Gastroesophageal reflux (GER), defined as the passage of gastric contents into the esophagus, occurs on a daily basis as a normal process in infants, children, and adults. Most episodes of physiologic reflux are transient, asymptomatic, and reach only the distal esophagus. Gastroesophageal reflux disease (GERD) is distinguished by reflux into the esophagus resulting in well-defined symptoms or medical problems (see Chapter 12). When children present with atypical complaints or extraesophageal symptoms, testing may be necessary to document the presence or absence of pathologic reflux, or the association between reflux events and specific symptoms.1

While endoscopy can be helpful in documenting acid damage to the esophageal mucosa in the form of erosions or ulcers, the majority of patients with symptoms of GERD do not have endoscopic or pathologic evidence of esophagitis.2 Tests designed to detect the presence of GER have been developed.

The first test utilized was esophageal pH monitoring, in which an electrode designed to detect changes in pH is used to assess the frequency and duration of acidic reflux present in the distal esophagus. Over the years the advantages, disadvantages, and limitations of traditional, catheter-based esophageal pH monitoring have become better defined, with a subsequent evolution of newer diagnostic techniques. Wireless methods to detect acidic contents in the esophagus have now become available (Bravo capsule). Additionally, we have seen the development of the technical possibility of measuring both acidic and non-acidic reflux with multichannel intraluminal impedance (MII). In the present chapter we will review the current techniques that are being used for the dynamic detection of reflux episodes.

Esophageal pH monitoring, which quantifies the frequency and duration of acidic reflux episodes, can be used to confirm abnormal esophageal acid exposure and/or correlate symptoms with acidic reflux episodes. Testing may be especially helpful in patients who present with atypical symptoms, patients with persistent symptoms despite pharmacologic treatment, or patients without evidence of mucosal damage on endoscopy. Other diagnostic approaches, such as barium contrast upper GI series or nuclear scintigraphy, have had variable sensitivity and specificity when compared to esophageal pH monitoring.1 Unacceptably high false-positive and false-negative rates seen with these radiographic tests have made esophageal pH monitoring preferable over barium contrast and scintigraphy studies for the diagnosis of reflux.

Equipment

Over the years, the methodology of esophageal pH monitoring in children has become relatively standardized as evidenced by the publication of professional practice guidelines.1,3 In addition to a catheter-based pH sensor, a portable data logger that records intraesophageal pH as well as events during the study such as symptoms, meals, position changes, and activity is required. As technology has improved and as electronic devices have become smaller, pH monitoring is now conducted on an ambulatory basis, even for pediatric patients.

Electrode Placement

The catheter-based pH electrode is placed through the nose into the distal esophagus.3 There are several types of pH electrodes available, including glass and antimony-based electrodes. Proper placement of the pH electrode relative to the lower esophageal sphincter (LES) is very important for accurate data. The closer the electrode is to the esophagogastric junction, the greater the acid exposure that is recorded. Similarly, as pH electrodes are placed at higher and higher locations in the esophagus above the LES, there is a linear decrease in acid exposure time, which ultimately decreases the sensitivity of the test. In adult studies, the pH electrode is typically inserted through the nose and positioned 5 cm above the superior margin of the LES.2 Ideally, localization of the LES is best achieved with stationary esophageal manometry prior to placement of the catheter. Because of the two required nasal intubations, this additional procedure is difficult in children, however, so manometry has not become standard procedure before pH-probe placement in children. As a result, there are other methods for localization of the pH electrode in children, including calculation of the esophageal length according to Strobel’s formula4 and fluoroscopy. If using Strobel’s formula [length from nares to LES (cm) = 5 + 0.252(height)], the tip should be placed at 87% of the distance between the nares and the LES (the rationale being that 5 cm is 13% of the standard adult esophageal length). Because of concerns about the accuracy of applying Strobel’s formula to children over 1 m in height, the Working Group of the European Society of Pediatric Gastroenterology and Nutrition recommended use of an X-ray or fluoroscopy to confirm placement of the pH electrode so that the sensor or tip lies over the third vertebral body above the diaphragm throughout the respiratory cycle.3 In general, in the absence of manometry, radiographic verification of catheter placement is necessary and adjustments can be made to the catheter location accordingly.

Recording Conditions

From a clinical standpoint, in order to best study GER, recording conditions should follow the patient’s normal daily routine as much as possible in terms of activity and diet. On the other hand, the monitoring protocol should be somewhat standardized to reduce variability and the effect of confounding factors, especially if the study is also being conducted for research purposes.

The optimal duration of monitoring should be at least 18 hours, including a day and a night period.3 Depending on the aim of the study, H2-blockers and proton-pump inhibitors should be stopped at least 3 or 7 days prior to the study, respectively.2 Performing a pH monitoring study while “on-medications” may be indicated if the investigator desires to evaluate the efficacy of acid-blocking medications in patients who have persistent symptoms despite maximal medical therapy. Depending on the clinical question in mind, prokinetics may also be stopped at least 48 hours before pH monitoring,5 although they do not interfere with reflux detection.

