Project topics and materials
Get Complete Project Material File(s) Now! »
CHAPTER 3 CONCLUSION AND RECOMMENDATIONS
Conclusion
There is an unacceptably high volume of MDOs within SA’s infant HIV testing programme. More than 17 000 specimens fail to yield either a positive or negative result per annum, with very poor follow-up testing performed among infants. Pre-analytical errors, which include samples rejected on account of poor sample quality, insufficient sample volume and clerical errors, comprise the bulk of MDOs. These can be addressed by utilizing routine laboratory data to direct in-service training. In contrast, analytical MDOs require addressing laboratory practice. Indeterminate results represent the majority of analytical MDOs, comprising approximately 3 000 HIV PCR results each year.
Indeterminate HIV PCR results are inconclusive instrument-positive results. As infants with perinatal HIV-infection require early diagnosis and prompt initiation of cART on account of rapid disease progression and early mortality, it is imperative to keep indeterminate results to a minimum (i.e. only infants with a high probability of having a false-positive result should be verified as indeterminate). Whereas the rate of indeterminate results (as a proportion of all instrument-positive results) was fairly stable between 2013 and 2015 at approximately 17%, this increased subsequent to the introduction of routine birth testing in June 2015. In 2017, among infants <7 days of age, 1 331/3 577 (37%) of instrument-positive results were verified as indeterminate. Importantly, more than half of those with an indeterminate birth test were found to be HIV-infected within a cohort of birth-tested infants followed-up in Johannesburg between 2014 and 2016, with indeterminate results associated with a significant delay in diagnosis and treatment initiation.105
Current standard operating procedures within the NHLS define valid instrument-positive results with a Ct >33.0 and/or RFI values <5.0 as indeterminate. Importantly, EID Ct value and plasma viral load are inversely correlated, with every one cycle increase in CAP/CTM EID Ct associated with a 0.3 log10 RNA decrease in viral load (95% CI: -0.3–-0.2). HIV-infected infants aged <1-month have been found to have markedly lower pre-treatment viral loads as compared with older ages (P <0.001). Indeed, some infants with confirmed in utero HIV-infection have even been found to be aviraemic at birth. Furthermore, age-adjusted viral load has significantly declined since 2010 (P <0.001), likely attributable to PMTCT practices, with more than a third of infants born to women on an EFV-based regimen found to have virologically suppressive levels of EFV at time of birth testing. Hence, the high proportion of indeterminate HIV PCR results at birth can partly be accounted for by low level viraemia. This highlights the importance of diagnosing HIV at low RNA levels, especially considering there is currently no consensus on what level of viraemia should be considered a true positive result in infants. Guidelines from the United States currently recommend a cut-off ≥5 000 RNA cps/ml in plasma as being diagnostic, although these recommendations are based on findings from older virological assays. More recent commercial assays are likely to be associated with an improved positive predictive value at lower viral load levels on Account of closed systems and the use of enzymes to reduce amplicon contamination.
Unless current NHLS verification practices are revised, increasing diagnostic uncertainty within EID services can be expected in the future. Although RFI was significantly associated with an irreproducible result on the original version of the CAP/CTM assay, this was not the case with the current CAP/CTM v2.0. By dropping RFI from verification criteria and utilising a Ct cut-off of <33.0 only, 97% of instrument-positive results on the CAP/CTM v2.0 assay would have been correctly identified as reproducible and the total number of indeterminate results would have been reduced by 30%. However, subsequent to the introduction of birth testing a higher proportion of HIV-infected infants with a Ct >33.0 can be expected. Reproducibility, in comparison to Ct value, represents a more accurate predictor of a positive HIV-status providing an opportunity to reduce the indeterminate rate further whilst maintaining high diagnostic accuracy. Furthermore, such a strategy provides the opportunity for simplified and standardised infant diagnostic practice across EID platforms. Evaluating the predictive value of reproducibility to differentiate clear positive from indeterminate results on different EID instruments should be considered a research priority.
Recommendations
- Distribution of consolidated laboratory reports on a weekly/ monthly basis, specifying the number and reasons for MDOs, to managers (including at healthcare facilities and laboratories) as a means of directing in-service training to reduce unnecessary rejections within the EID programme.
