Approach to chronic diarrhea in neonates and young infants (<6 months)

INTRODUCTION — Diarrhea in young infants (<6 months) is relatively common, generally mild, and often self-limited. Most cases in this age group are acquired, caused by infection or a food protein intolerance.

When diarrhea begins within days of birth and is persistent (lasting more than two weeks) or severe, several uncommon causes must be considered. These include congenital anatomic anomalies of the intestine and a group of rare genetic disorders collectively known as "congenital diarrheas and enteropathies" (CODEs) [1].

This topic review will outline a clinical approach to infants presenting with chronic diarrhea before six months of age and particularly during the first six weeks of life. The acquired causes will be outlined briefly, with links to more detailed discussions. The congenital causes, including CODEs, will be discussed in more detail.

Other aspects of chronic diarrhea in children are discussed in separate topic reviews:

●(See "Overview of the causes of chronic diarrhea in children in resource-abundant settings".)

●(See "Approach to chronic diarrhea in children >6 months in resource-abundant settings".)

●(See "Persistent diarrhea in children in resource-limited settings".)

OVERVIEW OF THE CAUSES OF DIARRHEA DURING INFANCY — In resource-abundant settings, a wide variety of disorders cause chronic diarrheas in neonates and young infants (table 1) [2]. Most frequently, chronic diarrhea is of allergic or infectious origin and is usually not associated with any significant long-term sequelae. Less common causes of diarrhea in infants include anatomic disorders such as gastroschisis or inflammatory disorders such as necrotizing enterocolitis, which can require surgical resection that may be complicated by short bowel syndrome and chronic diarrhea. In resource-limited settings, chronic diarrhea is usually associated with serial enteric infections and environmental enteric dysfunction [3]. (See "Persistent diarrhea in children in resource-limited settings".)

Congenital diarrheas and enteropathies (CODEs) are rare genetic disorders characterized by persistent and often severe diarrhea presenting in the first weeks of life, often associated with feeding intolerance, malabsorption, and growth failure [1]. Most CODEs are monogenic and can be categorized into genetic variants that directly affect the intestinal epithelium or those that affect the immune system, which secondarily impair epithelial function (table 2).

INITIAL EVALUATION

Assessment of the diarrhea — The first step is to determine the presence and severity of the diarrhea using either qualitative or quantitative measures, as appropriate to the clinical setting. Diarrhea in infancy is difficult to define based on stool frequency or consistency since the normal range for these parameters can vary greatly. In outpatient settings, the presence and severity of diarrhea can be inferred by the significant deviation from the daily stool pattern and by the level of dehydration, presence of electrolyte abnormalities, and poor weight gain. In inpatient settings, it may be possible to quantify the stool output, in which case, diarrhea can be defined as stool volume of more than 20 g/kg/day in a young infant. In children with ostomies, output of greater than 30 g/kg/day is considered excessive. (See "Pathogenesis of acute diarrhea in children".)

In infants, diarrhea is typically defined as acute if it lasts less than two weeks or chronic when it persists for longer than two weeks.

Clinical features that suggest a cause — The second step is to do a thorough history and physical examination to identify the most likely cause(s) of the diarrhea (table 1), focusing on the following considerations:

Acquired diarrheas

●Timing and severity – Acquired diarrheas can present at any time during infancy, usually after the first few weeks of life. They vary in severity but are usually mild or moderate. By contrast, diarrhea presenting immediately postnatally or requiring critical care intervention is more likely to represent a congenital diarrhea or enteropathy (CODE). (See 'Congenital diarrheas and enteropathies' below.)

●Triggers and associated features – Acquired diarrheas often have characteristic features that permit a provisional diagnosis. Important causes to consider in an infant are listed below; details of diagnosis and management are available in the linked topic reviews.

•Infections – Infections are the most common cause of diarrhea in either resource-abundant or resource-limited settings. Clues to enteric infections are sudden onset of symptoms, ill contacts, and sometimes fever or vomiting. While most infectious diarrheas resolve within two weeks and are therefore considered acute diarrheas, they occasionally persist, especially in immunocompromised individuals. In some cases, persistent diarrhea after an acute infection is caused by lingering mucosal injury with malabsorption, termed "postinfectious enteropathy."

Most infectious diarrheas are caused by viruses, including rotavirus, cytomegalovirus, adenovirus, and norovirus. Less common in resource-abundant settings, but a significant factor globally, are enteric bacterial pathogens, such as Salmonella enterica, Shigella spp, Campylobacter jejuni, and pathogenic Escherichia coli [4]. (See "Diagnostic approach to diarrhea in children in resource-abundant settings" and "Approach to the child with acute diarrhea in resource-limited settings".)

•Food-induced – Food protein-induced allergic proctocolitis is a common cause of bloody, loose stools in young infants [5]. The infant is otherwise healthy, and there is no associated vomiting or weight loss. In most cases, the triggering protein is cow's milk, ingested through either breast milk or cow's milk-based formula. It can be induced by soy protein as well. (See "Food protein-induced allergic proctocolitis of infancy".)

