Congenital infection by Toxoplasma gondii can lead to severe congenital defects such as hydrocephalus, mental retardation and retinochoroiditis, which can be present at birth or develop later in life.1 Early diagnosis is essential in order to determine the appropriate antibiotic therapy, but confirmatory tests are needed when specific IgMs or IgAs are not detected or to confirm potential false-positive results originating from maternal IgM or IgA leaks from the placenta in samples taken at birth or during the first 10 days of life.1 In these circumstances, the western blot test can help by distinguishing maternal antibodies from fetal–neonatal synthesized antibodies, and it can be applied to samples collected at birth or during the first 10 days of life, with sensitivities as high as 90%.2 However, western blot requires the application of different criteria, such as presence of specific bands of IgM or IgA in the child’s serum or densitometric differences between the IgG bands of mother and child.2 Consequently, there is a need to improve the diagnosis by evaluating alternative confirmatory assays that could increase sensitivity or reduce the need for multiple criteria for diagnoses in newborns.
The aim of this study was to assess the avidity and conventional and real-time polymerase chain reaction (PCR) assays for postnatal confirmatory diagnosis of congenital toxoplasmosis in children identified during prenatal and newborn screening programs as having potentially been infected.
MATERIALS AND METHODS
Informed written consent, according to regulation 008430 of 1993 of the Ministry of Health of Colombia, was obtained from all people who agreed to participate in the study. The University of Quindio Institutional Review Board (act number 14, June 23, 2009) approved the study. All mothers and parents or guardians who agreed to participate signed informed consent forms. Children who had positive results on neonatal screening for IgM or IgA anti-Toxoplasma or past histories of treatment for toxoplasmosis during pregnancy or compatible signs were invited to participate in diagnostic confirmation. Written informed consent for additional assessment and follow ups was obtained from the guardians responsible for those children who tested positive. Samples were obtained during the first Colombian multicenter evaluation of congenital toxoplasmosis newborn screening program.3 Additional cases were recruited during prenatal and newborn screening programs in Quindio (Colombia). Peripheral blood samples were collected from mothers and children during the infants’ first 3 months of life and were referred to the “Centro de Investigaciones Biomédicas” at the University of Quindio for confirmatory serological tests.
If the initial physical examinations were normal, parents received instructions about signs to look for and told that, if they saw evidence of those signs, they should promptly consult with physicians. The children were recalled for new physical examinations and samplings each month. Consequent to any alteration identified during the initial clinical assessment, the children were tested for complete blood counts, cerebrospinal fluid analyses and ophthalmologic assessments. All infants identified by prenatal and neonatal screening were followed up until their infection status had been established. A diagnosis of congenital toxoplasmosis was confirmed, as described by the European Network on Congenital Toxoplasmosis (ENCT),4 on the basis of persistence of specific IgG antibodies beyond 12 months of age or by the presence of stable titers in absence of treatment or by the presence of compatible symptoms (cerebral calcifications, retinochoroiditis or hydrocephaly) and a specific IgG and IgM anti-Toxoplasma positive test in the mother and of IgG anti-Toxoplasma antibodies in her child’s serum or by the presence of 3 or more specific bands in IgM western blot or densitometric differences in IgG bands between mother and child by using a commercial western blot (LDbio, France), as previously described.2 Children who became anti-Toxoplasma IgG negative in the absence of treatment were considered to be noninfected.4
The IgG avidity test was performed with a house-made immunoenzymatic assay, as previously described.5 A receptor operating curve was calculated by using different avidity indexes to select the best for sensitivity and specificity in the diagnosis of congenital toxoplasmosis. A conventional PCR for B1 target was performed, as described previously.6 The real-time PCR TaqMan assay for B1 and 529-bp repeat element (RE) targets in peripheral blood samples of newborns was used, as described by our laboratory.7 The results of the conventional or real-time PCR (cycle thresholds or Ct values) were reported as positive or negative.
