When esophageal pH monitoring was introduced to clinical practice more than 30 years ago, we assumed that the information obtained from these records (essentially, variations on how often and how long the esophagus is acidified) would accurately define reflux disease and clarify the relationship between acid reflux and clinical symptoms. During the years, experience with esophageal pH records has added a great deal to what we know about reflux disease, but many questions remain. The recent introduction of esophageal impedance monitoring, a pH-independent method, to study the movement of fluids in the esophagus brings with it a new set of promises and a new set of problems.
Until now, we have assumed that retrograde flow of acid material from the stomach to the esophagus was the basic pathologic event of reflux disease. To be more specific, we assumed there was a point at which the frequency and duration of acid reflux events in the esophagus changed from a normal function to a pathologic process, and that the symptoms of the pathologic condition related in some way to the increased frequency and duration of acid reflux events. However, the situation has never been this straightforward. Trying to tie symptoms other than spitting and vomiting to pH-probe–detected reflux episodes has been particularly problematic. For example, in babies with spells of choking or colicky crying, a close association between pH-probe–detected acid reflux and these symptoms is not routinely found. Some spells coincide with reflux episodes, but many do not. Is this because the colic or choking is not related to reflux? Or is it because the pH probe does not measure the event that is truly precipitating the abnormal behaviors. Similar questions can be raised when looking at pH-probe data on the relationship between acid reflux episodes and apnea/bradycardia spells of premature infants or between wheezing, cough, dental erosion, sleep disturbance, and all the other myriad of symptoms attributed to reflux.
With the use of the impedance monitor, which measures fluid movement rather than luminal pH changes, perhaps we will learn more about what is really going on in the esophagus. The article by Wenzl et al. (1) in this issue begins this learning process. One drawback of the study is that the patients evaluated are not “normal” infants and children. Thus, the data obtained should not be generalized to all children because they may not represent normal behavior. However, in looking only at the pH-probe records from these patients, one finds a surprisingly low overall frequency of acid reflux episodes (only 270 episodes in 318 hours, or 0.8/hour). One wonders whether these children may indeed be “normal.” Regrettably, no information was obtained in the late postprandial period, during which the collected experience from conventional pH-probe studies suggests that the real pathologic events of acid reflux occur. Despite these drawbacks, the study indicates that the pH probe does not simultaneously detect the majority of reflux events defined by impedance monitoring, presumably because the fluid boluses are not acid. Indeed, the pH probe detected only 282 of the 1,887 (15%) episodes of fluid movement. The overall frequency of reflux detected by impedance monitoring was 1,887 episodes in 318 hours (5.9/hour), which seems very high. Another surprising finding was that only 54% of the pH-detected episodes seemed to be associated with retrograde fluid movement of sufficient volume to be detected by the impedence monitor. These findings beg for better explanation. Do we conclude that fluid boluses move up the esophagus much more often that we have previously believed? Are these frequent neutral pH boluses pathologic? If you aspirate a neutral reflux, is that bad for you? Do episodes of nonacid reflux occur during every transient lower esophageal sphincter relaxation that is not associated with acid reflux? Why do sudden drops in pH occur in the absence of detectable fluid boluses from the stomach? What might one see with this technique in a sleeping child with an empty stomach? The questions are many, and the answers may change the way we define gastroesophageal reflux disease.
Only a few studies have evaluated esophageal impedance monitoring in adults, and these studies also raise more questions than they answer. Vela et al. (2) used simultaneous pH and impedance monitoring to study postprandial reflux in 12 adult patients with reflux. They studied the subjects before and after omeprazole therapy. They found what many people with reflux know, namely, that omeprazole therapy reduced the number of reflux episodes of acid pH but did not reduce the total number of reflux episodes detected by impedance. In other words, the reflux still occurs, but it just doesn't hurt anymore. Sifram et al. (3) performed 24-hour simultaneous impedance and pH monitoring in patients with esophagitis and in controls. They found that the rate of reflux detected by impedance was equal in the two groups but that the proportion of acid reflux events (those detected simultaneously by pH and impedance) was greater in patients with esophagitis than in controls (45% versus 33%). These studies suggest that omeprazole controls reflux symptoms but does not modify the primary pathologic mechanism of reflux disease (whatever that may be). The second study strongly suggests caution in tossing out all we have learned about reflux from pH probes, because perhaps only acid reflux episodes are worth counting.
This technology is still a long way from routine clinical application. The equipment is fragile and expensive. The records must be interpreted by visual inspection and, therefore, are open to bias. Movement artifact cause major problems. There are inexplicable results, such as those of Wenzl et al., in that many pH-probe–detected episodes were not simultaneously seen by the impedance sensors. Is this because of a volume limit on impedance-detected boluses? Or does this phenomenon represent tiny changes that occur at a pH near 4.0 that are not associated with fluid movement and that artifactually increase the number of reflux episodes on pH recordings?
These concerns notwithstanding, the prospect of a better way or even an additional way to measure reflux other than the pH probe alone is exciting. We have always known that the pH probe gives an incomplete picture of the events in the esophagus. Now we have the potential to find out just how incomplete that picture really is and to see whether fluid boluses from the stomach at a pH greater than 4 are part of the pathology we call gastroesophageal reflux disease. As investigations with this new technology continue, the pediatric gastroenterology world must not fall into the traps we fell into many years ago as we investigated reflux with the pH probe. We must do the following: select carefully defined study populations so that conclusions will have validity; cooperate early in the research planning process so that criteria for evaluating these records are uniform from center to center; always have a masked observer evaluate the raw records to prevent unavoidable bias in looking at these records (better yet, have two masked observers evaluate the raw records so that there is a clear indication of how subjective the evaluations are likely to be); and perform these critical initial studies in the setting of an independently funded clinical research center so that the financial interests of companies making money on drug sales do not control data. Studies using esophageal impedance may significantly change the definition of what is normal and what is abnormal for one of the most common conditions we treat. I hope the new information will be reliable.
1. Wenzl TC, Moroder C, Trachterna M, et al. Esophageal pH monitoring and impedance measurement: a comparison of two diagnostic tests for gastroesophageal reflux. J Ped Gastroenterol Nutr
2. Vela MF, Camacho-Lobato L, Srinivasan R, et al. Simultaneous intraesophageal impedance and pH measurement of acid and nonacid gastroesophageal reflux: effect of omeprazole. Gastroenterology 2001; 120:1599–1606.
3. Sifram D, Holloway R, Silny J, et al. Acid, nonacid, and gas reflux in patients with gastroesophageal reflux disease during ambulatory 24-hour pH impedance recordings. Gastroenterology 2001; 120:1588–98.