Gastroenteritis in Children: Principles of Diagnosis and Treatment

Medical College of Wisconsin,
Milwaukee, Wisconsin
Waukesha Family Practice Residency Program,
Waukesha, Wisconsin

Gastroenteritis in children is a common reason for visits to family physicians. Most cases of gastroenteritis have a viral etiology and are self-limited. However, more severe or prolonged cases of gastroenteritis can result in dehydration with significant morbidity and mortality. This is often the scenario in third-world countries, where gastroenteritis results in 3 million deaths annually. A proper clinical evaluation will allow the physician to estimate the percentage of dehydration and determine appropriate therapy. In some situations, laboratory studies such as determination of blood urea nitrogen and serum electrolytes may be helpful. Stool studies are indicated if a child is having bloody diarrhea or if an unusual etiology is suspected, such as Escherichia coli O157:H7 or Cryptosporidium. Most children with gastroenteritis can be treated with physiologically balanced oral rehydration solutions. In children who are hypovolemic, lethargic and estimated to be more than 5 percent dehydrated, initial treatment with intravenous boluses of isotonic saline or Ringer's lactate may be required. Children with severe diarrhea need nutrition to restore digestive function and, generally, food should not be withheld.

Gastroenteritis among children in the United States is a common and usually nonfatal illness, although it results in 3 million deaths annually worldwide.1 A clean water supply, good nutrition, reasonable physical cleanliness and appropriate disposal of human waste has allowed the United States and other developed countries to control the transmission and virulence of infectious agents that cause diarrhea. Most cases of gastroenteritis in this country are self-limited and require minimal intervention. Occasional episodes of severe, life-threatening gastroenteritis may occur, however, necessitating aggressive therapeutic intervention.

Most family physicians will occasionally encounter a severely ill child with gastroenteritis who needs aggressive fluid management. Inappropriate use of oral liquids, such as broths with a very high sodium content (up to 250 mEq per L) or soft drinks with a very low sodium content (1 to 2 mEq per L) may transform a mild isotonic dehydration into a more complex hypertonic or hypotonic dehydration.2 An understanding of the common etiologies of gastroenteritis, along with the principles of fluid and electrolyte management, is important for all physicians who care for children.

This article reviews common etiologies of gastroenteritis in children, appropriate assessment of hydration status and approaches to fluid replacement. Specific educational points and principles of treatment are emphasized. The history and physical examination are the cornerstone of diagnosis and assessment and, along with occasional laboratory tests, should guide therapy.



Types of Gastroenteritis According to Etiology



Physical examination


Laboratory test


Viruses (rotavirus most common), Norwalk virus, other viruses

Vomiting often present before diarrhea; large-volume, watery stools, usually not frequent or bloody; rotavirus infection usually occurs in winter

Usually varies, not toxic; may be mildly dehydrated; small children may be very ill

Common in infants and young children

Rotozyme (rapid test), electrolytes, BUN, creatinine and other tests as needed (i.e., urinalysis)

5% dehydration:

1. Usually can be treated with oral liquids.

2. Breast feeding may be continued.

3. Prolonged withholding of food not appropriate unless severe, protracted vomiting is present.

Shigella, Salmonella, Campylobacter, other bacteria, Entamoeba histolytica

Bloody stools common (50% of patients); may have bacterial infection without blood; more frequent, small-volume bowel movements; vomiting less common

Variable; may have a high fever

All age groups

Stool culture; stool examination for leukocytes (>5 white blood cells per high-power field in a stool smear); CBC

5 to 10% dehydration:

1. Depending on age, reliability of family, oral hydration may be most appropriate; for some patients, hospitalization is important.

Escherichia coli O157:H7

Associated with bloody diarrhea and hemolytic uremic syndrome in children; occurs in clusters of patients who have ingested contaminated foods

May be quite toxic;always evaluate mental status

More common in children <4 years of age

Capability of serotyping is important; electrolytes, BUN, creatinine, CBC, platelets; clotting studies; peripheral smear; LDH

10% dehydration:

1. Treat vigorously with isotonic intravenous fluids and supportive therapy.



Clostridium difficile;

Giardia lamblia;

Vibrio cholera;


BUN=blood urea nitrogen; CBC=complete blood cell count; LDH=lactate dehydrogenase.


