|
|
||||||||
| ABSTRACT |
|
|
|---|
| INTRODUCTION |
|
|
|---|
The prevalence of chronic T. cruzi infection in women, the transmission rate, and the morbidity and mortality of congenital infection vary largely according to the areas under study.28 Information on the effects of such chronic infection on pregnancy outcome, fetal growth, and health of uninfected babies born of infected mothers remains contradictory. Indeed, some studies mention that maternal infection induces an increased risk of pregnancy loss or prematurity,911 whereas other do not show any effect.8,1214
Although Bolivia is the most highly endemic country of Latin America for T. cruzi infection,15 few data have been reported on congenital Chagas disease,1618 and no information is available on the other consequences of maternal T. cruzi infection in this country. To get more information of such consequences on pregnancy and the health of infected and uninfected newborns, and their evolution during the last decade, we have compared the results of two epidemiologic and clinical surveys performed in the same Bolivian maternity clinic. In both studies, the prevalences of maternal infection, as well as obstetrical and clinical status of mothers, transmission rates, and incidences of congenital infection, Apgar scores, maturity and clinical patterns, hematologic and biochemical parameters, and mortality data of T. cruzi-infected and non-infected newborns of infected mothers were compared with data of uninfected babies born of uninfected mothers.
| MATERIALS AND METHODS |
|
|
|---|
Biologic diagnosis of T. cruzi infection.
Maternal infection was assessed by T. cruzi-specific serologic tests: indirect hemagglutination (using a commercially available kit; Polychaco, Buenos Aires, Argentina) and/or immunofluorescence.19 Titers
1:16 and
1:40, respectively, were considered positive for these two tests. Blood of newborns was collected from the umbilical cord, or in some cases by peripheral puncture before the 30th day of life. Congenital infection was diagnosed by microscopic examination of the buffy coat from blood collected in four microhematocrit heparinized tubes (each containing 50 µL of blood), as described elsewhere,20 and/or hemoculture of 2 mL of blood for 28 weeks.21 The combination of such diagnostic methods gave results similar to those obtained by a polymerase chain reaction.22 The presence of parasite(s) in one or more microhematocrit and/or hemoculture tubes defined a congenital infection.
Clinical examination. Each mother enrolled in the study was questioned about her age, previous pathologies, obstetrical antecedents, present symptoms, and last menstruation date. Abortion history was considered when fetuses were released before the sixth month of pregnancy. The state of amniotic membranes was noted. Premature rupture of membranes (PROM) was diagnosed when membranes were broken before admission or before the onset of contractions, whatever the timing of rupture.
Newborn weights, lengths, and head circumferences were measured at birth. A physical examination was performed at the 24th hour following delivery according to classic procedures. Apgar scores at one and five minutes, the general appearance in search of congenital malformations or deformations, body temperature, hepatomegaly (when
2 cm below the right costal margin), splenomegaly (whatever the spleen size under the left costal margin), the occurrence of anasarca (palpebral, genital, or leg edema alone were not considered), jaundice, ascitis, petechiae (whatever their localization), and meconium staining of the umbilical cord, nails, or skin were investigated. The diagnosis of respiratory distress syndrome (RDS) was considered when at least one of the following signs was present: tachypnea, throbbing of the ala nasi, expiratory grunting, intercostal retraction, and/or facial or systemic cyanosis (peripheral cyanosis was not considered). A neurologic examination included assessment of tone, level of alertness, Moro and other primary neonatal reflexes, deep tendon reflexes, spontaneous motor activity, pupil diameter (in search of mydriasis or miosis), bulging of fontanelles, and convulsions.23 The determination of gestational age was based on physical signs and neurologic characteristics of newborns24 related to the data obtained from the maternal last menstruation date. In some newborns, it was possible to perform chest radiographies to determine cardio-thoracic indexes, and/or abdominal echographies.
