Prospective Study of Congenital Toxoplasmosis Screening with Use of IgG Avidity and Multiplex Nested PCR Methods

Abstract

Acute infection with Toxoplasma gondii during pregnancy can cause congenital toxoplasmosis. The aim of this study was to evaluate whether screening with the use of IgG avidity and multiplex nested PCR methods was effective to detect a high-risk pregnancy. In a prospective study, serum T. gondii IgG avidity was measured in consecutive 146 pregnant women testing positive for T. gondii antibody and either positive or equivocal for IgM. Multiplex nested PCR for T. gondii DNA on amniotic fluid, maternal blood, and umbilical cord blood were performed with informed consent. A total of 51 (34.9%) women presented with low IgG avidity (<30%), 15 (10.3%) presented with borderline avidity (30 to 35%), and 80 (54.8%) presented with high avidity (>35%) indices. Amniotic fluid obtained at amniocentesis or birth yielded positive PCR results in nine women with low IgG avidity indices. Of these nine women, three had congenital toxoplasmosis. None of women with high or border line IgG avidity indices had a positive PCR result in the amniotic fluid or congenital toxoplasmosis. No congenital toxoplasmosis was detected in women whose amniotic fluids yielded negative PCR results. Ingestion of raw or undercooked meat was found to be the main risk factor for acute T. gondii infection. Congenital toxoplasmosis screening with a combination of IgG avidity in the maternal blood and multiplex nested PCR in the amniotic fluid was useful for detecting a high risk pregnancy and diagnosing congenital toxoplasmosis.

INTRODUCTION

Approximately one-third of the world's population is infected by Toxoplasma gondii, an obligate intracellular protozoan belonging to the phylum Apicomplexa, order Coccida. T. gondii infection is most frequently caused by ingestion of raw and undercooked meat, which carries tissue cysts, by consuming infected water and food, or by accidental intake of contaminated soil (5). Vertical transmission from a recently infected pregnant woman to her fetus may lead to congenital toxoplasmosis that causes choroidoretinitis, intracranial calcification, hydrocephalus, and mental retardation of the infant.

Serologic tests for the detection of IgM antibodies are commonly performed for the diagnosis of acute acquired T. gondii infection. However, it is well recognized that false-positive results may occur with many of these tests (14) and that T. gondii IgM antibodies may persist for many months or even years following the acute infection (1, 6). These complicate the appropriate interpretation of a positive T. gondii IgM result, especially in pregnant women. Therefore, to determine whether the positive IgM result reflects the likelihood of a recently acquired infection, confirmatory testing by additional tests such as IgG avidity is necessary (14, 21). T. gondii IgG avidity assay has been developed for the serologic diagnosis of acute acquired infection, and this measures the antigen-binding avidity of T. gondii-specific IgG antibodies. The avidity is the strength with which sera bind to multivalent antigens. During microbial infections, hypervariable regions of specific IgG antibodies mature, leading to an increase in serum avidity for a mix of antigens. Patients with acute infection exhibit a low avidity index, whereas subjects with past immunity have a high avidity index of T. gondii IgG (9). A low IgG avidity index assists in diagnosing acute acquired T. gondii infection and identifying pregnancies that are at a high risk for congenital toxoplasmosis. A study has demonstrated that IgG avidity assay (Platelia) has 100% sensitivity and 92.7% specificity for the detection of acute T. gondii infection (3).

Prenatal diagnosis of congenital toxoplasmosis is based on detection of T. gondii DNA in the amniotic fluid by PCR. The prenatal PCR analysis of the amniotic fluid has replaced cordocentesis, followed by culture and serologic analysis of fetal blood, and has been used since the early 1990s in order to lower the risk of fetal demise and to improve sensitivity (7, 10, 17). Thalib et al. reported 71% sensitivity and 98% specificity of the amniotic fluid PCR for the detection of congenital toxoplasmosis (20).

No epidemiological study has determined the accurate incidence of congenital toxoplasmosis in the Japanese population. Screening for T. gondii infection during pregnancy is not uniformly performed at all maternity hospitals or clinics in Japan, and the majority of facilities omit this screening. In 2005, we commenced a prospective study of a new screening system for T. gondii infection, in which pregnant women with a positive or border line for T. gondii IgM undergo T. gondii IgG avidity measurement, and multiplex nested PCR is tested with informed consent.