Food protein-induced enterocolitis syndrome (FPIES) is an uncommon gastrointestinal food hypersensitivity that manifests as profuse, repetitive vomiting, sometimes with diarrhea, leading to dehydration and lethargy in the acute setting or weight loss and failure to thrive in a chronic form [6,7]. Symptoms of classic FPIES usually begin in early infancy, within one to four weeks following introduction of cow's milk or soy protein. In extreme cases, symptoms may begin within the first days of life. FPIES may present later in infancy if the exposure is delayed, eg, when the trigger is a solid food. (See "Food protein-induced enterocolitis syndrome (FPIES)".)

•Prematurity or medical illness – Necrotizing enterocolitis usually presents in hospitalized premature infants with symptoms of watery or bloody diarrhea, abdominal distension, feeding intolerance, and temperature instability [8,9]. Abdominal radiographs show dilated bowel loops, often with intramural gas (pneumatosis intestinalis). Occasionally, necrotizing enterocolitis develops in full-term infants, usually in the setting of an underlying medical illness (eg, congenital heart disease or sepsis). (See "Neonatal necrotizing enterocolitis: Clinical features and diagnosis".)

If necrotizing enterocolitis, intestinal atresias, or other congenital bowel malformations require resection of substantial portions of bowel, this may lead to short bowel syndrome, with associated malabsorption and diarrhea. (See "Management of short bowel syndrome in children".)

Any medication or herbal supplement given to the infant should be reviewed as a possible cause of diarrhea. As examples, antibiotics or products with high concentrations of fructose commonly induce diarrhea [10].

Congenital diarrheas and enteropathies — CODEs are rare genetic disorders with diverse mechanisms, characterized by persistent and often severe diarrhea presenting in the first weeks of life (table 2).

●Timing and severity – Infants with CODEs typically have severe diarrhea from birth; although, in some cases, diarrhea may start after the neonatal period, particularly when caused by immune dysregulation. By contrast, acquired diarrheas due to infectious and allergic conditions generally have a symptom-free postnatal period of at least a few weeks before the diarrhea develops. In some CODEs, the volume of diarrhea may be so high that less experienced parents, caregivers, and clinicians may confuse watery diarrhea for urine, delaying recognition and appropriate consultation. In these cases, the infant may come to medical attention because of dehydration or a persistent metabolic acidosis.

●Other features – Other clinical features that raise the possibility of CODEs include polyhydramnios, multisystemic disease (eg, dysmorphism or other congenital anomalies or immune deficiencies), and consanguinity or membership in a higher-risk population (eg, Navajo) [11-16].

Other congenital disorders that may present with diarrhea — Congenital abnormalities of the intestine sometimes present with diarrhea in young infants. These usually can be diagnosed by fluoroscopic imaging, guided by the characteristic clinical features of each disorder (table 1):

●Hirschsprung disease – Infants with Hirschsprung disease occasionally present with Hirschsprung-associated enterocolitis, a potentially life-threatening illness with a sepsis-like picture including fever, vomiting, diarrhea, and abdominal distension. It can also present with subacute diarrhea and poor weight gain. More often, Hirschsprung disease presents with symptoms of distal intestinal obstruction, including bilious emesis, abdominal distension, delayed passage of meconium or stool, and constipation. An important clue is explosive release of gas and stool during digital examination of the rectum ("squirt sign"). Infants with Down syndrome have a high risk of having Hirschsprung disease. (See "Congenital aganglionic megacolon (Hirschsprung disease)", section on 'Suspected enterocolitis'.)

●Intestinal malrotation with intermittent volvulus – Intestinal malrotation with intermittent volvulus occasionally presents with diarrhea, which may be bloody, and intermittent abdominal pain. However, obstructive symptoms and vomiting are more common presenting features. (See "Intestinal malrotation in children" and "Lower gastrointestinal bleeding in children: Causes and diagnostic approach", section on 'Malrotation with midgut volvulus'.)

●Pseudo-obstruction – Chronic intestinal pseudo-obstruction (CIPO) is characterized by impaired intestinal motility without anatomic obstruction, leading to diffuse bowel dilation, abdominal distension, (usually) vomiting and constipation, and (sometimes) diarrhea [17]. Fluoroscopic imaging often reveals very dilated bowel loops and malrotation (similar to congenital short bowel syndrome). It is rare in infants and is caused by congenital defects in myopathies, neuropathies, or mitochondrial disorders. Some cases are associated with dilated but unobstructed bladder (known as megacystis, microcolon, and hypoperistalsis syndrome) [18,19]. (See "Chronic intestinal pseudo-obstruction: Etiology, clinical manifestations, and diagnosis", section on 'Genetic'.)

