Gastrointestinal (GI) problems are common and the decision of when to refer one is not always easy. On the one hand, problems like chronic diarrhea, constipation, vomiting, and abdominal discomfort are seen in our offices everyday, and we can handle most of these most of the time. Many of these problems are also self-limited, idiopathic, or have psychosocial components and are not life-threatening. On the other hand, psychosocial problems can be the most difficult to manage. Also, the pediatrician does not do endoscopie examinations, and these are the route to definitive diagnoses for many gastrointestinal disorders. Also, of course, the subspecialists1 depth of knowledge and experience can provide answers that elude us. Incidentally, we are talking about referral to a pediatrie gastroenterologist, not an adult gastroenterologist. HMOs often refer to the latter, and this does not work very well unless the child has cancer, which is almost never.
There are a few principals that apply to the decision of when to refer. When confronted with gastrointestinal signs or symptoms, ask yourself the following questions:
* Is failure to thrive present? If the answer is yes, the odds of needing help go up considerably. Think about the purpose of the gastrointestinal system. It provides nutrition. So growth failure implies a significant impairment of function, and your chances of needing a consultation for the patient with chronic diarrhea, abdominal pain, signs of malabsorption, or vomiting, in the presence of growth failure, are much greater than they are if the patient is thriving. The exception is the infant with non-organic failure to thrive and a history of mild loose stools or spitting. This association is not uncommon and resolution of the feeding problem (such as by a trial of feeding in the hospital) will usually resolve diarrhea, vomiting, and growth failure, confirming the non-organic etiology.
* Patients with either significant or persistent GI hemorrhage - hematemesis (vomiting blood, indicating a bleed proximal to the jejunum), melena (black, tarry stools from bleeding above the distal ileum), or hematochezia (red blood in stools from a distal bleed or massive bleed anywhere) will almost always need a referral. There are a few exceptions, however. Guaiacpositive or some gross blood mixed with stool is common in acute gastroenteritis, so wait to see if bleeding disappears as the diarrhea resolves, unless the bleeding is severe or prolonged, before referral. Look for rectal fissures by gentiy (do not produce these) spreading the folds between your gloved thumbs to look underneath, going around the clock of the anus. Bleeding from fissures is mild and primarily streaked on the stool.
Otherwise, you will probably need help at some point. You can tell whether the bleeding is from the duodenum or above by passing a nasogastric tube to see if blood is in the stomach. You can do a scan for a Meckel's diverticulum, or look for lesions that might be bleeding by upper or lower tract radiologie procedures, and you can do serology for Helicobacter pylori infection. However, no matter what you find, a more definitive diagnosis is generally available by upper or lower endoscopy and/or you will need subspecialty help for management. Besides, remember that GI hemorrhage can be life-threatening and deceptive because you cannot see it while it is happening.
* How long has the patient had signs or symptoms of gastrointestinal problems? If this has gone on for months, especially if you and other physicians have not been able to solve the problem, your threshold for referral should be low.
The indications to refer will change as technology and knowledge expand what we can do without endoscopie examinations. For example, ultrasonographic examination for (bleeding) juvenile colonie polyps after distending the colon with a saline enema is a promising alternative to colonoscopy for diagnosis.1 However, you will still need endoscopy if the polyp must be removed; but, science is moving rapidly and many disorders that require referral today will be managed by the general pediatrician in the future.
A good way to approach individual patients with gastrointestinal problems is to ask these three questions in order:
* Is the sign or symptom a variation of normal?
* If not, is the problem self-limited and of no risk to the patient?
* If not, do I need a referral now or should I try to manage the problem first?
The following two common problems illustrate this process. Of course, all evaluations begin with a history and physical examination with further investigation of any clues you uncover.