Sensitivity and specificity and confidence intervals (CIs) at 95% were calculated according to applicable tables in the first sample obtained during the first 3 months of life through the use of GraphPad Prism software, version 5.01 (GraphPad Software, Inc, LaJolla, CA). True positives were cases with criteria of congenital infection and true negatives were children with a disappearance of IgG-specific antibodies during follow ups in the absence of treatment, as described by the ENCT.4
Twenty-three children were defined as infected congenitally by the persistence of IgG anti-Toxoplasma titers during follow-up or by the other criteria proposed by the ENCT. Thirty-one children were uninfected as determined by the disappearance of IgG during follow-up in the absence of treatment. Eleven (47.8%) infected cases were symptomatic, ranging from a severe multisystemic form (hydrocephaly plus severe distress respiratory syndrome), unique ocular retinochoroiditis, to mild unique symptoms (transient splenomegaly). Three symptomatic children died before the first month of life (13%). Three cases were from children who had amniotic fluid PCR-positive tests during prenatal examinations. In 7 infected cases (30.4%), the mothers had received prenatal treatment (4 with spiramycin and 3 with sulfadoxine–pyrimethamine combination). In 25 cases of uninfected children (80%), the mothers had received spiramycin during pregnancy. The median age at the moment of first sampling was 31 days in infected newborns (range 0 to 93 days) and 41 days in uninfected newborns (range 6 to 93 days).
Not all infected children could be tested for all PCR assays due to low amounts of blood samples. In 7 of 10 infected newborns (sensitivity 70%; 95% CI: 34–90) and in 1 of 17 uninfected newborns (specificity 94.1%; 95% CI: 71–99), conventional PCRs were positive. In 4 of 11 infected children (sensitivity 36.3%; 95% CI: 10–69) and in 1 of 33 uninfected children, PCRs for the RE target were positive (specificity 96.9%; 95% CI: 84–99). No children, infected or uninfected, were positive for real-time PCR for the B1 target. PCR did not add additional confirmatory criteria: in all positive-infected children by conventional PCR or real-time PCR for the RE target, other immunological criteria were present such as densitometry or western blot IgM.
The receptor operating curve, with a range of cutoffs of 10, 20, 30, 40 and 50, was calculated for IgG avidity assay. The best performance of IgG avidity assay as a confirmatory test for congenital toxoplasmosis was an avidity index ≤40% (Fig. 1). This cutoff has a sensitivity of 60.8% (95% CI: 38–80) and a specificity of 100% (95% CI: 88–100). Avidity ≤40% was present in 2 cases, whereas other immunological criteria were negative (no IgM or IgA nor IgG densitometric differences).
The use of western blot and ENCT criteria achieved an overall sensitivity of 73.9% (17 of 23 confirmed infected cases). When avidity assays were included, overall sensitivity increased to 82.6% (19 of 23 confirmed infected cases), adding 2 negative cases by western blot. PCR did not add additional cases.
Prenatal treatment reduced significantly positivity of confirmatory test (including avidity assay) being the overall sensitivity of 75% (3/7) in prenatally treated newborns and of 100% (16/16) in untreated newborns (odds ratio: 0.19; 95% CI: 0.06–0.6; P = 0.0145).