Biologic investigations and serologic analysis of other infections in newborns and their mothers. In newborns, blood hematocrit rates, hemoglobin amounts, and white blood cell, neutrophil, eosinophil, lymphocyte, monocyte, and reticulocyte counts, as well as plasmatic levels of direct and indirect bilirubin, aspartate aminotrasferase, alanine aminotransferase, alkaline phosphatase, urea, and creatinine were determined by classic tests of clinical biology. IgM antibodies directed against most pathogens frequently infecting neonates, the so-called TORCH (Toxoplasma gondii, rubella virus, cytomegalovirus, and herpes simplex virus) pathogens,25 were also investigated in newborn plasma. The commercially available kits Eti-toxoK-M reverse plus, Eti-cytok-M reverse plus, and Eti-rubek-M reverse plus (all from DiaSorin, Saluggia, Italy) were used for the detection of IgM antibodies directed against T. gondii, cytomegalovirus, and rubella virus, respectively. IgM antibodies to Treponema pallidum and human immunodeficiency virus type 1 (HIV-1) and HIV-2 antibodies was also determined using the kits Trepo-spot IF (BioMérieux, Marcy LEtoile, France) and Murex HIV-1.2.0 (Abott/Murex Biotech, Limited, Dartford, United Kingdom), respectively. Investigation of Plasmodium-specific antibodies by immunofluorescence were performed in T. cruzi-infected mothers.26
Statistical analysis. Results are expressed as the mean ± SEM or in percentages. The Student t-test or Kruskal-Wallis non-parametric test were used to compare means or medians. Chi-square or Fishers exact tests, with the correction of Bonferroni for multiple comparisons, were used to compare proportions. A multiple logistic regression was used to analyze the effects of groups on obstetrical antecedents adjusting for maternal age. Sensitivity and specificity of clinical signs observed in congenital Chagas disease were also estimated. Positive and negative predictive values were computed from these values,27 which considered an estimated incidence of 1% of congenital infection.
| RESULTS |
|
|
|---|
|
|
|
Comparisons of the M+B+ and M+B- groups in both cohorts showed that M+B+ mothers were younger (mean ± SEM age = 23.7 ± 0.7 versus 26.4 ± 0.3; P < 0.05) and displayed lower numbers of previous pregnancies (mean ± SEM = 1.8 ± 0.2 versus 2.6 ± 0.1; P < 0.05).
The comparison of data of both cohorts in the more recent survey B showed a significant decrease of the number of previous pregnancies in the M-B- and M+B- groups, as well as a significant increase of the proportion of primiparity in M-B-mothers (P < 0.05). No significant differences were noted in the other maternal parameters.
Effects of T. cruzi maternal infection on Apgar scores and maturity parameters of newborns.
Mean Apgar scores monitored at one and five minutes, as well as mean maturity parameters in the three neonate groups in both cohorts are shown in Table 4
according to the sex of the newborn. Apgar scores at one minute and/or five minutes, gestational ages, birth weights and lengths, and head circumferences were significantly decreased in M+B+ versus M-B- newborns in cohorts A and/or B, and in males and/or females (0.001 < P < 0.05). Figure 1
shows that M+B+ newborns in cohorts A and/ or B displayed higher frequencies of Apgar scores < 7 at one minute, low birth weight (LBW < 2,500 grams), prematurity (gestational age < 37 weeks), or prematurity/dysmaturity (gestational age < 37 weeks and birth weight < 2,500 grams) than M+B- or M-B- newborns (0.01 < P < 0.0001). Interestingly, prematurity was observed in 45.5% and 11.1% of M+B+ babies born from mothers with PROM in cohorts A and B, respectively. In both cohorts, data of M+B- neonates were similar to those of M-B- newborns (Table 4
and Figure 1
). These results indicate that maternal T. cruzi infection affects intrauterine growth and maturity of congenitally infected fetuses, but not of non-infected fetuses.
|
|
Clinical pattern of newborns of T. cruzi-infected mothers.