MATERIALS AND METHODS

Patients.

This prospective study was performed in the cities of Sapporo in Hokkaido and Kobe in Hyogo prefecture, Japan, and conducted with informed consent from all of the subjects. During the period between April 2005 and November 2009, pregnant women with positive tests for T. gondii antibody, together with those positive or equivocal for T. gondii IgM, were included in the present study. A total of 146 pregnant women who tested positive for T. gondii antibody (hemagglutination [HA]) in the peripheral blood obtained at 9 to 14 weeks of gestation (GW) and also tested positive or equivocal for T. gondii IgM a few weeks later in the former hospital were referred to the NTT East Sapporo Medical Center, Hokkaido University Hospital or Kobe University Hospital for further assessment of acute T. gondii infection. The positive tests were confirmed by T. gondii antibody (TOXO-HA; Japan Lyophilization Laboratory, Tokyo, Japan) and T. gondii IgM (Plateria Toxo IgM; Bio-Rad, Tokyo, Japan) measurements. All 146 women underwent serum T. gondii IgG avidity measurements according to the congenital toxoplasmosis screening method as described below.

Congenital toxoplasmosis screening method.

The scheme used for congenital toxoplasmosis screening method in the present study is indicated in Fig. 1. The 146 pregnant women positive for T. gondii antibody and with a positive or equivocal test result for T. gondii IgM underwent serum T. gondii IgG avidity measurements. When the IgG avidity index was <30%, acute infection during pregnancy or periconception period was strongly suspected, and the women received cyclic administration of acetylspiramycin (1.2 g/day for 3 weeks, followed by an interval of no medication for 2 weeks) until delivery. Women with a 30 to 35% IgG avidity index (borderline) were advised to undergo acetylspiramycin therapy. When the IgG avidity index was >35%, chronic infection was suspected, women did not receive the therapy unless the women desired medication.

Fig. 1.

Congenital toxoplasmosis screening method with use of the IgG avidity and multiplex nested PCR methods. Pregnant women who tested positive for T. gondii antibody in the first trimester and positive or equivocal for T. gondii IgM were enrolled in the present study. They underwent T. gondii IgG avidity measurements and multiplex nested PCR for T. gondii DNA in the peripheral blood. Women with low or borderline avidity indices or determined to be positive by PCR received acetylspiramycin therapy. If women who underwent amniocentesis had positive PCR results in the amniotic fluid, they received also pyrimethamine and sulfadoxine. At birth, T. gondii IgM measurement and PCR testing in the umbilical cord blood were performed, and PCR tests in the amniotic fluid were performed after informed consent was obtained.

Multiplex nested PCR for T. gondii DNA in the peripheral blood was performed in addition to the IgG avidity measurement when the women agreed. When the blood tested positive for multiplex nested PCR, the women received the acetylspiramycin therapy regardless of the IgG avidity results. Amniocentesis followed by multiplex nested PCR for the amniotic fluid was performed with informed consent when the women desired it because of low or medium levels of IgG avidity or anxiety. If the amniotic fluid tested positive in the multiplex nested PCR, the women received, in addition to acetylspiramycin therapy, pyrimethamine at 25 to 50 mg/day and sulfadoxine at 500 to 1,000 mg/day until 28 GW.

At birth, the T. gondii IgM level in the umbilical cord blood was measured in all subjects except in cases of induced abortion, and multiplex nested PCR for the umbilical cord blood and/or amniotic fluid was performed when informed consent was obtained. The presence of congenital infection was assessed on all infants by ophthalmofundoscopy, cerebral ultrasonography, and physical and neurological examinations. A head computed tomographic (CT) scan was carried out if necessary.

IgG avidity measurements.