Instructions for feeding during the study should again represent a balance between maintaining a degree of standardization and recreating normal circumstances with minimal restrictions. Although a strict standardized diet is generally not necessary, acidic drinks and foods should be minimized and documented in the diary. Investigators may consider excluding meal times from analysis, since it is difficult to distinguish episodes of GER from pH changes secondary to swallowing acidic foods during periods of eating, one of the major limitations of pH monitoring. The study period should include at least three discrete meals with minimal snacking in between meals. Given the buffering effect of formula or food, an interval of 3–4 hours in between meals is recommended. Very hot and cold beverages should be avoided, since extreme fluctuations in temperature can affect the sensitivity of the pH electrode. Chewing gum or hard candy should also be avoided because these activities increase saliva production and therefore induce additional swallowing and peristalsis that can affect the detection of GER. Documentation of patient position and activity (e.g., sleeping) during the study should also be recorded since the effect of body position on different patterns of GER has been well-reported in infants, children, and adults.1

Definitions and Criteria

Once a pH monitoring study is completed, data are downloaded from the data logger and analyzed with validated software on a computer. A pH of 4 is generally accepted as the optimum cutoff in both children and adults. This pH threshold was based on early observational studies that showed that perfusion of acid into the esophagus, resulting in a drop in intraesophageal pH below 4, corresponded with symptoms of heartburn.6 In adults, using a cutoff of pH 4 also provided the best discrimination in terms of sensitivity and specificity between subjects with proven reflux disease and asymptomatic controls, when compared to other threshold values.1,2 The identification of a reflux episode typically starts when the pH drops below the cutoff limit, lasts for at least 5 seconds, and ends when the pH returns to the cutoff limit.7 Several parameters based on this pH cutoff have been studied, including the total number of reflux episodes, number of reflux episodes lasting >5 minutes, duration of the longest reflux episode, and the reflux index, which is the percentage of time of the entire duration of the investigation during which the pH <4. Complex scoring systems, such as the ones proposed by Johnson and DeMeester for adults8 and by Jolley et al. for children,9 incorporate several of these parameters. The reflux index, however, is generally considered the single most important variable in clinical practice for both adults and children.1,2

Normal Ranges

Once reflux episodes have been identified in a pH monitoring study, often the next step is to determine where on the continuum between physiologic reflux and pathologic GERD a particular patient should be placed. Normative data are therefore needed to guide interpretation of pH monitoring results. Although published pediatric data are rather limited, few would debate the occurrence of asymptomatic episodes of reflux in normal infants and children. In a study of 509 infants 1–11 months old, the upper limit of normal for the number of acidic reflux episodes was 73 episodes daily, and the upper limit of normal for the reflux index pH <4 was 11.7%.10 Normative values for older children are relatively scarce, limited by the difficulty in obtaining data from truly healthy and asymptomatic volunteers. In some cases, “normals” were obtained from children hospitalized for GER evaluations who turned out to be asymptomatic during the time of pH monitoring11 or were found to have other causes for their gastrointestinal symptoms.12 As a result, caution should be used when comparing results. Overall, these studies suggest that physiologic acidic reflux is a common occurrence in infants during the first year of life, with decreased acid exposure found in older children and adults (Table 11–1). Based on the available data, the North American Society for Pediatric Gastroenterology, Hepatology and Nutrition (NASPGHAN) developed the following recommendations in the assessment of normal versus abnormal pH monitoring studies in infants and children: an upper limit of normal of the reflux index up to 12% in the first year of life and up to 6% thereafter.1

Diagnostic Accuracy

Although endoscopy and histology remain the gold standard for diagnosis of esophagitis, in the absence of erosive or biopsy-proven esophagitis, there is no gold standard for the definition of GERD, since not all patients with GERD will have esophagitis. In comparing adults with endoscopic esophagitis with normal volunteers, esophageal pH monitoring was found to have 77–93% sensitivity and 85–97% specificity.1,2 In children, estimated sensitivity of pH monitoring to predict esophagitis has been similar to adults, ranging from 83% to 100%.1

For the subset of symptomatic patients without evidence of esophagitis, diagnosis using reflux monitoring is arguably even more crucial in distinguishing between reflux disease and functional heartburn. However, when acid exposure times were compared between normal adult controls and patients with GERD symptoms but negative endoscopic findings, pH testing was able to discriminate between the two groups, but with considerable overlap in the percentage of time that the pH <4; the sensitivity was decreased to only 61–64%, and specificity was 85–91%.2 The clinical utility of pH monitoring in children with non-erosive reflux disease has not been well studied. Although esophagitis is clearly associated with abnormal acid exposure in children, the severity of esophagitis or symptoms has not been found to correlate with the severity of reflux as measured by pH monitoring.1,2

