Viral Agents of Gastroenteritis Public Health Importance and Outbreak Management

This report was prepared by: Charles W. LeBaron, M.D. Navid P. Furutan, M.D.* Judy F. Lew, M.D. James R. Allen, Ph.D. Vera Gouvea, Ph.D. Christine Moe, Ph.D. Stephan S. Monroe, Ph.D. *Department of Pediatrics, Emory University Summary Each year, infectious gastroenteritis causes greater than 210,000 children in the United States to be hospitalized and 4-10 million children to die worldwide. Since the mid-1970s, knowledge has increased dramatically concerning the viral agents that are responsible for much of this public health burden. Rotavirus, the most common cause of diarrhea among children, infects virtually every child in the United States by the age of 4 years and causes potentially lethal dehydration in 0.75% of children less than 2 years of age. Other recently identified pathogens include the enteric adenoviruses, calicivirus, astrovirus, and the Norwalk family of agents. Conclusive diagnosis of these viruses requires electron microscopic examination of stool specimens, a laboratory technique that is available only at a few large centers, including CDC. Stool samples from an outbreak that are submitted to CDC for detection of viral pathology should be collected in bulk from 10 ill persons during their first 48 hours of illness, while feces are still liquid, and should be stored at 4 C (not frozen). Acute- and convalescent-phase serum samples should be collected from the same persons, plus from an equal number of controls, during the first week of illness and 3 weeks thereafter. Control measures for outbreaks of viral gastroenteritis should focus on the removal of an ongoing common source of infection (e.g., an ill food handler or the contamination of a water supply) and on the interruption of person-to-person transmission that can perpetuate an outbreak in a population after the common source has been removed. Because improvements in environmental hygiene may not be accompanied by reductions of endemic diarrhea caused by viruses, immunization may play an important role in future control; vaccine trials for rotavirus are in progress. In anticipation of vaccine development and use, CDC recently began national surveillance for the viral agents of gastroenteritis. Health-care facilities involved in the detection of rotavirus or the other viral agents of diarrhea can participate. INTRODUCTION

The public health burden of infectious diarrhea is substantial, particularly among children, both in the United States and worldwide. Each year in the United States greater than 210,000 children less than 5 years of age are hospitalized for gastroenteritis for an average of 4.5 days, at an annual inpatient cost of almost $1 billion (1). During the period 1973-1983, an average of 500 children in the United States died from diarrhea each year (2). Also, 25 work or school days/100 children are lost each year as a result of acute gastroenteritis (3), and approximately 14% of children in the United States are treated by a physician for rotavirus diarrhea alone (CDC, unpublished data). Worldwide, 3-5 billion cases of diarrhea occur, causing 5-10 million deaths annually (4).

Until the 1970s, diagnostic techniques for infectious diarrhea were limited to bacteria and protozoa, and an etiologic agent could be identified in a limited proportion of cases. Investigators had hypothesized, however, that viruses might account for many of the cases of unknown etiology. In 1972, in the examination of stool specimens, electron microscopy identified the Norwalk agent, the most common viral cause of gastroenteritis outbreaks among adults. In 1978, the same technique was used to detect rotavirus, the most common cause of severe diarrhea in children. Since that time, knowledge about these and other more recently discovered pathogens has increased dramatically. Unfortunately, diagnostic technology is insufficiently developed to permit determining the disease burden of each of the known viral pathogens. Even when diagnostic efforts are pursued aggressively, an agent cannot be identified for almost half of diarrheal cases (5).

Current priorities in enteric viral research include: 1) improving diagnostic capabilities for known pathogens to determine their endemic importance and their role in outbreaks; 2) identifying new agents for the 50% of diarrhea cases that are still of unknown etiology; and 3) determining the modes of transmission and the means to prevent disease, including the characteristics of natural immunity and effective vaccines.

As a result of rapid progress in these areas, etiologic identification may soon be possible in most cases of diarrhea, and within a few years a rotavirus vaccine may be licensed. Thus, public health and other health-care professionals should be familiar with the viral agents of gastroenteritis, their role in endemic disease, methods of diagnosis, and measures to manage outbreaks.

This document was prepared by the Viral Gastroenteritis Section, Respiratory and Enterovirus Branch (REB), Division of Viral and Rickettsial Diseases (DVRD), Center for Infectious Diseases (CID), and is intended primarily for the use of state and local health departments who investigate outbreaks. It may also be of use in academic settings, research groups, or other groups interested in studying outbreaks of gastroenteritis. PATHOGENS: EPIDEMIOLOGIC AND CLINICAL FEATURES Rotavirus Endemic disease

Rotavirus is the most common cause of severe diarrhea among children. In the United States, approximately 3.5 million cases occur each year. A child has a 2% lifetime chance of being hospitalized for rotavirus diarrhea, which accounts for 35% of diarrheal hospital stays (1) and an estimated 75-125 childhood deaths annually (2). Worldwide, an estimated 140 million cases occur each year, causing almost 1 million deaths (6).

In the United States, the peak incidence of rotavirus diarrhea is among children 6 months-2 years of age, although in developing countries younger infants may be affected. By 4 years of age, most persons have been infected and are immune to the severe dehydrating syndrome, but a high inoculum or lowered immunity can produce milder illness among older children or adults. One-third of parents whose children are infected with rotavirus become ill (7), and rotavirus diarrhea can occur among travelers to developing nations, the elderly, and persons with debilitative or immunosuppressive conditions. In the United States, rotavirus activity is concentrated in the cooler months of the year (October-April). Clinical syndrome

In general, the incubation period is approximately 2 days and is followed by vomiting for 3 days and watery diarrhea for 3-8 days. Fever and abdominal pain occur frequently. Usually, there are no sequelae other than temporary intolerance of lactose; however, without adequate fluid replacement the syndrome can result in severe dehydration and death. Modes of transmission

A person with rotavirus diarrhea may excrete approximately 1 trillion infectious particles/milliliter of stool. Since the infective dose in a child can be as few as 10 particles (8), person-to-person transmission probably perpetuates endemic disease. Infectivity does not parallel the presence of symptoms. Asymptomatic rotavirus excretion has been reported among half of children the day before diarrhea starts and among one-third during the week after symptoms end (9). Many children can shed rotavirus and never become ill (10,11). When food or water is contaminated to the extent that it overcomes adult immunity, rotavirus diarrhea among adults may be relatively common. In Thailand, 5% of adult gastroenteritis has been found to be caused by rotavirus (12). Outbreak characteristics

Nosocomial rotavirus among pediatric populations is common; in one study, all children hospitalized for greater than 2 weeks during rotavirus season ultimately shed the virus (13). Rotavirus at day-care centers, in both endemic and outbreak form, is also common (9,14,15). Outbreaks in neonatal units are frequently reported, but infection among full-term infants is usually benign, perhaps because maternal antibody transferred during the third trimester protects against illness for the first 3-6 months of life; premature infants are at higher risk. Among adults, an outbreak arising from rotavirus contamination of a municipal water supply has been reported (16), and foodborne transmission was suspected in two other outbreaks involving banquets (CDC, unpublished data). Immunity

Infection generally confers long-term immunity to serious gastroenteritis (17), but asymptomatic or minimally symptomatic reinfection can occur throughout life. Immunity may wane among the elderly, rendering them susceptible again to severe disease. Atypical rotaviruses