●Congenital short bowel syndrome – This is a rare congenital defect that usually presents during the first few days of life with generalized malabsorptive diarrhea and dilated bowel loops despite lack of abdominal distension, often with bile-stained vomiting and failure to thrive [20]. Fluoroscopic imaging often reveals malrotation (similar to CIPO), as well as the abnormally shortened small bowel. Genetic and in utero developmental causes have been identified. The presentation can be very similar to many CODE disorders.

EVALUATION FOR SUSPECTED CONGENITAL DIARRHEAS AND ENTEROPATHIES — If the initial evaluation suggests an acquired diarrhea, further evaluation depends on the suspected type (see 'Acquired diarrheas' above). If the initial evaluation suggests the possibility of a congenital diarrhea and enteropathy (CODE), an approach to further evaluation is described below.

Initial tests — For most patients with a suspected CODE, the initial evaluation includes a battery of serum tests, stool tests, fluoroscopic upper gastrointestinal series, and upper endoscopy and sigmoidoscopy (with colonoscopy for selected patients). A typical sequence of testing is outlined below, but the order may vary depending on the patient's presentation and clinical suspicion for a particular disorder (table 2).

Serum testing — Initial testing for all infants with a suspected CODE should include:

●Complete blood count with differential – Low hemoglobin suggests iron deficiency anemia (eg, due to blood loss), leukocytosis suggests an infectious diarrhea, and low total lymphocyte counts suggest a possible T or B cell deficiency.

●Electrolytes and trace elements (sodium [Na⁺], potassium [K⁺], chloride [Cl^-], bicarbonate [HCO₃^-], and glucose) – Hypokalemic metabolic acidosis (low K⁺ and HCO₃^-) is the most common electrolyte pattern associated with most diarrheas. However, congenital chloride diarrhea presents with a hypochloremic metabolic alkalosis [15]. Some of these disorders are associated with hyperglycemia or diabetes that develop during infancy (Mitchell-Riley syndrome [RFX6 gene mutation]) or later in life (immune dysregulation, polyendocrinopathy, enteropathy, X-linked [IPEX] syndrome, intestinal anendocrinosis [NEUROG3 gene mutation]).

●Systemic inflammatory biomarkers – Elevated levels of C-reactive protein or erythrocyte sedimentation rate suggest an acute or chronic inflammatory condition, especially if the diarrhea is bloody. In a young infant with chronic diarrhea, there are several possible causes of inflammatory diarrhea, which are often categorized as very early-onset inflammatory bowel disease, a congenital immunodeficiency, or both. (See "Clinical presentation and diagnosis of inflammatory bowel disease in children", section on 'Very early-onset inflammatory bowel disease'.)

●Liver function tests (alanine aminotransferase [ALT], aspartate transaminase [AST], alkaline phosphatase) – The primary purpose of measuring AST and ALT is to screen for cholestatic liver disease, which can cause chronic diarrhea due to bile acid insufficiency. In infants receiving parenteral nutrition, abnormalities in these tests are often due to intestinal failure-associated liver disease (see "Intestinal failure-associated liver disease in infants"). However, some monogenic CODEs are also associated with liver disease, including microvillus inclusion disease, trichohepatoenteric syndrome, familial hemophagocytic lymphohistiocytosis, and some autoimmune enteropathies.

●Albumin – Low serum albumin associated with diarrhea may suggest insufficient caloric absorption or excessive protein loss, indicative of protein-losing enteropathy. (See "Protein-losing gastroenteropathy".)

The following tests also may be included in the initial evaluation if suggestive clinical features are present (table 2) or may be selected later in the evaluation based on the type of diarrhea.

●If fat malabsorption is suspected (due to grossly fatty diarrhea, positive fecal fat screening, abnormal serum lipids, or clinical features of cystic fibrosis):

•Lipid panel (triglycerides, cholesterol, high-density lipoprotein, low-density lipoprotein [LDL]) – Very low LDL suggests abetalipoproteinemia. Elevated triglycerides and cholesterol suggest cholestatic liver disease. (See "Low LDL-cholesterol: Etiologies and approach to evaluation".)

•Fat-soluble vitamins (vitamin A, 25-hydroxyvitamin D, vitamin E [alpha-tocopherol], prothrombin time [or international normalized ratio]) – Abnormally low concentrations of fat-soluble vitamins in the absence of cholestasis suggests either exocrine pancreatic insufficiency or impairment of enterocyte absorption of fat-containing nutrients (eg, due to short bowel syndrome or disorders of fat transport and metabolism, such as abetalipoproteinemia).

●If an immunodeficiency is suspected, the following tests should be included. Signs and symptoms of an immunodeficiency may include bloody diarrhea, recurrent severe infections, desquamating rash, diabetes mellitus, syndromic appearance, other congenital anomalies, complete blood count abnormalities, an inflammatory infiltrate on endoscopic biopsies, abnormal results of tests for severe combined immunodeficiency on newborn screening, or family history of immunodeficiency (table 3). (See "Recognition of immunodeficiency in the first three months of life".)

