Relationship between duration of gastroesophageal reflux disease (GERD) and lower esophageal sphincter (LES) resting pressure. The length of history of GERD was significantly inversely related to LES resting pressure.
Relationship between duration of gastroesophageal reflux disease (GERD) and the amplitude of esophageal peristalsis at 5 cm (A), 10 cm (B), and 15 cm (C) proximal to the lower esophageal sphincter (LES). The longer the history of GERD, the lower the amplitude of esophageal peristalsis.
Relationship between esophageal transit time and duration of gastroesophageal reflux disease (GERD) (A) and amplitude of peristalsis at 5 cm proximal to the lower esophageal sphincter (LES) (B). Prolongation of esophageal transit time was significantly related to the length of history of GERD and significantly inversely related to the amplitude of peristalsis.
Chrysos E, Prokopakis G, Athanasakis E, Pechlivanides G, Tsiaoussis J, Mantides A, Xynos E. Factors Affecting Esophageal Motility in Gastroesophageal Reflux Disease. Arch Surg. 2003;138(3):241-246. doi:10.1001/archsurg.138.3.241
There are conflicting data concerning the effect of gastroesophageal reflux disease (GERD) on esophageal motor function.
Duration of GERD might affect severity of symptoms, grade of esophageal mucosal injury, and esophageal motor behavior.
Retrospective study of a defined cohort.
Two referral centers, one of them academic, for esophageal gastrointestinal motility disorders.
One hundred forty-seven patients with documented GERD.
Main Outcome Measures
Symptoms, grade of mucosal injury on esophagoscopy, esophageal manometry, ambulatory esophageal pH monitoring, and esophagogram.
Patients with GERD had significantly decreased lower esophageal sphincter resting pressure (P = .02), lower amplitude of esophageal peristalsis at all levels of measurement (P<.001), and more delayed esophageal transit (P = .007) compared with control subjects. Patients with dysphagia, severe esophagitis, and Barrett esophagus presented with a longer history of the disease, significantly worse esophageal motor function (P<.01), and more prolonged esophageal transit than patients without the above features of the disease. Impairment of esophageal peristalsis and lower esophageal sphincter resting pressure were significantly inversely related to the duration of the disease (P<.001). Also, delay of esophageal transit was significantly related to the duration of the disease (P = .002) and inversely related to the amplitude of esophageal peristalsis (P<.001). Unlike the manometric variables, the extent of reflux, as assessed by ambulatory 24-hour esophageal pH monitoring, was not related to the duration of the disease.
A long history of GERD is more commonly associated with presence of dysphagia, delayed esophageal transit, severe esophagitis, presence of Barrett esophagus, and impaired esophageal motility.
IMPAIRMENT OF esophageal motility is a common finding in patients with gastroesophageal reflux disease (GERD).1 At present, it remains unclear whether peristaltic dysfunction in patients with GERD is a primary pathogenetic factor or develops secondary to the mucosal injury, which is induced by the persistent reflux of gastric juice across a mechanically defective lower esophageal sphincter (LES). It has been shown that loss of esophageal contractility rarely occurs in the absence of mucosal injury of the esophagus.2 In addition, according to the findings of one study by our group,3 a very high amount of reflux is, in turn, associated with severe impairment of esophageal body motility, as that expressed by the amplitude and strength of esophageal body peristalsis.
Shiino et al4 asked whether the duration of GERD symptoms is related to the severity of esophageal dysmotility. They found that the duration of GERD has little, if any, influence on esophageal body motor behavior and LES function. That observation seems to contradict the aforementioned hypothesis, according to which impairment of esophageal motor function is the result of mucosal injury secondary to persistent reflux of gastric contents into the esophageal lumen.1- 3 The aim of the present study was to reassess any possible relationship of the duration of GERD to esophageal motor dysmotility, taking into account at the same time the severity of the disease, as expressed by the amount of reflux and the degree of esophageal injury.