Rotaviruses affecting humans were once thought to be limited to one antigenic family termed Group A, whereas other antigenic groups (B-E) were thought to be strictly zoonotic. In 1982, however, an epidemic of Group B rotavirus affected millions of persons in China (including adults, children, and neonates) (18), and since then outbreaks have recurred, although affecting fewer persons. Studies of immunoglobulin pools from Shanghai suggest that the Chinese population had been exposed to this pathogen in the past (19). Since Group B rotavirus is a common diarrheal pathogen for swine, and since all rotaviruses have a segmented genome (similar to influenza) capable of antigenic changes through reassortment of genes, investigators have hypothesized that this human Group B epidemic arose by a reassortment that allowed the swine virus to propagate in the human gut (20). Thus far, the Group B epidemic is not known to have spread beyond China, despite serologic evidence of human infection in Southeast Asia. Group C rotaviruses are also primarily swine pathogens but have been detected among humans in all parts of the world. Outbreaks have occurred in Japan and England, but the importance of Group C rotaviruses in endemic disease is unknown (21,22). Although both Group B and Group C rotaviruses in humans have been reported in the United States, serum studies suggest that exposure has been minimal in the past. With the U.S. population almost wholly susceptible, the extent to which these atypical rotaviruses represent a potential public health hazard in the United States is unclear. Adenovirus Endemic disease

Adenoviruses are widely recognized causes of respiratory, ocular, and genitourinary infections. However, serotypes 40 and 41 (previously called fastidious enteric adenoviruses) primarily affect the gut, contributing to 5%-20% of hospitalizations for childhood diarrhea in developed countries (23,24). Peak incidence is among children less than 2 years of age, but older children and adults may be infected, with or without symptoms. Infections occur throughout the year with no clear peaks (25). Other serotypes of adenovirus, particularly 31, have also been associated with diarrhea (26). Clinical syndrome

Incubation is between 3 and 10 days, with illness lasting greater than or equal to 1 week, longer than for other enteric viral pathogens (23,27). Diarrhea is more prominent than vomiting or fever, and respiratory symptoms are often present. Modes of transmission

Person-to-person transmission is presumably the principal mechanism for the spread of infection. Asymptomatic shedding has been documented, but generally infectivity parallels symptomatic disease (25). Food and water have not been reported as vehicles. Outbreak characteristics

Reported outbreaks have tended to occur in hospitals or day-care settings, and all have involved children. Adult contacts were infrequently affected (28,29). Immunity

Long-term immunity is thought to be acquired during childhood infection. Calicivirus Endemic disease

A British study suggested that approximately 3% of children hospitalized for diarrhea excrete calicivirus (30), and a U.S. study found approximately the same percentage (2.9%) for children with diarrhea in day-care centers (31). On the basis of antibody-prevalence studies of pooled immunoglobulin and serum samples from many parts of the world, most persons appear to have been infected by age 12, and the peak acquisition takes place between 3 months and 6 years (32,33). Seasonality is unknown. Clinical syndrome

The incubation period is 1-3 days, with illness lasting an average of 4 days. Vomiting and diarrhea are common, with upper respiratory symptoms and fever occurring less frequently. Infections in the elderly have also been documented. Modes of transmission

Person-to-person transmission is presumed to be essential for endemic disease, but contaminated shellfish, cold foods, and drinking water have been implicated as vehicles (34). Outbreak characteristics

Of seven calicivirus outbreaks reported in the literature since 1979, all occurred in institutional settings. Four outbreaks affected children: one in an orphanage and another at a school in Japan (35,36), and one in an infant-mother hospital unit and another at a school in England (37,38). Three outbreaks involved the elderly in nursing-home settings in England and Japan (39-41). Attack rates ranged from 50% to 70%. No calicivirus outbreak in the United States has been reported. Immunity

In the reported outbreaks, mothers of infected infants were rarely infected, suggesting that young adults retain effective immunity from earlier exposures, although the outbreaks among the elderly suggest that this immunity may wane with age. Astrovirus Endemic disease

Studies of hospitalized children suggest that astroviruses may account for 3%-5% of admissions for diarrhea (30). Children less than 7 years of age are principally affected, although adults can be infected and suffer mild disease. Antibody to all five serotypes of astrovirus was present in a study of pooled American gamma globulin from the United States (CDC, unpublished data), and 75% of British children have acquired antibody by the age of 10 years (42). Australian data suggest a winter peak (43). Clinical syndrome

The incubation period is between 24 and 36 hours, with illness lasting 1-4 days. Gastrointestinal symptoms are nonspecific, consisting of vomiting, diarrhea, fever, and abdominal pain. Astrovirus in ducks has been associated with hepatitis, but hepatic involvement has not been reported for humans. Modes of transmission

Although most transmission is probably person-to-person among children, contaminated water and shellfish have given rise to outbreaks in Britain (44). Asymptomatic shedding has been documented (45). Outbreak characteristics

Institutional settings account for four reported astroviral outbreaks from Britain and Japan: a kindergarten (46), a pediatric ward (47), and two nursing homes (39,48). In the kindergarten, a 50% attack rate was reported, with secondary transmission of the illness to families occurring in one-third of the cases. Immunity

The characteristics of immunity to astrovirus are unknown. Since reported outbreaks have involved only children and the elderly, young adults may have resistance to infection. Norwalk-like Viruses Endemic disease

The Norwalk virus is the representative agent of a heterogenous group of viruses, also called small round structured viruses (SRSVs) or the Norwalk-like family of agents. The antigenic interrelationships among the many members of this class are complex, and the agents are usually identified by the locale where an outbreak occurred (e.g., Hawaii, Snow Mountain, Montgomery County, Taunton, Amulree, Sapporo, and Otofuke). Although Norwalk-like viruses may play a role in endemic diarrhea, diagnostic technology is not yet sufficiently developed to assess this aspect of their disease burden. In the United States, illness is most commonly reported among persons of school age and older. Levels of antibody specific to the Norwalk agent are low during childhood but reach 50% by middle age (49). In developing countries, antibodies are acquired at an earlier age; peak incidence of illness may also occur among younger age groups than in developed nations (50). Clinical syndrome

The incubation period is 24-48 hours, and the mean duration of illness is 12-60 hours. Nausea is prominent, with vomiting, non-bloody diarrhea, and abdominal cramps occurring in most cases. These symptoms are experienced by all age groups, but diarrhea is relatively more prevalent among adults, whereas a higher proportion of children experience vomiting. From 25%-50% of affected persons also report headache, fever, chills, and myalgias. Adults have died during illness caused by Norwalk-like viruses, presumably from electrolyte imbalance. Late sequelae have not been reported, but the elderly often report persistence of constitutional symptoms for up to several weeks. Mode of transmission

Routes of transmission that have been documented include water, food (particularly shellfish and salads), aerosol, fomites, and person-to-person contact. Infectivity can last for as long as 2 days after resolution of symptoms (51). Presymptomatic shedding has been suspected on epidemiologic grounds but not proven in volunteer studies. Outbreak characteristics

Numerous reports have described the course of outbreaks caused by Norwalk-like agents, usually involving adults and older children. The settings are diverse and include banquets, cruise ships, geriatric facilities, psychiatric wards, emergency rooms, cafeterias, recreational lakes, swimming pools, campgrounds, football teams, hotels, schools, dormitories, fast food restaurants, and others (52). Norwalk-like agents probably create a low background level of infection in a community until an infected individual contaminates a common source, and an explosive outbreak occurs. Although secondary cases can multiply the number of persons affected, outbreaks are generally limited to 1-2 weeks unless transmission is facilitated by a closed environment (e.g., a nursing home) or prolonged by renewal of the susceptible population (e.g., a new set of passengers on a cruise ship). Immunity

Studies of volunteers have documented the paradox that persons with the highest preexisting levels of Norwalk antibodies are at highest risk of developing symptomatic infection (49). Most persons' antibody levels against Norwalk virus rise after infection; these titers normally peak by the third week and persist until approximately the sixth week, after which they decline. Although preexisting antibody levels correlate with risk of symptomatic illness upon exposure to the virus, acutely elevated antibody levels appear to correlate with resistance to reinfection. The nature of resistance and susceptibility to the Norwalk-like agents is poorly understood. Other Viruses

Several other agents, listed below, have been implicated in viral gastroenteritis, but their public health importance is not yet clear. Pestivirus

