The Gastroesophageal Reflux Disease-Health-Related Quality of Life (GERD-HRQL) scale is a reliable, valid, responsive, and practical measure of symptom severity in patients with GERD. This type of scale is needed to determine effects of treatments and their comparison. This study defines the relationship between the GERD-HRQL score and the physiologic parameters of esophagogastroduodenoscopy, esophageal manometry, and 24-hour esophageal pH monitoring. Fifty-five patients referred for surgical evaluation of GERD answered the GERD-HRQL, a nine-item ordinal-scaled questionnaire. They were evaluated with esophagogastroduodenoscopy, esophageal manometry, and 24-hour pH monitoring. The relationships among these results were determined by linear regression analysis. There were no correlations between lower esophageal sphincter (LES) and any of the pH monitoring parameters (all r < 0.25, P > 0.2), esophagitis grade (r = -0.21, P = 0.2), nor any individual GERD-HRQL item score nor total score (all r < 0.2, P > 0.11). There were correlations between all the pH monitoring parameters and esophagitis grade (all r > 0.6, P < 0.001), but not with any of the GERD-HRQL item scores or total score (r < 0.3, P > 0.15). Six of the nine items scores and the total GERD-HRQL score correlated with esophagitis grade (all r > 0.4, P < 0.01). LES pressure is a poor indicator of symptom severity, the amount of reflux, and esophageal mucosal damage. pH monitoring-measured reflux and GERD-HRQL-measured symptom severity correlate well with mucosal damage. If the goals of GERD treatment are to relieve symptoms and reverse mucosal damage, the GERD-HRQL score and 24-hour pH monitoring are better outcome measures than the LES pressure. In an era of cost containment, the GERD-HRQL may be an adequate outcome measure.

Gastroesophageal reflux is a common problem in the United States, with up to 40 per cent of individuals experiencing at least one episode of heartburn or regurgitation a month.[ 1] Most of these are transient episodes not requiring specific interventions. However, many individuals experience significant symptoms of heartburn and seek medical attention for symptom relief. In addition, severe complications of reflux can occur, such as stricture, ulceration, Barrett's esophagus and possibly adenocarcinoma.

The pathophysiologic process that produces symptoms and pathologic changes is an abnormal amount and duration of contact of gastric juice (and possibly duodenal juice) in the esophagus. The cause of this pathophysiology has generally been accepted to be an incompetent lower esophageal sphincter (LES) and/or poor esophageal clearance of gastric juice. This conception of the pathophysiology of gastroesophageal reflux disease (GERD) led a few pioneering surgeons to develop operations based on wrapping the lower esophagus with the stomach. However, results of these early operations were inconsistent, with the procedures falling into disfavor, prompting multiple modifications of the technique.[ 3]

The cause of these surgical failures was not ill-conceived operations, but rather poor patient selection. Antireflux operations will only be effective if symptom-producing pathologic reflux caused by a hypotensive LES is indeed the patient's problem. To better identify patients suffering from GERD who could be helped by surgery, several tests have been developed.[ 4] These include esophagogastroduodenoscopy (EGD), which provides a visual assessment of mucosal damage and samples four histologic evaluation; esophageal manometry, which can provide data on the function of the upper and lower esophageal sphincters as well as peristalsis; and 24-hour esophageal pH monitoring, which provides information on the number and duration of reflux episodes.

The purpose of this study was to define the relationship between these physiologic measures of reflux and patient perceived symptoms.
Materials and Methods

All patients referred for consideration of antireflux surgery underwent a standardized evaluation. This included a thorough history and physical examination. Patients who had symptoms and signs suggestive of diagnoses other than GERD were not included in this study. Patients were then asked to complete the Gastroesophageal Reflux Disease-Health-Related Quality of Life (GERD-HRQL) questionnaire to determine severity of symptoms (Table 1).[ 5]

The GERD-HRQL is a nine-item Likert-type questionnaire. (It has been revised to include a 10th item; see Table 1.) Each item is scored from 0 to 5, with each score based on a descriptive anchor. These descriptive anchors are important so that the responses are standardized among patients at the same level of severity. The individual item scores are then added to come to the total GERD-HRQL score. The best possible score is 0 (asymptomatic in all items) and the worst possible score is 45 (incapacitated in all items). This instrument has been previously tested and found to be a valid, reliable, practical, and responsive measure of GERD.

