1402608033

Emory Internal Medicine Residency: COVID-19 Visual Series

CON/ID-19 Antibody Testing: A Cochrane Review

A Cochrane Review assesses the current evidence on antibody testing for COVID-19.

total studies included

COVID-19 antibody test accuracy studies ofany design

$ 75%-

1$ 50%-

72.2%

(6 3 5 to 79.5]

based on smaller

15,976 total samples

30.1%

- 95% Confidence lnterval (Cl)

sample sizes & fewer studies

COVID+

COVID-

Symptom

onset

Insufficient

evidence

(21.4 to 40.7]

53.37%

46.63%

doys

- 1 to 7 -

doys

■ 8 to 14 ■

doys

-15 to 21-

doys

-21 to 35-

doys

-35 and on-

Most studies defined COVID+ by positive RT-PCR, though some included

clinically defined RT-PCR-negative cases

259

Samples:

©Ali tests were specific:

608

692

152

153

mmmm

CLIA

ChemiLuminescence ImmunoAssays ELISA

Enzyme-Linked ImmunoSorbent Assays CGIA

Colloidal Gold Immunochromatographic Assays

or greater specificity for all tests with Cl of no morę than ±1%

© Some differences between test technologies

For IgG/IgM, from the limited evaluations studied,

CLIA methods (97.5% (94.0 to 99.0]) appeared morę sensitive than ELISA (90.7% (83.3 to 95.0D or CGIA-based assays (90.7% (82.7 to 95.2]) There were also differences for IgG but no differences for IgM.

LIMITATIONS

-    Few non-hospitalized patients with milder disease included

-    Most studies at high or unclear risk of bias

-    Too few studies to determine accuracy beyond 35 days

EVIDENCE ASSESSED	RESULTS
_	O Sensitivity of all antibody tests are timing dependent:

100%-

—91.4%—

[87.0 to 94.4)

—96.0%—

190.6 to 98.31

Antibody tests are likely useful for detecting COVID-19 infection if used >15 days after symptom onset. The utility of these tests if used >35 days or in asymptomatic or mild cases is uncertain.

Deeks, Jonathan J., et al. 2020.

"Antibody Tests for Identification of Current and Past Infection with SARS-CoV-2." Cochrane Database of Systematic Reviews, no. 6.

https://www.cochranelibrary.com/cdsr/doi/10.1002/14651858.CD013652/full

Content: T. Tyler Daugherty (MS4, @daughertyler) Editing: Caroline Coleman (MD) and Jennifer Spicer (MD, MPH) Review: Helen Wakeford (MA, MPhil) and Liz Bickerdike (MSc)