Speeding triage of COVID-19 patients

Researchers in the U.K. have developed two early-detection models for COVID-19 to aid in the rapid triage of patients, one for patients who present at the emergency department (ED) and a second for the subset of patients who are admitted to the hospital. To do so, they trained linear and nonlinear machine learning classifiers to distinguish COVID-19 patients from prepandemic controls, using routinely collected electronic health record data such as laboratory test, blood gas measurements and vital signs. The models were used to assess a total of 115,394 patients, 72,310 of whom were admitted to the hospital. According to the results, the ED model demonstrated 77.4% sensitivity and 95.7% specificity for COVID-19 among all patients attending the hospital. For the subset of patients admitted to hospital, the admissions model had 77.4% sensitivity and 94.8% specificity. Both models demonstrated high negative predictive values across a range of prevalences. When prospectively validated in all patients presenting or admitted to Oxford University Hospitals over a two-week period, the ED model achieved 92.3% accuracy and the admissions model, 92.5%, compared with polymerase chain reaction (PCR) test results. “Our model performed effectively as a screening test for COVID-19, excluding the illness with high-confidence by use of clinical data routinely available within 1 hour or presentation to hospital,” the authors wrote. The added the approach is “rapidly scalable, fitting within the existing laboratory testing infrastructure and standard of care of hospitals in high-income and middle-income countries.” They published their work online Dec. 11, 2020, in The Lancet Digital Health.

New tests distinguishes COVID-19 and flu

Florence, Italy-based Menarini Diagnostics SrL has launched a new test that can identify if a patient is positive for COVID-19 or influenza A or B in 20 minutes, based on a single swab. The CE-marked test is performed on the VitaPCR platform, a point-of-care instrument being used in hundreds of European facilities. To conduct the test, a places a nasal or oropharyngeal swab in a vial containing a liquid. After being shaken, the contents are poured into another vial containing a reagent. That vial is closed and inserted into the VitaPCR system, initiating the analysis. The kit is currently available in Austria, Belgium, Germany, Greece, Italy, Luxembourg, the Netherlands, Portugal, Spain and the U.K.

Predicting heart failure

Researchers at the University of Eastern Finland and Kuopio University Hospital have identified new inflammatory biomarkers and cell energy metabolites associated with an increased risk of incident heart failure, hospitalizations with an overnight stay and a confirmed diagnosis of heart failure. The team measured conventional risk factors and inflammatory biomarkers. and measured systemic metabolism via NMR analysis in 10,106 men participating in the Metabolic Syndrome in Men study without HF at baseline. Over an eight-year follow-up, levels of certain biomarkers were linked to the development of HF in 172 subjects. These new biomarkers included adiponectin, high-sensitivity C-reactive protein, glycoprotein acetyls, alanine, phenylalanine, glycerol and pyruvate, which fuels the heart muscle. After adjusting for age, BMI, diabetes and statin medication, only adiponectin, pyruvate and urinary albumin excretion rate remained statistically significant. Detection of the novel metabolic and inflammatory biomarkers for HF suggests “early activation of respective pathways in the pathogenesis” of the disease, the researchers said. The study was published online Dec. 5, 2020, in ESC Heart Failure.

Expanding cryptosporidiosis testing in developing regions

Cryptosporidiosis is a common cause of diarrhea in children in low-income areas, but it can be difficult to diagnose. Researchers from Norway, Denmark and Ethiopia evaluated two methods for detecting Cryptosporidium species, light-emitting diode fluorescence microscopy with auramine-phenol staining (LED-AP), which is currently used for tuberculosis testing, and a lateral flow test that does not require refrigeration and was developed for Cryptosporidium spp detection. The prospective diagnostic accuracy study took stool samples from children under age five who had either diarrhea or dysentery and tested them for Cryptosporidium spp using LED-AP and lateral flow test strip. The accuracy was then verified against a reference standard comprising quantitative immunofluorescent antibody test, ELISA and quantitative PCR. According to the team, LED-AP’s sensitivity for cryptosporidiosis was 88% and its specificity was 99%. The sensitivity and specificity of the lateral flow test strip were 89% and 99%, respectively. Given these results, they concluded that LED-AP “should be considered as a dual-use technology that can be easily integrated with existing laboratory infrastructures in low-resource settings.” The lateral flow test strip “provides an alternative that does not require microscopy, although purchase cost of the test strip is unknown as it is not yet available on the market.” Their work appeared in the Dec. 3, 2020, online issue of The Lancet Infectious Diseases.