A recent study showed that on an Arizona Indian reservation, 23% of specimens from children less than 2 years of age with gastroenteritis of unknown etiology were antigen-positive for pestivirus, compared with 3% of controls. Illness was relatively mild, and duration was 3 days (53). Antibody studies of serum samples from Arizona, Maryland, and Peru suggested that 30%-50% of children and adults had been infected, with peak exposure occurring at less than 2 years of age (54). Picobirnavirus

Reports from Brazil documented human cases of diarrhea caused by picobirnavirus, which had been thought to be a cause of diarrhea only in animals (55). The importance of this pathogen is unknown. Parvovirus

Parvovirus-like particles have been identified by electron microscopy in stool specimens of both well and ill persons in Britain (56). The relationship of these particles to disease is unclear, but they have been associated with shellfish-related outbreaks of gastroenteritis (57). Enteroviruses

Enteroviruses cause a wide spectrum of disease, in which gastroenteritis plays a minor role (58). Although the entry of polio, coxsackie, echo, or other enteroviruses through the gut may cause incidental mild diarrheal symptoms, the spread of the virus through the bloodstream to other organs (e.g., central nervous system, heart, pleura, pancreatic islets) produces major disease manifestations. Although reports have linked some enteroviruses to illnesses in which diarrhea was the sole symptom, an outbreak or case of gastroenteritis should not be attributed to an enterovirus merely because it was isolated in the stool of an affected person. Torovirus

Toroviruses are known causes of diarrhea among cattle, and identification in human specimens has been reported (59). Antibodies to the Breda strain of toroviruses were not present, however, in 100 human sera from Britain, and the torovirus' importance as a cause of human disease is unknown (60). Coronavirus

Coronaviruses are well-established causes of diarrhea in animals and respiratory disease in humans. These viruses have been identified in the stool of persons with gastroenteritis (usually children less than 2 years of age), but human controls have been found to shed them with higher frequency (61,62), raising doubt about their etiologic role in human diarrhea. Coronaviruses have been detected most frequently in the southwest; one group reported that more than two-thirds of diarrheal stools examined by electron microscopy over an 8-year period contained such viruses, although no comparison was made with specimens from well persons (63). Worldwide, coronaviruses have been detected at highest rates in situations of poor sanitation. METHODS OF VIRAL DETECTION Antigen Detection

Commercial antigen-detection kits for rotavirus are widely available, inexpensive, and permit rapid viral diagnosis. Only small amounts of stool are required for the tests, and samples may be frozen before testing. Kits vary widely in range of sensitivities (70%-100%) and specificities (50%-100%) (64,65). Newborns and breast-feeding children have particularly high false-positive rates. Such kits are most useful for childhood diarrhea during the normal rotavirus season; they have less diagnostic value in situations in which rotavirus is probably rare, as in community outbreaks involving adults or in outbreaks of pediatric diarrhea outside the rotavirus season. Confirmatory testing should be performed in any case in which rotavirus disease would be unusual (e.g., among children in the summer or among adults at any time) as well as periodically to validate the reliability of the assay employed.

A commercial kit for enteric adenoviruses is also available, but because adenoviral diarrhea affects mainly children less than 2 years of age and because outbreaks involving adults have never been reported, the diagnostic value outside the preschool-age group is also limited.

Antigen-detection systems have been used for research on calicivirus, Norwalk, Snow Mountain agent, and astrovirus. Rapid assays for these and other agents are under development at the Viral Gastroenteritis Section, REB, DVRD, CID, CDC. Such techniques would allow the testing of large numbers of samples in a short period of time, an essential condition for determining the contribution of specific viruses to the incidence of diarrhea among populations and for identifying an etiologic agent during an outbreak. Antibody Detection

Persons infected with a viral agent of gastroenteritis will usually have a rise in antibodies to that virus.

For the Norwalk agent (the most commonly identified agent of outbreaks involving persons greater than 4 years of age), approximately half of adult Americans have preexisting IgG antibodies to the virus, so that a single specimen is insufficient to document recent infection. But if at least half of affected persons in an outbreak have a fourfold rise in specific antibody titers, the Norwalk agent can be designated as etiologic. Titers may begin to rise by the fifth day after onset of symptoms, peak at approximately the third week, and often begin to fall by the sixth week. Hence, the acute-phase serum should be drawn within the first week and the convalescent-phase serum during the third to sixth weeks. Diagnostic assays for IgM and IgA antibodies to Norwalk virus have been used on an experimental basis (66,67).

One disadvantage of serologic diagnosis is that patients are often reluctant to have serum drawn a month after a brief, self-limited illness. Furthermore, because this class of viruses cannot be cultivated, the supply of antigen for antibody testing is limited to a few research laboratories and cannot be offered for routine screening. In addition, antibodies can be detected to the Norwalk virus only, not the full spectrum of Norwalk-like agents that may cause disease. In a survey of 100 gastroenteritis outbreaks thought to be of viral origin submitted to CDC from 1985 to 1988, the Norwalk agent was identified by antibody rise (i.e., half or more of the persons showed a fourfold rise) in approximately 20% of outbreaks. Approximately 40% showed partial rises (less than half of the persons with a fourfold rise), suggesting that an antigenically related agent may have been involved. The remaining 40% showed no titer rises at all, indicating that an agent completely distinct from Norwalk virus caused the outbreak (68).

Adequate supplies of antigen are essential to any virus-testing system. Although rotavirus and astrovirus have been cultivated, the Norwalk family of viruses has proved resistant, and it may be several years before a panel of molecular diagnostic assays for known enteric viruses is available. Electron Microscopy

Under an electron microscope, a virus can be identified by its characteristic morphology in a stool specimen. The technique is highly specific but requires substantial resources. Since an electron microscope scans a field of approximately 1 millionth of a milliliter, there must be at least 1 million viruses/milliliter of stool for a detection to be made. Such levels of excretion are normally present only during the first 48 hours of viral diarrhea.

With a specialized technique called immune electron microscopy (IEM), the sensitivity of normal transmission electron microscopy can be improved 10-100 times. In one technique, the grid to be examined is coated with convalescent-phase serum before the stool specimen is applied; a high titer of virus-specific antibody tends to hold aggregates of homologous virus in the field, thereby enhancing diagnostic yield. Because reagents are scarce, this technique for diagnosing viral gastroenteritis is limited to a few centers in the United States. Other Techniques Culture

Rotavirus, enteric adenoviruses, and astrovirus can be cultured in research centers, but the techniques are not well suited for routine diagnosis. The other known major viral enteric pathogens cannot yet be cultivated. Enteroviruses can be cultivated but are not thought to be important causes of diarrhea. Electropherotyping

Rotavirus multiplies in the gut with such efficacy that, during infection, its genome dominates the ribonucleic acid (RNA) content of stool. When this RNA is extracted from stool and run on a gel, an electric field will cause the separate rotavirus genes to migrate in characteristic patterns. The presence of these patterns is diagnostic of rotaviral infection, and pattern variations provide insight into strain differences. Sensitivity is comparable with antigen detection ( greater than 90% in the first few days of illness), and specificity is 100%; in addition, atypical rotaviruses (Groups B and C) can be detected, which is not possible with antigen-detection systems for rotavirus A. Electropherotyping has long been used as the principal diagnostic technique for rotavirus in many nations, but in the United States it is used mainly as a research tool. Hybridization probes

Dot-hybridization assays have been developed for rotavirus that are considerably more sensitive than conventional antigen-detection techniques and at least equally specific (69,70), but they are currently available in only a few research centers. Hybridization assays have also been developed for adenoviruses, but they are less sensitive than antigen-detection techniques (26). Polymerase chain reaction (PCR)

Enzymatic amplification of a viral gene to raise its concentration in a specimen to the level of detectability would represent an ideal diagnostic technique for viral gastroenteritis, because at present agents can be identified only when they are excreted at maximal levels (millions of viruses/milliliter of stool). The development of PCR reagents, however, requires cloning and sequencing of viral genes. Cloning and sequencing have been accomplished only with rotavirus, but PCR techniques are being actively developed for the other agents. ENDEMIC CONTROL

Although essential in outbreak management, improved environmental hygiene (i.e., food, water, and sanitation) may be ineffective in endemic control of some of the known viral agents of gastroenteritis (e.g., rotavirus), perhaps because person-to-person transmission is the principal mechanism for the spread of infection. As a result, the population-based attack rate for these agents is thought to be the same (100%) in developed and developing countries, although disease caused by known agents tends to be acquired earlier in developing countries. The risk of death is highest in areas where medical care is least available and malnutrition is most prevalent.

