Vomiting in infants and children

Author :

David J. Hackam, MD, PhD

Attending Pediatric Surgeon, Children’s Hospital of Pittsburgh

Associate Professor of Surgery

University of Pittsburgh School of Medicine

Co-Director, Fetal Diagnosis and Treatment Center, Children’s Hospital of Pittsburgh

2008-07-28

Vomiting in infants and children :  Pyloric stenosis, gastro-esophageal reflux and other players
Highlights:

• Vomiting in infants is extremely common
• The most common cause is gastro-esophageal reflux. Other causes include pyloric stenosis, and malrotation with volvulus. These are explained below.
• Treatment options include simple changes in feeding habits and a combination of specific drugs or surgery as needed; these are effective in most cases. 

The vomiting infant: a normal part of life or a serious problem?

   One of the undeniable truisms in life is that infants will vomit. In fact, the activity of vomiting comprises a third of the entire physical repertoire of the average infant (along with eating and stooling). And it is because infant vomiting is so common that it is important to differentiate between normal vomiting – as occurs in almost all babies to some degree – and abnormal vomiting, which may be indicative of a potentially severe underlying disorder, is so important. This knol will focus on the causes of infant vomiting and the different treatment options available.

Signs of “healthy” versus “worrisome” vomiting in infants.

   As mentioned above, all babies vomit to some degree, and the ability to distinguish “healthy” from “worrisome” vomiting is therefore of major importance. Figure 1 is an anatomic picture of the various types of infant vomiting that are discussed in this knol. In order to determine the degree of seriousness of a particular infant that is vomiting, one needs to characterize what the vomit looks like (sorry, there’s no easy way to “sugar-coat” the importance of that) and how sick the baby is. Vomit that looks like feeds and comes up immediately after a feeding is almost always gastro-esophageal reflux (also called “reflux”). This may or may not be of concern, as is described below. Vomiting that occurs a short while after feeding, or indeed vomiting that projects out of the baby’s mouth may be indicative of a condition called pyloric stenosis, which requires surgery to treat.  By contrast, vomit that has any green color in it is always worrisome. This may be reflective of a twist in the intestine (a condition called volvulus), an underlying infection, or some other cause for a blockage in the intestine¹. A more detailed description of these conditions, as well as a discussion of the treatment options, is provided below.

1. Gastro-esophageal reflux in infants:

   Gastro-esophageal reflux (also called “GER”) refers to the condition in which food (either breast milk or infant formula) forcibly travels from the stomach “back up” into the esophagus, as is shown in Figure 1. All infants have some degree of reflux. Infants swallow a lot of air – during feeding, during crying and during babbling – which leads to vomiting (also called emesis) when belched up.

Many babies are susceptible to “over-feeding,” particularly in cases in which the first sign of crying or irritation is met with a full nipple (attached to either a breast or a bottle). In cases when the infant is crying for reasons unrelated to hunger, over-feeding can actually make a fussy baby even fussier, leading to more swallowed air, and more vomiting. In cases in which the reflux is the baby’s only disorder, the reflux itself doesn’t usually make the baby sick. They may be fussy and crying due to the irritation from the acidic vomit, but this is not a cause for serious concern if they can usually be soothed, they are interactive, and they don’t appear to look sick.

If, however, a baby is thought to have reflux and looks sick, is inconsolable, won’t make eye contact or can't be soothed, a more serious underlying disorder should be considered, and medical attention is required. This is absolutely true in cases in which the vomit becomes even slightly green-tinged (see below). And in cases in which the infant has other medical problems – such as a neurologically impaired child or a child with breathing problems – the reflux can lead to problems of aspiration (having the vomit travel down into the lungs), which can cause pneumonia.

The following discussion pertains to reflux as it occurs in otherwise healthy infants that vomit.

 
a) When is reflux in babies worrisome?

   The good news is that most of the time, reflux in infants is a self-limited condition that causes no problems (other than a great deal of laundry). However, sometimes, reflux can be a problem, and so it is important to be able to distinguish simple from pathological reflux, so called gastro-esophageal reflux disease, or GERD.  

The first sign that reflux is a problem may be that the baby isn’t gaining weight. This indicates that he or she simply isn’t keeping enough food down to grow and thrive. A second sign of problematic reflux occurs when the vomit has some blood in it, either bright red or more commonly a rusty color. Bloody vomit typically indicates that the lining of the stomach or lower esophagus has been irritated by the constant vomiting (although an underlying bleeding disorder should also be considered if the bleeding is prolonged, and can be ruled out by taking some simple blood tests). However, if the baby is gaining weight, and if the vomit is never bloody (or green – see below), then the reflux will typically resolve with little or no intervention.

b) How is the diagnosis of reflux in infants established?