As shown in Figure 2
, 42.9% and 54.5% of the congenitally infected newborns in cohorts A and B, respectively, showed at least one of the signs/syndromes listed in Table 5
. However, they were more rarely found in the non-infected M-B- (A = 11.1%, B < 1%) and M+B- (A = 15.5%, B = 6.9%) groups (P < 0.0001). The association of at least two of the severe signs frequently associated with prematurity (LBW, Apgar score < 7 at one minute, RDS, or anasarca) was observed in 50.0% and 18.4% of M+B+ babies in cohorts A and B respectively, versus < 1% and 7.3% in M+B- and M-B- babies of both cohorts (P < 0.001). Fever, jaundice, convulsions, and developmental anomalies were not observed in T. cruzi-infected babies. When the babies of both cohorts were considered, the frequencies of signs/syndromes among the congenital cases could be classified as RDS > hepatomegaly > splenomegaly > neurologic signs (other than convulsions) > anasarca, petechiae (Table 5
). Meconium staining was observed in similar proportions in the three groups of babies in both cohorts (913%; P > 0.05). Chest radiographies performed in 37 infected babies in cohort B showed cardiomegaly with a cardiothoracic index > 0.55 in four (10.8%) of them. Abdominal ultrasound echographies performed in 41 infected newborns in cohort B confirmed liver and/or spleen enlargements detected by physical examination, and showed the homogenous density of such organs and any other anomalies. As shown in Table 6
, analysis of the sensitivity, specificity, and predictive values of LBW, Apgar scores < 7 at one minute, and signs/syndromes in Table 5
, either taken separately or considering the association of at least two of the severe signs mentioned earlier, showed these clinical data to be highly specific (90100%), although poorly sensitive (827%). Their positive predictive values varied strongly and only the occurrence of anasarca, petechiae, and splenomegaly showed positive predictive values
20%. Comparison of such clinical data between cohorts indicates a higher frequency of symptomatic babies and neonates with severe signs in the M+B+ group, as well as in the M+B-, and/or M-B-groups in cohort A than in cohort B (0.001 < P < 0.05; Figure 2
).
|
|
|
Hematologic and biochemical patterns of newborns of T. cruzi-infected mothers.
The mean hematocrit rates and hemoglobin levels of M+B- and M+B+ babies of both cohorts were within normal ranges and similar to those observed in the local control M-B-groups. When both cohorts were considered together, although also remaining in the normal range, a significant reduction in the number of white blood cells was observed in congenital cases compared with the M+B- and M-B- groups (Table 7
). Such reduction resulted from a decrease in the levels of neutrophils and monocytes, but not in the level of lymphocytes, whereas the levels of eosinophils and reticulocytes remained similar to those of the controls. Parameters exploring liver (plasmatic direct and indirect bilirubin, asparatate aminotransferase, alanine aminotransferase, and alkaline phosphatase) and renal functions (plasmatic urea and creatinine) were investigated in both cohorts and remained within the normal range of values in congenitally T. cruzi-infected babies.
|
The data relative to the seven fatal cases of congenital infection observed in both cohorts (5 in cohort A and 2 in cohort B) are shown in Table 8
. Infection was detected at birth in all of these cases. Four of them (in cohort A) displayed extremely LBW and were premature, and/or had symptoms of respiratory distress. Three (in cohort A) displayed hepatomegaly and two (one in each cohort) presented anasarca and ascitis, corresponding to fetal hydrops. The latter was not due to blood group incompatibility, since their hematocrit values and hemoglobin and bilirubin levels were within normal ranges. Chest radiographs performed in cases A659, A1209, and B5572 did not show cardiomegaly. A neurologic sign (spontaneous motor activity) was detected only in case A456.
