Comparing predictive performance of pulmonary embolism risk stratification tools for acute clinical deterioration.

Objectives: Existing pulmonary embolism (PE) risk scores were developed to predict death within weeks, but not more proximate adverse events. We determined the ability of 3 PE risk stratification tools (simplified pulmonary embolism severity index [sPESI], 2019 European Society of Cardiology guidelines [ESC], and PE short-term clinical outcomes risk estimation [PE-SCORE]) to predict 5-day clinical deterioration after emergency department (ED) diagnosis of PE. Methods: We analyzed data from six EDs on ED patients with confirmed PE. Clinical deterioration was defined as death, respiratory failure, cardiac arrest, new dysrhythmia, sustained hypotension requiring vasopressors or volume resuscitation, or escalated intervention within 5 days of PE diagnosis. We determined sensitivity and specificity of sPESI, ESC, and PE-SCORE for predicting clinical deterioration. Results: Of 1569 patients, 24.5% had clinical deterioration within 5 days. sPESI, ESC, and PE-SCORE classifications were low-risk in 558 (35.6%), 167 (10.6%), and 309 (19.6%), respectively. Sensitivities of sPESI, ESC, and PE-SCORE for clinical deterioration were 81.8 (78, 85.7), 98.7 (97.6, 99.8), and 96.1 (94.2, 98), respectively. Specificities of sPESI, ESC, and PE-SCORE for clinical deterioration were 41.2 (38.4, 44), 13.7 (11.7, 15.6), and 24.8 (22.4, 27.3). Areas under the curve were 61.5 (59.1, 63.9), 56.2 (55.1, 57.3), and 60.5 (58.9, 62.0). Negative predictive values were 87.5 (84.7, 90.2), 97 (94.4, 99.6), and 95.1 (92.7, 97.5). Conclusions: ESC and PE-SCORE were better than sPESI for detecting clinical deterioration within 5 days after PE diagnosis.

Evaluating automated electronic case report form data entry from electronic health records.

Background: Many clinical trials leverage real-world data. Typically, these data are manually abstracted from electronic health records (EHRs) and entered into electronic case report forms (CRFs), a time and labor-intensive process that is also error-prone and may miss information. Automated transfer of data from EHRs to eCRFs has the potential to reduce data abstraction and entry burden as well as improve data quality and safety. Methods: We conducted a test of automated EHR-to-CRF data transfer for 40 participants in a clinical trial of hospitalized COVID-19 patients. We determined which coordinator-entered data could be automated from the EHR (coverage), and the frequency with which the values from the automated EHR feed and values entered by study personnel for the actual study matched exactly (concordance). Results: The automated EHR feed populated 10,081/11,952 (84%) coordinator-completed values. For fields where both the automation and study personnel provided data, the values matched exactly 89% of the time. Highest concordance was for daily lab results (94%), which also required the most personnel resources (30 minutes per participant). In a detailed analysis of 196 instances where personnel and automation entered values differed, both a study coordinator and a data analyst agreed that 152 (78%) instances were a result of data entry error. Conclusions: An automated EHR feed has the potential to significantly decrease study personnel effort while improving the accuracy of CRF data.

Quality of life 1 month after acute pulmonary embolism in emergency department patients.

OBJECTIVE: The Pulmonary Embolism Quality-of-Life (PEmb-QoL) questionnaire assesses quality of life (QoL) after pulmonary embolism (PE). We aimed to determine whether any clinical or pathophysiologic features of PE were associated with worse PEmb-QoL scores 1 month after PE. METHODS: In this prospective multicenter registry, we conducted PEmb-QoL questionnaires. We determined differences in QoL domain scores for four primary variables: clinical deterioration (death, cardiac arrest, respiratory failure, hypotension requiring fluid bolus, catecholamine support, or new dysrhythmia), right ventricular dysfunction (RVD), PE risk stratification, and subsequent rehospitalization. For overall QoL score, we fit a multivariable regression model that included these four primary variables as independent variables. RESULTS: Of 788 PE patients participating in QoL assessments, 156 (19.8%) had a clinical deterioration event, 236 (30.7%) had RVD of which 38 (16.1%) had escalated interventions. For those without and with clinical deterioration, social limitations had mean (±SD) scores of 2.07 (±1.27) and 2.36 (±1.47), respectively (p = 0.027). For intensity of complaints, mean (±SD) scores for patients without RVD (4.32 ± 2.69) were significantly higher than for those with RVD with or without reperfusion interventions (3.82 ± 1.81 and 3.83 ± 2.11, respectively; p = 0.043). There were no domain score differences between PE risk stratification groups. All domain scores were worse for patients with rehospitalization versus without. By multivariable analysis, worse total PEmb-QoL scores with effect sizes were subsequent rehospitalization 11.29 (6.68-15.89), chronic obstructive pulmonary disease (COPD) 8.17 (3.91-12.43), and longer index hospital length of stay 0.06 (0.03-0.08). CONCLUSIONS: Acute clinical deterioration, RVD, and PE severity were not predictors of QoL at 1 month post-PE. Independent predictors of worsened QoL were rehospitalization, COPD, and index hospital length of stay.

