Rotavirus infection is most common in winter months in temperate climates. In the USA, the annual winter peak spreads from west to east (Fig. 257-1). Unlike the spread of other winter viruses, such as influenza, this wave of increased incidence is not due to a single prevalent strain or serotype. Typically, several serotypes predominate in a given community for 1 or 2 seasons, while nearby locations may harbor unrelated strains. Disease tends to be most severe in patients 3-24 mo of age, although 25% of the cases of severe disease occur in children >2 yr of age, with serologic evidence of infection developing in virtually all children by 4-5 yr of age. Infants younger than 3 mo are relatively protected by transplacental antibody and possibly breast-feeding. Infections in neonates and in adults in close contact with infected children are generally asymptomatic. Some rotavirus strains have stably colonized newborn nurseries for years, infecting virtually all newborns without causing any overt illness.

Figure 257-1 Peak rotavirus activity by month in the USA for July 1996–June 1997. This pattern is typical of the annual rotavirus activity each year.

(From the Centers for Disease Control and Prevention: Laboratory-based surveillance for rotavirus—United States, July 1996-June 1997, MMWR Morb Mortal Wkly Rep 46:1092–1094, 1997.)

Rotavirus and the other gastrointestinal viruses spread efficiently via a fecal-oral route, and outbreaks are common in children’s hospitals and child-care centers. The virus is shed in stool at very high concentration before and for days after the clinical illness. Very few infectious virions are needed to cause disease in a susceptible host.

The epidemiology of astroviruses is not as thoroughly studied as that of rotavirus, but they are a common cause of mild to moderate watery winter diarrhea in children and infants and an uncommon pathogen in adults. Hospital outbreaks are common. Enteric adenovirus gastroenteritis occurs year-round, mostly in children younger than 2 yr. Nosocomial outbreaks occur but are less common than with rotavirus and astrovirus. Calicivirus is best known for causing large, explosive outbreaks among older children and adults, particularly in settings such as schools, cruise ships, and hospitals. Often a single food, such as shellfish or water used in food preparation, is identified as a source. Like astrovirus and rotavirus, caliciviruses are also commonly found in winter infantile gastroenteritis.

Pathogenesis

Viruses that cause human diarrhea selectively infect and destroy villus tip cells in the small intestine. Biopsies of the small intestines show variable degrees of villus blunting and round cell infiltrate in the lamina propria. Pathologic changes may not correlate with the severity of clinical symptoms and usually resolve before the clinical resolution of diarrhea. The gastric mucosa is not affected despite the commonly used term gastroenteritis, although delayed gastric emptying has been documented during Norwalk virus infection.

In the small intestine, the upper villus enterocytes are differentiated cells, which have both digestive functions, such as hydrolysis of disaccharides, and absorptive functions, such as the transport of water and electrolytes via glucose and amino acid co-transporters. Crypt enterocytes are undifferentiated cells that lack the brush border hydrolytic enzymes and are net secretors of water and electrolytes. Selective viral infection of intestinal villus tip cells thus leads to (1) decreased absorption of salt and water and an imbalance in the ratio of intestinal fluid absorption to secretion and (2) diminished disaccharidase activity and malabsorption of complex carbohydrates, particularly lactose. Most evidence supports altered absorption as the more important factor in the genesis of viral diarrhea. It has been proposed that a rotavirus nonstructural protein (NSP4) functions as an enterotoxin.

Viremia may occur often in severe, primary infections, but symptomatic extraintestinal infection is extremely rare in immunocompetent persons—although immunocompromised patients may rarely experience hepatic and renal involvement. The increased vulnerability of infants (compared with older children and adults) to severe morbidity and mortality from gastroenteritis viruses may relate to a number of factors, including decreased intestinal reserve function, lack of specific immunity, and decreased nonspecific host defense mechanisms such as gastric acid and mucus. Viral enteritis greatly enhances intestinal permeability to luminal macromolecules and has been postulated to increase the risk for food allergies.

Clinical Manifestations

Rotavirus infection typically begins after an incubation period of <48 hr (range 1-7 days) with mild to moderate fever as well as vomiting, followed by the onset of frequent, watery stools. All 3 symptoms are present in about 50-60% of cases. Vomiting and fever typically abate during the 2nd day of illness, but diarrhea often continues for 5-7 days. The stool is without gross blood or white blood cells (WBCs). Dehydration may develop and progress rapidly, particularly in infants. The most severe disease typically occurs among children 4-36 mo of age. Malnourished children and children with underlying intestinal disease such as short-bowel syndrome are particularly likely to acquire severe rotavirus diarrhea. Rarely, immunodeficient children experience severe and prolonged illness. Although most newborns infected with rotavirus are asymptomatic, some outbreaks of necrotizing enterocolitis have been associated with the appearance of a new rotavirus strain in the affected nurseries.