Reproducibility

In pediatrics, results from studies looking at the intrasubject reproducibility of esophageal pH results are variable. Vandenplas et al. studied 30 infants and children over two consecutive 24-hour periods; the Pearson correlation coefficients for the reflux index and number of reflux episodes between day 1 and day 2 were 0.95 and 0.98, respectively.13 In contrast, the Spearman correlation coefficient for the reflux index reported by Mahajan et al. was only 0.62 between day 1 and day 2, and for the number of reflux episodes it was 0.71.14 In yet another study that consisted of two consecutive 24-hour pH monitoring studies, 9 out of 30 children had discordant (normal versus abnormal) results between the 2 recording days, yielding an overall reproducibility of 70%.15 Similar studies in adults have reported slightly higher degrees reproducibility ranging from 77% to 89%.2 Overall, there appears to be some degree of day-to-day variability among patients; whether these differences are clinically significant is debatable. Esophageal pH monitoring results in isolation of clinical history should certainly be interpreted with caution, and consideration should be given for repeat testing when the clinical picture is unclear.

Symptom Correlation

For patients with typical symptoms of reflux, such as heartburn or regurgitation, clear findings of erosive disease on endoscopy, or histologic evidence of esophagitis, the diagnosis of GERD can usually be made without further diagnostic testing. Care should be taken in those patients that may have eosinophilic esophagitis. On the other hand, for patients with non-specific symptoms such as irritability and crying in infants, or extraesophageal symptoms of cough or wheezing, pH monitoring studies may be indicated to assess the relationship of symptoms with reflux episodes. There is currently no consensus on the time window that should be used to link a recorded symptom with a preceding or simultaneous reflux event. Based on analysis of time windows of various durations, most investigators recommend looking for symptom correlation during a 2-minute window beginning 2 minutes before onset of symptoms, although both longer and shorter intervals of between 30 seconds and 5 minutes have been proposed.2

Several statistical methods have also been proposed over the years to better quantify the association of symptoms and reflux episodes; there are no conclusive data, however, proving one index to be superior to the others. The symptom index (SI) is defined as the percentage of symptom episodes that are related to reflux16:

In adults, a SI score of scores ≥50% suggests a relationship between symptom and reflux.2,16 One drawback of using this method is that it does not take into account the total number of reflux episodes. For example, if a subject was found to have an abnormally elevated number of reflux episodes but only one or two symptoms that happened to coincide with an episode of reflux by chance, the relationship between symptoms and reflux may be overestimated by the SI.

The symptom sensitivity index (SSI), which was subsequently developed, is the percentage of reflux episodes that are associated with symptoms2,16:

An arbitrary cutoff of 10% or higher is commonly used to indicate a significant association between symptoms and reflux episodes. One disadvantage is that the SSI is more likely to be positive when the number of symptom episodes is high.

Most recently, the symptom association probability (SAP) was developed; this method statistically compares pH data temporally related to symptoms with pH data obtained during symptom-free periods and expresses the likelihood that the patient’s symptoms are related to reflux.2 Each 24-hour study is divided into consecutive 2-minute periods; these periods and the 2-minute periods preceding the onset of symptoms are then evaluated for the occurrence of reflux. Fisher’s exact test is then applied to calculate the probability (p) that reflux and symptom episodes are unrelated. SAP is calculated as (1 – p) 100%. By statistical convention, SAP ≥95% is considered positive.

Both SAP and SSI were recently shown to be significantly related to symptomatic response to high-dose omeprazole, albeit with a significant number of discordant cases. Diaz et al. also showed that SAP is an independent predictor of the success of antireflux surgery.17 Although patients with a positive relationship between symptoms and reflux have been shown to more likely respond to medical and surgical therapy, further prospective validation studies are needed. Ultimately, these indices can be helpful in evaluating the relationship between symptoms and reflux, but they do not directly take into account other factors that also influence the perception of symptoms, such as inherent esophageal mucosal sensitivity, duration of acid exposure, and proximal extent of reflux.2