The data obtained from esophageal endoscopy, manometry, and 24-hour ambulatory pH monitoring, as well as those from an esophagogram and esophageal transit assessment of 147 patients with documented GERD, were analyzed in relation to the duration of their symptoms. Duration of the disease was defined as the interval between the first time the patient sought medical advice and the onset of the study. Patients' symptoms included heartburn, regurgitation, dysphagia, and respiratory symptoms such as hoarseness, chronic cough, and symptoms suggesting endogenous asthma. Dysphagia was assessed according to the DeMeester scoring system,5 and only patients with grades II and III dysphagia were included in the dysphagia subset. All patients were referred either to the Laboratory of Gastrointestinal Motility of the Department of General Surgery, University Hospital of Heraklion, Heraklion, Greece, or to the Unit of Gastroenterology and the Second Department of General Surgery, Athens Naval and Veterans Hospital, Athens, Greece, for symptomatic and laboratory assessment before surgery. Among several regimens of conservative treatment, the patients all had been receiving proton pump inhibitors (PPIs) for at least 2 years, either continuously or intermittently. Patients with primary esophageal motor disorders, short esophagus, and paraesophageal hernia, as well as those with systemic diseases that might affect esophageal motility (collagen diseases, neuromuscular diseases, and diabetic neuropathy), were excluded from the study. Another 38 subjects of matched age and sex served as control subjects, provided they had never experienced GERD-related symptoms, had no upper gastrointestinal tract (including biliary) disease or surgery, and had never received PPIs, histamine2-blockers, or prokinetic treatment. The Postgraduate Committee of the Medical School of the University of Crete, Heraklion, approved the protocol of the study.
At esophagoscopy, the severity of esophagitis was assessed according to the Savary-Miller criteria6: grade I, esophageal mucosal edema and reddening; grade II, limited esophageal mucosal erosions not circumferentially distributed; grade III, circumferentially distributed erosive lesions; and grade IV, peptic ulcer or stricture. Barrett esophagus was documented on histologic examination as the presence of intestinal epithelium (goblet cell metaplasia) at the lower esophagus at a distance of either more than 3 cm (long Barrett) or 3 cm or less (short Barrett) from the endoscopically identified gastroesophageal junction.
Barium swallow study served to image the presence and length of hiatal hernia and to assess any peptic stricture. In addition, the gross esophageal transit time was assessed after the swallow of a 15-mL mouthful of barium sulfate–bread bolus in the erect position. The subject was instructed to swallow on command (every 20 minutes) until complete passage of the bolus into the stomach. Esophageal transit was defined as the time between first entry of the bolus into the tubular esophagus and the complete passage of the entire bolus through the cardia into the stomach.
Standard esophageal manometry was performed with an 8-lumen catheter. All lumens were constantly perfused with distilled water by a low-compliance perfusion system (Arndorfer Medical Specialties, Inc, Greendale, Wis). Pressure transducers were incorporated in each perfusion line and connected to a polygraph that served as amplifier (Synectics Medical, Stockholm, Sweden). Manometric tracings were depicted on the screen of an online computer and stored for retrospective analysis with the use of a dedicated program (Polygram version 4.2, Upper GI Edition; Gastrosoft Inc, Stockholm). From those tracings, the resting LES, LES relaxation at swallow, and amplitude of esophageal peristalsis at 5, 10, and 15 cm proximal to the LES were calculated according to standard methods.7
The device for the ambulatory 24-hour esophageal pH monitoring consisted of a transnasally introduced electrode, placed 5 cm proximal to the manometrically defined LES, and a digital data recorder (Synectics Medical). All patients under examination had a standard diet of 3 main meals. A reflux episode was defined as a drop in esophageal pH below 4 that lasted for more than 4 seconds. The calculated variables reported herein are the total composite score (DeMeester scoring system) and the total percentage of time with esophageal exposure to pH of less than 4.5,7 The presence of GERD was documented by abnormal results of esophageal pH monitoring, which was defined as a total composite score above 14.72 and esophageal intraluminal pH of less than 4 for more than 4% of the recording time during 24-hour pH measurement.5 All patients included in the study fulfilled the criterion of an abnormal result of esophageal pH monitoring as mentioned already. Before manometry and esophageal pH monitoring, any medication with gastrokinetic and antisecretory properties was discontinued for at least 3 and 7 days, respectively.
Unless otherwise stated, data are expressed as the median and range. Statistical analysis included linear correlation of the duration of symptoms with the LES pressure and the amplitude of esophageal peristalsis at 5, 10, and 15 cm proximal to the LES. Correlation of the acid reflux measures with the amplitude of esophageal peristalsis and the LES resting pressure, and with the duration of symptoms, was also performed in the same manner. Mann-Whitney test for unpaired values was applied to compare the various clinical and laboratory measures between patients with and without dysphagia and those with and without Barrett esophagus. Also, the same test was applied to assess differences in the manometric, esophageal pH monitoring, and esophageal transit data between different degrees of esophagitis. P values of less than .05 were considered statistically significant.
