critical Care Medicine Discovery Viral Infection and Respiratory Illness Universal Study coronavirus disease 2019 registry sought to develop and implement prespecified best practices combined with grassroots efforts from clinical sites worldwide in order to develop clinically useful knowledge in response to a pandemic.