Serum T. gondii IgG avidity was measured by using T. gondii antigen-coated wells (Enzygnost Toxoplasmosis/IgG; DADE Behring, Marburg, Germany). Briefly, diluted serum with sample buffer was added to the antigen-coated wells in duplicate. After the antigen-antibody reaction, one of the wells was treated with phosphate-buffered saline containing 0.1% bovine serum albumin-0.05% Tween 20 (reaction buffer), and the other well was treated with reaction buffer containing 8 M urea. After 30 min of incubation at room temperature, anti-T. gondii IgG was detected by horseradish peroxidase-conjugated anti-human IgG. Thereafter, these wells were stained with substrate reagent (Tetramethylbenzidine Plus Substrate-Chromogen; Dako, Carpinteria, CA), and the optical density at 450 nm (OD₄₅₀) was measured. The avidity index (AI) was calculated by using the following formula: AI (%) = [(8 M urea-treated OD₄₅₀ values)/(untreated OD₄₅₀ values)] × 100. The coefficients of variation were below 10% in the intra-assay and below 15% in the interassay.

Multiplex nested PCR methods.

In order to assess the presence of T. gondii DNA, multiplex nested PCRs for maternal blood, umbilical cord blood, and amniotic fluid were performed with informed consent. The primer pairs were designed for different four genes of T. gondii, including B1 (GenBank accession number AF179871), cyclin-dependent kinase (cdk; AJ534295), SAG5E (AY363043), bradyzoite surface antigen 4 (BSR4; AF394603). These four genes were simultaneously amplified in one tube reaction by multiplex PCR using a multiplex PCR kit (Qiagen GmbH, Hilden, Germany). As a nested PCR method, the primer pairs in the second multiplex PCR were designed so that each of the PCR products should be 5 bp smaller than that of the first multiplex PCR. These primer sequences are shown elsewhere (16).

Figure 2 shows the products of the first/second PCRs as follows: BSR4, 170/165 bp; cdk, 283/278 bp; B1, 425/420 bp; and SAG5E, 527/522 bp. When there were more than ∼100 gene copies, these four bands could be detected after the first PCR. After the second PCR, the four bands could be detected if there were more than ∼10 gene copies of T. gondii. Thus, T. gondii gene copy numbers in samples were estimated semiquantitatively.

Fig. 2.

Products of multiplex nested PCR for Toxoplasma gondii. According to phased dilution of T. gondii DNA in reaction mixtures, four bands of PCR products (BSR4, cdk, B1, and SAG5E) are indicated. To detect four visible bands simultaneously, approximately 100 gene copies in the first PCR and approximately 10 gene copies in the second PCR were necessary as minimal gene copy numbers. NC, negative control; PC, positive control.

Risk factors for T. gondii infection.

When the 146 women underwent serum T. gondii IgG avidity measurements, information about risk factors for T. gondii infection—including ingestion of raw or undercooked meat, gardening or intake of soil, the presence of a pet or stray cat in the neighborhood, travel abroad, and/or a cold or lymph node swelling—was collected. The frequencies of these risk factors were compared among acute (IgG avidity < 30%), borderline (30 to 35%), and chronic (>35%) groups by using a χ² test.

RESULTS

Of the 146 pregnant women who tested positive for T. gondii antibody and either positive or equivocal for T. gondii IgM, 51 (34.9%) had low IgG avidity (<30%), 15 (10.3%) had borderline avidity (30 to 35%), and 80 (54.8%) had high avidity (>35%) results.

Of 110 pregnant women who underwent multiplex nested PCR for T. gondii DNA in the peripheral blood, 5 (4.5%) had positive tests. Of 27 women who underwent prenatal amniocentesis and the PCR test at 16 to 30 GW, 5 (18.5%) had positive tests. At birth, the amniotic fluids for the PCR test were obtained from 67 women, and 7 (10.4%) had positive tests. Table 1 shows a total of nine women whose amniotic fluid obtained at prenatal amniocentesis or at birth yielded positive PCR results for T. gondii DNA. Cases 1, 6, and 7 had positive tests for both amniotic fluids of the prenatal amniocentesis and at birth. Cases 3, 4, 5, and 8 had positive tests for amniotic fluid at birth but not of the prenatal amniocentesis. Case 9 had a positive test for amniotic fluid of the prenatal amniocentesis but not at birth. Case 2 had a positive test for amniotic fluid of the prenatal amniocentesis, but the PCR test at the induced abortion was not permitted. All nine women had low IgG avidity results. The umbilical cord blood for the PCR test was performed for 100 women, and only one (case 8 in Table 1) had a positive test (1 to several gene copy numbers). T. gondii IgM in the umbilical cord blood was measured at birth; however, no one had a positive result.

