Astroviruses were first observed by electron microscopy (EM) in stool specimens from infants with gastroenteritis (Appleton et al., 1975; Madeley et al., 1975). However, this method was relatively insensitive and studies based on this method of detection indicated that astroviruses were a rare cause of gastroenteritis (Glass et al., 1996). For example, hospital based EM studies indicated that enteric adenoviruses were the second most common cause of viral diarrhea in young children and that incidence rates of astrovirus infection never exceed 4% (Kapikian, 1993). The development of more advanced methods of detection such as enzyme linked immunosorbent assays (ELISA) and reverse transcriptase polymerase chain reaction (RT-PCR) have revealed that astroviruses are second only to rotavirus as the most common cause of viral gastroenteritis in children worldwide (Glass et al., 1996; Matsui et al., In: Field's Virology; Cruz et al., 1992; Herrmann et al, 1991; Lew et al., 1991). Furthermore, the imminent licensing of a rotavirus vaccine (Tucker et al., 1998) may lead to astrovirus becoming the primary cause of viral gastroenteritis among young children in the near future. Although numerous studies have pointed to the medical importance of astrovirus, very little is understood about the epidemiology of this virus.

This paper will examine the findings of previous studies on the epidemiology of astrovirus, discuss questions raised by the epidemiological information provided by these studies, and identify areas for future study.

 

BACKGROUND

Molecular Biology

Astroviruses are small, nonenveloped icosahedral viruses. Astroviruses are 28-30 nm in diameter with a smooth margin and a star-like EM appearance. The genome consists of plus-sense, single-stranded RNA that it approximately 6,800 nucleotides in length excluding the 3' poly (A) tail. To date, seven serotypes of human astrovirus have been identified (Matsui et al., "Astroviruses", In: Field's Virology).

Pathogenesis

The pathogenesis of human astrovirus infection has not been well studied in humans. The correlation of diarrhea with fecal shedding of astrovirus and the identification of viral particles in the intestinal epithelial cells suggests that viral replication takes place in the intestinal tissue in humans (Phillips et al., 1982). The viral pathogenesis of animal astroviruses has been studied more extensively. Studies of bovine astrovirus have determined that the virus infects the M cells and adsorptive enterocytes overlying the dome villi of the Peyer's patches in the small bowel (Woode et al., 1984). Studies of lamb astrovirus revealed that a mild transient diarrhea occurred following infection of mature enterocytes in the apical two-thirds of the villi and the subepithelial macrophages (Snodgrass et al., 1979). While the above in vivo experiments suggest that astrovirus enters cells through the apical surface, studies of viral pathogenesis in cell culture indicate that human astroviruses enter through the basolateral surface instead of the apical surface (Willcocks et al., 1990).

Clinical Manifestations of Astrovirus Infection

Following a 1-4 day incubation period, astrovirus infection typically presents as watery diarrhea that resembles a mild form of rotavirus gastroenteritis. Astrovirus diarrhea is principally seen in young children 6 months to 2 years of age and may be associated with anorexia, fever, vomiting, abdominal pain. Although astrovirus diarrhea does not normally result in significant dehydration or hospitalization, individuals suffering from poor nutritional status, immunodeficiency, severe mixed infection, or underlying gastrointestinal disease are at risk for the development of complications (Blacklow et al., 1991; Greenberg et al., 1992).

Immunity

Immunity to astrovirus infection is not well understood. Young children and the institutionalized elderly are usually the populations that develop symptomatic infection, suggesting that antibody is acquired early in childhood, provides protection through adult life, and wanes late in life (Glass et al., 1996). A recent study by Koopmans and colleagues (1998) determined that there is no heterologous protection between the seven human astrovirus serotypes.