SPITTING OR VOMITING IN AN INFANT
Spitting or effortless régurgitation is common and usually normal; vomiting is more likely to be abnormal. Spitting, alias reflux, usually is a tablespoon or less in volume, happens most often during the 40 minutes after feeding, and consists of whatever the child has ingested (no blood or bile). It is troubling to families, but if the child is growing and developing normally, the history and physical examination are unremarkable (no complication of reflux - see below), and spitting only occurs a few times after each feed, it should be considered a normal variant and managed with reassurance. If spitting is excessive or the child vomits an ounce or more, move on to the most common causes of this by asking yourself three questions: Is the baby being overfed? Is the feeder very anxious? Is the vomiting due to inadequate burping? Overfeeding is a very common cause of vomiting and is diagnosed by a history of feeding volumes and the growth chart. Assume overfeeding is the diagnosis when a vomiting or spitting baby is upwardly crossing percentile curves. The treatment is concrete instructions that limit the amount of formula: "Your baby has a 4-oz stomach, so if you put 5 oz in there, you are going to get 1 oz back." The association between an anxious mother and a vomiting baby is also common. Many of these parents are hyperactive, bounce the baby excessively, or pat too vigorously for burping. However, exactly why the vomiting occurs is not clear. The treatment for an anxious mom is calm, relaxed feeding and reassurance: ask her to get someone else to handle other children or noxious distractions, turn the lights down, get in the most comfortable chair, turn on some quiet music (no acid rock), and make feeding last at least 15 minutes with lots of "love talk." Burp after every ounce, even if the baby screams when the bottle is pulled away. If a burp is heard during vomiting, the diagnosis is incomplete burping. My mother showed me a trick for "difficult burners" that is easy to demonstrate in the office. Sit the baby on your knee with one hand on the back, the other on the abdomen, and gently (no retinal hemorrhages) rock the baby side to side with your ear close to the stomach. If you hear a bubble gurgling in the stomach, burping is incomplete. This maneuver also encourages the air to come out.
When should you do a pH probe or an upper gastrointestinal radiologie evaluation in a spitting or vomiting infant? A good deal of money is wasted doing these studies to diagnose reflux in a baby who is spitting. Spitting equals reflux so if you see the stuff coming out of the mouth, you know reflux is present. Reflux is common in young babies and almost always disappears as they develop. So before you go further in the evaluation, ask yourself if the reflux is causing a problem (other than the mess it makes). Does the baby have failure to thrive, chronic wheezing, or pneumonia that could come from aspiration; are apnea or acute life- threatening events occurring (especially while awake); are there signs of esophagitis such as irritability after reflux; or are anemia or guaiac-positive stools present? Although unproven, some believe chronic otitis media or sinusitis may also be a complication of reflux. If a baby with reflux has one of these problems, the reason to do a pH probe is to determine whether the apnea, wheezing, screaming spells, and so on happen at the same time esophageal pH drops below 4. However, identifying this temporal relationship is notoriously difficult, making the pH probe of limited value in infants with apnea. In contrast, the pH probe can help diagnose and characterize reflux in an older child with reflux-type problems. The main reason to do an upper gastrointestinal evaluation in a refluxing infant is to rule out partial obstruction by following the barium through the upper tract. Hillemeier recently published a review of reflux management for all ages.2
If you do not have an answer after considering these questions, you will probably need to refer the patient. Remember, other things cause persistent vomiting. Do not forget that metabolic disorders, urinary tract disease, milk, soy, or other food allergy, and others can present this way.
It is normal for an infant to grunt, cry, or turn red when having a stool so these are not necessarily a sign of constipation, even though parents interpret them as such. Instead, this diagnosis rests on the hard consistency of stools. In infants, this translates into small balls of difficult to crush stool and is usually due to inadequate fluid intake. This can be treated by adding water to each day's feedings. Feeding whole milk that is unheated is also a common cause of hard stools in infants. Heating or using evaporated whole milk softens the casein and the stools. Look for rectal fissures, especially if stooling seems painful. Constipation leads to rectal fissures and rectal fissures may encourage a child to retain feces. Rectal fissures are treated by stopping rectal temperatures or the insertion of anything into the rectum (for example, an enema or a finger to break up stool), softening the stools and applying Vaseline around the anus after gentle sitz bath-type cleaning. Constipation in older children is usually responsive to increasing fiber in the diet. In this issue, Gremse and Sacks have an excellent discussion of irritable bowel syndrome, a relatively benign cause of constipation (and other problems) in older children, and Lewis and Rudolph describe fecal retention and the management of encopresis. Finally, be on the lookout for parents who are obsessed with their own and their child's stool functions. They may be treating constipation that is not there.
This editorial's title is not about stooling. Rather, it refers to the decision that the "gatekeeper" primary care pediatrician and the quarterback share^ - do it myself or refer to another?
1. Nagita A. Amemoto K, Yoden A, Yamazaki T, Mino M, Miyoshi H. Uhrasonographic diagnosis of juvenile colonie polyps. J Pediatr. 1994;24:535-540.
2. Hillerncier AC. Gastroesophageal reflux: diagnostic and therapeutic approaches. Pediatr Clin North Am. 1996; 43: 197-209.