Table 1.

Nine cases whose multiplex nested PCR of amniotic fluids tested positive for Toxoplasma gondii^a

Case	Maternal Toxoplasma status		Result of amniotic fluid-PCR at:		Age; outcome
	IgM index	IgG avidity (%)	Amniocentesis, estimated gene copy no.	Birth, estimated gene copy no.
1	2.8 (17)	23 (28), 71 (8 mo postpartum)	+ (28), 10-100	+ (38), 10-100	5 yr 8 mo; congenital toxoplasmosis, positive IgM and IgG at 18 mo, 1 yr of medication, no sequela
2	0.9 (13)	20 (17)	+ (17), 10-100	ND (21)	Termination of pregnancy
3	4.6 (15)	4 (15), 20 (21)	− (18)	+ (38), 10-100	2 yr 9 mo; positive IgM at 4 mo and positive IgG at 12 mo, no medication, no sequela
4	3.0 (13)	7 (15), 14 (36)	− (23)	+ (39), 1 to several	2 yr 5 mo; no infection
5	2.0 (16)	13 (18)	− (26)	+ (39), 1 to several	2 yr 1 mo; no infection
6	1.5 (19)	29 (20)	+ (25), 1 to several	+ (39), 1 to several	16 mo; medication, no infection
7	1.4 (13)	29 (15)	+ (18), 1 to several	+ (39), 10-100	14 mo; no infection
8	1.3 (20)	7 (20)	ND	+ (37), ≥100	14 mo; no infection
9	2.0 (12)	8 (13)	+ (17), 1 to several	− (38)	12 mo; no infection

^aThe time points, expressed as the gestational week number, are indicated in parentheses. ND, not determined.

Of the nine women whose amniotic fluids had positive PCR tests, three were diagnosed as having congenital toxoplasmosis (cases 1, 2, and 3 in Table 1). Case 1 was previously reported (16). Briefly, serum T. gondii antibody titers (320×) at 12 GW increased to 5,120× at 25 GW. Acetylspiramycin therapy from 22 weeks until delivery and pyrimethamine-sulfadoxine therapy from 25 to 28 GW were administered. A male neonate weighing 2,916 g was born at 38 GW by cesarean section. The maternal blood and amniotic fluids, but not the umbilical cord blood or cerebrospinal fluid of the neonate, tested positive using multiplex nested PCR. No abnormalities were detected by physical and ophthalmofundoscopic examinations, whereas a head CT scan of the neonate revealed three independent intracranial calcifications. The infant underwent a therapy with pyrimethamine and sulfadiazine for 1 year. The serum titers of T. gondii antibodies (passive hemagglutination [PHA], IgG, and IgM) were all less than the cutoff values between 5 and 12 months after birth, but all increased up to positive levels (PHA, 2,560×; IgG, 240 IU/ml; IgM, 1.5 index) 18 months after birth. At present, he is 5 years and 8 months old without physical or neurological abnormality. In case 2, the serum T. gondii antibody titer was 320× at 11 GW. She underwent acetylspiramycin therapy from 14 GW, and amniocentesis for the PCR test was performed at 17 GW because of the low IgG avidity result. The amniotic fluid, but not the maternal blood, tested positive for the PCR. This pregnancy ended in an induced abortion at 21 GW. The parents did not permit an autopsy or further examinations. In case 3, the serum T. gondii antibody titer was 2,560× at 15 GW. The mother underwent the acetylspiramycin therapy from 15 GW until delivery. The amniotic fluid obtained at 18 GW tested negative; however, the amniotic fluid at birth tested positive for the PCR. A male neonate weighing 3,220 g was vaginally delivered at 38 GW. Ophthalmofundoscopy, cerebral ultrasonography, a head CT scan, and physical or neurological examinations revealed no abnormality. T. gondii IgM index values were as follows: 0 for the umbilical cord blood and 1.0 (positive) at 4 months, 0.8 (positive) at 6 months, and 0.2 (negative) at 1 year of the peripheral blood. The T. gondii IgG serum titer at age 1 year was 240 IU/ml. At present, the 2-year 9-month-old boy has no physical or neurological abnormality without medication. The amniotic fluid obtained at 18 GW tested negative for the PCR in this case; therefore, a later fetal infection was suspected.

