Sei sulla pagina 1di 1

Poster number: WEPED891

THE PERFORMANCE OF VIROLOGICAL TESTING FOR EARLY INFANT


DIAGNOSIS OF HIV: A SYSTEMATIC REVIEW
D. Mallampati , A. Hannaford , N. Sugandhi , J. Markby , M. Penazzato
1 2 3 4 4

1
Harvard University, Harvard Medical School, Boston, USA, 2Mount Sinai, Icahn School of Medicine, New York, USA, 3Mount Sinai medical center, Institute of advanced medicine, New York, USA , 4World Health Organization, HIV Department, Geneva, Switzerland

BACKGROUND AND OBJECTIVES METHODS


Scale up of more effective PMTCT interventions requires review of the existing testing algorithm to The search strategy used, aimed to consider studies published from 2009 (date of the most recent WHO
optimize infant testing in the context of wider exposure to ARVs as a result of maternal ART and guidelines on infant testing) and included the following search terms: HIV, HIV-1, HIV-2, AIDS, NAAT/NAT,
infant prophylaxis. Knowledge of the performance of virological assays at different time points and PCR, whole blood, plasma, DBS, newborns, infants, children. PubMed, Embase, Cochrane Library, and LILACS
in the context of ARV exposure is critical to inform such revision.This systematic review informed the as well as conference proceedings from CROI, ICASA, IAS, and the International Workshop on HIV Pediatrics
revision of the World Health Organization (WHO) infant testing algorithm by assessing diagnostic were consulted. Studies were included if investigating performance of virological assays, against a standard
accuracy for virological testing at birth and at 6 weeks in the context of ARV exposure. comparator, in infants exposed to HIV and exposed to maternal ARVs or post-natal prophylaxis.

Two independent reviewers conducted the screening and a third reviewer was consulted to resolve
discordance. Retrieval of missing information was sought by contacting authors. Summary estimate for
performance were calculated. In order to assess the risk of bias the QUADAS-2 tool was used and the overall
assessment of the quality of evidence was performed by using the GRADE approach.

RESULTS
A total of 2203 records were screened with final selection of 5 manuscripts. Three studies were included to Articles identified through Abstract identified through
assess the accuracy of PCR on DBS specimens and in the context of ARV exposure. The pooled sensitivity electronic database (n=2,223) conference proceedings (n=20)
and specificity were 99.4% (98.27, 100) and 99.63% (99.11, 100) respectively. The risk of bias was judged
Papers after removal Excluded because:
as low yet the quality of the evidence, by using the GRADE approach, was considered low due to poor
of duplicates (n=1,968) Not population
generalizability and small sample sizes. of interest: 3
Papers added through No/inappropriate
Two studies were identified to assess PCR performance at birth compared to at 6 weeks of age. The other sources (n=1) comparator: 22
Study in another
calculated pooled sensitivity and specificity were 69.3% (61.1-77.4) and the specificity is 99.91% (99.55- Papers after screening language: 2
100) respectively. The risk of bias in these studies was judged low but the GRADE quality of evidence for by manuscript (n=5) Incomplete data: 4
sensitivity was estimated to be low due to poor generalizability and small sample sizes.
Birth testing 6-week PCR w ARV
(n=2) exposure (n=3)

A. Virological testing at Birth (compared to 6-week PCR) B. Virological testing at 6 weeks with DBS (compared to whole blood)

Patient Population Reference


Study ARV Exposure Index Test Reference Test Study ARV Exposure Index Test
Characteristics Characteristics Standard
Intrapartum prophylaxis: Leelawiwat Cohort of 162 paired Amplicor HIV-1 DNA v1.5
Amplicor or Cobas Monotherapy of short-course ZDV OR
none, zidovudine, dual therapy, triple et al 2009 samples at 2 months. Whole blood
on plasma (RNA) combo short course ZDV and sdNVP NucliSens RNA
therapy (Thailand) Non-breastfeeding
HIV RNA PCR
Burgard et al Cohort of 1293 non- Intrapartum prophylaxis: Lilian et al Cross-sectional study of
on plasma at 6 DNA PCR on
2012 (France) breastfeeding infants None, IV ZDV, sdNVP + IV ZDV, 2010 (South 125 infants (4-8 weeks). sdNVP to mother and infant NucliSens RNA
HIV DNA PCR on PBMC months whole blood
sdNVP/other Africa) Unknown breastfeeding
(3 different methods)
Neonatal prophylaxis: ZDV, dual Maternal Regimen: NVP pre partum
therapy, triple therapy Cross-sectional within followed by ZDV/3TC post partum (3),
Yapo et al Biocentric (DNA) Biocentric
cohort of 71 infants ZDV/3TC/NVP (13), TDF/FTC/NVP, D4T/3TC/
Lilian et al Cohort of 710 mother- Maternal AZT from 28 weeks with HIV DNA PCR on DB5
2013 (Cote NVP (9), ZDV/3TC/ABC (1) DNA Kit
sdNVP, infant AZT for 1 week, if DNA PCR on (4-8 weeks). 49/71 were
2010 (South infant-pairs who were CAP/CT on DBS dIvore) on cell pellets
suboptimal maternal prophylaxis, whole blood at 6w breastfeeding Infant prophylaxis: ZDV/3TC/NVP (11), TDF/ Amplicor DNA v1.5
Africa) breast-feeding Aptima on DBS
infant recieved 28 days AZT FTC/NVP, ZDV (3), unspecified (3) HIV RNA Cell Kit