Can right ventricular assessments improve triaging of low risk pulmonary embolism?

OBJECTIVES: Identifying right ventricle (RV) abnormalities is important to stratifying pulmonary embolism (PE) severity. Disposition decisions are influenced by concerns about early deterioration. Triaging strategies, like the Simplified Pulmonary Embolism Severity Index (sPESI), do not include RV assessments as predictors or early deterioration as outcome(s). We aimed to (1) determine if RV assessment variables add prognostic accuracy for 5-day clinical deterioration in patients classified low risk by sPESI, and (2) determine the prognostic importance of RV assessments compared to other variables and to each other. METHODS: We identified low risk sPESI patients (sPESI = 0) from a prospective PE registry. From a large field of candidate variables, we developed, and compared prognostic accuracy of, full and reduced random forest models (with and without RV assessment variables, respectively) on a validation database. We reported variable importance plots from full random forest and provided odds ratios for statistical inference of importance from multivariable logistic regression. Outcomes were death, cardiac arrest, hypotension, dysrhythmia, or respiratory failure within 5 days of PE. RESULTS: Of 1736 patients, 610 (35.1%) were low risk by sPESI and 72 (11.8%) experienced early deterioration. Of the 610, RV abnormality was present in 157 (25.7%) by CT, 121 (19.8%) by echocardiography, 132 (21.6%) by natriuretic peptide, and 107 (17.5%) by troponin. For deterioration, the receiver operating characteristics for full and reduced random forest prognostic models were 0.80 (0.77-0.82) and 0.71 (0.68-0.73), respectively. RV assessments were the top four in the variable importance plot for the random forest model. Echocardiography and CT significantly increased predicted probability of 5-day clinical deterioration by the multivariable logistic regression. CONCLUSIONS: A PE triaging strategy with RV imaging assessments had superior prognostic performance at classifying low risk for 5-day clinical deterioration versus one without.

Development and validation of a prognostic tool: Pulmonary embolism short-term clinical outcomes risk estimation (PE-SCORE).

OBJECTIVE: Develop and validate a prognostic model for clinical deterioration or death within days of pulmonary embolism (PE) diagnosis using point-of-care criteria. METHODS: We used prospective registry data from six emergency departments. The primary composite outcome was death or deterioration (respiratory failure, cardiac arrest, new dysrhythmia, sustained hypotension, and rescue reperfusion intervention) within 5 days. Candidate predictors included laboratory and imaging right ventricle (RV) assessments. The prognostic model was developed from 935 PE patients. Univariable analysis of 138 candidate variables was followed by penalized and standard logistic regression on 26 retained variables, and then tested with a validation database (N = 801). RESULTS: Logistic regression yielded a nine-variable model, then simplified to a nine-point tool (PE-SCORE): one point each for abnormal RV by echocardiography, abnormal RV by computed tomography, systolic blood pressure < 100 mmHg, dysrhythmia, suspected/confirmed systemic infection, syncope, medico-social admission reason, abnormal heart rate, and two points for creatinine greater than 2.0 mg/dL. In the development database, 22.4% had the primary outcome. Prognostic accuracy of logistic regression model versus PE-SCORE model: 0.83 (0.80, 0.86) vs. 0.78 (0.75, 0.82) using area under the curve (AUC) and 0.61 (0.57, 0.64) vs. 0.50 (0.39, 0.60) using precision-recall curve (AUCpr). In the validation database, 26.6% had the primary outcome. PE-SCORE had AUC 0.77 (0.73, 0.81) and AUCpr 0.63 (0.43, 0.81). As points increased, outcome proportions increased: a score of zero had 2% outcome, whereas scores of six and above had ≥ 69.6% outcomes. In the validation dataset, PE-SCORE zero had 8% outcome [no deaths], whereas all patients with PE-SCORE of six and above had the primary outcome. CONCLUSIONS: PE-SCORE model identifies PE patients at low- and high-risk for deterioration and may help guide decisions about early outpatient management versus need for hospital-based monitoring.