Illustrative Cases

Case 1
A four-year-old boy presented with a three-day history of bloody diarrhea with no vomiting and a temperature up to 40°C (104°F). Over the previous 24 hours the child had had watery, bloody bowel movements every 20 minutes. His fluid intake consisted of soft drinks, and urination was infrequent. One other family member had nonbloody diarrhea.

On physical examination, the patient's temperature was 39.7°C (103.6°F), his respiratory rate was 30 and his pulse rate was 130 when supine and 160 when upright. The patient weighed 15 kg (33 lb) and had a blood pressure measurement of 80/60 mm Hg. He was alert but irritable, with a dry mouth and no tearing. His skin turgor and capillary refilling were normal. Bowel sounds were increased, and the abdominal examination was otherwise normal.

Laboratory results showed the following electrolyte levels: sodium, 118 mEq per L (118 mmol per L); potassium, 3.2 mEq per L (3.2 mmol per L); chloride, 90 mEq per L (90 mmol per L); bicarbonate, 14 mEq per L (14 mmol per L); creatinine, 0.8 mg per dL (70 µmol per L), and blood urea nitrogen, 20 mg per dL (7.1 mmol per L). The child was eventually diagnosed with a bacterial infection caused by a Shigella species.

Case 2
A four-week-old female twin born at 37 weeks of gestation had large, watery diarrheal stools five to six times per day for three to four days, with some vomiting. She was taking formula and no clear liquids. Her twin sister had a similar but less severe illness. The infant's neonatal course was normal. At two weeks of age she weighed 2.36 kg (5 lb, 3 oz).

On physical examination, the infant weighed 2.05 kg (4 lb, 8 oz), with a respiratory rate of 40 and a heart rate of 120. Her temperature was 37.0°C (98.7°F), and her blood pressure measurement was 30/0 mm Hg. The patient was lethargic and weakly responsive, with cool extremities. Her skin turgor was decreased, and capillary refilling time was about 3.0 seconds.

Results of laboratory tests included the following: sodium level, 145 mEq per L (145 mmol per L); potassium level, 4.2 mEq per L (4.2 mmol per L); bicarbonate level, 12 mEq per L (12 mmol per L); blood urea nitrogen level, 25 mg per dL (8.9 mmol per L); and creatinine level, 0.8 mg per dL (70 µmol per L). The illness was probably viral in origin.

Case 3
A three-year-old boy presented with abdominal pain, fever and bloody diarrhea. No other family members were ill. The child had four to five bloody diarrheal stools per day, with no vomiting and decreased urination.

On physical examination, the child had a temperature of 39.6°C (103.4°F), a respiratory rate of 20 per minute, a heart rate of 100 and a blood pressure of 70/30 mm Hg. The patient was alert and irritable with normal skin turgor, normal capillary refill and a normal abdominal examination. Rare petechiae were present. The electrolyte, blood urea nitrogen and creatinine levels were normal.

Case 4
An 18-month-old girl presented with a three-day history of fever and frequent, bright red, bloody diarrheal stools. She was taking fluids, including some broth. There was no vomiting, and no other family members were ill. The child had recently traveled to Mexico.

On physical examination, the child had a temperature of 39.6°C (103.4°F), a heart rate of 140, a respiratory rate of 25 and blood pressure of 90/60 mm Hg. The child was irritable but not lethargic with good capillary refill but decreased skin turgor. Bowel sounds were increased with no guarding or rebound and some generalized tenderness.

Results of laboratory tests included the following: sodium, 155 mEq per L (155 mmol per L); blood urea nitrogen, 20 mg per dL (7.1 mmol per L); and creatinine, 0.9 mg per dL (80 µmol per L). The stool culture was negative after 24 hours.


Diagnosis and Assessment of Hydration

Table 1 summarizes pertinent items relating to common etiologies of gastroenteritis. Most life-threatening viral diarrheal illnesses occur in children under three months of age or in those who may be compromised by prematurity, malnutrition or poverty. Viral gastroenteritis in older children is generally self-limited and usually requires only oral rehydration therapy.