|
| DISCUSSION |
|
|
|---|
Our estimation of the rates of materno-fetal transmission of parasites used a sensitive procedure with four microhematocrit tubes, which limited the possibilities of undiagnosed cases, as previously validated by a polymerase chain reaction.22 Nevertheless, the transmission rate of 56% observed in the main maternity clinic of Cochabamba is lower than previously reported in other studies in Santa Cruz, Bolivia.1618 This might be related to the use of histopathologic examination of placentas to assess congenital infection, instead of parasitologic or molecular detection of parasites in neonatal blood, as used in our work. This probably leads to an overestimation of congenital cases, since placentas of uninfected babies born of infected mothers are also susceptible to display parasites.14,28 Indeed, the transmission rates of congenital T. cruzi infection we have observed are comparable to those reported in Argentine,8,29,30 but higher than those reported in Brazil31 and Paraguay,32 and lower than those in Chile.33,34 The reasons for such differences remain unknown. In addition to possible differences in the sensitivity of the diagnostic procedures used, the strain of parasites or some peculiar immunologic features of mothers might contribute to such geographic variations.
The clinical signs observed in the congenitally infected babies in our cohorts globally agree with those mentioned in previous studies in other countries.5,7,8,35 Our complementary serologic investigations clearly eliminated possible co-infections between T. cruzi and most TORCH pathogens, as well as Treponema and HIV in the M+B+ babies. The absence of severe meningoencephalitis among our cases also rules out possible co-infection with HIV, since the latter is frequently associated with the co-infection T. cruzi/HIV.36 These data, as well as the significantly higher frequency of LBW, prematurity, hepatomegaly, splenomegaly, RDS, petechiae, and anasarca in M+B+ babies than in the M+B- and M-B- groups of the same maternity clinic validated such signs as currently associated with congenital Chagas disease, rather than being associated with TORCH co-infections.25 However, these clinical signs are observed only in 4354% of the congenital infections and display low positive predictive values. This indicates that they are poor markers of congenital T. cruzi infection, highlighting the need to assess the diagnosis of infection through the detection of parasites.
The higher frequency of PROM observed in mothers of congenitally infected babies might be related to the frequent chorioamnionitis detected in their placentas (Lambot MA and others, unpublished data). We also noted that in comparison with controls, infected newborns displayed significantly lower levels of leukocytes with reductions of neutrophil and monocyte counts, but not of lymphocyte counts, although such cell counts remained within physiologic ranges. This might be related to the relative expansion of CD8 T lymphocytes and the higher production of interferon-
that we have previously documented in newborns congenitally infected with T. cruzi,37 since this cytokine is known to support myelosuppressive activities.38
An interesting result of our comparative study of both surveys separated by 79 years is the observation of a significant decrease of frequencies of severe and mortal forms of congenital Chagas disease over time. Indeed, during these years, a sensitive improvement of the economic situation occurred in Bolivia, as attested by the doubling of the gross national product from $3,651,000 (U.S. dollars) in 1993 to $7,744,000 in 2000 (data from Instituto Nacional de Estadistica de Bolivia, http://www.ine.gov.bo). Although the economic situation or educational level of mothers enrolled in both cohorts have not been studied, our observation of a lower birth rate in M-B- and M+B- mothers in the more recent survey suggests that the economic improvement had induced a concomitant reduction of poverty.39 Moreover, the reduced frequencies of symptoms, altered Apgar scores, and LBW, and the decrease in neonatal mortality, which was also observed in the more recent survey in babies of uninfected (M-B-) as well as of infected (M+B- and M+B+) mothers, support the notion of an improvement in materno-fetal health over time. Another possible effect of the socioeconomic improvement might be the qualitative amelioration of houses. This improvement, which was associated with the recent increasing efforts of the Bolivian public health authorities in controlling vector bugs in dwellings, might have reduced the intradomiciliary vectorial density. Whether a reduction of reinfection rates during pregnancy also induces a reduction of morbidity and mortality of congenital Chagas disease remains to be studied.
