Latest COVID-19 research looks at predictive biomarkers, timing of testing, health care worker infections

Lactate dehydrogenase and other biomarkers appear to predict risk of critical illness and death, and testing soon after exposure is associated with high rates of false negatives, new studies showed.


Several recent studies identified predictors of worse outcomes in patients hospitalized with COVID-19.

A retrospective analysis from China, published by JAMA Internal Medicine on May 12, used a development cohort of 1,590 patients and a validation cohort of 710 patients to create a clinical risk score to predict critical illness. The score is based on 10 variables that were found to be independent predictors of need for ICU care (chest radiographic abnormality, age, hemoptysis, dyspnea, unconsciousness, number of comorbidities, cancer history, neutrophil-to-lymphocyte ratio, lactate dehydrogenase [LDH] level, and direct bilirubin level), and it has been translated into an online risk calculator.

A systematic review and meta-analysis, published by Clinical Infectious Diseases on May 14, identified some of the same predictors of critical illness and some different ones. It included 45 studies of 4,203 patients with COVID-19 and found that ICU admission was predicted by elevations in leukocyte count and levels of alanine aminotransferase, aspartate transaminase, LDH, and procalcitonin. Risk of developing acute respiratory distress syndrome (ARDS) was predicted by LDH, while mortality was associated with LDH and leukocyte count. The analysis also looked at associations between treatment and outcomes and found no significant benefit from lopinavir-ritonavir on risk of mortality or ARDS. Corticosteroid use was associated with higher risk of ARDS.

Another study, published by Nature Machine Intelligence on May 14, used a machine learning model and blood samples from 485 patients in China to identify predictors of mortality from COVID-19. It found three predictive biomarkers: LDH, lymphocyte count, and high-sensitivity C-reactive protein. The results provide “a simple and intuitive clinical test to precisely and quickly quantify the risk of death,” potentially enabling early intervention, the study authors said. They also suggested that lymphocytes may serve as a potential therapeutic target.

Another recent study, published by Clinical Infectious Diseases on May 15, looked at COVID-19 infections and outcomes among health care workers in China. Out of a total of 2,457 health care worker cases, 52.06% were nurses, 33.62% were doctors, and 14.33% were other staff. The case infection rate was 2.22% among nurses and 1.92% among doctors. Although the case infection rate of health care workers was significantly higher than that of non-health care workers (2.10% vs. 0.43%), their case fatality rate was significantly lower (0.69% vs. 5.30%). The results should spur protective measures for health care workers beyond an adequate supply of personal protective equipment, the authors said. “Other measures should be considered, including nutritious food supply, adequate rest time, and societal, familial, and psychological support.”

Finally, testing for SARS-CoV-2 soon after exposure results in a high false-negative rate, according to an analysis published by Annals of Internal Medicine on May 13. It included seven studies with 1,330 patients (some hospitalized, some not) that provided data on reverse transcriptase polymerase chain reaction (RT-PCR) test performance by time since symptom onset or exposure. The study found that with a typical time of symptom onset on day 5, the probability of a false-negative result in an infected person decreases from 100% on day 1 to 67% on day 4, 38% on day 5, and 20% on day 8. It then increases from 21% on day 9 to 66% on day 21. Given that the false-negative rate was lowest on day 8, the results suggest clinicians “should consider waiting 1 to 3 days after symptom onset to minimize the probability of a false-negative result,” the study authors said. They noted that “if testing is done immediately after exposure, the pretest probability is equal to the negative posttest probability, meaning that the test provides no additional information about the likelihood of infection” and that “if clinical suspicion is high, infection should not be ruled out on the basis of RT-PCR alone, and the clinical and epidemiologic situation should be carefully considered.” The latest edition of Annals On Call also addressed testing for COVID-19, as did a recent Q&A in ACP Hospitalist Weekly.