Is the role of proximal acid reflux in respiratory symptoms prevalent in infants?
S.G. Jolley (Las Vegas), D.G. Johnson (Salt Lake City)
In infants and children with gastroesophageal reflux disease (GERD), the relationship of any respiratory symptoms to the GERD is often difficult to ascertain. The respiratory complaints associated with GERD in children include apnea, choking, recurrent pneumonia, wheezing, stridor, cyanosis, and chronic coughing. Acute respiratory events have been grouped together and called apparent life-threatening events (ALTE). Respiratory symptoms may be totally related, partially related, or totally unrelated to the child's GERD . Extended (18-24 hours) esophageal pH monitoring of the distal esophagus has been an important tool in sorting out the relationship of respiratory symptoms to GERD in children. Since it has been presumed that aspiration of gastric contents is the major mechanism in reflux-related respiratory symptoms, attention has been focused on the monitoring of proximal esophageal pH in infants and children [2, 3].
In 1981, we reported a group of infants with and without major respiratory symptoms who had GERD documented by extended esophageal pH monitoring . These infants were part of a larger cohort of 66 infants and children (ages two weeks to four years) with documented GERD who were followed prospectively and analyzed retrospectively to determine the relationship between respiratory symptoms and GERD, and to determine the presence of any findings on the esophageal pH recording which may distinguish reflux-related from reflux-unrelated respiratory symptoms. All patients had extended monitoring of distal esophageal pH (middle left atrium level = T7-8 level), but 23 patients also had simultaneous monitoring of proximal esophageal pH at the T3 level. Glass pH electrodes (Microelectrodes Inc., Londonderry, NH, No. 502 ) requiring an external reference electrode (Beckman Instruments Inc., Fullerton, CA, No. 40249) were used to monitor esophageal pH by a method described previously . Two sets of electrodes and pH meters were used in patients with simultaneous monitoring of both the proximal and distal esophagus (Figure 1). The position of the pH electrodes was verified and adjusted using a chest radiograph. The children were separated into groups with (n = 34) and without
(n = 32) major respiratory symptoms. As described previously, 4 patients with respiratory symptoms were further categorized as having reflux-related (n = 19), reflux-unrelated
(n = 8), or indeterminate (n = 7) respiratory symptoms based on follow-up extended esophageal pH monitoring and the response of the respiratory symptoms to the treatment directed exclusively at the GERD.
Figure 1. A diagram and radiograph depicting the upper (T3 level) and lower (T7-8 level) pH probe position for simultaneous monitoring of esophageal pH in children.
A role for monitoring proximal esophageal pH was examined in this study by looking at four parameters during the first two hours postcibal and after the first two hours postcibal, and according to whether patients were awake, sleeping, upright (seated), or supine. The four parameters examined in the proximal esophagus were the frequency of reflux episodes (episodes/12 hours), the mean duration of acid (pH < 4) reflux episodes, the percentage of time that the esophageal pH was less than 4, and the percentage of acid reflux episodes detected in the distal esophagus which were also detected in the proximal esophagus. Proximal esophageal pH records were available in six patients with reflux-related, five patients with reflux-unrelated, and two patients with indeterminate respiratory symptoms.
Results are expressed as the mean ± standard error of the mean. Linear correlation was used to compare the relationship of proximal to distal esophageal pH parameters. Comparison of groups was performed using the one-way analysis of variance.
This study's important findings, which were reported previously , came from the monitoring of distal esophageal pH and can be summarized as:
1) children with respiratory symptoms have a longer mean duration of reflux episodes during sleep (ZMD) than children without respiratory symptoms;
2) children with reflux-related respiratory symptoms have a prolonged ZMD (> 7.5 minutes in the first two hours postcibal, > 3.8 minutes after the first two hours postcibal), whereas children with reflux-unrelated respiratory symptoms have a normal ZMD (Figure 2);
Figure 2. The mean duration of sleep reflux episodes after the first two hours postcibal (ZMD) in the distal esophagus of children with reflux-related, reflux-unrelated, and indeterminate respiratory symptoms. Note the prolonged ZMD in children with reflux-related respiratory symptoms. Children with GERD and no respiratory symptoms, and asymptomatic controls (normal range) are included for comparison (from ).
3) all children with acid reflux episodes immediately preceding acute respiratory events had a prolonged ZMD and reflux-related respiratory symptoms;
4) acute respiratory events from GERD are precipitated by the prolonged clearance of a single reflux episode (Figure 3) or the clustering of more than one reflux episode during sleep (Figure 4);
5) aspiration of acidic gastric contents does occur in at least some patients with respiratory symptoms caused by GERD (Figure 4);
6) only 37% of children with reflux-related respiratory symptoms have an acute respiratory event immediately following an acid reflux episode during 18-24 hours of esophageal pH monitoring.