However, despite a substantial reduction in the frequency of the severe form of congenital Chagas disease in the last decade in Bolivia, congenital infection with T. cruzi remains an important problem of public health in this country. Indeed, severe morbidity still occurs in 18% of infected babies detected at birth in the more recent survey in 2001. Moreover, such data were collected in the neonatal period and no information is available on the further development of congenital infection in asymptomatic undiagnosed and untreated babies later in childhood or at adult age. In addition, even if the general incidence of infection decreases (as shown by the reduction of maternal prevalence between both surveys) as a consequence of national programs controlling intradomiciliary vectors and blood banks, there is an important pool of infected women (17% of the female population of Cochabamba in 2001) who are likely to transmit parasites to their fetuses. Moreover, such T. cruzi congenital transmission is likely to have larger epidemiologic consequences since it occurs from one generation to another,40 allowing a vector-independent, uncontrolled spreading of the parasite for a long period of time.
A challenging result of our comparison of both surveys relates to the absence of reduction of the materno-fetal transmission rates of parasites, whereas neonatal morbidity and mortality decreased significantly over time. This suggests that parasite transmission depends on individual factors specific to each mother, who are susceptible to modify their capacity to control such transmission at the placental or systemic levels. The fact that both sets of twins were congenitally infected, which is consistent with previous observations,5,32,41 and that parasite-transmitting mothers (M+B+) were younger and displayed fewer numbers of previous pregnancies compared with non-transmitting mothers (M+B-) in both cohorts, argue for the role of such individual maternal factors.
An encouraging result confirmed in both surveys is that maternal chronic T. cruzi infection, when there is no parasite transmission, has no effect on pregnancy outcome, maturity, and general health of newborns. This agrees with previous studies in Brazil,1214 whereas the association of abortion histories with maternal T. cruzi seropositivity was reported in Argentine and Chile.911 The reasons for such a difference are unknown. Birth weighs and gestational ages were similar in M+B- and M-B- Bolivian babies, indicating that maternal chronic infection without parasite transmission does not induce premature delivery or fetal growth retardation, which is consistent with most of the previous reports from various countries.10,11,14
In conclusion, our study suggests that a decrease of poverty may reduce the morbidity and mortality, but not the transmission rate, of congenital T. cruzi infection. The latter remains an important risk for the babies of chronically infected mothers and a serious public health problem in Bolivia. Congenital T. cruzi infection is frequently associated with severe alterations in growth and maturity and neonatal death. This strongly argues for the development in Bolivia of programs aiming to detect infection soon after birth and to treat newborns to limit the consequences of this important route of parasite transmission.
Received July 14, 2003. Accepted for publication October 10, 2003.
Acknowledgments: We thank Marisol Cordova and the staff of the German Urquidi maternity clinic (Cochabamba, Bolivia) for the management of patients; Miguel Guzman, Myrian Huanca, Rudy Parrado, and Marco Antonio Solano (Centro Universitario de Medicina Tropical, Universidad Mayor de San Simon, Cochabamba, Bolivia) for the serologic and parasitologic diagnosis of patients; and Corinne Liesnard (Erasmus Hospital, Brussels, Belgium) for performing the serologic analysis of TORCH-related antibodies. We are indebted to Bruno Dujardin, Christine Kirkpatrick, and Anne Pardou for their critical review of the manuscript. Cristina Alonso-Vega is a fellow of the Association pour la Promotion de lÉducation et la Formation à lÉtranger (Communauté Française de Belgique). Part of this study has been presented at the International Colloquium "Infeccion Congenita por Trypanosoma cruzi: Desde los Mecanismos de Transmision Hasta Una Estrategia de Diagnostico y Control" held on November 68, 2002, in Cochabamba, Bolivia.
Financial support: This study was supported by the Conseil Interuni-versitaire de la Communauté Française de Belgique, the Community and Child Health Chagas Control Program (United States Agency for International Development, the Centre de Recherche Interuni-versitaire en Vaccinologie sponsored by the Région Wallonne and GlaxoSmithKline (Rixensart, Belgium), and the Fonds National de la Recherche Scientifique Médicale (Belgium, convention 3.4595.99).