•Serum immunoglobulin G and A (IgG and IgA) level – Hypogammaglobulinemia suggests either a primary humoral immunodeficiency or a globulin loss, as may occur in a protein-losing enteropathy. The results should be interpreted using age-specific reference ranges provided by the laboratory since there is a physiologic nadir during the first six months of life (figure 1). (See "Primary humoral immunodeficiencies: An overview".)

•Specific T, B, and natural killer (NK) cell subset analysis by flow cytometry – To evaluate for primary immunodeficiencies. (See "Flow cytometry for the diagnosis of inborn errors of immunity".)

●If an eczematous dermatitis is present (picture 1):

•Evaluate for IPEX syndrome. (See "IPEX: Immune dysregulation, polyendocrinopathy, enteropathy, X-linked".)

●If acral or periorificial dermatitis (picture 2) is present:

•Zinc concentration – A low zinc concentration suggests acrodermatitis enteropathica, a recessively inherited defect in intestinal zinc absorption. (See "Zinc deficiency and supplementation in children", section on 'Acrodermatitis enteropathica'.)

Stool testing — Stool testing is a key component of the diagnostic evaluation for infantile diarrheas. The first step is to evaluate for the most common infectious diarrheas, if these have not already been excluded. This testing should include standard assessment of enteric bacterial pathogens by culture and/or by polymerase chain reaction. Viral pathogens such as rotavirus, cytomegalovirus, norovirus, and adenovirus should be tested by polymerase chain reaction and/or enzyme-linked immunosorbent assay (ELISA). Cytomegalovirus and adenovirus can also be assessed on the intestinal biopsy if performed later in the evaluation.

The following tests require special stool-collection procedures and should be performed during feeds with a full caloric challenge, as well as during fasting (unless the diarrhea abates completely during fasting). The infant's dietary intake at the time of testing should be recorded in detail to permit accurate interpretation of the stool test results.

Collection of accurate stool samples can be challenging because the diaper rapidly absorbs stool water and stool is often mixed with urine in the diaper. To avoid loss of stool water in an absorbent diaper, either reverse the diaper so that the nonabsorbent surface is on the inside or line the diaper with a nonabsorbent surface, such as an adhesive wound dressing. In addition, the stool must be separated from urine to allow for accurate stool sampling and volume measurement. For boys, this usually can be accomplished by using an adhesive bag to catch the urine. For girls (or for boys if the adhesive bag is not successful), a urine catheter may be needed but should only be used for a few days to reduce the risk a of secondary urinary tract infection. The stool samples should be packaged and sent for testing as soon as possible after excretion.

Samples of stool should be sent for the following tests:

●Electrolytes – This is a key test in evaluating the nature of the diarrhea and for calculation of the stool osmotic gap. Results are interpreted as follows (see "Pathogenesis of acute diarrhea in children", section on 'Diarrhea classification'):

•High stool Na⁺ or Cl^- suggests an electrolyte transport-related diarrhea. These include congenital chloride diarrhea, congenital sodium diarrhea, hormone diarrhea (due to VIPoma or serotonin-secreting tumor), or primary bile acid diarrhea (due to a defect in ileal bile acid reuptake transporters [eg, SLC10A2]). In congenital chloride diarrhea, stool Cl^- is typically >120 mM [15]. In congenital sodium diarrhea, stool Na⁺ is typically >145 mM.

•Stool osmotic gap – The stool osmotic gap can be calculated from the stool electrolytes using this calculator (calculator 1) or the following formula:

Results are expressed in mOsm/kg stool weight.

-A stool osmotic gap >100 mOsm/kg is high, suggesting diet-induced (osmotic) diarrhea, usually due to unabsorbed carbohydrate moieties

-A stool osmotic gap <50 mOsm/kg is low, suggesting electrolyte transport-related (secretory) diarrhea

-Intermediate results (50 to 100 mOsm/kg) are equivocal; they may be caused by a mixed mechanism for diarrhea or artifact of the stool collection

●Osmolarity – Stool osmolarity should be measured if there is suspicion of improper collection (eg, delay in collection or contamination with urine or water). This is also valuable for any infant with profuse watery diarrhea to validate the other stool tests. Proper stool collection is confirmed if the stool osmolality is isoosmolar to serum (approximately 290 mOsm). Stool osmolarity that is well above this threshold suggests contamination with urine (or delay in sample processing, with evaporation of water or fermentation). Stool osmolarity below this threshold suggests contamination with water (eg, factitious diarrhea).

●Reducing substances and pH – These tests serve as an initial screen for carbohydrate malabsorption. They must be interpreted in the context of the infant's diet and age. Neonates normally have some degree of carbohydrate malabsorption because of immaturity of the disaccharidase enzymes (eg, lactase) on the intestinal brush border.

•Reducing substances >0.5 percent suggests possible malabsorption of disaccharides or monosaccharides (although up to 0.75 percent can be normal in neonates and young infants).