A stool specimen should be examined for white blood cells in any child who appears toxic with high fever and diarrhea. The finding of white blood cells should prompt further investigation to rule out invasive bacterial disease. The presence of gross blood in the diarrheal stool also suggests a more serious infection, so children with bloody diarrhea should undergo a rectal swab or stool culture.2 Other laboratory tests are optional and are dictated by the severity of illness. If Escherichia coli is identified in a patient with bloody diarrhea, serotyping should be done to specifically identify the O157:H7 strain, which can cause hemolytic uremic syndrome, a potentially fatal illness.3 The petechiae in the patient in illustrative case 4 are an indication of a consumptive coagulopathy, which may occur with E. coli O157:H7 infections.




Methods for Assessing Degree of Dehydration

Degree of dehydration




5 to 10%


Capillary refilling time8

0.8 seconds

1.5 seconds or less

1.5 to 3.0 seconds

>3.0 seconds

Elevated BUN (normal BUN: 8 to 25 mg per dL9)

<10 mg per dL

10 to 20 mg per dL

21 to 25 mg per dL

>25 mg per dL

Skin turgor9


Slightly decreased


Decreased (pinch retracts slowly >2 seconds)

Deep and rapid, acidotic breathing (pH <7.35; decreased bicarbonate)9


Slightly increased respiratory rate

Increased respiratory rate

Breathing is deep and rapid

General state1

Infants and young children

Thirsty, alert, restless

Restless or lethargic: irritable to touch

Limp, drowsy; may be comatose; cyanotic extremities (cold sweats)

Older children

Thirsty, alert, restless

Thirsty, alert, postural hypotension

Usually conscious; wrinkled skin at fingers and toes; cyanotic extremities (cold sweats)


Slight increase; normal strength

Rapid and weak

Rapid, feeble, sometimes impalpable

Systolic blood pressure



May be unrecordable

BUN=blood urea nitrogen.

NOTE: Clinical findings are most reliable when the examination is performed by two or more different physicians.7 Other helpful but not specific signs include sunken fontanel, absent tears, decreased urine output, sunken eyes, thirst and dry mucous membranes.

Information from references 1 and 7 through 9.


In most children with dehydration greater than 5 percent, serum electrolytes, blood urea nitrogen and creatinine levels should be measured. Other tests that may be indicated are listed in Table 1. If unusual causes of gastroenteritis are suspected (i.e., travel to areas at risk, affected contact or immunosuppression), three fresh stool specimens should be evaluated for routine culture, ova, parasites and Cryptosporidium species. The child in illustrative case 4 was infected with Entamoeba histolytica, which was identified in only one of three stool specimens. Other unusual causes of gastroenteritis include Clostridium difficile, Cryptosporidium species,4 Giardia lamblia,5 Vibrio cholera6 and Cyclospora.4
description of image}
Children with significant diarrhea and secondary fluid loss often develop metabolic acidosis manifested by a low bicarbonate level.
description of image}

Table 21,7-9 offers a guide for assessing the state of hydration in children. Recent weight change documented with the same scale is very helpful in assessing fluid loss.1,8,9 Accurate weights may be harder to obtain in younger children, therefore the technique is important. The comparison of weights in illustrative case 2 was very helpful, since it indicated more than 10 percent dehydration, an indication that was consistent with the clinical assessment of dehydration in this patient. The assessment of mental status is also important in the evaluation of an ill child. A child who is happy and playful usually has no significant hydration problem. Increasing disturbances in mental status, from restlessness to irritability and then to lethargy, indicate a general overall worsening in the patient's condition and should help guide the aggressiveness of fluid resuscitation.