The evaluation of the association between reflux and aerodigestive disorders has been challenging. Traditional 24-hour esophageal pH monitoring with a probe placed in the distal esophagus has not been shown to be sensitive for supraesophageal symptoms.18 Dual-probe pH monitoring adds a second pH electrode for measurement of pH changes in the proximal esophagus. There are several limitations, however, with dual-probe pH monitoring. At the present time, there is no consensus on the best location for proximal probe placement. Various studies report probe placement below the upper esophageal sphincter (UES), at or above the UES, or even above the esophagus in direct contact with the pharynx. There is also no agreement on the definition of a proximal reflux event; conventionally, it has been defined as a drop in pH below 4, based on the pH threshold for distal esophageal reflux. Recent data, however, have suggested that non-acidic or weakly acidic reflux with pH between 4 and 7 may also play a clinically significant role in aerodigestive disease,16,19 and there have been proposals to revise the pH criteria for proximal reflux. These various issues have ultimately made the establishment of normative values for proximal reflux elusive. Studies employing dual-probe pH monitoring in children and adults have had mixed results in terms of sensitivity and specificity for extraesophageal manifestations of reflux as well as intrasubject reproducibility.18 Newer diagnostic modalities, such as oropharyngeal or nasopharyngeal pH monitoring, and the non-invasive measurement of exhaled breath condensates for the detection of aspiration of gastric contents are currently under investigation. At the current time, however, the clinical advantage of dual-probe pH monitoring in children is not yet clearly proven; more research is needed before these new methodologies can become part of the routine evaluation of children with extraesophageal manifestations of reflux.

Limitations of Catheter-Based pH Monitoring

Nasally passed pH catheters can be uncomfortable and cumbersome for the patient. In some cases, patients may restrict their daily routines and activity levels during the performance of the test, leading to the potential for “false-negative” outcomes.2,20 To avoid this possibility and to increase patient’s compliance, wireless methods for measuring intraesophageal pH have been developed, as discussed below.

There are also patients with intractable symptoms who remain refractory despite aggressive medical therapy or have normal results on traditional esophageal pH monitoring studies. These patients have been shown to ultimately benefit from fundoplication,16 suggesting that pH monitoring may be missing some important information, most notably non-acidic reflux. The advent of multichannel intraluminal impedance with pH (MII-pH) technology has allowed for the detection of non-acidic reflux and will also be discussed later.

The Bravo pH system (Medtronic, Shoreview, MN) consists of an antimony pH electrode contained within a small capsule that transmits pH data wirelessly to a portable receiver using radio telemetry. In adults, the capsule is securely attached to the mucosal wall of the distal esophagus 6 cm above the squamocolumnar junction, with placement confirmation by endoscopy. This location was selected based on the understanding that the proximal border of the high-pressure zone, representing the LES, is 1–1.5 cm above the squamocolumnar junction (Figure 11–1). Since the capsule remains securely pinned to the esophageal mucosa for at least 48 hours, there is less chance for displacement of the pH sensor than with a transnasal catheter with changes in body position, swallowing, or vomiting. The Bravo system also allows for prolonged monitoring over a 48-hour period, and is expelled from the body into the stool after 3–5 days. Compared to catheter-based technique, the Bravo system has been shown to have less effect on daily routines, diet, and activity levels.2,21

The wireless pH capsule currently samples data at 6-second intervals (0.17 Hz), which is slower than the 4-second intervals (0.25 Hz) used by the catheter-based equipment.2,21 The 95th percentile for distal esophageal acid exposure in normal adults using the wireless pH system was 5.3%, a value higher than values reported in several (although not all) catheter-based system studies.2,21 The higher acid exposure threshold reported in healthy controls using the wireless pH system may be the consequence of less restriction in daily activities. No similar information is available for control children.

In published studies of children older than 4 years old, distal esophageal pH monitoring with the Bravo capsule has been found to be well tolerated, with no significant complications other than mild chest discomfort.20,22–25 Case reports have mentioned the possibility of applying the Bravo capsule even in smaller children. The capsule has been significantly better tolerated than a transnasal catheter in terms of appetite, activity, and satisfaction.20 There is currently no consensus on proper location for placement of the capsule in children, with one group confirming position by fluoroscopy similar to ESPGHAN recommendations for pH catheters, for example,25 and another group using Strobel’s formula to estimate 87% of the length of the esophagus between incisors and the LES.20

In terms of reproducibility and accuracy, relatively few studies have been performed in children. Only one study to date has compared the Bravo capsule side-by-side with a simultaneous transnasal pH catheter in children. Croffie et al. found no significant difference in the reflux index obtained by the two devices on day 1; on day 2, however, the median reflux index recorded by the Bravo capsule was significantly higher compared to day 1 of both the capsule and catheter.20 The clinical significance of this is unclear, with only one patient having discordant (abnormal versus normal) results between the 2 days of recording. Gunnarsdottir et al. also studied the use of 48-hour monitoring with a wireless capsule in children and found no statistically significant difference in the fraction of time with pH <4 between the first 24 hours and the entire 48-hour recording.23 In contrast, in our own series of 145 Bravo studies in children, there were significantly higher values on day 1 versus day 2 for the number of long reflux episodes, duration of longest episode, and fraction of time with pH <4 in the upright position.22Figure 11–2 shows a representative tracing in a normal patient, while Figure 11–3 shows the results in a patient with pathologic acidic reflux.