Glucose, lactose, and fructose are reducing sugars. Therefore, this test accurately reflects carbohydrate malabsorption if the infant is breastfed or on a formula that contains these sugars (eg, any cow's milk-based formula). By contrast, reducing substances may be negative if the infant is on a formula that contains only sucrose (a nonreducing sugar). However, malabsorbed sucrose is sometimes degraded by colonic bacteria to glucose and fructose, resulting in a positive test for reducing substances.

•Low pH (<5.3) also suggests carbohydrate malabsorption; this finding is caused by fermentation of malabsorbed carbohydrates by colonic bacteria, resulting in an abundance of short-chain fatty acids.

●Alpha-1 antitrypsin – Elevated alpha-1 antitrypsin in the stool reflects intestinal protein loss; patients with this finding may have low serum albumin and IgG as well. Causes include a protein-losing enteropathy (eg, lymphangiectasia, diacylglycerol acyltransferase 1 [DGAT1] deficiency, or complement decay-accelerating factor [CD55] deficiency) or an inflammatory enteropathy (eg, immune dysregulation-associated enteropathies [autoimmune or immunodeficiency] or postinfectious enteropathy). Alpha-1 antitrypsin is a protein that is largely resistant to the action of intestinal proteases, unless it is secreted in the stomach.

●Fat – A qualitative or spot fecal fat (neutral or split) is a helpful initial screen for fat malabsorption, although the sensitivity and specificity of this test are limited. Elevation in neutral fat reflects increased unhydrolyzed mono-, di-, or triglyceride content in the stool and is consistent with pancreatic insufficiency. Elevation in split fat reflects increased free fatty acids in stool and correlates with fat malabsorption from injury to or loss of the mucosal surface.

If fat malabsorption is suspected (eg, due to fatty-appearing stools or positive spot test), a quantitative 72-hour fecal fat collection should be performed for a more accurate assessment of fat malabsorption. The test requires a complete collection of all stools passed during the collection period, as well as a record of dietary fat intake, to permit calculation of the percentage of fat absorbed. Normal neonates and young infants absorb 75 to 80 percent of dietary fat, and this proportion increases with age; a healthy six-month-old infant absorbs 80 to 95 percent of dietary fat.

●Elastase – Stool elastase should be measured to screen for pancreatic insufficiency. Elastase is secreted by the pancreas, and low values (<200 mcg/g) suggest pancreatic insufficiency. Abnormal results also may be seen in any type of high-volume diarrhea due to the effects of dilution.

●Inflammation – The following tests serve to screen for an inflammatory process:

•Occult blood – A positive test for occult blood (Hemoccult) indicates or confirms the presence of blood in the stool and could indicate an inflammatory process. However, positive tests are common in infants with severe diarrhea of any cause, due to skin irritation. Therefore, the results should only be used as supportive evidence for an inflammatory process.

•Inflammatory markers – Elevated calprotectin or lactoferrin are correlated with the presence of mucosal intestinal inflammation.

Imaging — Infants with suspected CODEs should be evaluated with a fluoroscopic upper gastrointestinal series with small bowel follow-through (UGI/SBFT). This is particularly important for infants presenting with any signs or symptoms of obstruction, including vomiting and abdominal distension, or with signs of proximal bowel obstruction on conventional abdominal radiograph. The main purpose is to evaluate for malrotation, congenital short bowel syndrome, and chronic intestinal pseudo-obstruction (CIPO). Congenital short bowel syndrome and CIPO are characterized by very dilated bowel loops and are often associated with malrotation. (See 'Other congenital disorders that may present with diarrhea' above.)

If distal bowel obstruction is suspected based on the conventional radiograph or other clinical features (eg, Hirschsprung disease), an unprepped contrast enema is indicated. (See "Congenital aganglionic megacolon (Hirschsprung disease)", section on 'Contrast enema'.)

Endoscopy/histology — Endoscopic evaluation of the bowel is almost always needed in the evaluation of an infant for a suspected CODE. Regardless of the nature of the diarrhea, evaluation of endoscopic biopsies is a key step, allowing prioritization of further testing (see 'Diagnostic algorithm' below). In general, at least an upper endoscopy (esophagogastroduodenoscopy) should be performed to assess small intestinal structure. Sigmoidoscopy is often performed as well to assess colonic changes, especially if there is any suspicion of colitis. Samples should be sent for histologic examination and a sample prepared for electron microscopy.

