Routine measurement of cardiovascular biomarkers in patients hospitalized with COVID-19 did not yield additional prognostic information, new research suggests.
The researchers, publishing the findings in a research letter in Circulation, conducted a prospective, observational study that included consecutive adult patients (n=135) hospitalized with COVID-19 in Norway between March 18 and May 4, 2020. The researchers extracted clinical information from electronic medical records and blood samples. They also used multivariable logistic regression models for adjustment of variables like age, sex, and race/ethnicity. A second model adjusted for CVD, BMI, eGFR and symptom duration; a third model adjusted for National Early Warning Scores (NEWS). The primary endpoint of interest was a composite of hospital mortality and admission to the ICU with a need for mechanical ventilation and lasting more than 24 hours.
According to the study results. 40 patients reached the primary endpoint, of whom 36 went to the ICU and eight died. The authors reported that NEWS score at admission, white blood cell count, CRP, ferritin, LD, D-dimer, NT-proBNP and high-sensitivity cardiac troponin were all higher in patients who achieved the primary endpoint. The link was attenuated, however, after adjustment for demographics, and further when adjusting for factors in the second and third models. Only ferritin and LD were associated with the primary study endpoint after adjustment.
“In our prospective study of unselected patients hospitalized with COVID-19, measurements of established cardiovascular biomarkers at admission did not provide prognostic information beyond that obtained from clinical characteristics and NEWS,” the authors wrote in the letter. “Although cardiovascular biomarker levels were higher among ICU-patients and non-survivors, this difference did not persist when accounting for clinical characteristics.”
The authors noted that the findings do contrast with some retrospective reports from China, and the study limitation that it only looked at prognostic value of biomarkers rather than diagnostic.
“Our findings do not support routine measurements of cardiovascular and inflammatory biomarkers on admission for prognostic purposes in patients hospitalized for COVID-19,” the authors concluded.
— John William McEvoy (@johnwmcevoy) September 14, 2020
⚠️Diverging results from study design and ⬆️ selection bias in retrospective studies ❓🤔 https://t.co/OTkuuNfQVr
— Peder Myhre (@pmyhre) September 12, 2020