Vital signs, including respiratory rate, add significantly to the assessment of hydration. Children with diarrhea and excessive fluid loss usually develop a non­anion-gap metabolic acidosis with a declining bicarbonate level. To compensate, the rate and depth of breathing increase (compensatory respiratory alkalosis). A postural change in heart rate is a useful clue in assessing the fluid state of children over four years of age. An increase greater than 20 beats per minute when moving from a lying to a standing position is an indicator of hypovolemia. If the postural changes are greater, the fluid depletion is likely to be greater. However, changes in postural blood pressure have not been found to be very useful in children under nine years of age.10


Candidates for Oral Rehydration Therapy
description of image}
Mild to moderate dehydration (<10%)
Age >4 months
No persistent vomiting
Unlikely to have an underlying cause other than viral gastroenteritis
Isonatremia is present
Serum bicarbonate level >18 mEq per L (18 mmol per L)

The history and physical examination, with an assessment of the child's hydration status, guide the fluid treatment of children. Worldwide and in the United States, oral rehydration solutions such as the World Health Organization solutions and other clear liquids are the cornerstone of treatment.

In 1985, the American Academy of Pediatrics (AAP) published a policy statement on the treatment of infants with acute diarrhea complicated by mild to moderate dehydration.11 The AAP recommended rapid rehydration in four to six hours with an oral glucose-electrolyte rehydration solution followed by diluted formula or milk. Contrary to widespread belief and practice, lactose-based milk does not have to be eliminated. In older infants or children, rice cereal, bananas, potatoes or other nonlactose, carbohydrate-rich foods should be offered shortly after successful rehydration. In a follow-up study of practicing family physicians and pediatricians, it was noted that these clinicians tended to take much longer than recommended to rehydrate their patients, often used a lactose-free formula and did not reinstitute feeding of full-strength formula soon enough.12


description of image}
A patient with a bicarbonate level less than 18 or with vomiting that occurs at least five times within a 24-hour period may benefit from an intravenous fluid bolus of 20 mL per kg given over 20 minutes.
description of image}

A recent meta-analysis of 13 studies on the safety of oral rehydration solutions noted that the failure rate of rehydration was only 3.6 percent. Significant evidence13 has now accumulated supporting the usefulness of oral rehydration solutions as the initial treatment of choice in low-risk infants and children, as defined in Table 3. Patients with depressed bicarbonate levels (13 to 18 mEq per L [13 to 18 mmol per L]) or those who vomit at least five times in a 24-hour period may benefit from an initial intravenous infusion of isotonic solution, 20 mL per kg over 20 minutes, followed by oral administration of fluids.14 Continued vomiting, large-volume diarrhea, or both, would constitute a reason for hospital admission. Factors that significantly contribute to the progression of dehydration include withdrawal of breast feeding, not giving oral rehydration solutions during diarrhea and vomiting more than two times a day.15

Comparison of Oral Rehydration Solutions
description of image}
Glucose (g per dL)
Sodium (mEq per L)
Potassium (mEq per L)
Chloride (mEq per L)
Commercial solutions
WHO solution 2.0 90 20 80
Hydra-Lyte 1.2 84 10 59
Rehydralyte 2.5 75 20 65
Pedialyte 2.5 45 20 35
Generic pediatric solution* 2.5 45 20 35
Lytren 2.0 50 25 45
Resol 2.0 50 20 50
Infalyte 2.0 50 20 40
Ricelyte Starch polymers 50 25 45
Home remedies (not recommended)
Jell-O (one-half strength) 8.0 6 to 17 0.2 --
Gatorade 5.0 24 3 17
Soft drinks 7.0 to 12.0 1 to 7 0.1 to 0.4 --
Apple juice 12.0 0.1 to 3.5 24 to 43 --
Broth -- 250 -- --

WHO=World Health Organization.

*--Similar to Pedialyte.

Information from references 1 and 2.

Oral Rehydration Therapy for Severe Diarrheal States, Including Cholera
description of image}
  1. For mild dehydration, give oral rehydration, 100 mL per kg for the first 4 hours, then re-evaluate. May give more if the patient desires.
  2. Give additional fluids after each liquid bowel movement (children up to 2 years: 50 to 100 mL; 2 to 10 years: 100 to 200 mL; over 10 years, as much as desired).
  3. Patient may drink water ad libitum but should continue intake of oral rehydration solution.
  4. If there is no dehydration, give oral rehydration solution to replace stool losses.
  5. If the patient is stable, give a two-day supply of oral rehydration solution.
  6. Patients who are unable to drink or severely dehydrated (10% or more), or both, need rapid fluid replacement (intravenous) when possible. Large amounts of fluids are necessary.