Vaccines may prove to be the most effective method of endemic control. Because rotavirus is the leading cause of diarrheal mortality in the world and because natural infection appears to induce lifetime immunity, rotavirus has been the main focus of efforts at vaccine development: within 10 years of its discovery in 1973, trials were already under way. Although four major serotypes affect humans, three prototype vaccines have each been of a single serotype, and the results have been inconsistent (71-75). These inconsistencies have been ascribed by some investigators to differences between the vaccine serotype and the strain of rotavirus in the community. A large, multicenter, 2-year, efficacy trial of a vaccine that incorporates all four major serotypes that affect humans is now under way in the United States. The endemic disease incidence and the nature of immunity for the other viruses is less well defined, and vaccine development for them must await advances in knowledge of their epidemiology and molecular virology. ENVIRONMENTAL PROTECTION

Although person-to-person transmission is an important aspect of endemic disease, the initiating event for most outbreaks of viral gastroenteritis is contamination of a common source. In contrast to bacterial pathogens, enteric viruses cannot multiply outside their host; hence, the original inoculum into the common source determines infectivity. Food Shellfish

Shellfish that grow in fecally contaminated water concentrate enteric viruses in their tissues, and even harvests meeting bacteriologic standards of hygiene may contain viral agents. In addition, depuration (a technique in which shellfish are flushed with clean water treated with ultraviolet light) is less effective in viral than in bacterial decontamination (76). Finally, steaming for as long as 10 minutes may fail to inactivate all viral agents (77). Although boiling shellfish will inactivate viruses, such preparation is not popular with consumers.

The difficulties of assuring virus-free shellfish and the common preference for eating them raw have contributed to their prominent role in gastroenteritis outbreaks. Approximately 50% of Norwalk-confirmed foodborne outbreaks reported to CDC from 1976 to 1980 involved shellfish (52). A continent-wide epidemic of gastroenteritis in Australia in 1978 was attributed to oysters (78). Shellfish were implicated in 103 outbreaks of viral gastroenteritis in New York State in 1982, and similar outbreaks have occurred in the Northeast since then (79). Of 13 outbreaks of Norwalk virus documented in Britain during the period 1984-1985, seven were related to shellfish (57). Ill food handlers

When foods other than shellfish are implicated in viral gastroenteritis outbreaks, the contamination has usually taken place near the point of consumption. An ill food handler was identified in nine of the 15 documented Norwalk outbreaks reported to CDC from 1985 to 1988 for which adequate epidemiologic data were available (CDC, unpublished data). Foods that require handling and no subsequent cooking (e.g., salads) constitute the greatest risk. Among Norwalk-confirmed foodborne outbreaks from 1976 to 1980 that were not attributable to shellfish, salad was the most commonly implicated food (52). Other aspects of foodborne transmission

The long list of foods implicated in outbreaks of viral gastroenteritis reflects the variety of foods that are handled by food-service personnel and the low infectious dose (10-100 particles) of most viral agents of gastroenteritis, rather than peculiar viral tropisms. In contrast to the factors important in amplifying bacterial contamination, practices such as leaving foods unrefrigerated or warming them for prolonged periods are not direct risk factors for increased viral transmission because the viruses do not multiply outside the human host. Such practices, however, may be indicators of poor food hygiene in general.

The Norwalk agent can remain infective even if frozen for years or heated to 60 C for 30 minutes (80); however, cooking temperatures at boiling or above are probably adequate to inactivate Norwalk and most other enteric viral pathogens. Water

Outbreaks of viral gastroenteritis have been associated with various sources of contaminated water, including municipal water, well water, stream water, commercial ice, lake water, and pool water.

The most recent U. S. Environmental Protection Agency guidelines (June 29, 1989) for municipal water systems recommend residual chlorine concentrations of greater than or equal to 0.2 milligrams/liter (mg/L) (81), and in many localities peak levels of 5 mg/L are administered. Studies have documented that the Norwalk agent can remain highly infective despite 30-minute exposure to concentrations of chlorine as high as 6.25 mg/L, and levels of 10 mg/L appear necessary to inactivate it (82). This resistance may explain why the Norwalk agents are prominent in outbreaks of waterborne disease. Of 96 waterborne outbreaks with sufficient data reported to CDC from 1976 to 1979, 23% met epidemiologic criteria of a Norwalk virus outbreak (83), and subsequent surveillance data on waterborne outbreaks have been consistent with this finding. Of 38 serologically confirmed Norwalk virus outbreaks between 1976 and 1980, 13 were waterborne (52). Rotavirus, for which only one waterborne outbreak has been documented in the United States, is more sensitive to chlorine than the Norwalk agent and is inactivated by a 30-minute exposure to 3.75 mg/L (82). All viral agents of gastroenteritis are thought to be inactivated by boiling for 10 minutes. Surfaces

Because rotavirus can survive for several days on nonporous materials in conditions of low temperature and humidity, fomites may contribute to its nosocomial transmission (84). A recent study of a Norwalk viral outbreak on a cruise ship implicated toilets shared between staterooms as a risk factor for infection, suggesting that surfaces contaminated by Norwalk particles from spattered or aerosolized material may play a role in transmission of Norwalk-like viruses (85). Data are lacking on the efficacy of disinfectants against Norwalk-like agents, but a number of germicidal chemicals have been shown in laboratory tests to be ineffective in reducing rotavirus activity (86-88). However, detergents do inactivate rotavirus (89) and should be used for laundering fecally contaminated linens and clothing. Thorough cleaning of environmental surfaces is required, as a minimum, to control spread of the viral agents of gastroenteritis. Hands

Hands that have been contaminated directly or from surfaces may be the most important means by which enteric viruses are transmitted. Because the active ingredients in some commercial handwashing preparations are ineffective against rotavirus (90), the use of special handwashing products is not indicated. Vigorous handwashing with soap, performed consistently at appropriate intervals, is necessary to control the spread of all enteric pathogens. Aerosols

Aerosolized or splattered Norwalk-like particles have been implicated in the transmission of gastroenteritis (91). Aerosolized rotavirus has caused diarrheal illness in mice (92), and airborne transmission of this agent among humans has been suspected. Studies are needed to address the efficacy of barrier precautions (e.g., face shields, respirators) in interrupting transmission of these agents. Zoonoses

Nearly all the agents of viral gastroenteritis in humans have related strains that can cause diarrhea in animal species. But these strains appear to be highly host-specific, and zoonotic transmission has not been documented as having an important role in human disease, either endemically or in outbreaks (93). TREATMENT

For most humans, viral gastroenteritis is a self-limited illness of a few days' duration, with virus replication restricted to the mucosa of the gut. The main risk is of dehydration and electolyte imbalance.

Children, in whom the risk of fluid loss is greatest, respond well to oral rehydration therapy (ORT). Hospitalization and treatment with intravenous fluids are required only for cases in which dehydration is severe, or in which the parent or caretaker cannot provide adequate oral rehydration. Analysis of geographic and demographic patterns of diarrheal mortality in the United States suggests that lack of access to medical care, rather than disease virulence, is a principal risk factor for death from gastroenteritis (2). Although infants with diarrhea may manifest subsequent mild lactose intolerance (often 10-14 days for rotavirus infection) (94), most infants completely recover. Breast milk may have a protective effect against bacterial or viral enteric infection, and most infants can be "fed through" an episode (95).