All patients were further evaluated with EGD. Patients who were found to have esophageal disease not related to GERD were excluded from further study. Reflux-induced esophagitis was graded according to Silverstein and Tytgat as follows: grade 0, no evidence of esophagitis, normal-appearing mucosa; grade I, equivocal injury with erythema but no clear break in mucosa; grade II, superficial erosions involving less than 10 per cent of the distal 5 cm of the esophagus; grade III, confluent erosions involving less than 50 per cent of the distal 5 cm of the esophagus; grade IV, circumferential involvement of the distal esophagus; and grade V, severe esophagitis with any degree of stricturing or deep ulceration.

Patients were initially managed medically. However, those who had symptoms intractable to medical management or complications of GIERD or who were interested in surgical management to discontinue medication were considered for surgical therapy. Patients who were candidates for antireflux surgery were further evaluated with esophageal manometry and 24-hour esophageal pH monitoring.

Esophageal manometry was carried out in the manner described by Stein et al. After placement of the multiport pressure tube, the average of three wet swallows was used to derive the LES pressure (in mm Hg), qualitative assessment of esophageal peristalsis and amplitude (in mm Hg), and upper esophageal sphincter relaxation and coordination.

Twenty-four-hour esophageal pH monitoring was also done in the manner recommended by Stein et al. After placement of the probe, patients kept a log of activities and symptoms. The following items were recorded: total number of reflux episodes in 24 hours, number of reflux episodes longer than 5 minutes, the longest reflux episode, total time pH was below 4, total time pH was below 4 in upright position, and total time pH was below 4 in supine position; these data were then used to calculate the total DeMeester score.

Data were analyzed using the True Epistat statistical computer program. Linear regression analysis was used to determine correlations among each individual item of the GERD-HRQL score, each individual measurement of the 24-hour pH monitoring, total De-Meester score, EGD-determined esophagitis grade, LES pressure, and esophageal peristalsis amplitude. Kruskal-Wallis analysis of variance was done to determine trends in the median scores of the GERDHRQL with esophagitis grade and median scores of the DeMeester score with esophagitis grade.
Results

A total of 55 patients were evaluated. Of these, 28 underwent surgical antireflux procedures. Fig. 1 demonstrates the statistically significant (P = 0.03) trend of higher median total GERD-HRQL scores (i.e., more symptomatic) with higher EGD-derived esophagitis grades. Using linear regression analysis, there is a statistically significant relationship (P < 0.001) between esophagitis grade and total GERD-HRQL (r = 0.53). Furthermore, a linear relationship exists between the esophagitis grade and the individual GERDHRQL items 3 through 8. This implies that mucosal damage is associated with worsening patient-perceived symptoms.

Fig. 2 shows the statistically significant (P = 0.0007) trend of higher median total DeMeester score with esophagitis grade. Similarly, by linear regression analysis, there were correlations between all of the parameters measured by 24-hour pH monitoring and EGD-determined esophagitis grade (Table 3), This implies that the amount of acid reflux as measured by pH monitoring is associated with the severity of mucosal damage.