Few reports have analyzed alternative methods to the western blot. One retrospective study evaluated anti-Toxoplasma IgG avidity in 32 infected and 20 uninfected newborns, but it was focused to the effect of treatment on avidity.8 Another report evaluated parasite DNA amplification by conventional PCR using a very low number of patients: 6 infected and 6 uninfected children.9 In another study, which was performed in Egypt, 80 preterm children were evaluated by PCR and avidity assay, but definitions of cases or controls did not include follow ups until the disappearance of specific IgGs.10 The results that we obtained during clinical validation indicated that an avidity test can identify newborns with congenital toxoplasmosis with a very good rate of specificity. Our cohort included a large range of clinical presentations and the largest series described until now of uninfected children. The authors of previous work evaluating avidity in congenital toxoplasmosis have been reluctant to use this criterion because of the theoretical possibility that avidity reflects the mothers’ IgG avidity.8 Our results indicated that noninfected children have significant higher avidity indexes than congenitally infected ones. Low IgG avidity is an interesting marker of congenital infection that merits inclusion as a criterion of congenital infection. We believe that low avidity in congenitally infected children reflects the new antibodies produced by the infected children and not the avidity of the mothers’ IgG. The children also have the capacity to produce IgG antibodies, and the low avidity in infected children reflects the immaturity of neo-synthesized antibodies. This fact is proved by the difference in bands that it is possible to see in some cases when comparing IgG results of western blot from mothers and their infected children.2
Surprisingly, real-time PCRs have poor performance compared with conventional PCRs. The same results were obtained by other authors.9 The nested approach for conventional PCR could explain the better results as compared with real-time PCR. However, molecular detection was not better than immunologically based tests; in addition, it presented some false-positive cases. No satisfactory explanation can be given at this time for this phenomenon, but it could suggest transient nonpersistent parasitemia. Laboratory contamination was discarded with negative controls during DNA extraction and PCR amplification. The presence of inhibitors such as hemoglobin, blood proteins and anticoagulants is a problem in blood-related products.6,7 In addition, it is critical to understand that the total amount of DNA in the template preparation and the DNA concentrations from the pathogen and its host, for example, in high concentrations of human DNA, do decrease the sensitivity of molecular assays designed to detect Toxoplasma.6,7
Our results show that prenatal treatment significantly reduces the sensitivity of confirmatory assays. In our country, the prenatal treatment of toxoplasmosis has increasingly posed problems for postnatal diagnosis; consequently, clinicians should be aware of the fact that negative confirmatory tests during the postnatal period do not discard the congenital infections and that children need to be followed until the disappearance of specific IgGs.
1. Gómez-Marin JE Protozoologia Médica: Protozoos Parásitos en el Contexto Latinoamericano. 20101st ed Bogotá Editorial Manual Moderno:65–87
2. Gallego-Marín C, Henao AC, Gómez-Marín JE. Clinical validation of a western blot assay for congenital toxoplasmosis
and newborn screening in a hospital in Armenia (Quindio) Colombia. J Trop Pediatr. 2006;52:107–112
3. Gómez-Marin JE, de-la-Torre A, Angel-Muller E, et al. First Colombian multicentric newborn screening for congenital toxoplasmosis
. PLoS Negl Trop Dis. 2011;5:e1195
4. Lebech M, Joynson DH, Seitz HM, et al. Classification system and case definitions of Toxoplasma gondii infection in immunocompetent pregnant women and their congenitally infected offspring. European Research Network on Congenital Toxoplasmosis
. Eur J Clin Microbiol Infect Dis. 1996;15:799–805
5. Torres E, Gomez Marin JE. Evaluación de una ELISA IgG de avidez para Toxoplasma para el diagnostico en el embarazo y correlación con IgM e IgA. Rev Col Ginecol Obstet. 2008;59:199–205
6. Ponce N, Gomez JE. Estandarización y validación clínica de la prueba de reacción en cadena de la polimerasa (PCR) para diagnostico de toxoplasmosis cerebral en pacientes infectados por el VIH. Infectio. 2003;7:8–14
7. Cardona N, Basto N, Parra B, et al. Detection of Toxoplasma DNA in the peripheral blood of HIV-positive patients with neuropportunistic infections by a real time PCR assay. J Neuroparasitol. 2011 DOI:10.4303/jnp/N110402. Available at: http://www.ashdin.com/journals/jnp/N110402.aspx
. Accessed March 30, 2013
8. Buffolano W, Lappalainen M, Hedman L, et al. Delayed maturation of IgG avidity in congenital toxoplasmosis
. Eur J Clin Microbiol Infect Dis. 2004;23:825–830
9. Sterkers Y, Ribot J, Albaba S, et al. Diagnosis
of congenital toxoplasmosis
by polymerase chain reaction on neonatal peripheral blood. Diagn Microbiol Infect Dis. 2011;71:174–176
10. Said RN, Zaki MM, Abdelrazik MB. Congenital toxoplasmosis
: evaluation of molecular and serological methods for achieving economic and early diagnosis
among Egyptian preterm infants. J Trop Pediatr. 2011;57:333–339