For adults, maintenance of good hydration is also important, particularly among the elderly and those receiving diuretic medication. In one study, bismuth subsalicylate reduced duration of Norwalk infection from 27 to 20 hours (96). SPECIAL HOSTS The Malnourished

The public health impact of diarrheal viruses is compounded in developing countries by the cycle of diarrhea and malnutrition: acute diarrhea converts a marginal nutritional status into undernourishment, thereby reducing resistance to infection, predisposing persons to chronic diarrhea, which leads to further undernourishment. Studies are needed to determine if interventions can interrupt this downward clinical spiral once it has started. Pregnant Women

Although dehydration and electrolyte imbalance from any cause can pose a risk to pregnancy, no evidence indicates that the viral agents of gastroenteritis constitute a particular threat. Other than in children with primary immunodeficiencies, viremic states from these agents are not known to occur in humans, and thus the risk of transplacental exposure, fetal demise, or malformation is probably low or nonexistent. Neonates

Data on neonatal infection and immunity are available principally for rotavirus, for which immunity appears to be transferred transplacentally from mother to fetus late in the third trimester, perhaps mainly in the final month of gestation. Consequently, for the term infant, rotaviral infection during the first month of life tends to be mild or asymptomatic, and only when maternal antibody levels wane between the third and sixth month of life does infection pose a significant risk of illness. Evidence from Australia suggests that neonatal infection with rotavirus may function as a vaccination--an asymptomatic infection producing subsequent immunity (17). The preterm infant, however, lacks adequate maternal antibody and appears to be at increased risk for early and symptomatic rotaviral infection. Data from Britain suggest that oral administration of immunoglobulin appears to protect against disease during outbreaks in nurseries and may shorten the duration of symptoms (97).

Outbreaks of necrotizing enterocolitis have occurred concurrently with outbreaks of rotavirus diarrhea (98,99). But annual patterns of mortality for necrotizing enterocolitis do not show the winter seasonality of rotavirus infections (100), and--if an association exists--other factors are probably important. The Elderly

Antibody levels to many viral pathogens, as well as total IgG levels, wane with age; therefore, the elderly are at risk for infections to which younger adults are resistant. Diuretic medications and debility can increase the risk of an adverse outcome in what otherwise might be a mild diarrheal episode. The role of rotavirus in deaths of the elderly from gastroenteritis is being investigated. Persons with Immunodeficiencies Acquired immunodeficiency syndrome (AIDS)

Chronic diarrhea is a common complication of AIDS, and lists of etiologic pathogens usually include Cryptosporidium, Isospora, atypical mycobacteria, Salmonella, and cytomegalovirus, among others. However, few studies have addressed the role of the viral agents of gastroenteritis among persons with AIDS. A recent antigen-detection study failed to detect rotavirus in the stool of 20 U.S. AIDS patients with diarrhea (101). In contrast, an Australian study that used electron microscopy (the diagnositc reference method) found rotavirus in 37% of specimens and adenovirus in 24% of specimens from 68 HIV-positive homosexual men with diarrhea (102). These viruses were detected more frequently than other microbial agents, were associated with higher degrees of immunocompromise, and were found less commonly in specimens from homosexual men who were HIV-negative or were HIV-positive but did not have diarrhea. Norwalk virus was detected in one instance, and no caliciviruses or astroviruses were identified. Immunity to most endemic viral agents of gastroenteritis appears to be acquired in childhood and may persist during early stages of adult AIDS. Children with AIDS may never acquire this immunity and thus may be at increased risk for persistent infection. Studies are needed to determine the role of viruses in diarrhea associated with pediatric AIDS. Other immunodeficiencies

Case reports have documented chronic diarrheal excretion of rotavirus, adenovirus, calicivirus, astrovirus, and other viruses among children with other immunodeficiencies, most notably severe combined immunodeficiency syndrome (SCIDS) (103,104). Rotaviral and adenoviral infections have been documented in one study involving recipients of bone marrow transplants, and such infections are associated with markedly increased mortality (105). Unusual forms of rotaviral infection can occur in persons with immunodeficiencies: antigen has been detected in serum, suggesting that viremia may occur (106), and unique genomic rearrangements have been noted among patients with T-cell deficiencies (107). The immunodeficient host and viral investigation

The effect of viral agents of gastroenteritis on persons with immunodeficiencies is of particular interest. Although immunization appears to offer the principal hope of endemic control, the nature of immunity to the viral agents of diarrhea is poorly understood, and lack of information in this area has hampered vaccine development. Interventions, such as immunoglobulin, that might prove successful in halting the chronic infection of immunocompromise might also prove useful in other situations, such as the chronic diarrhea of malnourishment. SPECIFIC SITUATIONS Nursing Homes and Residential Institutions

Protracted outbreaks of viral gastroenteritis have occurred in nursing homes and institutional residences. Risk factors in such settings include the enclosed living quarters in which most residents spend their time and the decreased personal hygiene among some residents because of incontinence, immobility, or reduced alertness. Although Norwalk-like agents are most commonly involved in nursing home outbreaks, viruses usually associated with childhood diarrhea, such ascalicivirus and astrovirus, have also been implicated etiologically in some outbreaks, suggesting that the waning immunity of the elderly may also be a predisposing factor. Cruise Ships and Camps

The close living quarters of ships and camp dormitories amplify opportunities for person-to-person transmission of viral agents. Gastroenteritis outbreaks would normally end after all susceptible persons have been infected; however, in such settings, new and uninfected populations usually arrive every 1 or 2 weeks, thereby renewing the epidemic. Norwalk viral outbreaks extending over five successive cruises have been documented (108). Pediatric Wards and Day-Care Facilities

Continuous close contact among unrelated children, some of whom may be ill, can accelerate the progression of endemic diarrhea through a small population into an outbreak. Nosocomial and day-care transmission of rotavirus during its peak season is particularly efficient and difficult to control. Calicivirus, adenovirus, astrovirus, and Norwalk-like particles are found more frequently in stool specimens from children in these settings than from children in other settings. CLINICAL CHARACTERISTICS OF VIRAL GASTROENTERITIS OUTBREAKS

In outbreaks of gastroenteritis, investigators often face the problem of having to take action before an etiologic agent can be identified. This problem particularly applies to outbreaks caused by viruses, since diagnosis can be delayed for months or longer. Distinguishing viral from bacterial or protozoal etiologies is sometimes difficult because of overlap in clinical syndromes. A study of 74 outbreaks of probable viral gastroenteritis that occurred from 1976 to 1980 showed that affected persons had the following symptoms: nausea, 79%; abdominal cramps, 71%; vomiting, 69%; diarrhea (never bloody), 66%; headache, 50%; fever, 37%; chills, 32%; myalgias, 26%; and sore throat, 18%. In greater than 90% of the 38 outbreaks in this survey documented to have been caused by the Norwalk virus, the average incubation period was 24-48 hours and the average duration of illness was 12-60 hours. Vomiting occurred among more than half of the persons in almost 90% of the outbreaks (52). Review of Norwalk-documented outbreaks reported to CDC from 1985 to 1988 confirms these patterns (CDC, unpublished data).