   Most of the time, the diagnosis of reflux in infants can be established by taking a careful history and performing a simple physical examination. The key elements of history include spitting up feeds, the absence of any red (bloody) or green (bile) feeds in an otherwise healthy infant. On physical examination, babies with reflux usually appear healthy and well hydrated, and do not exhibit any other worrisome findings as will be described in detail below. In cases where the diagnosis is unclear, an upper gastro-intestinal (upper-GI) series may be performed (as described below), not so much to ascertain the diagnosis of reflux, but to exclude the diagnosis of other conditions that can cause vomiting including malrotation or pyloric stenosis (see further details below). Additional studies including pH testing, endoscopy and impedance testing may also be obtained in selected cases². From a practical standpoint in an otherwise healthy child, the diagnosis of reflux can usually be made on history and physical, with an upper GI to exclude other more worrisome causes.

c) How is reflux in infants treated?

   In most cases, simple measures are successful, including thickening the formula with rice cereal, or switching from one formula to another. Specific feeding measures can help, such as feeding the baby in an upright position, and keeping him or her upright for half an hour or so after feeding. In more severe cases, particularly when the stomach is irritated and the vomit has become bloody, medications may be required³. These medications include different preparations that take away the acid, so that the vomiting isn’t quite so irritating, and the baby isn’t so uncomfortable. In cases in which these measures are not successful, surgery may be required. This typically involves an operation called  “fundoplication,” which involves wrapping the upper part of the stomach around the lower part of the esophagus, thereby creating a valve that prevents excess vomiting. A gastrostomy tube may be placed directly into the stomach at the time of surgery to facilitate feeding. Although this surgery is considered “permanent”, babies will eventually “outgrow” it as they get older, so the operation itself never needs to be reversed. Babies that require such an operation can still eat normally, and can even vomit if they really have to, but the extent of vomiting is significantly decreased⁴. Only a small minority of otherwise normal babies will ever require such an operation – unless the extent of reflux is very severe. In babies with other co-morbidities, such as neurological impairment or lung disease, the need for such operations is significantly greater, as such babies are at significantly greater risk of aspirating. However, for practical reasons, otherwise healthy babies with reflux that are growing and thriving usually respond to non-surgical approaches.

d) How do infants with reflux do overall?

   The good news is that most of the time, otherwise healthy babies out-grow their reflux, usually by approximately six months of age. This corresponds to the time that they are starting to take increasing amounts of solid food, and when they are spending more of their time in an upright position. If medication is required, infants typically will outgrow their need for such medications, often by approximately six to eight months of age.
   Surgery is highly effective for the treatment of reflux in appropriately selected patients. However, cases in which surgery is necessary represent a small subset of patients with reflux, and often have a number of other diseases. As such, the nature and extent of these other conditions (such as prematurity, cerebral palsy, spasticity, muscle problems or other “birth defects”) are more important determinants in predicting the overall outcome of these patients than is the reflux alone.

2. Pyloric stenosis

   Pyloric stenosis refers to a condition in which the outlet of the stomach – also termed the pylorus - grows excessively, such that the stomach does not empty properly. Pyloric stenosis develops acutely in previously healthy children. As a result, babies who develop pyloric stenosis vomit excessively. An illustration of a typical infant with pyloric stenosis is given in the case below.

a) How do infants with pyloric stenosis present? Consider the story of Baby James – a four week old baby boy.

   When James was born, his parents - Nick and Jenna - knew that their life would never be the same. And they were right. James was their first child, and after a surprisingly uneventful pregnancy that went to full term, Nick and Jenna were ready to welcome James into the world. And for the first three weeks of his life, James was a pretty easy addition to the family. He did what most babies do – he took his bottle, he pooped, and he slept. And just when they thought they were starting to figure out the whole “baby-thing,” James started to spit up. At first it was just once in a while, and so Nick and Jenna didn’t worry too much about it. But then it started to get worse. The “spit-up” always looked like his formula, and so James’ pediatrician suggested that a different formula be used, and that the formula be thickened with rice. But none of these measures seemed to help, and James just spat up more and more. Within a few days, James was worse. He was throwing up every feed, which was now actually starting to project out of his mouth. To make things more worrisome, James had started to act differently. He seemed less interactive, and he was having fewer and fewer wet diapers. Even though this was their first child, Nick and Jenna knew something was wrong. They took James to their pediatrician, who was concerned enough that he arranged for James to be seen in the emergency department of their nearest hospital. At the hospital, James was found to be quite dehydrated. An intravenous line was started, and James underwent an ultrasound of his abdomen, that revealed him to have pyloric stenosis. After additional intravenous fluids to correct his dehydration, James underwent an operation to open up the pylorus. He improved quickly, and was sent home from the hospital keeping his formula down, and overall doing great. After a few days of being at home, it was hard to tell that James had ever been sick.

b) Who gets pyloric stenosis?