There were 95 male and 52 female patients, with a median age of 51 years (range, 20-78 years), who participated in the study. The duration of the reflux symptoms ranged from 3 to 52 years (median, 16 years). Twenty-seven patients (12 male, 15 female) reported grades II and III dysphagia. They all had been receiving conservative treatment with PPIs, mostly intermittently and at maintenance doses for 1.8 to 7.2 years. For patients with a GERD history of less than 10 years, there was positive correlation of the duration of the disease and the length of treatment with PPIs (P = .005).
At esophagoscopy, there were 32 patients (21.8%) without esophagitis, while 40 (27.2%) had grade I, 45 (30.6%) had grade II, and the remaining 30 (20.4%) had grade III esophagitis. Most of the patients with grade III esophagitis were those who had failed to fully comply with PPI maintenance treatment. Twenty patients had short-segment (6 [30%]) or long-segment (14 [70%]) Barrett esophagus. On the esophagogram, 114 patients (77.6%) had a sliding nonreducing hiatal hernia, while esophageal transit was significantly slower than in controls (18 seconds [7-55 seconds] vs 14 seconds [6-31 seconds]; P = .007). Compared with controls, patients had significantly lower LES resting pressure (12 mm Hg [3-41 mm Hg] vs 21 mm Hg [12-31 mm Hg]; P = .02) and amplitude of esophageal peristalsis at all levels of measurement (5 cm above LES: 58 mm Hg [11-114 mm Hg] vs 96 mm Hg [48-135 mm Hg], P<.001; 10 cm above LES: 50 mm Hg [10-93 mm Hg] vs 89 mm Hg [42-131 mm Hg], P<.001; 15 cm above LES: 34 mm Hg [6-81 mm Hg] vs 66 mm Hg [34-96 mm Hg], P = .03). All of the manometric measures (LES pressure and amplitude of esophageal peristalsis at all levels) were significantly inversely related to the duration of the disease (Figure 1 and Figure 2). Also, the delay of esophageal transit was significantly related to the duration of the disease (Figure 3A) and, as expected, inversely related to the amplitude of peristalsis at the distal esophagus (Figure 3B). Unlike the manometric measures, the extent of reflux at ambulatory 24-hour esophageal pH monitoring, as expressed by the DeMeester score of 69 (34-127) and the percentage of time with pH less than 4 (16% [4.5%-45%]), was not related to the duration of the disease. Also, none of the manometric variables nor those of esophageal transit were related to the duration of treatment with PPIs.
Patients with dysphagia tended to be significantly older and have a longer history of disease, lower amplitude of esophageal peristalsis, and lower LES resting pressure as compared with those without dysphagia. Furthermore, the former subset of patients presented with more severe esophagitis, greater incidence of Barrett esophagus, and slower esophageal transit than the latter one (Table 1).
Severity of esophagitis was significantly related to the duration of disease. Patients with grade III esophagitis had the longest history of reflux disease. Also, patients with grade II esophagitis had significantly lower amplitude of esophageal peristalsis at all levels, lower LES pressure, and greater extent of reflux than those with grade I or no esophagitis at all. Patients with grade III esophagitis presented with even more impaired esophageal peristalsis, lower LES pressure, and greater extent of reflux. Similarly, the more severe the esophagitis, the more prolonged was the esophageal transit (Table 2). No relationship between the extent of reflux as measured by esophageal pH monitoring and the severity of esophagitis was detected.
Patients with Barrett esophagus were older and had a longer history of disease and grater incidence of dysphagia than those without. Furthermore, the former group of patients had significantly more severe esophagitis, lower amplitude of esophageal peristalsis and LES resting pressure, greater extent of reflux, and more delayed esophageal transit than the latter one (Table 3).