Outbreaks of viral gastroenteritis reported to CDC are usually community based, involving adults or older children; thus, the characteristics of these outbreaks are those of the Norwalk-like viruses that affect these age groups. Outbreaks involving preschool children are more likely to be caused by the agents of endemic childhood diarrhea (e.g., rotavirus, adenovirus, calicivirus, or astrovirus) and are often in institutional settings, such as pediatric wards or day-care centers. Although clinical syndromes differ somewhat among viral pathogens, common features of viral outbreaks involving this younger age group include a high rate of vomiting, an absence of blood in the stool, and a duration of illness that is usually less than or equal to 1 week. OUTBREAK CONTROL MEASURES

Most outbreaks of viral gastroenteritis are self-limited; however, certain factors create risks of intense or prolonged transmission that may require aggressive intervention. These risk factors include a closed environment (e.g., nursing home), a constantly renewing population of susceptible persons (e.g., children at camp), or persons at special risk (e.g., the elderly). Whatever the initial source of the outbreak, subsequent viral transmission is often person-to-person, with both direct fecal-oral and airborne transport probably involved. Although interruption of this transmission may be difficult, the following measures may be helpful in controlling the spread of infection. Identify and Eliminate a Common Source

For Norwalk virus outbreaks, an ill food handler is a likely source, although water, ice, and shellfish are other common sources. When a water supply is thought to be contaminated with Norwalk virus, shock chlorine concentrations (greater than or equal to 10 mg/L for 30 minutes or longer) may be helpful. Prevent Employee Transmission of Illness

In many settings, employees (e.g., health-care providers, staff of day-care centers) are at highest risk for transmitting disease because of their many contacts with ill persons. Any staff member with symptoms that suggest infection should be excluded from contact with potentially susceptible persons for at least 2 days after resolution of illness. This exclusion is particularly important for food handlers, who also should not be involved in preparing food for the same period. Prevent Employee Acquisition of Illness

Personnel coming into direct contact with ill persons should wear disposable plastic gloves. When contamination of clothing with fecal material is possible, personnel should also wear gowns. Hands, which are the most likely means by which viral spread occurs, should be washed after each contact. The recommended procedure is to rub all surfaces of lathered hands together vigorously for at least 10 seconds, with plain soap or an antimicrobial-containing product, and then thoroughly rinse the hands under a stream of water. Since spattering or aerosols of infectious material may be involved in disease transmission, wearing of masks should be considered, particularly by persons who clean areas grossly contaminated by feces or vomitus. Use Safeguards with Laundry

Soiled linens and clothes should be handled as little as possible and with minimum agitation to prevent microbial contamination of the air and of persons handling the linen. Laundry should be transported in an enclosed and sanitary manner (e.g., in a plastic bag if the laundry is wet or moist), promptly machine washed with a detergent in water at the maximum cycle length, and then machine dried (109). Clean Soiled Surfaces

Because environmental surfaces in certain settings have been implicated in the transmission of enteric viruses, bathrooms and rooms occupied by ill persons should be kept visibly clean on a routine basis. Surfaces that have been soiled, especially by feces or vomitus, should first be cleaned of visible material and then disinfected with an appropriate commercial germicidal product according to the manufacturer's instructions. Feces and vomitus collected during the cleaning procedure should be promptly disposed of in a manner that prevents transfer of this material to other surfaces or persons. Persons performing these tasks should wear appropriate protective barriers (e.g., utility gloves--and if splashing is anticipated, a mask or face shield and garments such as a uniform, jumpsuit, or gown to protect street clothing). Minimize Contact Between Well and Ill Persons

When possible, ill persons should be separated from well persons until at least 2 days after resolution of symptoms. If nosocomial rotavirus is involved, this period should be longer--at least until the ill person's stool is negative by antigen detection, which may be greater than or equal to 1 week. In certain settings (e.g., camp, cruise ship, or nursing home), the clinic may function as a focus of transmission; persons with complaints of gastroenteritis should be seen by medical care personnel in the patient's living quarters, or at least in a separate area of the clinic. Stop Renewal of Susceptible Population

In situations in which the epidemic is extended by periodic renewal of the susceptible population (e.g., camps and cruise ships), consideration may have to be given to interrupting this process until the outbreak has ended completely. SPECIMEN COLLECTION DURING OUTBREAKS

Electron microscopy is necessary for complete viral diagnosis of an outbreak of gastroenteritis. Most of the electron microscopes in the United States are devoted to research or fee-related clinical diagnostic activity. Very few state laboratories have electron microscopes that can be used in the evaluation of outbreaks. Consequently, many state and local health departments submit outbreak specimens to CDC. Because of advances in the field of viral diagnostics, recommendations for specimen collection have changed substantially, and many investigators in the field may not be aware of these changes.

The following guidelines are designed to help outbreak investigators make the best use of CDC's viral laboratory facilities. Proper specimen collection is also critical for the diagnosis of bacterial and parasitic causes of gastroenteritis (see Appendix). Guidelines for Collecting Specimens for Viral Diagnosis Stool

• Collection in the first 48 hours. Presently, viral diagnosis of a stool sample can be made only when the level of excretion is approximately 1 million particles/ml. For many viruses, this level of excretion is present only during the first 2 days of illness, and occasionally during the third. If specimens are not collected during the first 2-3 days of illness, an agent is unlikely to be detected. Thus, appropriate specimens should be collected as soon as an outbreak occurs. Specimen collection should not await the results of epidemiologic and other investigations, since delay will almost certainly preclude a viral diagnosis. If information gathered subsequently indicates that a viral etiology is unlikely, the specimens can be discarded before the cost of testing is incurred.

• Ten diarrheal bulk specimens. Bulk samples (enough to fill a large stool cup) are preferred, and only those specimens loose enough to assume the shape of their containers are likely to yield positive results. Serial specimens from persons with acute, frequent, high-volume diarrhea are particularly useful. The smaller the specimen and the more formed the stool, the lower the diagnostic yield. Rectal swabs are of little or no value. Specimens from at least 10 ill persons should be collected to maximize the chance that a diagnosis can be made. (The diagnostic yield is low when specimens from less than 10 persons are submitted.)

• Storage at +4 C. Because freezing may destroy the characteristic viral morphology that permits a diagnosis by electron microscopy, specimens should be kept at +4 C. Paired Serum Specimens (essential for diagnosis)

• Timing Acute: during the first week of symptoms

Convalescent: third to sixth week

• Number 10 pairs from ill persons (the same persons submitting stool specimens)10 pairs from well persons

• Quantity Adults: 10 ml

Children: 3 ml

• Storage Tubes containing no anticoagulant (tubes with red tops) should be used for collection. Sera should be spun off and frozen. If a centrifuge is not available, a clot should be allowed to form, and the serum should be decanted and frozen. If this step cannot be taken, the whole blood should be refrigerated, not frozen. Other Specimens

Viruses causing gastroenteritis cannot normally be detected in vomitus, water, food, or environmental samples. Although British researchers report electron microscope detection of virus in shellfish, no successful effort has yet been reported in the United States. Consultation

At any time during the course of an outbreak, the Viral Gastroenteritis Section, REB, DVRD, CID (telephone 404-639-3577), is available for advice and assistance.

Since the viruses involved in most outbreaks of gastroenteritis reported to CDC cannot be cultivated, the antigen and antibody reagents used in diagnosis are not easily renewable. Thus, CDC cannot routinely screen persons and must limit use of these reagents. The following information would help determine the best use of laboratory resources:

• Setting of the outbreak (e.g., nursing home, restaurant)

• Date of onset of outbreak

• Number of persons exposed and number ill

• Population characteristics (e.g., elderly, children)

• Incubation period

• Duration of illness

• Symptoms

• percent vomiting

• percent with diarrhea

• percent with bloody diarrhea

• other (e.g., fever, myalgia, malaise)

• Adverse outcomes (e.g., hospitalizations, deaths)

• Suspected common source (e.g., shellfish, other food, food handler, water, other)

• Ongoing problems Sending Specimens for Viral Diagnosis

If, after consultation, viral diagnostic services are considered to be useful, specimens may be shipped to the Viral Gastroenteritis Section, REB, DVRD, CID, following these guidelines:

• Each specimen should be labeled (waterproof) with a unique identifier firmly attached and should be accompanied by CDC Specimen Information Form 50.34.

• Stool specimens should be shipped as soon as they can be batched. Individual containers should be verified as leakproof and then enclosed in a plastic bag. The entire collection should be bagged in plastic and placed in a padded, insulated box with refrigerant packs.

• Frozen acute- and convalescent-phase serum samples should be batched and sent in a single shipment. Waterproof, padded, insulated boxes should be used, with dry ice added to maintain freezing. Whole blood samples should not be frozen, and refrigerant packs should be used instead of dry ice.

• Final notification should be made by telephone (404-639-3577) just before shipping.