The clinical course of astrovirus appears to be similar to that of rotavirus, with the notable exception that the disease tends to be milder, with less significant dehydration. Adenovirus enteritis tends to cause diarrhea of longer duration, often 10-14 days. The Norwalk virus has a short (12-hr) incubation period. Vomiting and nausea tend to predominate in illness associated with the Norwalk virus, and the duration is brief, usually consisting of 1-3 days of symptoms. The clinical and epidemiologic picture of Norwalk virus often closely resembles so-called food poisoning from preformed toxins such as Staphylococcus aureus and Bacillus cereus.

Diagnosis

In most cases, a satisfactory diagnosis can be made on the basis of the clinical and epidemiologic features. Enzyme-linked immunosorbent assays (ELISAs), which offer >90% specificity and sensitivity, are available for detection of group A rotavirus, caliciviruses, and enteric adenovirus in stool samples. Latex agglutination assays are also available for group A rotavirus and are less sensitive than ELISA. Research tools include electron microscopy of stools, RNA polymerase chain reaction analysis to identify G and P antigens, and culture. The diagnosis of viral gastroenteritis should always be questioned in patients with persistent or high fever, blood or WBCs in the stool, or persistent severe or bilious vomiting, especially in the absence of diarrhea.

Laboratory Findings

Isotonic dehydration with acidosis is the most common finding in children with severe viral enteritis. The stools are free of blood and leukocytes. Although the WBC count may be moderately elevated secondary to stress, the marked left shift seen with invasive bacterial enteritis is absent.

Differential Diagnosis

The differential diagnosis includes other infectious causes of enteritis, such as bacteria and protozoa. Occasionally, surgical conditions such as appendicitis, bowel obstruction, and intussusception may initially mimic viral gastroenteritis.

Treatment

Avoiding and treating dehydration are the main goals in treatment of viral enteritis. A secondary goal is maintenance of the nutritional status of the patient (Chapters 55 and 332).

There is no routine role for antiviral drug treatment of viral gastroenteritis. Controlled studies have shown no benefit from antiemetics or antidiarrheal drugs, and there is a significant risk for serious side effects with both types of agents. Antibiotics are similarly of no benefit. Immunoglobulins have been administered orally to both normal and immunodeficient patients with severe rotavirus gastroenteritis, but this treatment is currently considered experimental. Therapy with probiotic organisms such as Lactobacillus species has been shown to be helpful only in mild cases and not in dehydrating disease.

Supportive Treatment

Rehydration via the oral route can be accomplished in most patients with mild to moderate dehydration (Chapters 55 and 332). Severe rehydration requires immediate intravenous therapy followed by oral rehydration. Modern oral rehydration solutions containing appropriate quantities of sodium and glucose promote optimum absorption of fluid from the intestine. There is no evidence that a particular carbohydrate source (rice) or addition of amino acids improves the efficacy of these solutions for children with viral enteritis. Other clear liquids, such as flat soda, fruit juice, and sports drinks, are inappropriate for rehydration of young children with significant stool loss. Rehydration via the oral (or nasogastric) route should be done over 6-8 hr, and feedings begun immediately thereafter. Providing the rehydration fluid at a slow, steady rate, typically 5 mL/min, reduces vomiting and improves the success of oral therapy. Rehydration solution should be continued as a supplement to make up for ongoing excessive stool loss. Initial intravenous fluids are required for the infant in shock or the occasional child with intractable vomiting.

After rehydration has been achieved, resumption of a normal diet for age has been shown to result in a more rapid recovery from viral gastroenteritis. Prolonged (>12 hr) administration of exclusive clear liquids or dilute formula is without clinical benefit and actually prolongs the duration of diarrhea. Breast-feeding should be continued even during rehydration. Selected infants may benefit from lactose-free feedings (such as soy formula and lactose-free cow’s milk) for several days, although this step is not necessary for most children. Hypocaloric diets low in protein and fat such as BRAT (bananas, rice, cereal, applesauce, and toast) have not been shown to be superior to a regular diet.