EPIDEMIOLOGY

Transmission

Astroviruses are transmitted from person to person by the fecal-oral route. Fecal-oral transmission has been verified by numerous volunteer studies. In 1979, Kurtz and colleagues examined filtrate from a child with mild gastroenteritis by EM and determined that it contained a large number of astrovirus particles. The filtrate was administered to eight volunteers by mouth. One volunteer developed diarrheal illness and shed large amounts of astrovirus in the feces and one experienced mild constitutional symptoms and a lower level of shedding. The original filtrate was subsequently given to nine more volunteers, two of whom shed virus. Thirteen of 16 inoculated subjects experienced an increase in titer of homologous antibody in their serum. Ultimately the authors concluded that astroviruses produce a transmissible infection of low pathogenicity for adults. Midthun and colleagues (1993) performed a similar volunteer study. Subjects were inoculated with a bacterium-free filtrate prepared from a stool specimen of an individual who had been ill during the 1978 Marin County outbreak of gastroenteritis which was attributed to human astrovirus serotype 5. Nineteen adult volunteers received an oral administration of the filtrate. One developed gastrointestinal illness and nine others experienced serologic responses.

Contaminated water and food have occasionally been implicated as the cause of astrovirus outbreaks. On June 21, 1991 there was an outbreak of acute gastroenteritis identified from students and staff of the 10 primary and 4 junior high schools in Katano City, Osaka Prefecture, Japan. All of the schools had separate water supplies but the schools lunches came from three central kitchens which all got their food from a common supplier. Oishi and colleagues (1994) examined food samples from the three central kitchens and fecal samples from 38 patients from 10 different schools. Astrovirus was detected in 10 of the fecal samples. Shigella species, Salmonella species, C. jejuni, S. aureus, V. parahemolyticus, B. cereus, Y. enterocolitica, and various types of Escherichia coli were not detected in either the food samples or the stool samples.

Utagawa and colleagues (1994) identified astrovirus as the etiologic agent of gastroenteritis in three separate outbreaks. In the first outbreak, 22 (67%) of 33 people who dined at a restaurant in the Nagano Prefecture later developed gastroenteritis. The median incubation period was 36 hours and the median duration of illness was 72 hours. Symptoms included vomiting (46%), diarrhea (73%), fever (46%), and abdominal pain (36%). Two of the nine specimens that were collected were found to have SRSV-like particles by EM, but these samples were used up during examination. Three of the remaining seven were positive for astrovirus by EIA. The second outbreak involved 1,419 kitchen staff and students from a school in the Yamaguchi Prefecture who became ill after eating contaminated school lunches. The median incubation period was 44 hours and the median duration of illness was 72 hours. Symptoms experienced by the individuals in the Yamaguchi outbreak also included vomiting (70%), abdominal pain (65%), fever (64%), and diarrhea (40%). Of the 28 fecal specimens obtained for examination (22 from kitchen staff and 6 from school children), six specimens were positive by EIA and all of the childrens' specimens were EM and EIA negative. In the third outbreak, 45 of 68 high schools students became ill within 24 hours after eating at a ski lodge. Again, symptoms included vomiting (78%), fever (73%), diarrhea (69%), and abdominal pain (69%). Of the six specimens examined, one was positive for astrovirus by EM and all specimens were positive for astrovirus by EIA.

Prevalence

Estimates of the prevalence of astrovirus infection vary widely depending on the population and the method of detection used. The findings of the major studies of astrovirus prevalence are summarized in Table 1. Using a Northern (RNA) dot blot technique, Palombo and colleagues (1995) detected astrovirus in 4.2% of fecal samples from children in Melbourne, Australia who had been hospitalized for acute gastroenteritis.

Using an ELISA capable of detecting the human astrovirus serotypes 1 to 5, Herrmann and colleagues (1991) performed controlled studies in Thailand comparing the incidence of astrovirus, rotavirus, and enteric adenovirus among pediatric outpatients with gastroenteritis. Astroviruses were detected in 8.6% of the children experiencing gastroenteritis. Rotavirus and enteric adenovirus were detected in 19% and 2.6% of those with gastroenteritis, respectively. Similarly, using the same ELISA, Cruz and colleagues (1992) detected astrovirus in 38.6% of rural Guatemalan children 0 to three years of age.

More recent studies by Koopmans and colleagues (1998) and Maldonado and colleagues (1998) have indicated an extremely high prevalence of astrovirus among certain populations. In the Koopmans study (1998), the overall percentage of persons with neutralizing antibodies to astrovirus ranged from 10-91% depending on the astrovirus serotype.