• All shipments should be sent by overnight mail, due to arrive on a weekday, addressed to the Viral Gastroenteritis Section, REB, DVRD, CID, Mailstop G04, Centers for Disease Control, Atlanta, GA 30333. SURVEILLANCE

Few states maintain surveillance for the viral agents of gastroenteritis. In 1989, CDC began surveillance of rotavirus in anticipation of vaccine development and use. Postcard reports of monthly detections are submitted by almost 100 centers in the United States. Regular summaries of detection patterns will soon be issued to contributors, state public health departments, and other interested persons. Plans are under way for surveillance of the other agents of gastroenteritis, in collaboration with laboratories that use electron microscopy. States and centers interested in participating in these surveillance activities are invited to contact the Viral Gastroenteritis Section, REB, DVRD, CID, CDC.

References

1. Ho M, Glass RI, Pinsky PF, Anderson L. Rotavirus as a cause of

diarrheal morbidity and mortality in the United States. J Infect Dis 1988;158:1112-6.

2. Ho M, Glass RI, Pinsky PF, et al. Diarrheal deaths in American children: are they preventable? JAMA 1988;260:3281-5.

3. National Center for Health Statistics. Current estimates from the health interview survey. Series 10, no. 85, 1972. DHEW publication (HRA) 74-1512. Washington, DC: Government Printing Office, 1973.

4. Walsh JA, Warren KS. Selective primary health care: An interim strategy for disease control in developing countries. N Engl J Med 1979;302:967-74.

5. Kotloff KL, Wasserman SS, Steciak JY, et al. Acute diarrhea in Baltimore children attending an outpatient clinic. Pediatr Infect Dis J 1988;7:753-9.

6. Institute of Medicine. Prospects for immunizing against rotavirus. Washington, DC: National Academy Press, 1985.

7. Wenman WM, Hinde D, Feltham S, Gurwith M. Rotavirus infection in adults: results of a prospective family study. N Engl J Med 1979;301:303-6.

8. Ward RL, Bernstein DI, Young EC, Sherwood JR, Knowlton DR, Schiff GM. Human rotavirus studies in volunteers: determination of infectious dose and serological response to infection. J Infect Dis 1986;154:871-80.

9. Pickering LK, Bartlett III AV, Reves RR, Morrow A. Asymptomatic excretion of rotavirus before and after rotavirus diarhhea in children in day care centers. J Pediatr 1988;112:361-5.

10. Champsaur H, Questiaux E, Prevot J, et al. Rotavirus carriage, asymptomatic infection, and disease in the first two years of life. Part I: Virus shedding. J Infect Dis 1984;149:667-74.

11. Champsaur H, Henry-Amar M, Goldszimdt D, et al. Rotavirus carriage, asymptomatic infection, and disease in the first two years of life. Part II: Serological response. J Infect Dis 1984;149:675-82.

12. Echeverria P, Blacklow NR, Cukor GG, Vivulbandhitkit S, Changchawalit S, Boonthai P. Rotavirus as a cause of severe gastroenteritis in adults. J Clin Microbiol 1983;18:663-7.

13. Eiden JJ, Verleur DG, Vonderfcht SL, Yolken RH. Duration and pattern of asymptomatic rotavirus shedding by hospitalized children. Pediatr Infect Dis J 1988;7:564-9.

14. Bartlett III AV, Reves RR, Pickering LK. Rotavirus in infant-toddler day care centers: epidemiology relevant to disease control strategies. J Pediatr 1988;113:435-41.

15. Pickering LK, Bartlett III AV, Woodward WE. Acute infectious diarrhea among children in day care: epidemiology and control. Rev Infect Dis 1986;8:539-47.

16. Hopkins RS, Gaspard GB, Williams FPJ, Karlin RJ, Cukor G, Blacklow NR. A community waterborne gastroenteritis outbreak: evidence for rotavirus as the agent. Am J Public Health 1984;74:263-5.

17. Bishop RF, Barnes GL, Cipriani E, Lund JS. Clinical immunity after rotavirus infection: a prospective longitudinal study in young children. N Engl J Med 1983;309:72-6.

18. Hung T, Chen G, Wang C, et al. Waterborne outbreak of rotavirus diarrhea in adults in China caused by a novel rotavirus. Lancet 1984;i:1139-42.

19. Hung T, Chen G, Wang C, et al. Seroepidemiology and molecular epidemiology of the Chinese rotavirus. In: Bock G, Whelan J, eds. Novel diarrhoea viruses. (Ciba Foundation Symposium; 128). Chichester, UK: John Wiley & Sons, Ltd, 1987:49-62.

20. Bridger JC. Non-group-A rotaviruses. In: Viruses and the gut: proceedings of the Ninth BSG-SK&F International Workshop. Welwyn Garden City, UK: Smith Kline & French Laboratories, Ltd, 1988:79-81.

21. Bridger JC, Pedley S, McCrae MA. Group C rotaviruses in humans. J Clin Microbiol 1986;23:760-3.

22. Penaranda ME, Cubitt WD, Sinarachatanant P, et al. Group C rotavirus infections in patients with diarrhea in Thailand, Nepal, and England. J Infect Dis 1989;160:392-7.

23. Uhnoo I, Wadell G, Svenson L, Johansson ME. Importance of enteric adenoviruses 40 and 41 in acute gastroenteritis in infants and young children. J Clin Microbiol 1984;20:365-72.

24. Moffet HL, Schulenberger HK, Burkholder ER. Epidemiology and etiology of severe infantile diarrhea. J Pediatr 1968;72:1-14.

25. Kotloff KL, Losonsky GA, Morris JG, Wasserman SS, Singh-Naz N, Levine MM. Enteric adenovirus infection and childhood diarrhea: an epidemiologic study in three clinical settings. Pediatrics 1989;84:219-25.

26. Hammond GW, Hannan C, Yeh T, Fischer K, Mauthe G, Straus SE. DNA hybridization for diagnosis of enteric adenovirus infection from directly spotted human fecal specimens. J Clin Microbiol 1987;25:1881-5.

27. Wood DJ, Longhurst D, Killough RI, David TJ. One-year prospective cross-sectional study to assess the importance of group F adenovirus infection in children under 2 years admitted to hospital. J Med Virol 1988;26:429-35.

28. Chiba S, Nakata S, Nakamura I, et al. Outbreak of infantile gastroenteritis due to type 40 adenovirus. Lancet 1983;2:954-7.

29. Richmond SJ, Caul EO, Dunn SM, Ashley CR, Clarke SK, Seymour NR. An outbreak of gastroenteritis in young children caused by adenoviruses. Lancet 1979;i:1178-81.

30. Ellis ME, Watson B, Mandal BK, et al. Micro-organisms in gastroenteritis. Arch Dis Child 1984;59:848-55.

31. Matson DO, Estes MK, Glass RI, et al. Human calicivirus-associated diarrhea in children attending day care centers. J Infect Dis 1989;159:71-7.

32. Sakuma Y, Chiba S, Kogasaka R, et al. Prevalence of antibody to human calicivirus in general population of northern Japan. J Med Virol 1981;7:221-5.

33. Cubitt WD, McSwiggan DA. Seroepidemiological survey of the prevalence of antibodies to a strain of human calicivirus. J Med Virol 1987;21:361-8.

34. Cubitt WD. Caliciviruses. In: Farthing MJ, ed. Viruses and the gut: proceedings of the ninth BSG-SK&F international workshop. Welwyn Garden City, UK: Smith Kline & French Laboratories Ltd, 1988:82-4.

35. Chiba S, Sakuma Y, Kogasaka R, et al. An outbreak of gastroenteritis associated with calicivirus in an infant home. J Med Virol 1979;4:249-54.

36. Oishi I, Maeda A, Yamazaki K, Minekawa Y, Nishimura H, Kitaura T. Calicivirus detected in outbreaks of acute gastroenteritis in school children. Biken J 1980;23:163-8.

