Diagnostic accuracy of point-of-care nucleic acid-based testing for HIV infection among infants, children

A systematic review and meta-analysis found that point-of-care (POC) nucleic acid-based testing (NAT) for HIV-1/HIV-2 infection had high sensitivity and specificity among children and infants who were exposed to HIV. These findings were published in the Cochrane Database of Systematic Reviews.

Publication databases were searched for studies of HIV testing in children and infants aged 18 months and younger. A total of 12 studies comprising 15,120 children or infants were included. All studies used POC NAT testing and were conducted in Sub-Saharan Africa between 2014 and 2020.

The studies were of cross-sectional (n=11) or unclear (n=1) designs. Most studies were assessed as having a decreased risk of bias, however, 1 study was found to have an increased risk for selection bias. In addition, the researchers noted concerns regarding the applicability of the index test because 5 studies conducted POC testing in laboratory settings only.

Of note, most participants (71%) underwent testing at 14 weeks of age or younger, and most studies (n=10) were conducted in the field instead of a laboratory setting. Tests were conducted using whole blood (n=11) or dried blood (n=4) specimens. The specific tests were Alere q HIV-1/HIV-2 qualitative test (n=6), Xpert HIV-1 qualitative test (n=6), and SAMBA HIV-1 qualitative test (n=3).

Of the 3 diagnostic tests, sensitivities ranged between 83% and 100% and specificities between 99% and 100%. The pooled analysis showed an overall sensitivity of 98.6% (95% CI, 96.1%-99.5%) and overall specificity of 99.9% (95% CI, 99.7%-99.9%).

After stratification by patient age, tests performed at birth had a sensitivity of 99.0%, and tests performed at either 12 or 18 months of age and younger had sensitivities of 96.6% and 97.9%, respectively.  On sensitivity analysis of the 3 diagnostic tests, sensitivity was 99.2% for the Xpert test, 96.6% for the Alere test, and 97.3% for the SAMBA test. Tests performed in the field had a sensitivity of 98.7% vs 97.4% for tests performed in laboratory settings. Assessment of whole blood and dried blood specimens had a sensitivity of 98.4% and 97.7%, respectively.

This analysis was limited in its ability to assess bias risk as many of the included studies incompletely reported their methods.

The researchers noted that POC NAT for HIV-1/HIV-2 infection was highly sensitive and specific among children and infants who were exposed to HIV infection. In addition, “these tests could therefore complement or replace laboratory-based viral assays,” the researchers concluded.

By Jessica Nye, PhD

Reference

Ochodo EA, Guleid F, Deeks JJ, Mallett S. Point-of-care tests detecting HIV nucleic acids for diagnosis of HIV-1 or HIV-2 infection in infants and children aged 18 months or less. Cochrane Database Syst Rev. 2021;8(8):CD013207. doi:10.1002/14651858.CD013207.pub2