Format

Send to

Choose Destination
Virol J. 2015 May 14;12:75. doi: 10.1186/s12985-015-0306-4.

Dilution testing using rapid diagnostic tests in a HIV diagnostic algorithm: a novel alternative for confirmation testing in resource limited settings.

Author information

1
Médecins Sans Frontières, Amsterdam, The Netherlands. Leslie.shanks@amsterdam.msf.org.
2
Médecins Sans Frontières, London, UK. ruby.siddiqui@london.msf.org.
3
Ethiopian Health and Nutrition Research Institute, Addis Ababa, Ethiopia. almaz_abe1@yahoo.com.
4
Médecins Sans Frontières, Amsterdam, The Netherlands. erwan.piriou@amsterdam.msf.org.
5
London School of Hygiene and Tropical Medicine, London, UK. Neil.Pearce@lshtm.ac.uk.
6
London School of Hygiene and Tropical Medicine, London, UK. cono.ariti@lshtm.ac.uk.
7
Médecins Sans Frontières, Amsterdam, The Netherlands. jmasiga.jm@gmail.com.
8
Ethiopian Health and Nutrition Research Institute, Addis Ababa, Ethiopia. libse77m@gmail.com.
9
Médecins Sans Frontières, Amsterdam, The Netherlands. jos_wazome@yahoo.co.uk.
10
Médecins Sans Frontières, Amsterdam, The Netherlands. Koert.Ritmeijer@amsterdam.msf.org.
11
Médecins Sans Frontières, Amsterdam, The Netherlands. derryck.klarkowski@gmail.com.

Abstract

BACKGROUND:

Current WHO testing guidelines for resource limited settings diagnose HIV on the basis of screening tests without a confirmation test due to cost constraints. This leads to a potential risk of false positive HIV diagnosis. In this paper, we evaluate the dilution test, a novel method for confirmation testing, which is simple, rapid, and low cost. The principle of the dilution test is to alter the sensitivity of a rapid diagnostic test (RDT) by dilution of the sample, in order to screen out the cross reacting antibodies responsible for falsely positive RDT results.

METHODS:

Participants were recruited from two testing centres in Ethiopia where a tiebreaker algorithm using 3 different RDTs in series is used to diagnose HIV. All samples positive on the initial screening RDT and every 10th negative sample underwent testing with the gold standard and dilution test. Dilution testing was performed using Determine™ rapid diagnostic test at 6 different dilutions. Results were compared to the gold standard of Western Blot; where Western Blot was indeterminate, PCR testing determined the final result.

RESULTS:

2895 samples were recruited to the study. 247 were positive for a prevalence of 8.5 % (247/2895). A total of 495 samples underwent dilution testing. The RDT diagnostic algorithm misclassified 18 samples as positive. Dilution at the level of 1/160 was able to correctly identify all these 18 false positives, but at a cost of a single false negative result (sensitivity 99.6 %, 95 % CI 97.8-100; specificity 100 %, 95 % CI: 98.5-100). Concordance between the gold standard and the 1/160 dilution strength was 99.8 %.

CONCLUSION:

This study provides proof of concept for a new, low cost method of confirming HIV diagnosis in resource-limited settings. It has potential for use as a supplementary test in a confirmatory algorithm, whereby double positive RDT results undergo dilution testing, with positive results confirming HIV infection. Negative results require nucleic acid testing to rule out false negative results due to seroconversion or misclassification by the lower sensitivity dilution test. Further research is needed to determine if these results can be replicated in other settings.

TRIAL REGISTRATION:

ClinicalTrials.gov, NCT01716299 .

PMID:
25972188
PMCID:
PMC4432962
DOI:
10.1186/s12985-015-0306-4
[Indexed for MEDLINE]
Free PMC Article

Supplemental Content

Full text links

Icon for BioMed Central Icon for PubMed Central
Loading ...
Support Center