Use of dynamic microsimulation to predict disease progression in patients with pneumonia-related sepsis

Optimizing the First Response to Sepsis: An Electronic Health Record-Based Markov Decision Process Model

Sepsis is considered a medical emergency where delays in initial treatment are associated with increased morbidity and mortality, yet there is no gold standard for identifying sepsis onset and thus treatment timing. We leverage electronic health record (EHR) data with clinical expertise to develop a continuous-time Markov decision process (MDP) optimal stopping model that identifies the optimal first intervention action (anti-infective, fluid, or wait). To study the impact of initial treatment of patients at risk for developing sepsis, we define the delayed treatment population who received delayed treatment upon admission or during hospitalization and serves as an approximation of the natural history of sepsis. We apply the optimal first treatment policy to sample patient visits from the nondelayed treatment population. This analysis indicates the average risk of death could be reduced by approximately 2.2%, the average time until treatment could be reduced by 106 minutes, and the average severity of the treatment state could be reduced by 15.5% compared with the treatment they received in the hospital. We study the properties of the optimal policy to define an easily interpretable initial treatment heuristic that considers a patient’s organ dysfunction, location, and septic shock status. This generalizable framework can inform personalized treatment of patients at risk for sepsis.

A critical appraisal of the prognostic predictive models for patients with sepsis: Which model can be applied in clinical practice?

BACKGROUND

Sepsis is associated with high mortality and predictive models can help in clinical decision-making. The objective of this study was to carry out a systematic review of these models.

METHODS

In 2019, we conducted a systematic review in MEDLINE and EMBASE (CDR42018111121:PROSPERO) of articles that developed predictive models for mortality in septic patients (inclusion criteria). We followed the CHARMS recommendations (Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modelling Studies), extracting the information from its 11 domains (Source of data, Participants, etc). We determined the risk of bias and applicability (participants, outcome, predictors and analysis) through PROBAST (Prediction model Risk Of Bias ASsessment Tool).

RESULTS

A total of 14 studies were included. In the CHARMS extraction, the models found showed great variability in its 11 domains. Regarding the PROBAST checklist, only one article had an unclear risk of bias as it did not indicate how missing data were handled while the others all had a high risk of bias. This was mainly due to the statistical analysis (inadequate sample size, handling of continuous predictors, missing data and selection of predictors), since 13 studies had a high risk of bias. Applicability was satisfactory in six articles. Most of the models integrate predictors from routine clinical practice. Discrimination and calibration were assessed for almost all the models, with the area under the ROC curve ranging from 0.59 to 0.955 and no lack of calibration. Only three models were externally validated and their maximum discrimination values in the derivation were from 0.712 and 0.84. One of them (Osborn) had undergone multiple validation studies.

DISCUSSION

Despite most of the studies showing a high risk of bias, we very cautiously recommend applying the Osborn model, as this has been externally validated various times.

Simulation Training in Hemodynamic Monitoring and Mechanical Ventilation: An Assessment of Physician's Performance

Background

Simulation is to imitate or replicate real-life scenarios in order to improve cognitive, diagnostic and therapeutic skills. An ideal model should be good enough to output realistic clinical scenarios and respond to interventions done by trainees in real time. Use of simulation-based training has been tried in various fields of medicine. The aim of our study was to prospectively evaluate the effectiveness of simulation model “CRITICA”™ (MEDUPLAY systems) in training critical care physicians.

Materials and methods

The advanced intensive care unit (ICU) simulator “CRITICA”™ (MEDUPLAY systems) was developed as a joint collaboration between the Indian Institute of Science, Bengaluru and St John’s Medical College, Bengaluru. Two-day workshop was conducted. Intensive didactic and case-based scenarios were simulated to formally teach principles of advanced ICU scenarios. The physicians were tested on clinical scenarios in hemodynamic monitoring and mechanical ventilation displayed on the simulator. Assessment of the analytical thinking and pattern recognition ability was carried out before and after the display of the scenarios. Pre- and posttest scores were collected.

Results

The postsimulation test scores were higher than pretest scores and were statistically significant in hemodynamic monitoring and mechanical ventilation module. [Hemodynamic monitoring pre- and posttest scores 4.41 (2.06) vs 5.23 (2.22) p < 0.001] [Mechanical ventilation pre- and posttest scores 4 (2–5.5) vs 7.5 (6.5–8.5) p < 0.001]. A greater increase in posttest scores was seen in the mechanical ventilation module as compared to hemodynamic module. There was no effect of specialty or designation of a trainee on difference in pre- and posttest scores.