Authors addresses: Faustino Torrico, Cristina Alonso-Vega, Eduardo Suarez, Patricia Rodriguez, and Mary-Cruz Torrico, Centro Universitario de Medicina Tropical, Facultad de Medecina, Universidad Mayor de San Simon, Avenida Aniceto Arce 371, Casilla 3023, Cochabamba, Bolivia, Telephone/fax : 591-442-30009. Michèle Dramaix, Laboratoire de Statistiques Médicales, Ecole de SantéPublique, Université Libre de Bruxelles, 808 Route de Lennik, CP 598, B-1070 Bruxelles, Belgium, Telephone: 32-2-555- 4051, Fax : 32-2-555-4047. Carine Truyens and Yves Carlier, Laboratoire de Parasitologie, Faculté de Médecine, Université Libre de Bruxelles, 808 Route de Lennik, CP 616, B-1070 Bruxelles, Belgium, Telephone: 32-2-555-6255, Fax: 32-2-555-6128, E-mail: ycarlier{at}ulb.ac.be.
Reprint requests: Yves Carlier, Laboratoire de Parasitologie, Faculté de Médecine, U.L.B., 808 route de Lennik CP 616, B-1070 Bruxelles, Belgium.
| REFERENCES |
|
|
|---|
This article has been cited by other articles:
![]() |
L. S. Wilson, J. M. Ramsey, Y. B. Koplowicz, L. Valiente-Banuet, C. Motter, S. M. Bertozzi, and L. H. Tobler Cost-effectiveness of Implementation Methods for ELISA Serology Testing of Trypanosoma cruzi in California Blood Banks Am J Trop Med Hyg, July 1, 2008; 79(1): 53 - 68. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Bern, S. P. Montgomery, B. L. Herwaldt, A. Rassi Jr, J. A. Marin-Neto, R. O. Dantas, J. H. Maguire, H. Acquatella, C. Morillo, L. V. Kirchhoff, et al. Evaluation and Treatment of Chagas Disease in the United States: A Systematic Review JAMA, November 14, 2007; 298(18): 2171 - 2181. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Virreira, C. Truyens, C. Alonso-Vega, L. Brutus, J. Jijena, F. Torrico, Y. Carlier, and M. Svoboda Comparison of Trypanosoma cruzi Lineages and Levels of Parasitic DNA in Infected Mothers and Their Newborns Am J Trop Med Hyg, July 1, 2007; 77(1): 102 - 106. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. RIERA, A. GUARRO, H. E. KASSAB, J. M. JORBA, M. CASTRO, R. ANGRILL, M. GALLEGO, R. FISA, C. MARTIN, A. LOBATO, et al. CONGENITAL TRANSMISSION OF TRYPANOSOMA CRUZI IN EUROPE (SPAIN): A CASE REPORT Am J Trop Med Hyg, December 1, 2006; 75(6): 1078 - 1081. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. VIRREIRA, S. MARTINEZ, C. ALONSO-VEGA, F. TORRICO, M. SOLANO, M. C. TORRICO, R. PARRADO, C. TRUYENS, Y. CARLIER, and M. SVOBODA AMNIOTIC FLUID IS NOT USEFUL FOR DIAGNOSIS OF CONGENITAL TRYPANOSOMA CRUZI INFECTION Am J Trop Med Hyg, December 1, 2006; 75(6): 1082 - 1084. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. VIRREIRA, C. ALONSO-VEGA, M. SOLANO, J. JIJENA, L. BRUTUS, Z. BUSTAMANTE, C. TRUYENS, D. SCHNEIDER, F. TORRICO, Y. CARLIER, et al. CONGENITAL CHAGAS DISEASE IN BOLIVIA IS NOT ASSOCIATED WITH DNA POLYMORPHISM OF TRYPANOSOMA CRUZI Am J Trop Med Hyg, November 1, 2006; 75(5): 871 - 879. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. S. WILSON, A. M. STROSBERG, and K. BARRIO COST-EFFECTIVENESS OF CHAGAS DISEASE INTERVENTIONS IN LATIN AMERICA AND THE CARIBBEAN: MARKOV MODELS Am J Trop Med Hyg, November 1, 2005; 73(5): 901 - 910. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |