"Infant Reflux and GERD: Distinctions and Management" sponsored by NASPGHAN, CDHNF, and the AAP has just been released. This 10-minute video, available on the NASPGHAN Web site, is an informative and admirably low-key discussion of the symptoms and outcome of infants who reflux. The video is excellent and worth recommending to parents. One issue raised, but not solved, by the video is the distinction between infant reflux and GERD. The video says that infant reflux "crosses the line" and becomes GERD if the baby has "serious" symptoms such as food refusal, failure to thrive, emesis of blood, breathing problems or dysphagia. Where did this concept come from, and is it legitimate?
A recent study from our center (JPGN 2006;42:16-21) measured simultaneous esophageal pH and impedance in 34 babies with a clinical diagnosis of reflux bad enough to see a gastroenterologist. These infants all fit the video's criteria for GERD. Most were fussy babies with feeding problems, failure to thrive and pain behavior. Although not the primary objective of the study, only 4 of the infants had pathologic reflux by pH probe analysis. Although normal standards for impedance are not established, none of the babies appeared abnormal by this test either.
As illustrated by these patients, using symptom severity as the diagnostic criterion resulted in a huge overdiagnosis of GERD. Should we really use severe symptoms to diagnose infant GERD as suggested by the video? Or should we use the amount of reflux demonstrated by pH probe? Distinguishing simple infant reflux from GERD by symptoms is an untested and unvalidated idea. If GERD is a real disease and infant reflux is not a disease, then there should be an objective way to distinguish them so that the disease can be treated and the nondisease left alone.
The problem with diagnosing reflux disease at any age is that everybody refluxes. After the introduction of esophageal pH monitoring, it was generally agreed that pathologic reflux or reflux disease was present when esophageal acid exposure exceeded an upper limit, usually between 5% and 10% of a 24-hour period. Composite reflux scores were generated based on various combinations of episode frequency and duration.
Unfortunately, pH probe studies are not the sine qua non for the diagnosis of GERD. Neither the composite score nor any of its components vary on a continuum with the severity of most symptoms. No doubt, people with peptic esophagitis have abnormal pH scores, but the patient with grade IV esophagitis does not have a more abnormal test result than the patient with grade II. The variability of the test from day to day is high. Thus, the difference between 15% reflux time and 10% reflux time does not translate into a 30% difference in disease severity. Repeated studies cannot be used to monitor medical therapy. Even if patients agree to repeated pH probe analyses, effective acid blockade prevents detection of reflux episodes. Further complicating the use of pH probes to monitor therapy are data from simultaneous impedance and pH probe studies showing that acid blocker therapy does not change the total number of refluxes; it just reduces the number of acid events while increasing the nonacid. (Gastroenterology 2001;129:1599-1606)
Truth be told, diagnosis and treatment of GERD have never been based solely on quantification by pH probe. No one recommends a fundoplication for a thriving colicky infant with an abnormal pH probe. Some, however, might recommend a fundoplication in a handicapped child with repeated aspiration pneumonias who has a normal study. I don't feel confident that reflux is the cause of a baby's crying when the pH probe reveals a symptom correlation index of 50%. What's the other 50% caused by? Treatment decisions in all of these situations are driven by clinical judgment and symptom severity, not just quantitative test results.
So, what about using symptoms to diagnose infant GERD as suggested by the video? Symptom severity often does not correlate with pH probe results. Severity is a subjective assessment defined in great part by parent and physician concern. One person's reflux-induced food refusal is another person's variable appetite. One person's reflux-induced apnea is another's appropriate airway protection during vomiting. One person's reflux-induced colic is another's fussy baby. And so on.
It is important to distinguish diagnostic testing from mechanism of disease. Mechanism refers to the abnormal process producing disease. So far, none of the "tests" for reflux have confirmed a mechanism. pH monitoring has revealed some unique characteristics of patients with abnormal tests that suggest possible mechanisms. Reflux during sleep is a major factor causing the abnormal total duration of acid exposure. Prolonged reflux episodes, also usually occurring at night, are common in abnormal records. These 2 findings suggest but do not prove that the mechanism of GERD might be abnormal acid clearance resulting from decreased swallowing frequency, poor buffer production, or decreased arousal from sleep. The concept of reflux as an esophageal motility disease has not been validated. Esophageal motor abnormalities in reflux are nonspecific, inconsistent from patient to patient, and almost always secondary to acid exposure, not the primary abnormality. Motility agents do not cure GERD or uniformly control symptoms, especially in infants (J Pediatr 1988;112:483-487). There was some hope that excess numbers of spontaneous LES relaxations would be found in people with GERD. Unfortunately we have not discovered this association either. The lack of a mechanism for GERD is the primary cause of our inability to make it better.
Here is the heretical part. Is it possible that the amount of acid reflux measured by an infant's pH probe is not the disease-producing mechanism of GERD? Could it be that the signs and symptoms are the result of abnormal host sensory integration and response? Alternatively, could they be a result of abnormal esophageal repair or regenerative mechanisms? Abnormalities of either of these processes could be the reason that infant reflux turns into GERD in a particular infant with or without a pathologic level of acid exposure.
The final problem with the symptom-based diagnosis of GERD is that, with the possible exception of esophagitis, none of the symptoms relate to outcome. Spontaneous recovery occurs in most infants with reflux regardless of symptom severity. As the infant's need for large-volume, frequent feedings diminishes, he spits less. As he develops more effective reactions to emesis, he is less likely to cough or turn blue around the mouth. As his self-settling skills develop, he spends less time fretting.
We are all encountering more parents who have been convinced by internet testimonials, drug company infomercials, and physicians that their babies have a disease. They expect a cure. Defining infant GERD as "reflux with bad symptoms" perpetuates a shaky disease concept and is likely to increase parental anxiety, increase the demand for acid blocker therapy, and possibly increase the demand for surgery in healthy infants. Unfortunately, the quantitative measure of reflux provided by pH probe is not a satisfactory alternative and also does not address the need for identifying the mechanism of reflux disease.
I don't doubt the existence of GERD. I would only suggest caution in the quick clinical diagnosis of GERD in infancy. If there is such a disease, we don't know the mechanism, and we certainly don't know the cure. We should stay open to the possibility that infant reflux and GERD are similar entities with differing severity of symptoms, depending on the responsiveness of the infant. We should stop promising a medical cure until there is one.
Judith M. Sondheimer, MD
University of Colorado Health Sciences Center and the
Children's Hospital, Denver, CO