Sociodemographic and clinical features predictive of SARS-CoV-2 test positivity across healthcare visit-types

Background Despite increased testing efforts and the deployment of vaccines, COVID-19 cases and death toll continue to rise at record rates. Health systems routinely collect clinical and non-clinical information in electronic health records (EHR), yet little is known about how the minimal or intermediate spectra of EHR data can be leveraged to characterize patient SARS-CoV-2 pretest probability in support of interventional strategies. Methods and findings We modeled patient pretest probability for SARS-CoV-2 test positivity and determined which features were contributing to the prediction and relative to patients triaged in inpatient, outpatient, and telehealth/drive-up visit-types. Data from the University of Washington (UW) Medicine Health System, which excluded UW Medicine care providers, included patients predominately residing in the Seattle Puget Sound area, were used to develop a gradient-boosting decision tree (GBDT) model. Patients were included if they had at least one visit prior to initial SARS-CoV-2 RT-PCR testing between January 01, 2020 through August 7, 2020. Model performance assessments used area-under-the-receiver-operating-characteristic (AUROC) and area-under-the-precision-recall (AUPR) curves. Feature performance assessments used SHapley Additive exPlanations (SHAP) values. The generalized pretest probability model using all available features achieved high overall discriminative performance (AUROC, 0.82). Performance among inpatients (AUROC, 0.86) was higher than telehealth/drive-up testing (AUROC, 0.81) or outpatient testing (AUROC, 0.76). The two-week test positivity rate in patient ZIP code was the most informative feature towards test positivity across visit-types. Geographic and sociodemographic factors were more important predictors of SARS-CoV-2 positivity than individual clinical characteristics. Conclusions Recent geographic and sociodemographic factors, routinely collected in EHR though not routinely considered in clinical care, are the strongest predictors of initial SARS-CoV-2 test result. These findings were consistent across visit types, informing our understanding of individual SARS-CoV-2 risk factors with implications for deployment of testing, outreach, and population-level prevention efforts.

Role of echocardiography to avoid misclassification of pulmonary hypertension associated with left heart disease as precapillary pulmonary hypertension

Background The proportion of patients (pts) diagnosed with pulmonary arterial hypertension (PAH) at a more advanced age and/or with more risk factors for left ventricular diastolic dysfunction is increasing. Therefore, it can be challenging to differentiate PH associated with left heart disease (PH-LHD, PHpost) from other precapillary forms of PH (PHpre). Purpose We analyzed the performance of the Opotowsky (OS), D'Alto (DS), and simplified D'Alto (sDS) echocardiographic scores according to the pretest probability (before right heart catheterization – RHC) of PH-LHD in pts with suspected PH submitted to RHC to identify the hemodynamic phenotype. Methods 37 consecutive stable pts (3/2018–3/2020) with a tricuspid regurgitation peak velocity >2.8 m/s were prospectively included (21F, 49±17 yrs). Blinded transthoracic echocardiography was performed within 2 hours of RHC. We assessed OS (−2 to 2 points) and DS/sDS (0 to 34/7 points). We estimated cardiac index (thermodilution) and hemodynamic parameters using standard formulas. If PA occlusion pressure (PAOP) cannot properly be measured at end-expiration, we assessed left ventricular end-diastolic pressure (LVEDP). PH was defined as a mean PA pressure (mPAP) ≥25 mmHg. PAOP/LVEDP >15 mmHg defined PHpost. If the PAOP/LVEDP was between 13–15 mmHg in an I pt, a volume challenge was done. We categorized pts according to the pretest probability of PH-LHD proposed in the 6th WSPH based on the combination of 7 noninvasive variables (age, presence of CV comorbidities, presence of current or paroxysmal atrial fibrillation, prior cardiac intervention, presence of structural LHD, presence of left bundle branch/LV hypertrophy or RV strain in ECG, presence of left atrial dilatation/grade >2 mitral flow in Echo). The individual average probability was calculated by assigning a score of 1, 2, and 3 for each variable (1 = low (L), 2 = intermediate (I), and 3 = high (H) probability) rounding the average of the sum of values allocated for each variable to the nearest integer. Nonparametric ROC plots assessed the performance of echo-scores. Results All pts had PH. 19 pts showed PHpost, 10/19 with PVR >3Wu (Combined PHpost). All scores were lower in PHpost compared to PHpre pts (p<0.05) (Table 1). ROC area was >0.9 with a similar Youden index (0.83) among the three scores (p<0.05) (Figure 1). 17 PHpost with H pts were correctly identified by all scores (94–100%). In 15 PHpre with L pts OS performed better than DS/sDS (93 vs. 80%). In 3 PHpre and 2 PHpost with I pts, DS/sDS performed better than OS (100 vs. 80%). Conclusion The use of simple echo-scores could facilitate the screening of the hemodynamic phenotype in pts with PH, regardless of the pretest probability of PH-LHD. D'Alto scores might have some advantage compared to OS to classify the intermediate pretest probability of PH pts correctly. FUNDunding Acknowledgement Type of funding sources: Public hospital(s). Main funding source(s): Centro Cardiovascular Universitario. Hospital de Clínicas. Facultad de Medicina. Universidad de la República Table 1. Echo & Hemodynamic Data Figure 1. ROC curves of Echo scores