Interpretation — Initial evaluation of hematoxylin and eosin (H&E)-stained biopsy sections should focus on overall intestinal epithelial architecture, namely villus-to-crypt ratio, abundance of the epithelial cell types and microscopic structure, and immune cell composition in the lamina propria and intraepithelial compartments [21]. Findings of note for infants with suspected CODEs are (table 2):

●Normal villus and crypt architecture and ratio – Normal villus architecture is seen in several different conditions in which diarrhea is caused by intracellular abnormalities that affect digestion or absorption of nutrients. These include specific carbohydrate malabsorption disorders or defects in electrolyte transport, such as congenital chloride or sodium diarrheas. Normal villus and crypt architecture are also seen in diarrhea caused by defects of enteroendocrine cell development or function (eg, enteric anendocrinosis [NEUROG3 deficiency], proprotein convertase type 1 [PCSK1] deficiency, or Mitchell-Riley syndrome [RFX6]).

●Abnormal villus and crypt architecture and villus-to-crypt ratio – Villi that are short or flattened may be associated with either crypt hyperplasia or crypt hypoplasia; these findings may be seen in defects of enterocyte structure, vesicular trafficking, differentiation, and immune-mediated conditions. Specifically, villus blunting and apoptosis at the crypt base suggests an immune-mediated disorder. Epithelial crowding and disorganization at the top of the villus is found in tufting enteropathy and is associated with crypt hyperplasia.

●The relative abundance and distribution of cells in the stem cell-based compartment and differentiated cells of the epithelium can be appreciated by H&E and by cell type-specific staining. Selective depletion of differentiated cell types may be associated with autoimmune enteropathies and endocrinopathies. Enteroendocrine cell loss is seen in enteric anendocrinosis (NEUROG3 deficiency).

●Abnormal abundance or absence of mononuclear cells in the lamina propria and intraepithelial compartments suggest an immune-mediated disorder. A dominance of eosinophilic infiltrates within the mucosa and lamina propria, associated with villus blunting, may be an indicator of eosinophilic gastroenteritis.

●Electron microscopy is performed to assess the presence, relative size, and location of the microvilli and to identify intracellular microvillus inclusions (characteristic of microvillus inclusion disease) or abnormal vesicular structures suggestive of disorders of intracellular trafficking (eg, due to mutations in the STX3 gene, or TTC37). For practical reasons, electron microscopy is usually ordered for all patients with suspected CODEs, but it is most relevant for patients with abnormal villus architecture on H&E stain and lack of inflammatory infiltrate. One practical option is to prepare a sample for electron microscopy but defer the actual test until the H&E results are available.

Immunohistochemical staining — Further evaluation of the endoscopic biopsies depends upon the suspected disorder (table 2). In particular, infants with watery diarrhea and abnormal villus architecture should be evaluated with special stains and electron microscopy to assess for microvillus inclusion disease or tufting enteropathy.

●If the H&E stain shows normal villus architecture, stain for enteroendocrine cells (chromogranin A) to evaluate for enteric anendocrinosis (NEUROG3 deficiency).

●If the H&E stain showed abnormal villus architecture, perform the following [22,23]:

•CD10/villin immunostaining (to evaluate for microvillus inclusion disease)

•MOC-31 immunostaining (to evaluate for congenital tufting enteropathy)

•Frozen sections-staining with oil red O if lipid trafficking disorders are under consideration

•Electron microscopy to evaluate for abnormalities of the microvilli if not already performed (see 'Interpretation' above)

●If an immunodeficiency is suspected clinically or if there is an abundance or paucity of immune cells on the initial biopsy, further evaluation may include further specific markers for immune cell subtypes.

Diagnostic algorithm — The evaluation for CODEs is initiated following the exclusion of more commonly acquired diarrhea [1]. (See 'Acquired diarrheas' above.)

For an infant with a suspected CODE, the sequence and interpretation of the initial serum and stool tests and endoscopic findings is facilitated by categorizing the stool by its general appearance into one of three categories, which help to narrow the diagnostic possibilities (table 2):

●Watery stool is characterized by a high liquid content, often with very little form, which can be mistaken for urine.

●Fatty stools are usually foul smelling, can have a bulky or "fluffy" appearance, are pale in color, and/or are spot fecal fat-positive.

●Bloody stools contain gross blood mixed in with stools. The presence of large volumes of bright red blood or melena should trigger evaluation of vascular or anatomic gastrointestinal bleeding as well as bacterial infection.

Genetic testing can be performed after the diagnostic possibilities are narrowed down by initial testing. Alternatively, these tests can be performed earlier in the evaluation, especially if the infectious work-up is negative and there are clear factors to suspect a monogenic diarrheal disease, such as significant consanguinity, a family history of gastrointestinal disease in infancy, and clinical indicators such as diarrhea severity and neonatal onset. (See 'Genomic testing' below.)

Evaluation of watery diarrhea

●Trial of fasting – Following documentation of diarrheal output with normal feeding, evaluation should include a trial of withholding feeds for a minimum of 24 hours, with assessment of stool output and stool electrolytes (if possible), preferably while receiving intravenous fluids.

•If diarrheal volume is unchanged or is minimally changed following fasting, this suggests electrolyte transport-related diarrhea (also known as secretory diarrhea). These disorders are usually associated with elevated stool electrolytes with a low osmotic gap. They include congenital chloride diarrhea, congenital sodium diarrhea, hormone diarrhea (due to VIPoma or serotonin-secreting tumors), or primary bile acid diarrhea. (See 'Stool testing' above.)

•If diarrheal volume significantly improves during fasting, this suggests diet-induced diarrhea (also known as osmotic diarrhea).

•Other types of diarrhea have mixed mechanisms, and the fasting trial may result in no change, improvement, or equivocal results. This is true for disorders with epithelial structural defects (eg, microvillus inclusion disease, tufting enteropathy).

●If diet-induced diarrhea is present, the next step is to determine if this is due to unabsorbed carbohydrates, which are the most common cause. Carbohydrate malabsorption is supported by the presence of reducing substances in the stool and a low pH. Further support comes from a trial of feeds with a carbohydrate-free formula (eg, Ross Carbohydrate Free); significant improvement on this diet supports the diagnosis of carbohydrate malabsorption. If the child is clinically stable, the dietary trial should be done with bolus feeds of full-strength formula administered at a full caloric load.

If diarrhea persists on a carbohydrate-free diet, consider the disorders associated with a generalized malabsorption, which include enteric endocrinopathies (eg, enteric anendocrinosis or dysendocrinosis) and congenital short gut. In this setting, challenges with glucose (eg, Pedialyte) alone would also exacerbate the diarrhea.

●If carbohydrate malabsorption is present, the next step is to determine if there is selective malabsorption of a specific carbohydrate:

•The specificity of carbohydrate absorption can be evaluated by dietary challenges with various disaccharides (sucrose, lactose, and maltose) and monosaccharides (glucose and fructose) and assessed by changes in stool volume. Breath hydrogen testing also could identify the specific carbohydrate, but the use of this test is limited because accurate breath collection is very challenging in infants [24].

•If an endoscopy is performed, disaccharidase activity assays for lactase, sucrase, maltase, palatinase, and trehalase can be measured in biopsies of the proximal small bowel in an attempt to diagnose disaccharidase deficiency. However, these enzymatic assays can be unreliable due to poor sampling or specimen handling. In the setting of inflammation or villus atrophy, the loss of mucosal epithelium results in secondary and generalized disaccharidase deficiency.

●If there is no clear evidence of a selective carbohydrate malabsorption, the next steps are:

•Esophagogastroduodenoscopy and flexible sigmoidoscopy with biopsies for histologic analysis if not contraindicated by the clinical status of the infant [25]. Biopsies should include samples for both routine histology and electron microscopy. (See 'Endoscopy/histology' above.)

•In addition, the patient should be evaluated for a possible protein-losing enteropathy, which is suggested by elevated levels of alpha-1 antitrypsin in the stool, often in conjunction with low serum albumin, low IgG, and lymphopenia.

A protein-losing enteropathy may be caused by a compromised epithelial barrier, as in an immune dysregulation-associated enteropathy (autoimmune or immunodeficiency-related); lymphangiectasia; or certain enterocyte-specific monogenic disorders, including DGAT1 or CD55 deficiency. (See 'Initial tests' above.)

●Alternatively, the endoscopy can be performed earlier in the diagnostic sequence (before the dietary trials). Early endoscopy and biopsy helps to narrow the diagnostic categories by identifying abnormalities in villus architecture and determining if an inflammatory infiltrate is present. These findings often streamline the remaining steps in the diagnostic evaluation, including dietary trials and genetic testing.

Evaluation of fatty diarrhea — The first step in evaluation of fatty diarrhea is to confirm the presence of excessive fat in the stool by spot testing for neutral and split (hydrolyzed) fat. If possible, a 72-hour stool fat collection should be performed for quantitative evaluation of fecal fat. Stool elastase also should be measured.

●The combination of elevated fecal fat and low fecal elastase suggests exocrine pancreatic insufficiency, such as cystic fibrosis, although fecal elastase can be falsely low (false-positive) with high-volume diarrhea. Pancreatic insufficiency is confirmed by the responsiveness of diarrheal symptoms to enzyme replacement therapy [26]. (See "Cystic fibrosis: Assessment and management of pancreatic insufficiency", section on 'Fecal elastase'.)

●The combination of elevated fecal fat and normal fecal elastase suggests intestinal fat malabsorption due to mucosal abnormalities or a disorder of fat transport and metabolism, such as chylomicron retention disease and abetalipoproteinemia. These disorders of fat transport and metabolism are also characterized by fat-laden enterocytes in histologic sections and serum lipid abnormalities, especially low LDL.

Patients with suspected or confirmed fat malabsorption also should be evaluated for associated fat-soluble vitamin deficiencies. (See 'Serum testing' above.)