None of the women with high or borderline IgG avidity indices had a positive PCR test results for T. gondii DNA in the amniotic fluid or congenital toxoplasmosis. No congenital toxoplasmosis was detected in women whose amniotic fluids yielded negative PCR results. Of five women with positive PCR results in the peripheral blood, only one (case 1) had a low IgG avidity index. The other four women had high IgG avidity indices, negative PCR results in the amniotic fluid, and no congenital toxoplasmosis.

Table 2 presents these findings according to IgG avidity indices. The frequency of “ingestion of raw or undercooked meat” in women with low or borderline IgG avidity index (n = 66) was significantly (P < 0.05) higher than that in women with high IgG avidity index (n = 80).

Table 2.

Frequencies of risk factors for Toxoplasma gondii infection during pregnancy

Risk factor	Frequency (%) of risk factor for Toxoplasma IgG avidity
	<30% (n = 51)	30∼35% (n = 15)	>35% (n = 80)
Ingestion of raw or undercooked meat	51.0	60.0	37.5a
Gardening or intake of soil	19.6	20.0	16.3
Presence of a pet or stray cat in the neighborhood	11.8	20.0	17.5
Travel abroad	11.8	26.7	7.5
A cold or lymph node swelling	5.9	6.7	11.3
None	17.6	20.0	33.8

^aP < 0.05 compared to the frequency in women with ∼35% IgG avidity.

DISCUSSION

In this prospective study of congenital toxoplasmosis screening using IgG avidity and multiplex nested PCR methods, we for the first time demonstrated an incidence of congenital toxoplasmosis (2.1%) in 146 Japanese women testing positive for T. gondii antibody and either positive or equivocal for T. gondii IgM. Of the 146 women, 51 (34.9%) presented with low IgG avidity index, and we determined that the 51 women acquired a primary infection of T. gondii during their pregnancies. None of the women with high or borderline IgG avidity indices had a positive PCR test result for T. gondii DNA in the amniotic fluid or congenital toxoplasmosis. Conversely, all nine women with positive PCR test results for T. gondii DNA in the amniotic fluid had low IgG avidity indices (Table 1). Hence, it followed that T. gondii DNA in the amniotic fluid was once present in 9 (17.6%) of the 51 women who had primary T. gondii infections during pregnancy, and an incidence of congenital toxoplasmosis in women with the primary infection was 5.9% (3/51) in the Japanese population. These data as important references can be used in clinical practices for congenital toxoplasmosis screening. However, we could not confirm the definite infection in case 2, because the parents did not consent to any further examination after the induced abortion. The quality of our PCR method and the the false-positive and false negative rates are still unclear. A false-positive PCR result at the prenatal amniocentesis in case 2 cannot be excluded.

Maternal medications might have beneficial influence on a condition of fetal T. gondii infection. In the present study, only one case had a positive PCR result for the umbilical cord blood. None yielded a positive T. gondii IgM result for the umbilical cord blood. Therefore, for the diagnosis of congenital toxoplasmosis, the PCR test for T. gondii DNA in the amniotic fluid was more sensitive than the PCR test or IgM measurement in the umbilical cord blood. However, we encountered three women whose the PCR results in the amniotic fluid changed from negative at amniocentesis to positive at birth and one woman whose PCR results in the amniotic fluid changed from positive at amniocentesis to negative at birth. These changes of the PCR results in the amniotic fluid might be caused by too-early prenatal amniocentesis (case 3 in Table 1), a very low T. gondii load (cases 4, 5, and 9), no effect of acetylspiramycin therapy (cases 3, 4, and 5), an effect of the therapy (case 9), and/or problems with the PCR tests.