Two clinical tests assessing long thoracic nerve function to determine C5 and C6 root graft eligibility in patients with brachial plexus injury

OBJECTIVE Identifying roots available for grafting is of paramount importance prior to reconstructing complex injuries involving the brachial plexus. This is traditionally achieved by combining input from both clinical examinations and imaging studies. In this paper, the authors describe and evaluate two new clinical tests to study long thoracic nerve function and, consequently, to predict the status of the C5 and C6 roots after global brachial plexus injuries. METHODS From March 2020 to December 2020, in 41 patients undergoing brachial plexus repair, preoperative clinical assessments were performed using modified C5 and C6 protraction tests, C5 and C6 Tinel’s signs, and MRI findings to predict whether graft-eligible C5 and C6 roots would be identified intraoperatively. Findings from these three assessments were then combined in a logistic regression model to predict graft eligibility, with overall predictive accuracies calculated as areas under receiver operating characteristic curves. RESULTS In the 41 patients, the pretest probability of C5 root availability for grafting was 85% but increased to 92% with a positive C5 protraction test and to 100% when that finding was combined with a positive C5 Tinel’s sign and favorable MRI findings. The pretest probability of C6 root availability was 40%, which increased to 84% after a positive C6 protraction test and to 93% when the protraction test result concurred with Tinel’s test and MRI findings. CONCLUSIONS Combining observations of the protraction tests with Tinel’s sign and MRI findings accurately predicts C5 and C6 root graft eligibility.

Infection and infectivity: Utility of rapid antigen tests for the diagnosis of COVID-19

Detection of SARS-CoV-2 proteins is commercially available in the form of lateral-flow rapid antigen test for the point-of-care diagnosis of COVID-19. This platform has been validated for symptomatic and asymptomatic individuals, for diagnosis or screening, and as part of single or sequential diagnostic strategies. Although in general less sensitive than amplification techniques, antigen tests may be particularly valid during the first days of symptoms and to detect individuals with greater viral load, thereby with enhanced chances of viral transmission. The simplicity of antigen tests make them very suitable to discard infection in settings with low pretest probability, and to detect infection in case of higher chances of having COVID-19.

Using body temperature and variables commonly available in the EHR to predict acute infection: a proof-of-concept study showing improved pretest probability estimates for acute COVID-19 infection among discharged emergency department patients

Objectives Obtaining body temperature is a quick and easy method to screen for acute infection such as COVID-19. Currently, the predictive value of body temperature for acute infection is inhibited by failure to account for other readily available variables that affect temperature values. In this proof-of-concept study, we sought to improve COVID-19 pretest probability estimation by incorporating covariates known to be associated with body temperature, including patient age, sex, comorbidities, month, and time of day. Methods For patients discharged from an academic hospital emergency department after testing for COVID-19 in March and April of 2020, we abstracted clinical data. We reviewed physician documentation to retrospectively generate estimates of pretest probability for COVID-19. Using patients’ COVID-19 PCR test results as a gold standard, we compared AUCs of logistic regression models predicting COVID-19 positivity that used: (1) body temperature alone; (2) body temperature and pretest probability; (3) body temperature, pretest probability, and body temperature-relevant covariates. Calibration plots and bootstrap validation were used to assess predictive performance for model #3. Results Data from 117 patients were included. The models’ AUCs were: (1) 0.69 (2) 0.72, and (3) 0.76, respectively. The absolute difference in AUC was 0.029 (95% CI −0.057 to 0.114, p=0.25) between model 2 and 1 and 0.038 (95% CI −0.021 to 0.097, p=0.10) between model 3 and 2. Conclusions By incorporating covariates known to affect body temperature, we demonstrated improved pretest probability estimates of acute COVID-19 infection. Future work should be undertaken to further develop and validate our model in a larger, multi-institutional sample.