Information from A manual for the treatment of diarrhoea. Geneva, Switzerland: World Health Organization, 1990. Publication WHO/CDD/SER/80.2 Rev. 2. 1990.


Children with gastroenteritis lose 40 to 70 mEq per L (40 to 70 mmol per L) of sodium and 10 to 20 mEq per L (10 to 20 mmol per L) of potassium in diarrheal stools. Appropriate oral rehydration solutions should come close to matching the sodium and potassium content of the diarrheal stools, especially in children with gastroenteritis who are under one year of age or who are otherwise compromised. Oral rehydration solutions such as Pedialyte, Infalyte or the equivalent (Table 4) effectively treat children less than two years of age with mild to moderate (3 to 9 percent) dehydration secondary to gastroenteritis.16 Broths should be avoided in the treatment of diarrhea because of their high sodium content. Rice-based oral rehydration solution seems to offer some benefit in the treatment of high-output diarrhea, especially that associated with cholera.17 Table 518 lists some guidelines for treatment of mild to severe diarrheal states (including cholera) where oral rehydration therapy may be the only therapy available.8

Isotonic solutions such as normal saline and Ringer's lactate are usually the correct fluids for the initial treatment of dehydration that requires intravenous fluids (Tables 1 and 6). This would be the correct type of fluid for treatment of the children in the illustrative cases. The fluid is administered in boluses of 15 to 30 mL per kg every 20 minutes until hypovolemia is corrected and some urine flow is established. At that time, continued maintenance and further deficit replacement can be continued at a slower rate. Adjustments can then be made in tonicity of the fluid, and appropriate potassium may be added to the infusion as needed.

In children who are hypovolemic with signs of more than 5 percent dehydration with mental status changes, the initial fluid resuscitation should be aggressive, using boluses of isotonic fluid. Intravenous access should be initiated quickly while the child is being assessed. In children with normal cardiac and renal function, it is most important to give adequate amounts of fluid. Overhydrating is preferred to underhydrating. Any excess fluid will be eliminated eventually in the urine. If it is difficult to start an intravenous line, fluids can be given orally by a drip method or, if the situation is life-threatening, by an interosseous route until intravenous access is obtained.

Principles and Calculation of Fluid Replacement and Maintenance of Intravenous Therapy in Dehydration
description of image}
  1. In hypovolemic children, boluses of isotonic fluid (15 to 30 mL per kg) should be given over 20 minutes.
  2. Fluid boluses should be repeated every 20 minutes until hypovolemia is corrected and urine output established.
  3. Isotonic fluid (normal saline or Ringer's lactate) is usually the correct initial fluid.
  4. After correction of hypovolemia, further intravenous therapy proceeds at a slower pace.
  5. Replacement of fluid in 24 hours=deficit + maintenance + ongoing losses.
  6. Deficit fluid=(percentage of dehydration) x (body weight in g)=mL of fluid needed.
  7. Twenty-four hour maintenance fluid equals 100 mL per kg for the first 10 kg, 50 mL per kg for the next 10 kg, and 25 mL per kg for every kg over 20 kg. (Sodium=3 to 5 mEq per kg per day; potassium=2 to 4 mEq per kg per day.)

The calculation of total fluid needs is based on the deficit, maintenance and ongoing losses. One of several accepted methods for the calculation of fluids is listed in Table 6. After the initial correction of hypovolemia, fluid correction proceeds at a slower pace. Generally, one half of the daily requirement is given over eight hours, with the remainder given over 16 hours. If the child has a hypertonic dehydration, isotonic fluid replacement is continued in order to not correct the hypernatremia too quickly, as that may cause cerebral edema and seizures. The correction of the hypernatraemia should take place slowly over one to two days, depending on its severity. After the deficit is corrected, maintenance fluid may be given as one-half normal saline.