Evaluation of bloody diarrhea — The presence of gross blood implies significant colitis, and further evaluation should include stool inflammatory markers (calprotectin or lactoferrin), colonoscopy, and esophagogastroduodenoscopy. The presence of inflammatory changes in histologic sections should precipitate further investigation for immune dysregulation-associated enteropathies, which include autoimmune enteropathies and primary immunodeficiencies. Some of these disorders are classified as infantile very early-onset inflammatory bowel disease. (See "Approach to the child with recurrent infections" and "Primary humoral immunodeficiencies: An overview" and "Laboratory evaluation of the immune system" and "Genetic testing in patients with a suspected primary immunodeficiency or autoinflammatory syndrome" and "Clinical presentation and diagnosis of inflammatory bowel disease in children", section on 'Very early-onset inflammatory bowel disease'.)

It should be noted that not all forms of immune dysregulation-related enteropathies are associated with bloody stools. Some of these disorders present with watery diarrhea or with diet-induced characteristics (improves with fasting, low stool Na⁺, and high osmolar gap), and others present with an intermediate osmolar gap [27].

Genomic testing — Advances in sequencing technologies have allowed diagnosis of new and unknown genetic causes of diarrhea. In cases where there is a high suspicion of a CODE diagnosis but the specific etiology requires confirmation, either targeted genetic testing (Sanger sequencing) or whole-exome sequencing (WES) can identify the genetic cause and potentially direct management. Trio WES that includes the proband and biologic parents is considerably more informative and should be encouraged whenever possible. (See "Next-generation DNA sequencing (NGS): Principles and clinical applications".)

In cases of a suspected CODE where the diagnosis based on clinical evaluation is unclear, WES should be performed to identify a possible causative genetic mutation. However, WES may not detect genetic defects in genes due to technical reasons (eg, regions with poor coverage, large insertions and deletions, and mutations in regulatory and splice or intronic regions). Moreover, interpretation of WES can be challenging because it provides data regarding "variants" in multiple genes, many of which are unexpected, outside the gene(s) of interest and of unknown significance. These findings require interpretation by a clinical geneticist and functional confirmation by tertiary centers before a diagnosis can be confirmed. In cases where a monogenic disorder is likely despite a normal or inconclusive WES, whole-genome sequencing should be considered.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Chronic diarrhea".)

SUMMARY AND RECOMMENDATIONS

●Definitions – During infancy, stool volumes of >20 g/kg/day are considered diarrhea. If symptoms persist for more than two weeks, it is classified as chronic. (See 'Assessment of the diarrhea' above.)

●Acquired diarrheas – Acquired forms of chronic diarrhea result from infectious, allergic, or underlying congenital anomalies or medical disorders (table 1). If the latter require resection of significant portions of the bowel, they may lead to short bowel syndrome, with associated malabsorption and chronic diarrhea. Some cases of persistent or recurrent infectious diarrhea are caused by a primary immunodeficiency. Food protein-induced allergic proctocolitis (due to cow's milk protein) is a common benign cause of bloody stools in young infants. (See 'Acquired diarrheas' above.)

●Congenital diarrheas and enteropathies (CODEs) – CODEs are uncommon but serious causes of chronic diarrhea in neonates and young infants. Most CODEs are monogenic and can be categorized into genetic variants that directly affect the intestinal epithelium and impair absorption of nutrients or electrolyte flux, or those that affect the immune system, which, due to inflammation, secondarily impair epithelial function (table 2).

•Clinical presentation – Most CODEs present with severe diarrheal symptoms that begin within the first weeks of life, although some CODEs, particularly those caused by immune dysregulation, may present after the neonatal period. Other clinical features that raise the possibility of a CODE include membership in a population with a high frequency of certain genetic variants or consanguinity, multisystemic disease (eg, dysmorphism or other congenital anomalies or immune deficiencies), and polyhydramnios. (See 'Congenital diarrheas and enteropathies' above.)

•Evaluation – For most patients with a suspected CODE, the initial evaluation includes a battery of serum tests, stool tests, and upper endoscopy (with colonoscopy for selected patients). A few of these tests may be selected later in the evaluation based on the type of diarrhea.

-Standard serum testing includes complete blood count, electrolytes, erythrocyte sedimentation rate or C-reactive protein, liver function tests, and albumin. Additional testing for selected patients includes a lipid panel, immunoglobulin G and A (IgG and IgA), and T and B cell subsets. (See 'Serum testing' above.)

-Standard stool tests consist of electrolytes (to calculate an osmotic gap), reducing substance and pH, alpha-1 antitrypsin, elastase, calprotectin, and culture and/or polymerase chain reaction techniques to rule out bacterial or other infectious causes. Some of these tests require special collection procedures to ensure an accurate result. (See 'Stool testing' above.)

The results of these tests inform the diagnostic steps used to determine the most likely type of CODE. For watery diarrhea, trials of fasting or carbohydrate-free formula help to narrow the diagnostic possibilities. Once the likely category of CODE is identified, functional testing or genomic testing can be used to confirm the specific diagnosis. (See 'Diagnostic algorithm' above and 'Genomic testing' above.)