Risk of bias Applicability Risk of bias Applicability


Proportion of studies with low, high, or unclear risk Proportion of studies with low, high, or unclear concernes Proportion of studies with low, high, or unclear risk Proportion of studies with low, high, or unclear concernes
of bias (n=2) about applicability (n=2) of bias (n=3) about applicability (n=3)

Flow and Timing Reference Test Flow and Timing Reference


QUADDAS-2 Domains

QUADDAS-2 Domains

QUADDAS-2 Domains

QUADDAS-2 Domains

Standard
Low Risk Low Concern Low Risk Low Concern
Reference Test Reference Test
High Risk Index Test High Concern High Risk Index Test High Concern
Index Test Unclear Unclear Index Test Unclear Unclear
Risk Concern Risk Concern
Patient: Selection Patient: Selection Patient: Selection Patient: Selection

0 0.2 0.4 0.6 0.8 1.0 0 0.2 0.4 0.6 0.8 1.0 0 0.2 0.4 0.6 0.8 1.0 0 0.2 0.4 0.6 0.8 1.0

ROC curve
Sensitivity Sensitivity
Lilian et al (DNA) 68.42 (51.30, 82.50)
(Receiver Operating Characteristic)
Leelawiwat et al (Amplicor DNA) 99.56 (96.20, 99.60) Leelawiwat et al (Amplicor DNA) 99.77 (97.96, 99.79)
Lilian et al (CAP/CTM) 76.32 (59.80, 82.50)
Lilian et al (APTIMA) 76.32 (59.80, 82.50) 1.0 Leelawiwat et al (NucliSens RNA) 99.56 (96.20, 99.60) Leelawiwat et al (NucliSens RNA) 99.77 (97.96, 99.79)
Burgard et al (Amplicor/COBAS) 56.67 (37.40, 74.50)
Burgard et al (DNA) 56.67 (37.40, 74.50) 0.8 Yapo et al (Biocentric DNA) 96.29 (71.41, 96.91) Yapo er al (Biocentric DNA) 99.62 (96.71, 99.65)
Subtotal 69.26 (61.10, 77.42)
Yapo et al (Amplicor DNA) 96.29 (71.41, 96.91) Yapo et al (Amplicor DNA) 99.62 (96.71, 99.65)
Sensitivity

Sensitivity 0.6
Lilian et al (DNA) 100.00 (94.60, 100.00) Yapo et al (Generic RNA) 96.29 (71.41, 96.91) Yapo et al (Generic RNA) 99.62 (96.71, 99.65)
Lilian et al (CAP/CTM) 100.00 (94.80, 100.00) 0.4
Lilian et al (APTIMA) 98.11 (93.30, 99.80) Lilian et al (NucliSens RNA) 98.30 (85.99, 98.51) Lilian et al (NucliSens RNA) 95.10 (90.35, 98.30)
Burgard et al (Amplicor/COBAS) 100.00 (99.10, 100.00) 0.2 Subtotal 99.43 (98.27, 100.60) Overall 99.63 (99.11, 100.15)
Burgard et al (DNA) 99.77 (98.70, 100.00)
Subtotal 99.91 (98.55, 100.27) 0.0 0 100 0 100
0 100 1.0 Specificity 0.0 *ROC not performed due to limitations in data points

CONCLUSION
Our systematic review shows that there is currently no evidence to suggest that virological assays on DBS have poor performance when infants are exposed to ARVs. However only few subjects in the studies
were infants exposed to triple maternal ART and postnatal prophylaxis.
The performance of PCR at birth demonstrated low sensitivity and high specificity. However, this may reflect the inability of PCRs to detect intrapartum infections rather than a lack of accuracy of the assays
used. Sensitivity of PCR at birth may therefore vary depending on the transmission dynamics that are influenced by the PMTCT intervention provided.
Further research to assess accuracy of PCR at different time-points and in the context of more effective PMTCT interventions is urgently needed.

REFERENCES
Burgard, M., et al (2012). Performance of HIV-1 DNA or HIV-1 RNA Tests for Early Diagnosis of Perinatal HIV-1 Infection during Anti-Retroviral Prophylaxis. The Journal of Pediatrics, 160 (1): 60-66.
Leelawiwat, W., et al (2009). Dried blood spots for the diagnosis and quantitation of HIV-1: stability studies and evaluation of sensitivity and specificity for the diagnosis of infant HIV-1 infection in Thailand. J Virol Methods, 155(2), 109117.
Lilian, R. R., et al (2010). Early diagnosis of human immunodeficiency virus-1 infection in infants with the NucliSens EasyQ assay on dried blood spots. Journal of Clinical Virology: The Official Publication of the Pan American Society for Clinical Virology, 48(1), 403.
Lilian, R. R., et al (2012). Early diagnosis of in utero and intrapartum HIV infection in infants prior to 6 weeks of age. Journal of Clinical Microbiology, 50(7), 23737.
UNAIDS. (2013). GLOBAL REPORT: UNAIDS Report on the global AIDS epidemic 2013.
Yapo, V., et al (2013). Evaluation of dried blood spot diagnosis using HIV1-DNA and HIV1-RNA Biocentric assays in infants in Abidjan, Cte dIvoire. The Pedi-Test DBS ANRS 12183 Study. Journal of Virological Methods, 193(2), 43945.

Potrebbero piacerti anche