| We tested the hypothesis that Cache Valley Virus (CVV), an endemic North American bunyavirus, may be involved in the pathogenesis of human neural tube defects. This investigation followed a 1990 and 1991 south Texas outbreak of neural tube defects with a high prevalence of anencephaly and the demonstration in 1987 that in utero infection by CVV was the cause of outbreaks of central nervous system and musculoskeletal defects in North American ruminants. Sera from 74 women who gave birth to infants with neural tube defects in south Texas from 1993 through early 1995 were tested for CVV neutralizing antibody. All tested sera did not neutralize CVV. These data suggest that CVV is not involved in the induction of human neural tube defects during nonepidemic periods but do not preclude CVV involvement during epidemics. Other endemic bunyaviruses may still be involved in the pathogenesis of neural tube defects or other congenital central nervous system or musculoskeletal malformations. |
Anencephaly, a severe defect of the central nervous system, results from failure of the anterior portion of the neuropore to close during embryonic development (1-4). A subsequent severe defect of the cranial vault produces brain malformation with an exposed mass of nervous tissue on the exterior of the head. Infants with this condition are stillborn or die shortly after birth. Anencephaly is one of several defects caused by failure of the neural tube to close. These defects, including spina bifida, encephalocele, encephalomyelocele, and myeloschisis, are often grouped under the classification neural tube defects. Public concern regarding a cluster of such defects (primarily anencephaly) in Brownsville,Texas, in 1991, has resulted in an investigation and a neural tube defect surveillance and intervention project in Texas.
The prevalence rate of anencephaly and other neural tube defects in the United States has been declining steadily (4) but is estimated to be 6 per 10,000 births; anencephaly has an approximate prevalence rate of 3 per 10,000 births (3). A previous study of anencephaly in Texas showed that from 1981 to 1986 the incidence rate was 3.8 to 4.3 cases per 10,000 births (1); the prevalence rate of this defect was higher among mothers with Hispanic surnames (particularly those with three or more previous live births or a history of stillbirths) and among women living in east and south Texas. The annual prevalence rate of anencephaly in south Texas from 1981 through 1986 was approximately 4.9 per 10,000 live births. A study of the cluster of cases in Cameron County (a Texas-Mexico border county of which Brownsville is the largest city) showed that the prevalence rates of neural tube defects increased in 1990 to 1991 largely because of the increase in cases of anencephaly (5,6). The rate of neural tube defects in this area increased from 14.7 per 10,000 during 1986 to 1989 to 27.1 per 10,000 births in 1990 to 1991, with the rates of anencephaly rising from 9.6 to 19.7 per 10,000 births.
Because of the make-up of the study population in south Texas, some socioeconomic and religious bias may be reflected in the reported prevalence of neural tube defects in Cameron County. A recent U.S. study in which data on neural tube defect prevalence included both cases diagnosed at birth and cases diagnosed prenatally (7) found that the prevalence rate of these defects may be reduced by 30% to 70% when only data on defects diagnosed at birth are included. In the Cameron County study population, neural tube defects may not have been diagnosed prenatally, and termination of pregnancy would not be common. Still, the data related to the cluster of cases at the Texas border show an obvious increase in anencephaly.
Many neural tube defects have a high serious prevalence rate and are associated with more illness than other congenital defects. However, despite extensive studies, we know little about their etiology. Evidence suggests that they are etiologically heterogeneous and may be caused by a wide variety of fetal insults, such as maternal hyperthermia, folic acid deficiency, and maternal anticonvulsant (valproate) therapy during the embryonic period of neural tube closure (8-11).
In 1987, just before the observed increase in prevalence of human anencephaly in south Texas, an outbreak of severe congenital malformations of the central nervous system and musculoskeletal system of lambs occurred in San Angelo, Texas (12). It was later recognized that the ovine problem was caused by in utero infection of lambs by Cache Valley Virus (CVV). This insect-borne bunyavirus had been known to commonly infect ruminants in North America (13) but had not been recognized to cause severe clinical consequences in its host. Experimentally, it was determined that infection of the dam in early gestation and transplacental infection of the ovine fetus produce a variety of severe brain malformations and arthrogryposis multiplex congenital, an anomaly characterized by limbs fixed in contracture (12). Central nervous system malformations associated with experimental and spontaneous CVV infection include hydrocephalus, hydranencephaly, porencephaly, micrencephaly, and micromelia. After the syndrome was characterized, outbreaks of CVV-induced malformations in ruminants were diagnosed throughout North America. There was no epidemiologic evidence that CVV was related to human malformations in San Angelo; however, because work by Calisher and Sever (14) had linked CVV to congenital cases of human macrocephaly in the United States, it was decided to test the hypothesis that CVV infection was related to congenital human anencephaly.
Sera from 74 women who had given birth to infants with neural tube defects (36 with spina bifida, 34 with anencephaly, and four with encephalocele) in south Texas, in 1990 to 1991 were examined for a possible link between these defects and CVV. The sera were screened by a standard, microtiter serum dilution neutralization test (15) at final dilutions of 1:2, 1:4, 1:8, and 1:16. The virus used in all tests was the prototype CVV (strain 6V-633, provided by the Centers for Disease Control and Prevention, Ft. Collins, CO) that had been passaged one time in Vero cells after receipt. Controls included sera from women of undetermined CVV status who gave birth to healthy infants in south Texas [8], sera collected from sheep before CVV infection [3], and normal macaque [4], horse [1] and bovine [1] sera. Positive controls included CVV-convalescent-phase ovine sera [3] and CVV antibody-positive sera from a horse and a cow. No serum neutralization activity for CVV was detected in sera from women with healthy infants or from those with infants who had neural tube defects. If CVV infection had been present in these women in early gestation, antibody to CVV would have persisted in their sera.
Before this study, the relationship between CVV and human neural tube defects was unknown. A fundamental problem in establishing a causal relationship between an agent and a low frequency event or malformation by using sero-epidemiology is being able to test an adequate number of matched controls, especially when there is some evidence of causal relationship. In this study, there was no evidence that CVV was related to the cases in Texas. Had CVV antibodies been detected, it would have been necessary to test numerous sera from site- and age-matched women who gave birth to healthy infants to establish a statistically significant correlation between CVV infection and neural tube defects in Texas infants.
Anencephaly, spina bifida, and encephalocele have not been associated with spontaneous or experimental fetal CVV infection in animals. However, several bunyaviruses can cause other congenital malformations (e.g., hydrocephalus, porencephaly, and hydranencephaly) of the central nervous system in infected livestock (12). CVV or other North American bunyaviruses may be involved in the pathogenesis of several, as yet undiagnosed, syndromes of congenital malformations in humans, particularly those of the musculoskeletal or central nervous system (16,17). A recent study correlated the occurrence of both microcephaly and macrocephaly with antibody to Tenshaw virus (another Bunyamwera serogroup bunyavirus) and CVV in mothers of affected infants (14). In addition, in two cases of macrocephaly, rising titers to CVV were measured in two women, indicating a recent infection by CVV. It would seem valid to continue to investigate the relationship of CVV and other arboviruses to human congenital defects and fetal and embryologic death. Because of the wide variety of defects induced by these viruses, laboratory models of fetal infection by the Bunyaviridae would facilitate the understanding of mechanisms of human viral teratogenesis.
J. F. Edwards,* and K. Hendricks†
* Texas A&M University, College Station, Texas, USA; †Texas Department of Health, Austin, Texas, USA
References
URL: http://www.cdc.gov/ncidod/EID/vol3no1/edwards.htm
Posted: January 10, 1997