Conclusion

Simulator-based training in hemodynamic monitoring and mechanical ventilation was effective. Comparison of routine classroom teaching and simulator-based training needs to be evaluated prospectively.

How to cite this article

Havaldar AA, Krishna B, Sampath S, Paramasivam SK. Simulation Training in Hemodynamic Monitoring and Mechanical Ventilation: An Assessment of Physician’s Performance. Indian J Crit Care Med 2020;24(6):423–428.

Microsimulation Model Using Christiana Care Early Warning System (CEWS) to Evaluate Physiological Deterioration

While physiological warning signs prior to deterioration events during hospitalization have been widely studied, evaluating clinical interventions, such as rapid response team (RRT) activations, based on scoring systems remains an understudied area. Simulation of physiological deterioration patterns represented by scoring systems can facilitate testing different RRT policies without disturbing care processes. Christiana Care Early Warning System (CEWS) is a scoring system developed at the study hospital to detect the physiological warning signs and inform RRT activations. The objective of this study is to evaluate CEWS-triggered RRT policies based on patient demographics and policy structures. Using retrospective data derived from a subset of electronic health records between December 2015 and December 2016 (6000 patients), we developed a microsimulation model with integrated regression analysis to compare RRT policies on subpopulations defined by age, gender, and comorbidities to find score thresholds that result in the lowest percent of time spent above critical CEWS values. Policies that rely on average scores were more sensitive to threshold changes compared to policies that rely on current value and change in the CEWS. Policy using score threshold 10 provided the lowest percentage of time under the critical condition for majority of subpopulations. The proposed model is a novel framework to simulate individual deterioration patterns and systematically evaluate RRT policies based on their impact on health conditions. Our work highlights the importance of integration of data-driven models into personalized care and represents a significant opportunity to inform biomedical and health informatics research on designing and evaluating EWS-based clinical interventions.

Clinical course of sepsis, severe sepsis, and septic shock in a cohort of infected patients from ten Colombian hospitals

Background

Sepsis has several clinical stages, and mortality rates are different for each stage. Our goal was to establish the evolution and the determinants of the progression of clinical stages, from infection to septic shock, over the first week, as well as their relationship to 7-day and 28-day mortality.

Methods

This is a secondary analysis of a multicenter cohort of inpatients hospitalized in general wards or intensive care units (ICUs). The general estimating equations (GEE) model was used to estimate the risk of progression and the determinants of stages of infection over the first week. Cox regression with time-dependent covariates and fixed covariates was used to determine the factors related with 7-day and 28-day mortality, respectively.

Results

In 2681 patients we show that progression to severe sepsis and septic shock increases with intraabdominal and respiratory sources of infection [OR = 1,32; 95%IC = 1,20-1,46 and OR = 1.21, 95%CI = 1,11-1,33 respectively], as well as according to Acute Physiology and Chronic Health Evaluation II (APACHE II) [OR = 1,03; 95%CI = 1,02-1,03] and Sequential Organ Failure Assessment (SOFA) [OR = 1,16; 95%CI = 1,14-1,17] scores. The variables related with first-week mortality were progression to severe sepsis [HR = 2,13; 95%CI = 1,13-4,03] and septic shock [HR = 3,00; 95%CI = 1,50-5.98], respiratory source of infection [HR = 1,76; 95%IC = 1,12-2,77], APACHE II [HR = 1,07; 95% CI = 1,04-1,10] and SOFA [HR = 1,09; 95%IC = 1,04-1,15] scores.

Conclusions

Intraabdominal and respiratory sources of infection, independently of SOFA and APACHE II scores, increase the risk of clinical progression to more severe stages of sepsis; and these factors, together with progression of the infection itself, are the main determinants of 7-day and 28-day mortality.

CURB-65 and SMRT-CO in the prediction of early transfers to the intensive care unit among patients with community-acquired pneumonia initially admitted to a general ward

BACKGROUND

No study has assessed the ability of pneumonia severity scores to identify the risk for early intensive care unit (ICU) transfer in patients with community-acquired pneumonia (CAP) admitted to general wards (GW). We aimed to compare the ability of CURB-65 (confusion, urea level, respiratory rate, blood pressure, and age ≥65 years) and SMRT-CO (systolic blood pressure, multilobar chest radiography involvement, respiratory rate, tachycardia, confusion, and oxygenation) scores to predict early ICU transfers in these patients.

DESIGN

Retrospective, case-control study. Cases were defined as patients admitted to GW with CAP that required ICU transfer within 48 hours. Controls were defined as patients admitted to GW with CAP that did not require ICU transfer. CURB-65 and SMRT-CO scores were calculated on presentation to emergency department (ED), and upon admission to GW. Composite scores were calculated combining data from ED and GW. Sensitivities, specificities, likelihood ratios, and areas-under-the-curve (AUC) were calculated for each score.

RESULTS

From 2003 to 2009, 115 cases and 345 controls were identified. Both groups had similar baseline characteristics. Composite scores combining data from ED and GW had better sensitivity and AUC than scores calculated only with ED or GW data (P < 0.001). A composite SMRT-CO score ≥2 had 76.5% (95% CI, 67.7 to 83.9) sensitivity, 67.5% (95% CI, 62.3 to 72.4) specificity, and 0.81 (95% CI, 0.77 to 0.85) AUC. A composite CURB-65 score ≥3 had 36.5% (95% CI, 27.7 to 46.0) sensitivity, 86.3% (95% CI, 82.3 to 89.8) specificity, and 0.66 (95% CI, 0.60 to 0.72) AUC to predict early ICU transfers. Composite SMRT-CO had higher sensitivity and AUC (P < 0.001) than composite CURB-65.

CONCLUSIONS

Composite SMRT-CO had a better combination of sensitivity and specificity than CURB-65 for predicting early ICU transfers. Prospective studies to confirm our findings are needed.

The pathogenesis of sepsis

Sepsis is a serious clinical condition that represents a patient's response to a severe infection and has a very high mortality rate. Normal immune and physiologic responses eradicate pathogens, and the pathophysiology of sepsis is due to the inappropriate regulation of these normal reactions. In an ideal scenario, the first pathogen contact with the inflammatory system should eliminate the microbe and quickly return the host to homeostasis. The septic response may accelerate due to continued activation of neutrophils and macrophages/monocytes. Upregulation of lymphocyte costimulatory molecules and rapid lymphocyte apoptosis, delayed apoptosis of neutrophils, and enhanced necrosis of cells/tissues also contribute to the pathogenesis of sepsis. The coagulation system is closely tied to the inflammatory response, with cross talk between the two systems driving the dysregulated response. Biomarkers may be used to help diagnose patients with sepsis, and they may also help to identify patients who would benefit from immunomodulatory therapies.

Using hierarchical dynamic Bayesian networks to investigate dynamics of organ failure in patients in the Intensive Care Unit

In intensive care medicine close monitoring of organ failure status is important for the prognosis of patients and for choices regarding ICU management. Major challenges in analyzing the multitude of data pertaining to the functioning of the organ systems over time are to extract meaningful clinical patterns and to provide predictions for the future course of diseases. With their explicit states and probabilistic state transitions, Markov models seem to fit this purpose well. In complex domains such as intensive care a choice is often made between a simple model that is estimated from the data, or a more complex model in which the parameters are provided by domain experts. Our primary aim is to combine these approaches and develop a set of complex Markov models based on clinical data. In this paper we describe the design choices underlying the models, which enable them to identify temporal patterns, predict outcomes, and test clinical hypotheses. Our models are characterized by the choice of the dynamic hierarchical Bayesian network structure and the use of logistic regression equations in estimating the transition probabilities. We demonstrate the induction, inference, evaluation, and use of these models in practice in a case-study of patients with severe sepsis admitted to four Dutch ICUs.

Between prediction, education, and quality control: simulation models in critical care

Today, computer-aided strategies in social sciences are an indispensable component of teaching programs. In recent years, microsimulation modeling has gained attention in its ability to represent predicted physiological developments visually, thus providing the user with a full understanding of the impacts of a proposed scheme. There are several microsimulation models in human medicine, and they can be either dynamic or static. If the model is dynamic the course of variables changes over time; in contrast, in the static case time constancy is assumed. In critical care there have been several approaches to implement microsimulation models to predict outcome. This commentary describes current approaches for predicting disease progression by using dynamic microsimulation in pneumonia-related sepsis.