It has been suggested that amplitude of esophageal contractions is not significantly altered in patients with GERD, even in the presence of severe inflammation, although impairment of LES tone is secondary to esophagitis. Furthermore, it has been reported that healing of macroscopic mucosal lesions with conservative treatment does affect esophageal body motility and that antireflux operations have little influence on esophageal motility pattern.8,9 In contrast to our observations, recent data indicate a strong relationship between esophageal exposure to acid and motility impairment. In particular, esophageal motility deteriorates and the prevalence of a mechanically defective LES increases, as mucosal injury progresses,2 although there are patients with severe mucosal lesions who still have a competent sphincter.2,10 The fact that the severity of reflux esophagitis and esophageal motility impairment are related to the duration and frequency of the reflux episodes, as shown in a previous study by our group,3 may be explained by the presence of acid-sensitive receptors in the esophageal mucosa, as proposed by Bontempo et al.11
According to the findings of the present study, which are in agreement with the latter hypothesis, there is a strong relationship between severity of mucosal injury on the one hand and presence of dysphagia and extent of esophageal motility impairment on the other, in GERD. In particular, dysphagia, although not commonly encountered, was detected almost exclusively in patients with severe esophagitis, while delay of esophageal transit was significantly related to the grade of mucosal injury. Furthermore, severity of mucosal inflammation was inversely related to the amplitude of esophageal peristalsis and the LES resting pressure. Our findings partly confirm those reported by Grande et al,12 according to which dysphagia is related to esophageal motor impairment, rather than to mucosal inflammation.
Despite the aforementioned data, the assumption that both extent and duration of gastroesophageal reflux may adversely influence esophageal motility, although attractive, is neither fully documented nor universally accepted. In regard to severity of GERD, the results reported by Shiino et al4 and those found in the present study agree in that the extent of acid reflux, as assessed on esophageal pH monitoring, was not related either to the duration and severity of symptoms or to the delay of esophageal transit, the grade of mucosal injury, the presence of Barrett esophagus, or even the impairment of esophageal peristalsis and LES resting pressure. Hence, it can be speculated that impairment of esophageal motor function may be detected independently of the extent of reflux present at a given time and that, since severe esophageal mucosal injury is established, further esophageal motor deterioration is not dependent on the extent of ongoing reflux.
In contrast to the findings of Shiino et al4 and Armstrong et al,13 those of the present study showed a strong relationship between duration of GERD symptoms and severity of esophageal motor impairment. We found a strong inverse correlation between the length of history of GERD and the grade of esophageal mucosal injury, as well as the amplitude of esophageal peristalsis and LES resting pressure. Furthermore, patients with dysphagia had a more prolonged history of the disease than those without. Also, the former subset of patients exhibited more severe esophagitis, greater esophageal motor impairment, and more delayed esophageal transit. In addition, GERD history was longer in patients with Barrett esophagus than in those without. Also, impairment of esophageal motility was greater and esophageal transit more delayed in the former group of patients.
Discrepancy between the assumptions concerning influence of GERD duration on esophageal motor function reported by Shiino et al4 and those exhibited in the present study might be explained by several differences in patient groups between the 2 studies. First, our group included exclusively patients with documented GERD who were referred for surgical management, in contrast to the patients in Shiino and coworkers' study, which also included patients who were referred to determine the presence of GERD. That means that patients in the present series had generally more severe reflux disease than those of Shiino et al.4 Indeed, mean acid reflux score was significantly higher in our series (70 vs 43). Second, more than half of our patients had severe mucosal inflammation on endoscopy, while the vast majority of patients in the Shiino et al series had either normal esophageal mucosa on endoscopy or no endoscopy at all. Finally, duration of GERD symptoms was significantly shorter in the Shiino et al series (median, 60 months) than in ours (median, 150 months).
Taking into account that the longer the duration of GERD, the more severe the esophageal body dysfunction, and considering that esophageal body motility is a major determining factor for the necessity and timing of surgical management in patients with GERD,2,10 we propose that patients with reflux who have a long history of the disease should be promptly offered antireflux surgery, instead of maintenance conservative treatment with PPIs. Esophageal body motor function was deteriorated by time in the present series, although all patients had been receiving long-term medical treatment. This means that, even if macroscopic mucosal lesions finally become healed during treatment with PPIs, functional improvement is not always achieved, as already reported by others.9,14
In conclusion, esophageal body motility and, consequently, esophageal clearance function are inversely related to duration of GERD. It seems that this inverse correlation exists in late rather than early stages of the disease, and that medical treatment fails to prevent esophageal motor dysfunction. We suggest that surgical treatment of patients with GERD be performed before severe esophageal body dysmotility develops, and independent of LES functional status.
Corresponding author and reprints: Evaghelos Xynos, MD, Department of General Surgery, University Hospital of Heraklion, Heraklion, Crete GR-711 10, Greece (e-mail: firstname.lastname@example.org).
Accepted for publication October 26, 2002.