Figure 3. Simultaneous monitoring of proximal and distal esophageal pH in a
4 month old infant with choking episodes and cyanosis. This child developed a choking episode and cyanosis during the prolonged clearance of a single acid reflux episode (pH < 4) during sleep and transmitted to the level of the proximal esophagus. There were no other choking episodes during the extended esophageal pH study in this patient (from ).
The data from monitoring of proximal esophageal pH is shown in Table I. With the exception of the mean duration of acid reflux episodes during sleep, the measured parameters in the proximal esophagus were usually lower in the reflux patients with respiratory symptoms as compared to the reflux patients without respiratory symptoms. The mean duration of acid reflux episodes during sleep was consistently longer in reflux patients with respiratory symptoms. Furthermore, when the ZMD in the distal esophagus was examined, the percentage of distal esophageal acid reflux episodes reaching the proximal esophageal pH electrode during sleep was substantially higher in patients with a prolonged ZMD than in patients with a normal ZMD, whether in the first two hours postcibal (70.6 ± 8.5% v. 9.9 ± 5.3%, p = .0001) or after the first two hours postcibal (56.5 ± 10.1 % v. 13.0 ± 6.6%, p = .001). In fact, a significant linear correlation existed for the mean duration of acid reflux episodes during sleep between the distal and proximal esophagus both for the first two hours postcibal (y = 1.27 + 0.62x, r = 0.8, p < .001) and after the first two hours postcibal (y = -0.01 + 0.53x, r = 0.76, p < .001). Similar to findings in the distal esophagus, the mean duration of acid reflux episodes during sleep in the proximal esophagus, after the first two hours postcibal, was longer in children with reflux-related respiratory symptoms (6.7 ± 2.4 minutes) than in children with reflux-unrelated respiratory symptoms (0.4 ± 0.4 minutes, p = .01 9).
Figure 4. Simultaneous recording of proximal and distal esophageal pH in a 6 month old child with a tracheostomy and apneic episodes despite the tracheostomy. There were no episodes of emesis from this child. During the extended esophageal pH study, this patient clustered two episodes of acid (pH < 4) reflux during sleep and immediately prior to a major apneic episode with cyanosis. Acidic material was suctioned from the tracheostomy during the apneic episode and compared to the normal pH of 6-7 for this patient's tracheostomy secretions. Note the transmission of acid reflux to the proximal esophagus in this child prior to the apneic episode. The apnea persisted for several minutes even after the tracheostomy was cleared of secretions (from ).
These findings indicated to us that most proximal esophageal pH parameters were predictable from pH monitoring in the distal esophagus. The detection of proximal esophageal pH was helpful in the documentation that acid reflux episodes reach the proximal esophagus in children who have acute reflux-related respiratory distress. It may also be helpful in patients whose rates of esophageal clearance vary dramatically between the proximal and distal esophagus (i.e. very prolonged proximal esophageal clearance compared to the distal esophagus), since the ZMD in these patients may be more prolonged in the proximal esophagus than expected from distal esophageal pH measurements. Examples of such patients are those with repaired esophageal atresia or central nervous system disorders . Because our normal controls were not studied with a pH electrode in the proximal esophagus , we could not determine the normal range for the ZMD in the proximal esophagus, and, therefore, determine if some patients with reflux-related respiratory symptoms could have a normal ZMD in the proximal esophagus and a prolonged ZMD in the distal esophagus. The accuracy of the ZMD, in the distal esophagus, for identifying correctly children with reflux-related and reflux-unrelated respiratory symptoms is 97% , and has been confirmed by other investigators .
The data from our study suggests that major reflux-related respiratory symptoms are an intermittent "low frequency" phenomenon. Episodes of respiratory distress in most children do not occur after every, or even most, episode of acid reflux. To do so probably would not be compatible with survival. Instead, this "low frequency" of acid reflux episodes precipitating respiratory distress in children becomes catastrophic and, seemingly, catches the child's protective mechanisms for eliminating refluxed acid "unprepared". Untreated children with a prolonged ZMD are at significant risk for the subsequent occurrence of sudden death from the GERD . Of interest is the observation that the prevalence of a prolonged ZMD increases with postconceptual age and decreases after six months of age in children with apnea or choking [7, 10]. The peak prevalence for a prolonged ZMD in children with apnea or choking is from term birth to six months of age. In children with recurrent pneumonia, wheezing or chronic coughing, the peak prevalence of a prolonged ZMD is from term birth to one year of age. The reasons behind the age-related prevalence of a prolonged ZMD in children is not understood fully.
In summary, the data presented here and by others  does not indicate that monitoring of proximal esophageal pH is a practical method for screening children with major respiratory symptoms from GERD. With few exceptions, monitoring of distal esophageal pH alone, with a determination of the ZMD, can provide the most useful information for determining the relationship of respiratory symptoms to GERD in children.
7. Halpern LM, Jolley SG, Tunell WP, et al. The mean duration of gastroesophageal reflux during sleep as an indicator of respiratory symptoms from gastroesophageal reflux in children. J Pediatr Surg 1991;26:686-690.