Children with severe diarrhea need adequate nutrition in order to restore their digestive abilities, to recover from their illness and to prevent development of so-called "starvation diarrhea."2 Unless they have severe vomiting, children should not be deprived of nutrition for longer than one to two days. Breast feeding should be continued. Special elemental formulas may be needed at times to provide this nutrition until clinical recovery is adequate.1

The Authors

is an associate professor and vice chair of graduate medical education in the Department of Family and Community Medicine at the Medical College of Wisconsin, Milwaukee. He is a graduate of Temple University School of Medicine, Philadelphia, and served a residency in family medicine and pediatrics at Peter Bent Brigham Hospital in Boston.

is an associate professor at the Medical College of Wisconsin and program director of the Family Practice Residency Program at Waukesha Memorial Hospital, Waukesha, Wis. Dr. Lewan received his medical degree from the University of Chicago School of Medicine and served a residency in family practice at the University of Illinois­Rockford.

Address correspondence to B. Clair Eliason, M.D., Department of Family and Community Medicine, Medical College of Wisconsin, 1000 N. 92nd St., Milwaukee, WI 53226. Reprints are not available from the authors.


  1. Nelson WE, Behrman RE, Kliegman RM, Arvin AM, eds. Nelson Textbook of pediatrics. 15th ed. Philadelphia: Saunders, 1996.
  2. DeWitt TG. Acute diarrhea in children. Pediatr Rev 1989;11:6-13 [Published erratum in Pediatr Rev 1989;11:124].
  3. MacDonald KL, Osterholm MT. The emergence of Escherichia coli O157:H7 infection in the United States. The changing epidemiology of foodborne disease [Editorial]. JAMA 1993;269:2264-6.
  4. Goodgame RW. Understanding intestinal spore-forming protozoa: cryptosporidia, microsporidia, isospora, and cyclospora. Ann Intern Med 1996; 124:429-41.
  5. Farthing MJ. Giardiasis. Gastroenterol Clin North Am 1996;25:493-515.
  6. Swerdlow DL, Ries AA. Cholera in the Americas: guidelines for the clinician. JAMA 1992;267:1495-9.
  7. DeFancisco A, Khan SA, Chakraborty J. Interrater agreement in the assessment of dehydration in infants [Letter]. J Trop Pediatr 1997;43:119-21.
  8. Saavedra JM, Harris GD, Li S, Finberg L. Capillary refilling (skin turgor) in the assessment of dehydration. Am J Dis Child 1991;145:296-8.
  9. Mackenzie A, Barnes G, Shann F. Clinical signs of dehydration in children. Lancet 1989;2(8663):605-7.
  10. Fuchs SM, Jaffe DM. Evaluation of the "tilt test" in children. Ann Emerg Med 1987;16:386-90.
  11. American Academy of Pediatrics Committee on Nutrition. Use of oral fluid therapy and posttreatment feeding following enteritis in children in a developed country. Pediatrics 1985;75:358-61.
  12. Bezerra JA, Stathos TH, Duncan B, Gaines JA, Udall JN Jr. Treatment of infants with acute diarrhea: what's recommended and what's practiced. Pediatrics 1992;90(1):1-4.
  13. Gavin N, Merrick N, Davidson B. Efficacy of glucose-based oral rehydration therapy. Pediatrics 1996;98:45-51.
  14. Reid SR, Bonadio WA. Outpatient rapid intravenous rehydration to correct dehydration and resolve vomiting in children with acute gastroenteritis. Ann Emerg Med 1996;28:318-23.
  15. Bhattacharya SK, Bhattacharya MK, Manna B, Dutta D, Deb A, Dutta P, et al. Risk factors for the development of dehydration in young children with acute watery diarrhoea: a case-control study. Acta Paediatr 1995;84:160-4.
  16. Cohen MB, Mezoff AG, Laney DW Jr, Bezerra JA, Beane BM, Drazner D, et al. Use of a single solution for oral rehydration and maintenance therapy of infants with diarrhea and mild to moderate dehydration. Pediatrics 1995;95:639-45.
  17. Gore SM, Fontaine O, Pierce NF. Impact of rice based oral rehydration solution on stool ouput and duration of diarrhoea: meta-analysis of 13 clinical trials. BMJ 1992;304:287-91.
  18. A manual for the treatment of diarrhoea. Geneva, Switzerland: World Health Organization, 1990. Publication WHO/CDD/SER/80.2 Rev. 2. 1990.