- Implement a national unique patient identifier, available for all newborn infants at time of delivery, that can be captured within the LIS.
- Prioritize comprehensive care packages for infants with positive HIV PCR result where Ct correlates with a plasma viral load ≥5 log10 copies/ml (i.e. CAP/CTM v2.0 Ct ≤23.0; Xpert Ct ≤31.0).
- Revise indeterminate criteria on CAP/CTM v2.0: Repeat all specimens which yield an instrument-positive HIV PCR result, with all reproducible results verified as positive and irreproducible results verified as indeterminate.
- Revised guidelines for clinical management of infants with indeterminate HIV PCR results as per Appendix C have been incorporated in to the revised 2018 SA National PMTCT Guidelines (Prof Ute Feucht, PMTCT Technical Working Group – personal communication, October 2018).
Future research
Evaluate the accuracy of result-reproducibility to differentiate positive from inconclusive ‘HIV-detected’ PCR results on different EID assays including point-of-care instruments and new technologies
- Evaluate novel EID methods to detect low–level HIV reservoir and viral load
- Cohort monitoring of infants with indeterminate PCR results to determine final HIV-status
- Evaluate sensitivity and specificity of plasma RNA tests to diagnose HIV at birth and 10-weeks of age
- Determine whether there is an association between HIV drug resistance and accuracy of EID assays (including pre-treatment viral load)
- Correlate ARV levels (including NRTI) with infant plasma viral load among HIV-infected infants prior to initiation of cART
Monitor PCR sensitivity and indeterminate results in the context of infants exposed to maternal dolutegravir. Dolutegravir is a potent ARV that concentrates in breast milk and is soon to be available within SA’s public health sector as a component of a fixed-dose combination first-line cART regimen
Declaration
Acknowledgements
Preface
Summary
Presentations and publications
List of abbreviations
Terminology
List of tables
List of figures
Chapter 1: Overview
1.1 Literature review
1.1.1 Introduction
1.1.2. Evolution of South Africa’s PMTCT programme
1.1.3. Mother-to-child transmission landscape in South Africa
1.1.4. Infant diagnostic services within South Africa
1.1.5. Infant HIV testing guidelines
1.1.6. Assays used for early infant diagnosis in South Africa
1.1.7. Result verification and indeterminate HIV PCR results
1.1.8. Diagnostic assay validation
1.1.9. Analytical sensitivity and diagnostic sensitivity
1.1.10 Infant exposure to antiretroviral drugs
1.1.11. Clinical implications of infant antiretroviral drug exposure
1.1.12. The impact of antiretroviral drugs on early infant diagnosis
1.2. Problem statement
1.3. Aim and Objectives
1.4. Overview of the research
1.4.1. Article 1
1.4.2. Article 2
1.4.3. Article 3
1.4.4. Article 4
1.4.5. Article 5
1.4.6. Article 6
1.5. Limitations
1.6. References
Chapter 2: Articles
2.1 Article 1: Missed diagnostic opportunities within South Africa’s early infant diagnosis program, 2010– 2015
2.2. Article 2: Recommendations for the management of indeterminate HIV PCR results within South Africa’s early infant diagnosis programme
2.3. Article 3: Declining Baseline Viremia and Escalating Discordant HIV-1 Confirmatory Results Within South Africa’s Early Infant Diagnosis Program, 2010–2016
2.4. Article 4: Early Infant Diagnosis HIV-1 PCR Cycle-threshold Predicts Infant Viral Load at Birth
2.5. Article 5: Differentiating clearly positive from indeterminate results: A review of irreproducible HIV-1 PCR positive samples from South Africa’s Early Infant Diagnosis Program, 2010–2015
2.6. Article 6: Non-nucleoside reverse transcriptase inhibitor levels among HIV-exposed uninfected infants at the time of HIV PCR testing – findings from a tertiary healthcare facility in Pretoria, South Africa
Chapter 3: Conclusion
3.1. Conclusions, recommendations and future research
Appendices
Appendix A: Research Ethics Committee Approval Certificate
Appendix B. Research Ethics Committee Approval Certificate – Amendment
Appendix C: Recommendations for the management of infants with an indeterminate HIV PCR result
Appendix D. The candidates role in articles presented in thesis
GET THE COMPLETE PROJECT