Using noncommercial ELISA and RT-PCR, Maldonado and colleagues (1998) identified astrovirus as the most common enteric pathogen in a population of rural Mayan infants living in the highlands of Chiapas, Mexico. In this population, little rotavirus or enteric adenovirus infection was detected in a three-year birth cohort of infants in the first year of life. Astrovirus was detected in 164 of 271 infants (61%) enrolled in an oral polio vaccine (OPV) immunogenicity study. This is the highest astrovirus prevalence reported among infants from any region in the world.

Prevalence of the Seven Human Astrovirus Serotypes

Numerous studies indicate that human astrovirus serotype 1 is the most predominant serotype worldwide. Table 2 summarizes the findings of the major studies on the prevalence of the seven human astrovirus serotypes. Lee and colleagues (1994) examined stool specimens from the Oxford Public Health Laboratory from children under six years of age who either had acute or recent gastroenteritis. Occasionally specimens from adults were also examined, but only when a viral agent of gastroenteritis was suspected. Specimens were first examined for viruses by electron microscopy and were then serotyped using immune electron microscopy (IEM). Of 291 astrovirus-positive stool specimens collected during the 17-year period from 1976-1992, 64.9% (n=189 specimens) contained astrovirus serotype 1, 11.3% (n=33 specimens) contained serotype 2, 9.3% (n=27 specimens) contained serotype 3, 11.3% (n=33 specimens) contained serotype 4, 2.1% (n=6 specimens) contained serotype 5, 0.3% (n=1 specimen) contained serotype 6, and 0.7% (n=2 specimens) contained serotype 7.

Noel and colleagues (1994) collected 162 astrovirus-positive stool samples from children with diarrhea from 1981-1993. The viruses isolated from these samples were typed using an enzyme immunoassay (EIA) and immune electron microscopy (IEM) and it was found that 66% (n=107) of the isolated viruses were serotype 1, 0.6% (n=1) were serotype 2, 6.2% (n=10) were serotype 3, 4.3% (n=7) were serotype 4, and 0% (n=0) were serotype 5. There was no information provided for astrovirus serotypes 6 and 7.

In 1995, Noel and colleagues used a typing enzyme immunoassay (TYPE-EIA) to determine the types of 64 astrovirus-positive specimens from nine collections from seven different countries. 52% (n=32 specimens) were astrovirus serotype 1, 11% (n=7 specimens) were astrovirus serotype 2, 16% (n=10 specimens) were astrovirus serotype 3, 10% (n=6 specimens) were astrovirus serotype 4, 2% (n=1 specimen) were astrovirus serotype 5, 2% (n=1 specimen) were astrovirus serotype 6, 0% (n=0 specimens) were astrovirus serotype 7.

In the Koopmans study, it was found that the overall percentage of persons with neutralizing antibodies to astrovirus was highest for astrovirus serotype 1 (91%), followed by 69% for serotype 3, 56% for serotype 4, 36% for serotype 5, 31% for serotype 2, 16% for serotype 6, and 10% for serotype 7.

Seasonal Distribution of Astrovirus Infection

In temperate regions, most astrovirus infections are detected in the winter while in tropical climates, most astrovirus infections are noted in during the rainy season (Matsui et al., In: Field's Virology; Maldonado, 1996). This temporal pattern of infection is similar to that of rotavirus infection (Maldonado, 1996).

Utagawa and colleagues (1994) examined the epidemiology and prevalence of astrovirus in six prefectural public health institutes in Japan from 1982 to 1992. The seasonal distributions of astrovirus from the sporadic cases examined by this study indicated that there was a peak in March and April which accounted for 65% of the cases. Sporadic cases did occur from November to May, but no case was detected from June to October.

Cruz and colleagues (1992) conducted a longitudinal study of astrovirus infections and diarrhea among rural ambulatory children under three years of age living in a rural community of Guatemala and found that the highest monthly detection rate of astrovirus in sick children occurred in May which was the start of the rainy season.

Maldonado and colleagues (1998) examined the epidemiology of astrovirus in a population of rural Mayan infants living in the highlands of Chiapas, Mexico and found that although astrovirus was present throughout the year, it peaked in March and May and decreased in September. Increased astrovirus prevalence correlated best with the beginning of the warm rainy season.

Although this pattern has been observed in a number of populations, the reasons behind the seasonal distribution of astrovirus infection have yet to be explained.

Age Distribution of Astrovirus Infection

To date, there have been a limited number of studies on the age distribution of astrovirus infection, but the few studies that have been conducted suggest that the majority of people acquire antibodies by the time they are five years of age.