On the other hand, the level of LES hypotension had no correlation with the severity of symptoms as measured by the GERD-HRQL scores, EGD-derived esophagitis grade, or the amount of acid reflux as measured by 24-hour esophageal pH monitoring. This implies that once the LES pressure becomes pathologically low, further deterioration of LES pressure does not lead to worsening of symptoms, esophagitis, and amount of acid reflux.
Discussion

GERD is a disease process of great variation. This is true for the severity of symptoms, with some patients having minimal symptoms requiring no treatment or simple lifestyle changes and others with symptoms so severe that surgical intervention is warranted. A previous study has shown that the severity of symptoms as measured by the GERD-HRQL score was a good predictor of patients who would eventually choose surgical intervention. However, severity of symptoms is not enough to recommend surgical intervention. First, other diseases may cause symptoms that may be interpreted as reflux, such as malignancies, peptic ulcer disease, and collagen vascular disorders affecting the esophagus, and nonesophageal diseases such as coronary artery disease and pulmonary disease. These can be effectively ruled out with a combination of history and physical examination, a barium upper gastrointestinal series and/or EGD. EGD is essential in determining the severity of mucosal damage by reflux. And this study as shown that the severity of mucosal damage does correlate with the severity of symptoms.

Because antireflux surgery is designed to increase LES pressure, it is important to determine that the patient has a pathologically hypotensive LES. In patients with symptoms suggestive or atypical of reflux disease, a normal LES pressure should lead the physician to suspect other disease processes. To better define the LES pressure, esophageal manometry was developed. This measures the LES, as well as the peristalsis of the esophageal body and the pressure and coordination of the upper esophageal sphincter. This study shows that although all patients with symptoms severe enough to be considered for surgery had pathologically low LES pressure, there was no statistical correlation between how low the LES pressure was and how severe symptoms were, the severity of mucosal damage, or the amount of reflux. This implies that the barrier function of the LES is an "on/off" phenomenon. That is, once a pathologically low pressure is reached, the gate is open (so to speak), and reflux occurs independent of pressure.

Twenty-four-hour esophageal pH monitoring measures reflux per se. The data collected include the total number of reflux episodes, the total number of episodes lasting more than 5 minutes, the episode of longest duration, the percentage of total time that pH < 4, the percentage of time in the upright position that pH < 4, and the percentage of time in the supine position that pH < 4. Using a special formulation, these data are factored together to obtain the DeMeester score, which is reflective of the amount of reflux. Twenty-four-hour pH monitoring correlates with the severity of mucosal damage as determined by EGD (Table 2, Fig. 2). This is satisfying in the sense that this confirms our concept of GERD. Nevertheless, there were no statistically significant correlations with symptom severity as determined by the GERD-HRQL. We believe this may be due to the wide variations seen in patient-perceived symptoms and the wide variation in the 24-hour pH monitoring scores.

Short of therapy for reflux-induced complications, it is ultimately the relief of symptoms that is the purpose of most treatments for GERD. However, objectively quantitating the degree of symptoms has been problematic. Traditionally, symptoms have been measured qualitatively as "mild, moderate, severe, etc." as demonstrated by a recent publication. The difficulty with this method is the great variation in the interpretation of the meaning of these words to individual patients and physicians. That is, what may be "mild" to one person may be "severe" to another. This is why the use of descriptive anchors in an ordinal scale is so important. Patients can then use the anchor to determine their level of symptom severity. The GERD-HRQL was developed in this manner and has been found to be a reliable; valid; and, most importantly, a practical and responsive instrument. This study shows that the GERD-HRQL correlates well with the degree of mucosal damage, but not with the 24-hour pH monitoring score nor with the degree of LES hypotension. Although the lack of correlation with LES pressure is not surprising, the lack of correlation with pH monitoring is surprising, especially given that both the GERD-HRQL and 24-hour pH monitoring both correlate with EGD-determined esophagitis grade. We suspect that this is due to enough statistical variation in the data so that significance cannot be reached. Nevertheless, other studies have also shown inconsistencies between physiologic measures of disease and patient-perceived symptoms.

Given these physiologic findings, this leads to the question of how best to measure treatment response. Given the correlation between mucosal damage with symptom severity and mucosal damage with the amount of reflux, it seems that the use of the GERDHRQL, 24-hour pH monitoring, or EGD would make the most sense. Moreover, in the present era of cost consciousness, the use of invasive testing needs to be well justified. Therefore, symptom severity, as measured by the GERD-HRQL, may be an adequate measure of symptom outcome in the treatment of GERD.

Last updated Jan 4/07