37. Cubitt WD, McSwiggan DA, Arstall S. An outbreak of calicivirus infection in a mother and baby unit. J Clin Pathol 1980;33:1095-8.

38. Cubitt WD, McSwiggan DA, Moore W. Winter vomiting disease caused by calicivirus. J Clin Pathol 1979;32:786-93.

39. Gary JJ, Wreghitt TG, Cubitt WD, Elliot PR. An outbreak of gastroenteritis in a home for the elderly associated with astrovirus type 1 and human calicivirus. J Med Virol 1987;23:377-81.

40. Humphrey TJ, Cruickshank JG, Cubitt WD. An outbreak of calicivirus associated gastroenteritis in an elderly persons home: a possible zoonosis? J Hyg 1984;93:293-9.

41. Cubitt WD, Moscovici OP PJ, Lebon PS AA. A new serotype of calicivirus associated with an outbreak of gastroenteritis in a residential home for the elderly. J Clin Pathol 1981;34:924-6.

42. Kurtz J, Lee T. Astrovirus gastroenteritis: age distribution of antibody. Med Microbiol Immunol 1978;166:227-30.

43. Grohmann G. Viral diarrhoea in Australia. In: Tzipori S, ed. Infectious diarrhoea in the young. Amsterdam: Excerpta Medica Elsevier, 1985:25-8.

44. Kurtz JB, Lee TW. Astrovirus: human and animal. In: Bock G, Whelan J, eds. Novel diarrhoea viruses. (Ciba Foundation Symposium; 128). Chichester, UK: John Wiley & Sons Ltd, 1987:92-107.

45. Ashley CR, Caul EO, Paver WK. Astrovirus-associated gastroenteritis in children. J Clin Pathol 1978;31:939-43.

46. Konno T, Suzuki H, Ishida N, Chiba R, Mochizuki K, Tsunoda A. Astrovirus-associated epidemic gastroenteritis in Japan. J Med Virol 1982;9:11-7.

47. Kurtz JB, Lee TW, Pickering D. Astrovirus associated gastroenteritis in a children's ward. J Clin Pathol 1977;30:948-52.

48. Oshiro LS, Haley CE, Roberto RR, et al. A 27-nm virus isolated during an outbreak of acute infectious non-bacterial gastroenteritis in a convalescent hospital: a possible new serotype. J Infect Dis 1981;143:791-5.

49. Blacklow NR, Cukor G, Bedigian MK, et al. Immune response and prevalence of antibody to Norwalk enteritis virus as determined by radioimmunoassay. J Clin Microbiol 1979;10:903-9.

50. Greenberg HB, Valdesuso JR, Kapikian AZ, et al. Prevalence of antibody to the Norwalk virus in various countries. Infect Immun 1979;26:270-3.

51. White KE, Osterholm MT, Mariotti JA, et al. A foodborne outbreak of Norwalk virus gastroenteritis: evidence for post-recovery transmission. Am J Epidemiol 1986;124:120-6.

52. Kaplan JE, Gary GW, Baron RC, et al. Epidemiology of Norwalk gastroenteritis and the role of Norwalk virus in outbreaks of acute nonbacterial gastroenteritis. Ann Intern Med 1982;96:756-61.

53. Yolken R, Leister F, Almeido-Hill J, Dubovi E, Reid R, Santosham M. Infantile gastroenteritis associated with excretion of pestivirus antigens. Lancet 1989;i:517-9.

54. Yolken R, Santosham M, Reid R, Dubovi E. Pestiviruses: major etiological agents of gastroenteritis in human infants and children? (Abstract). Clin Res 1988;36:80A.

55. Pereira HG, Fialho AM, Flewett TH, Teixeira JMS, Andrade ZP. Novel viruses in human faeces. Lancet 1988;ii:103-4.

56. Flewett TH, Davies H, Bryden AS, Robertson MJ. Diagnostic electron microscopy of faeces. II. Acute gastroenteritis associated with reovirus-like particles. J Clin Pathol 1974;27:608-14. 57. Appleton H. Small round viruses: classification and role in food-borne infections. In: Bock G, Whelan J, eds. Novel diarrhoea viruses. (Ciba Foundation Symposium; 128). Chichester, UK: John Wiley & Sons Ltd, 1987:108-25.

58. Birch CJ, Lewis FA, Kennett ML, Homola M, Pritchard H, Gust ID. A study of the prevalence of rotavirus infection in children with gastroenteritis admitted to an infectious disease hospital. J Med Virol 1977;1:69-77.

59. Beards GM, Green J, Hall C, Flewett TH, Lamouliatte F, Du Pasquier P. An enveloped virus in stools of children and adults that resembles the Breda virus of calves. Lancet 1984;i:1050-2.

60. Brown DWG, Beards GM, Flewett TH. Detection of Breda virus antigen and antibody in humans and animals by enzyme immunoassay. J Clin Microbiol 1987;25:637-40.

61. Kidd AH, Esrey SA, Ujfalusi MJ. Shedding of coronavirus-like particles by children in Lesotho. J Med Virol 1989;27:164-9.

62. Mathan M, Mathan VI, Swaminathan SP, Yesudoss S, Baker SJ. Pleomorphic virus-like particles in human faeces. Lancet 1975;i:1068-9.

63. Payne CM, Ray CG, Borduin V, Minnich LL, Lebowitz MD. An eight-year study of the viral agents of acute gastroenteritis in humans: ultrastructural observations and seasonal distrubution with a major emphasis on coronavirus-like particles. Diagn Microbiol Infect Dis 1986;5:39-54.

64. Dennehy PH, Gauntlett DR, Tente WE. Comparison of nine commercial immunoassays for the detection of rotavirus in fecal samples. J Clin Microbiol 1988;26:1630-4.

65. Thomas EE, Puterman ML, Kawano E, Curran M. Evaluation of seven immunoassays for detection of rotavirus in pediatric stool samples. J Clin Microbiol 1988;26:1189-93.

66. Erdman DD, Gary GW, Anderson LJ. Development and evaluation of an IgM capture enzyme immunoassay for diagnosis of recent Norwalk virus infection. J Virol Meth 1989;24:57-66.

67. Erdman DD, Gary GW, Anderson LJ. Serum Immunoglobulin: a response to Norwalk virus infection. J Clin Microbiol 1989;27:1417-8.

68. Glass RI, Monroe SS, Stine S, et al. Small round structured viruses: the Norwalk family of agents. In: Farthing MJG, ed. Viruses and the gut: Proceedings of the Ninth BSG-SK&F International Workshop. Welwyn Garden City, UK: Smith Kline & French Laboratories, 1988:87-90. References 69 through 109 may be obtained from the Viral Gastroenteritis Section, REB, DVRD, CID, Mailstop G04, Centers for Disease Control, Atlanta, GA 30333. APPENDIX

The following procedures are recommended for use by state and local health departments for investigation of outbreaks. Instructions are summarized for each category of pathogens.

State public health laboratories can provide diagnostic services for most bacterial and parasitic pathogens. No specimens should be sent to CDC for diagnostic testing unless the specific program concerned has been consulted.

Disclaimer   All MMWR HTML documents published before January 1993 are electronic conversions from ASCII text into HTML. This conversion may have resulted in character translation or format errors in the HTML version. Users should not rely on this HTML document, but are referred to the original MMWR paper copy for the official text, figures, and tables. An original paper copy of this issue can be obtained from the Superintendent of Documents, U.S. Government Printing Office (GPO), Washington, DC 20402-9371; telephone: (202) 512-1800. Contact GPO for current prices.

Page converted: 08/05/98

HOME  |  ABOUT MMWR  |  MMWR SEARCH  |  DOWNLOADS  |  RSS |  CONTACT POLICY  |  DISCLAIMER  |  ACCESSIBILITY

Morbidity and Mortality Weekly Report Centers for Disease Control and Prevention 1600 Clifton Rd, MailStop E-90, Atlanta, GA 30333, U.S.A Department of Health and Human Services

This page last reviewed 5/2/01