A study by Kurtz and colleagues (1978) examined the prevalence of astrovirus antibody in 87 children less than 10 years of age living in the Oxford region of the United Kingdom. Astrovirus antibodies were detected in 7% of infants 6 to 12 months of age, 70% of school-aged children, and 75% of children 10 years of age. A group of nursing students were also tested and astrovirus antibodies were detected in 77% of these young adults.

More recently, Koopmans and colleagues (1998) used neutralizing tests to determine the seroprevalence rates of neutralizing antibodies in an age-stratified sample of the population in the Utrecht Province. The age-stratified overall seroprevalence of the seven human astrovirus serotypes was approximately 15% for individuals less than 1 year of age, 29% for individuals 1 to 4 years of age group, and 44% for individuals 5 years of age and older.

DISCUSSION

Many questions still remain regarding the epidemiology of astroviruses. In particular, the seasonal distribution of astrovirus infection remains largely unexplained even though it has been well-documented that most astrovirus infections are noted during the winter months in temperate climates and during the rainy season in tropical climates. While the high incidence noted during the winter months is extremely puzzling, it is likely that, in tropical climates, the rainy season creates conditions which increase the opportunities for transmission such as the breakdown of sanitation in developing nations. Interestingly, the seasonal distribution of astrovirus infection is different from that observed for most pathogens spread by the fecal-oral route yet it is similar to the pattern observed for rotavirus.

The few studies that have been done on the age distribution of astrovirus infection indicate that antibody is acquired before five years of age. Studies that look at astrovirus prevalence among children under the age of five who have been divided into separate one year age groups would help to clarify the issue of age distribution. Understanding more about the age distribution of this pathogen will be of particular importance if vaccination is to be one of the long term goals of astrovirus research.

Although previous studies on the prevalence of the seven human astrovirus serotypes have clearly indicated that serotype 1 is the most prevalent worldwide, these studies have also produced widely varying results which has mainly been due to differences in the sensitivities of the detection methods used by the investigators. A comparison of the various detection methods and their sensitivities would allow us to determine which procedure should be used as the standard detection method thereby making the results of subsequent studies would be more meaningful and easier to compare. Another major problem with the prevalence studies to date is that they examine the prevalence of astrovirus among extremely different populations. For example, a number of studies look at the prevalence of astrovirus among young children with gastroenteritis while others look at the prevalence of astrovirus among healthy individuals. Many of the studies of children with gastroenteritis were performed in order to determine whether or not astrovirus is an etiologic agent of gastroenteritis. Now that we know astrovirus is the second most common cause of viral gastroenteritis, studies that investigate the prevalence of the seven serotypes among healthy populations living in different regions of the world would greatly advance our understanding of the epidemiology of this pathogen. Finally, once use of the rotavirus vaccine begins, prevalence studies of astrovirus will become of increasing importance because they will allow us to determine whether or not the role of astrovirus infection in the morbidity and mortality of young children is changing and whether or not intervention is needed.

The medical importance of astroviruses results from the prevalence of this pathogen among the pediatric population and the fact that astrovirus gastroenteritis can result in complications and lead to hospitalization among individuals suffering from poor nutritional status, immunodeficiency, severe mixed infection, or underlying gastrointestinal disease. If the licensing and routine use of the rotavirus vaccine occurs in the future, as it appears it will, the impact of astrovirus infection on the morbidity of infants and children may become increasingly important. The development of an effective astrovirus vaccine would be a worthwhile goal, especially for the protection of young children, the institutionalized elderly, and those at risk for the development of complicated disease. A vaccine would be of greatest medical benefit to those living in developing countries where tends to be a higher rate of malnourishment.

Before we can begin to develop a vaccine, there are a number of issues regarding both immunity to and the epidemiology of astrovirus that need to be clarified. First, there is the issue of heterologous versus homologous protection. Koopmans and colleagues (1998) determined that there was no heterologous protection between the seven human astrovirus serotypes, but other studies need to confirm these findings before any solid conclusions can be made. If it is confirmed that there is no heterologous protection between the human astrovirus serotypes, it will probably be necessary to develop a "septavalent" vaccine that would provide protection from all seven serotypes. The other issue that will be important in the development of a vaccine is the prevalence of the seven serotypes. If future prevalence studies reveal that serotype 1 is significantly predominant then it might be possible to vaccinate against just one serotype. However, if studies reveal that serotype 1 is not as predominant as was previously believed, then it may be necessary to develop a vaccine that incorporates more than one serotype.