   The above case illustrates a typical child with pyloric stenosis. Pyloric stenosis occurs usually after a child is several weeks of age, and typically develops in otherwise healthy infants that had been growing and developing appropriately. And although James happened to be bottle fed, pyloric stenosis can occur in babies that are bottle fed, breast fed, or a combination of the two. Pyloric stenosis traditionally has been thought to occur with increased frequency in first born children, and in boys more commonly than girls⁵. In families with a relative that had been diagnosed with pyloric stenosis (a so called positive family history of pyloric stenosis), the diagnosis is significantly more common.

c) How is the diagnosis of pyloric stenosis established?

The diagnosis of pyloric stenosis can generally be established using a combination of careful history, physical examination, x-ray tests, and pyloric ultrasound. These are discussed below.

   i) History of presentation:

   The first step in establishing the diagnosis of pyloric stenosis is to obtain a careful history. In early stages, babies with pyloric stenosis may look quite similar to other babies with reflux. They spit up some or all of their feeds, and are otherwise quite well between feeds. However, unlike babies with reflux who generally remain stable vomiters, babies with pyloric stenosis get worse. The frequency of vomiting increases, so that they begin to vomit after every feed. They also will typically begin to vomit more forcefully, so that the vomit actually projects out of their mouths shortly after a feed – a phenomenon that is called “projectile vomiting.” Babies of course have no control over their vomiting, and will continue to eat with a voracious appetite shortly after vomiting. If left untreated, babies develop a predictable, downhill course. They become dehydrated, as manifest by a decline in the number of wet diapers. They may become lethargic, a sign of dehydration. The vomit itself may become bloody, a manifestation of the stomach irritation. In more advanced cases, they may develop persistent constipation – a sign that very little food is getting through their gastrointestinal tract.  And while healthy babies will typically pass a lot of gas per rectum, babies with pyloric stenosis do not, another sign that the stomach is blocked and little air is getting through. A comparison of pyloric stenosis with gastroesophageal reflux is shown in Table 1.

   ii) Physical examination

   Pyloric stenosis occurs when the muscular valve at the end of the stomach (“the pylorus”) grows too quickly, which causes a blockage (the term used for narrowing or blockage is “stenosis”). In experienced hands, the thickened muscle can be palpated when the baby is examined. The muscle itself feels like a small, smooth round olive in the upper part of the baby’s abdomen. This may be quite difficult to feel, particularly in infants that are quite well nourished with thick abdominal walls⁶. Techniques that can facilitate in the physical detection of the pyloric muscle include administration of a bottle of sugar water (which relaxes the abdominal musculature and soothes a crying child), as well as decompression of the stomach with a nasogastric tube (a plastic tube inserted through the nose). In all cases, a calm, reassuring, and patient approach is necessary.

   iii) X-ray tests.

   In many cases, however, it may be difficult to detect this pyloric “olive” by hand. In these cases, other diagnostic tests are used. Many physicians may obtain a plain x-ray of the abdomen. As shown in Figure 2, the abdominal x-ray in pyloric stenosis will show a large, air filled stomach, along with a relative paucity (but never a complete absence) of gas throughout the rest of the abdomen. In certain centers, an upper-gastro-intestinal contrast study may be obtained to make the diagnosis (a so called “upper-GI study”). For this test, infants are administered an oral contrast agent such as barium, and the course of this agent through the stomach and upper intestines is revealed by obtaining a series of abdominal x-rays.  Whereas in healthy infants the contrast agent will quickly pass from the stomach into the intestine, in babies with pyloric stenosis the contrast material is held up in the stomach for a much longer period of time. And when the material does eventually leave the stomach, the outline of the thickened pylorus is visualized as a narrowed irregular line of material, the so-called “string-sign” as shown in Figure 3.

   iv) Abdominal ultrasound:

   Although the x-ray tests described above may be helpful in establishing the diagnosis of pyloric stenosis, the definitive diagnosis is established through the use of an abdominal ultrasound. This test will show the thickened pylorus in virtually all cases, and offer the advantage of avoiding exposure to the potentially harmful radiation⁷ of x-rays.  The ultrasound is considered positive when the thickness is greater than 4mm, and is considered negative when thickness is less than 4mm (Figure 4). In cases in which the thickness is between 3 and 4 mm, the study is considered either borderline or inconclusive. In these cases, repeat testing may be recommended after 24-48 hours, or an upper GI study may be performed. It is important to point out that these criteria were established in babies between four and six weeks of age. In younger babies, the ultrasound may be considered positive when the thickness is between 3 and 4 mm ⁸. In our experience, a thickness less than 3mm will generally be negative, no matter how young the baby is. 
   There are several important considerations with the use of the ultrasound that should be mentioned. First, the results of the test are quite dependent upon the person who is actually performing the test. As such, it is important to ensure that the test be performed in an experienced center using high quality equipment to minimize the chance of false positive and false negative results. Secondly, care should be taken in making the diagnosis of pyloric stenosis in older infants, those greater than nine to 10 weeks of age. These children rarely have pyloric stenosis, yet may have a pyloric thickness of greater than 4mm simply because they are growing. Such children will more typically have reflux as a cause of their vomiting. In these cases, an upper GI study may again be quite helpful.

e) Treatment of pyloric stenosis.

   The treatment of pyloric stenosis involves correction of dehydration followed by surgery to “open up” the tightened muscle. All infants with pyloric stenosis are relatively dehydrated to some degree. Correction of this dehydration is absolutely critical prior to the performance of any surgery whatsoever. Much as was illustrated for James in the case described above, this will require intravenous fluids. During this time, it is important that infants with pyloric stenosis not be fed, as this will only serve to worsen the vomiting, and make the infant more uncomfortable. 

After the correction of dehydration, the definitive treatment of pyloric stenosis requires surgery to “open up” the pyloric muscle⁹. This operation may be performed either through an incision in the upper part of the baby’s abdomen, or through small little openings around the belly button. The steps in the operation are the same, and include making an incision in the thickened muscle of the pylorus, which is then split to expose its innermost layer (Figure 5). This operation requires a general anesthetic¹⁰.

f) How do infants with pyloric stenosis do after surgery?

   In most cases, infants recover quickly after surgery. Post-operative pain is well tolerated and may be controlled with oral acetaminophen or equivalent.  Infants are typically fed within six hours or so and generally are able to keep their feeds down without throwing up within a day or so of surgery. In cases in which infants have been vomiting for a relatively longer period of time before surgery, the time to achieve full feeds may be more prolonged. It is important to emphasize that some degree of patience in the post-operative management of patients with pyloric stenosis is necessary.  
   In rare cases (less than three percent of all patients), complications from surgery may occur. These can include bleeding at the time of surgery, infection in the incision, an inadequate opening in the pyloric muscle that doesn’t adequately correct the problem, or a cut into the pyloric muscle that is excessively deep, leading to leakage of stomach contents. This latter complication requires additional operative measures be taken to close the excessive opening, and results in a slightly longer post-operative course. Generally though, the outcome after any of these complications (which overall occur in approximately 1% of all patients) in experienced centers is excellent.

3. Other causes of infant vomiting.

a. Malrotation with volvulus.

   No discussion of infant vomiting would be complete without a mention of one of the most serious causes of vomiting in infants: malrotation with volvulus. The term malrotation refers to a defect in the way the intestines are formed, such that they lie on a narrow “stalk” and are inadequately attached within the abdomen. Moreover, intestines in infants with malrotation have congenital bands that cross over the upper parts, which can partially block the passage of feeds. Most worrisomely, infants with malrotation are at severe risk of having the inadequately attached intestines twist – a situation called a volvulus, which is a life threatening event. Infants that develop a volvulus look different from those with reflux or pyloric stenosis. First, the vomit is green – a manifestation of the twisted, blocked intestine. Secondly, the twist occurs relatively quickly, and so babies are generally healthy until they become acutely sick. Third, babies with volvulus are not well. Their abdomens become swollen, they become lethargic, and they are in obvious pain. Pain in babies with volvulus may be manifest by an abnormal cry, which occasionally may be more of a moan than a cry. The diagnosis of malrotation with volvulus can generally be made using an upper-GI test. Importantly, in cases in which a volvulus is considered likely, exploratory surgery may be performed to avoid a delay in diagnosis. The outcome of patients with malrotation and volvulus is determined by the extent of twisting that has occurred, and how healthy the intestines appear after they are untwisted. In cases in which the diagnosis has been delayed, the twisted intestines may lose their blood supply, which leads to their death – an event that can be fatal. It is for this reason that it is so important to be mindful of this diagnosis and to intervene in a timely manner, in particular in cases in which the vomit is even slightly bile-stained. Surgery for intestinal volvulus always requires urgent surgery, which involves untwisting the intestine. In cases in which the intestine has died, parts – or occasionally all – of the intestine will need to be removed. The extent of resection and the severity of the child’s illness prior to surgery determine the outcome. In general, patients operated on fairly quickly and suffering no loss of intestine can do well. The sicker the child preoperatively, and the more intestine that needs to be removed, the worse the outcome.

b. Infections/Sepsis as a cause of vomiting in infants.