The B1 gene has been used as a standard for T. gondii PCR analysis in previous studies (10, 18). In order to decrease the rate of false-positive results, we for the first time developed and used a multiplex PCR method so that the presence of T. gondii DNA could be confirmed when four bands appeared simultaneously. This nested PCR method enabled semiquantitation of the T. gondii gene copy numbers. It was reported that nested PCR and real-time PCR methods were equivalent in sensitivity and specificity for the detection of T. gondii DNA (2). A recent study demonstrated that PCR quantification of T. gondii DNA in the amniotic fluid can contribute to the early prognosis of congenital toxoplasmosis (18). For maternal infections acquired before 20 weeks of gestation, a parasite load greater than 100/ml of the amniotic fluid has the highest risk of severe fetal outcome, i.e., death or cerebral ventricular dilatation. Conversely, most cases with a parasite load less than 100/ml of the amniotic fluid are subclinical or mild, having cerebral calcification, choroidoretinitis, or no symptoms without ventricular dilatation (18). The results of semiquantification of T. gondii DNA in the present study (Table 1) seemed to be concordant with the results of the above-mentioned earlier study.

We found five women with positive PCR results in the peripheral blood, and all women received the therapy. Only one had a low IgG avidity index and congenital toxoplasmosis, but the other four women had high IgG avidity indices, negative PCR results in the amniotic fluid, and no congenital toxoplasmosis. A positive PCR result for T. gondii DNA in the maternal blood has rarely been reported in the literature, and the clinical significance of this was unclear. It may suggest primary infection and may reflect a reactivation of the T. gondii during pregnancy.

Tenter et al. (19) collected data on the nationwide T. gondii seroprevalence in women at child-bearing age (1990 to 2000). The positive rates of T. gondii antibody were 58% in Central European countries; 51 to 72% in Latin-American countries; 54 to 77% in West African countries; 4 to 39% in Southwest Asia, China, and Korea; 11 to 18% in Scandinavian countries; and 15% in the United States (12). A positive rate of T. gondii antibody in Japanese adults was reported as 16.4% (13), but to date no English literature reported the rate in Japanese pregnant women. We previously reported positive rates of T. gondii antibody (3.63%) and T. gondii antibody plus T. gondii IgM (0.60%) in 1,848 pregnant women in Japanese studies (15). Using these figures and data obtained in the present study, the incidence of congenital toxoplasmosis in Japan is estimated to be at least 0.0126% (1.26 per 10,000 births) under the condition that all pregnant women underwent congenital toxoplasmosis screening and therapy if necessary as performed in the present study. The estimated incidence of congenital toxoplasmosis in Japan was compatible with incidences (1 to 10 per 10,000 live births) reported in the United States (8) and Europe (5).

In the present study, 34.9% of women testing positive for T. gondii antibody and positive or equivocal for T. gondii IgM had primary infections of T. gondii due to their low IgG avidity index (<30%); additionally, 10.3% of the women had borderline avidity index values (30 to 35%). We could estimate that 0.21 to 0.27% of Japanese pregnant women acquired primary T. gondii infections in the first trimester. We did not evaluate seroconversion of the T. gondii antibody during the second and third trimesters. Therefore, the actual frequency of the primary infection during pregnancy must be more than these estimated figures in Japanese women.

We found that among risk factors for T. gondii infection, the frequency of “ingestion of raw or undercooked meat” in women with low or borderline IgG avidity indices was significantly higher than that in women with high IgG avidity indices. It was likely that “ingestion of raw or undercooked meat” was the main risk factor for acute T. gondii infection in Japanese pregnant women. This was concordant with results from previous studies in the Unites States (11) and Europe (5).

It was reported that pregnant women who were informed by their physician that they had a positive T. gondii IgM result requested induced abortion in up to 20% of cases (4, 14). This tragedy and that of such pregnant women who desire to continue their pregnancy but who will worry about its outcome for the remainder of gestation should be avoided by more careful evaluation. Congenital toxoplasmosis screening with combined assessments of IgG avidity in the maternal blood and multiplex nested PCR in the amniotic fluid was useful for detecting a high-risk pregnancy and diagnosing congenital toxoplasmosis.