As this review has shown, our understanding of astrovirus has increased greatly in recent years, but there is still much that we have to learn. Now that we know the basics of the epidemiology of this virus, we need to work on clarifying our understanding and eventually begin working towards the goal of vaccination.

 

Table 1. Summary of Major Astrovirus Prevalence Studies

Study	Region	Age of Study Population	Detection Method	Astrovirus Prevalence
Palombo et al. (1978)	Melbourne, Australia	<5 years of age	Northern (RNA) Dot Blot	4.2%
Herrmann et al. (1991) Cruz et al. (1992)	Bangkok, Thailand   Guatemala	<5 years of age   <3 years of age	ELISA   ELISA	8.6%   38.6%
Koopmans et al. (1998) Maldonado et al. (1998)	The Netherlands   Chiapas, Mexico	<1 year to 79 years of age <3 years of age	Neutralization Tests ELISA and RT-PCR	10-91%*   61%

* This wide range in prevalence values is a result of differences in the prevalence of the seven different human astrovirus serotypes as illustrated by Table 2.

 

Table 2. Summary of the Major Serotype Prevalence Studies

Prevalence of Astrovirus Serotype

Study	Detection Method	1	2	3	4	5	6	7
Lee et al. (1994) Noel et al. (1994) Noel et al. (1995)	IEM   EIA and IEM   EIA	64.9%   66%   52%	11.3%   0.6%   11%	9.3%   6.2%   16%	11.3%   4.3%   10%	2.1%   0%   2%	0.3%   -*   2%	0.7%   -*   0%
Koopmans et al. (1998)	Neutralization Tests	91%	33%	69%	56%	36%	16%	10%

* No data was available for the prevalence of serotypes 6 and 7 in this study.

 

REFERENCES

1. Appleton, H; Higgins, PG. Viruses and gastroenteritis in infants. The Lancet 1975; i: 1297.

2. Blacklow, NR; Greenberg, HB. Viral gastroenteritis. New England Journal of Medicine, 1991, 325: 252-264.

3. Cruz, JR; Bartlett, AV; Herrmann, JE; Caceres, P; Blacklow, NR; Cano, F. Astrovirus-associated diarrhea among Guatemalan ambulatory rural children. Journal of Clinical Microbiology, 1992, 30: 1140-1144.

4. Glass, RI; Noel, J; Mitchell, D; Herrmann, JE; Blacklow, NR; Pickering, LK; Dennehy, P; Ruiz-Palacios, G; de Guerrero, ML; Monroe, SS. The changing epidemiology of astrovirus-associated gastroenteritis: a review. Archives of Virology. Supplementum, 1996, 12:287-300.

5. Greenberg, HB; Matsui, SM. Astroviruses and caliciviruses: emerging enteric pathogens. Infectious Agents of Disease, 1992, I: 71-91.

5. Herrmann, JE; Taylor, DN; Echeverria, P; Blacklow, NR. Astroviruses as a cause of viral gastroenteritis in children. New England Journal of Medicine, 1991, 324: 1757-1760.

6. Kapikian, AZ. Viral gastroenteritis. JAMA, 1993, 269: 627-630.

7. Koopmans, MP; Bijen, MH; Monroe, SS; Vinjbe, J. Age-stratified seroprevalence of neutralizing antibodies to astrovirus types 1 to 7 in humans in The Netherlands. Clinical and Diagnostic Laboratory Immunology, 1998 Jan, 5(1):33-7.

8. Kurtz, J; Lee, T. Astrovirus gastroenteritis age distribution of antibody. Medical Microbiology and Immunology, 1978; 166: 227-230.

9. Kurtz, JB: Lee, TW; Craig, JW; Reed, SE. Astrovirus infection in volunteers. Journal of Medical Virology, 1979, 3: 221-230.

10. Lee, TW; Kurtz, JB. Prevalence of human astrovirus serotypes in the Oxford region 1976-1992, with evidence for two new serotypes. Epidemiology and Infection, 112: 187-193.