   Infants that vomit may be suffering from an underlying infection. Such infections may be related to viral gastroenteritis (“the viral stomach flu”), or the infection may be in some other part of the body such that the vomiting may simply be a response to that infection. In these cases, the vomit may be comprised of feeds, may be bilious, or may be bloody. Such patients will often develop a fever. Worrisomely, infants with infections as a cause of their vomiting tend to become sick quite quickly – both as a result of the underlying infection, as well as from the dehydration that can quickly occur. These babies should therefore always be seen quickly by a pediatrician, who will make the specific diagnosis and at the same time will rule out other causes of vomiting, such as pyloric stenosis or volvulus. Treatment includes correction of dehydration with intravenous fluids, as well as specific therapy if available for the individual specific infection. Potential infections include both viral causes (“stomach flu”) and infectious causes (such as salmonella, “food poisoning”). Most often, these conditions respond to intravenous treatment and a clear liquid diet, and resolve without surgical intervention.

c. Other miscellaneous causes of vomiting in infants.

   For the sake of completeness, there exist other causes of vomiting in infants. These include so-called “metabolic” causes, as well as other surgical causes, such as a defective development of the intestine, or an intestine that is stuck within a hernia. In addition, rarely, infants with cerebral problems – characterized by an increase in the pressure on the brain – can lead to vomiting. In each of these cases – all of which are far rarer than anything discussed above – the infants suffer from a variety of medical problems. These conditions can be diagnosed and treated by experienced caregivers.

4. Summary and conclusions:

   Vomiting in infants is extremely common, and is in many cases a normal part of being a baby. The most common cause of vomiting is gastro-esophageal reflux, which in otherwise healthy babies is typically managed through simple measures such as thickening the feeds, or occasionally the administration of medications. Generally infants with reflux outgrow the condition, and rarely is surgery indicated. In cases in which the vomiting is persistent or becomes forceful, the diagnosis of pyloric stenosis should be considered, which requires surgery for cure. When the vomit becomes bilious, more serious causes need to be excluded, including malrotation with volvulus, infection or other causes of intestinal blockage. In all cases, the vomiting child is at risk of developing dehydration, and so early – and often repeated – visits to the pediatric caregiver are entirely appropriate.

References:

1.         Hackam D, Newman K, Ford H. Pediatric Surgery. 7 ed. New York: McGraw-Hill; 2005.
2.         Vandenplas Y SS, Vieira MC, Hauser B. Will esophageal impedance replace pH monitoring? Pediatrics 2007;119:118-22.
3.         Khoshoo V ED, Thompson A, Rubin M. Are we overprescribing antireflux medications for infants with regurgitation? Pediatrics 2007;120:946-9.
4.         Goldin AB SR, Seidel KD, Flum DR. Do antireflux operations decrease the rate of reflux-related hospitalizations in children? Pediatrics 2006;118:2326-33.
5.         MacMahon B. The continuing enigma of pyloric stenosis of infancy: a review. Epidemiology 2006;17:195-201.
6.         Ozsvath RR P-AM, Leonidas JC, et al. Pyloric volume:  an important factor in the surgeon's ability to palpate the pyloric "olive" in hypertrophic pyloric stenosis. Ped Radiol 1997;27:175-7.
7.         Teele RL SE. Ultrasound in the diagnosis of idiopathic hypertrophic pyloric stenosis. N Engl J Med 1997;296:1149-50.
8.         Leaphart C, Borland K, Kane T, Hackam D. Hypertrophic pyloric stenosis in infants less than 21 days of age: Remodeling the path of surgical intervention. J Ped Surg 2008;in press.
9.         Safford SD PR, Safford KM, et al. A study of 11,003 patients with hypertrophic pyloric stenosis and the association between surgeon and hospital volume and outcomes. J Pediatr Surg 2005;40:967-73.
10.       Hernanz-Schulman M SL, Ambrosino MM, et al. Infantile hypertrophic pyloric stenosis. Radiology 2003;227:319-31.