11. Lew, JF; Moe, CL; Monroe, SS; Allen, JR; Harrison, BM; Forrester, BD; Stine, SE; Woods, PA; Hierholzer, JC; Herrmann, JE; Blacklow, NR; Bartlett, AV; Glass, RI. Astrovirus and adenovirus associated with diarrhea in children in day care settings. Journal of Infectious Diseases, 1991, 164: 673-678.

12. Madeley, CR; Cosgrove, BP. Viruses in infantile gastroenteritis in infants. The Lancet 1975, ii:124.

13. Maldonado, YA. Astrovirus infections in children. The Report on Pediatric Infectious Diseases, 1996, 6: 39-40.

14. Maldonado, YA; Cantwell, M; Old, M; Hill, D; de la Luz Sanchez, M; Logan, L; Millan-Velasco, F; Valdespino, JL; Sepulveda, J; Matsui, S. Population-based prevalence of symptomatic and asymptomatic astrovirus infection in rural Mayan infants. Journal of Infectious Diseases, In press.

15. Matsui, SM; Greenberg, HM. Astroviruses. In: Fields Virology, Third Edition. Fields, BN; Knipe, DM; Howley, PM (eds.). Raven Press, Ltd., New York, 1996, pp. 811-824.

16. Midthun, K; Greenberg, HB; Kurtz, JB; Gary, GW; Lin, FC; Kapikian, AZ. Characterization and seroepidemiology of a type 5 astrovirus associated with an outbreak of gastroenteritis in Marin County, California. Journal of Clinical Microbiology, 1993, 31: 955-962.

17. Noel, JS; Cubbitt, D. Identification of astrovirus serotypes from children treated at the Hospitals for Sick Children, London 1981-93. Epidemiology and Infection, 1994, 113: 153-159.

18. Noel, JS; Lee, TW; Kurtz, JB; Glass, RI; Monroe, SS. Typing of human astroviruses from clinical isolates by enzyme immunoassay and nucleotide sequencing. Journal of Clinical Microbiology, 1995 April, 33(4):797-801.

19. Oishi, I; Yamazaki, K; Kimoto, T; Minekawa, Y; Utagawa, E; Yamazaki, S; Inouye, S; Grohmann, GS; Monroe, SS; Stine, SE; et al. A large outbreak of acute gastroenteritis associated with astrovirus among students and teachers in Osaka, Japan. Journal of Infectious Diseases, 1994 Aug, 170(2):439-43.

20. Palombo, EA; Bishop, RF. Annual incidence, serotype distribution, and genetic diversity of human astrovirus isolates from hospitalized children in Melbourne, Australia. Journal of Clinical Microbiology, 1996 July, 34(7):1750-3.

21. Phillips, AD; Rice, SJ; Walker-Smith, JA. Astrovirus within human small intestinal mucosa. Gut, 1982, 23: A923-924.

22. Snodgrass, DR; Angus, KW; Gray, EW; Menzies, JD, Paul, G. Pathogenesis of diarrhoea caused by astrovirus infections in lambs. Archives of Virology, 1979, 60: 217-226.

23. Tucker, AW; Haddix, AC; Bresee, JS; Holman, RC; Parashar, UD; Glass, RI. Cost-effectiveness analysis of a rotavirus immunization program for the United States. JAMA, 1998, 279: 1371-1376.

24. Utagawa, ET; Nishizawa, S; Sekine, S; Hayashi, Y; Ishihara, Y; Oishi, I; Iwasaki, A; Yamashita, I; Miyamura, K; Yamazaki, S. Astrovirus as a cause of gastroenteritis in Japan. Journal of Clinical Microbiology, August 1994, 32(8):1841-5.

25. Willcocks, MM; Carter, MJ; Laidler, FR; Madeley, CR. Growth and characterization of human faecal astrovirus in a continuous cell line. Archives of Virology, 1990, 113: 73-81.

26. Woode, GN; Pohlenz, JF; Kelso Goulery, NE; Fagerland, JA. Astrovirus and breda virus infections of dome cell epithelium of bovine ileum. Journal of Clinical Microbiology, 1984, 19: 623-630.

Last modified: Tue Jun 16 12:36:39 PDT 1998