We reviewed the medical records and the findings from endoscopy and ambulatory 24-hour dual-probe pH monitoring in an unselected group of consecutive patients who were referred to our motility laboratory from January 2004 to June 2006. Six hundred and eighty-nine patients received both of these examinations due to extraesophageal symptoms such as asthma, chronic cough, hoarseness, globus sensation or chronic sinusitis, but 62 patients were excluded due to incomplete medical records or endoscopic information. So, a total of 627 patients were included: 475 men and 152 women, and with a mean age of 50.9±11.8 years.

The presence or absence of reflux esophagitis and hiatal hernia were determined, with atrophic gastritis graded by one endoscopist (Kim GH), according to the criteria listed below. The endoscopist was kept "blind" to the information of the ambulatory 24-hour dual-probe pH monitoring.

If esophagitis was present, it was graded according to the Los Angeles classification18).

Hiatal hernia was defined as a circular extension of the gastric mucosa above the diaphragmatic hiatus greater than 2 cm in axial length.

Atrophic changes in the gastric body on endoscopy were diagnosed on the basis of the atrophic area displaying discoloration with or without blood vessels transparency. The grade of atrophic gastritis was assessed endoscopically using the atrophic pattern system described by Kimura et al13, 14). This classification divides the extent of atrophy into a closed type (C-type) and an open type (O-type). The C-type indicates that the atrophic border remains on the lesser curvature of the stomach, while the O-type means that the atrophic border no longer exits on the lesser curvature but extends along the anterior and posterior walls of the stomach (Figure 1). The C-type is subdivided by where the atrophic border crosses: C1, the angulus on the lesser curvature; C2, the lower and middle parts of the corpus; or C3, the upper part of the corpus. The O-type is also subdivided by the location of the atrophic border, which is parallel to the vertical axis of the stomach: O1, on the lesser curvature; O2, on the anterior and posterior walls; or O3, on the greater curvature.

Prolonged ambulatory pH monitoring was performed immediately after the standard esophageal manometry19), with a single-use, mono-crystalline antimony, dual-site pH probe (Zinetics 24, Medtronic Inc., Minneapolis, MN, USA) with electrodes placed at the tip and 15 cm proximal to the tip. A cutaneous reference electrode placed on the upper chest was also used. All electrodes were calibrated in buffer solutions of pH 7 and then pH 1. The pH catheter was introduced transnasally into the stomach and then withdrawn back into the esophagus until the electrodes were 5 cm or 20 cm above the proximal margin of the LES (lower esophageal sphincter). Data were collected with a portable data logger (Digitrapper Mark III, Synetics Medical Co., Stockholm, Sweden) with a sampling rate of 4 seconds. The subjects were encouraged to eat their regular meals, but with restricted intake of food or drink with a pH below 4. All subjects had a diary in which they recorded meal times (start and end), body position (supine and upright), and symptoms, if any. Following the 24-hour monitoring period, the probe was removed, and the data was transferred to a computer for analysis with "Polygram for Windows Release" 2.04 (Synetics Medical Co., Stockholm, Sweden). In addition, the pH recordings were displayed on a screen for a detailed manual analysis and determination of the temporal relationship of pH declines registered at the two probe sites. For both sites, a decrease in pH below 4 not induced by eating or drinking was marked as the beginning of a reflux episode, which continued until a rise above pH 5. To be accepted as a gastropharyngeal reflux event, the decrease at the proximal probe had to be abrupt and simultaneous with a decrease in the esophagus, or to be preceded by a decrease in pH of a similar or larger magnitude in the distal probe. Thus, acid episodes induced by oral intake, aero-digestive tract residue and secretions, proximal probe movement, or loss of mucosal contact in which the proximal pH decline may precede the esophageal pH drop were not included as gastropharyngeal reflux episodes.

The variables assessed for gastroesophageal reflux in the distal probe were: the total percentage of time the pH was < 4, the percentage of time the pH was < 4 in the supine and upright positions, the number of episodes where the pH was < 4, the number of episodes where the pH was < 4 for ≥ 5 min, the duration of the longest episode the pH was < 4, and the Johnson-DeMeester score20). The variables assessed for gastropharyngeal reflux in the proximal probe were: the total percentage of time the pH was < 4, the percentage of time the pH was < 4 in the supine and upright positions, and the number of episodes where the pH was < 4.

For the diagnosis of gastroesophageal reflux disease in the distal probe, two different aspects were analyzed21, 22): (1) Total reflux time: the total proportion of the recorded time with pH < 4; a value of > 4% was considered abnormal. (2) Number of reflux episodes: the total number of pH episodes with pH < 4 during the recording; a value of > 35 episodes was considered abnormal. For the diagnosis of gastropharyngeal reflux disease in the proximal probe, we considered time at pH < 4 of more than 0.1% total, 0.2% upright, or 0% supine to be pathological. More than 4 reflux episodes was considered pathological23, 24).

The data is expressed as medians (ranges) unless otherwise noted. The Student's t-test was used to assess the statistical significance for age and body mass index according to the grade of atrophic gastritis. The Mann-Whitney test was used to assess the statistical significance for the parameters of ambulatory pH monitoring according to the grade of atrophic gastritis. The differences in gender, alcohol drinking, smoking, reflux symptoms, reflux esophagitis, hiatal hernia, gastroesophageal reflux disease and gastropharyngeal reflux disease, according to the grade of atrophic gastritis, were assessed using the χ² test or Fisher's exact test. p<0.05 was considered statistically significant. Statistical calculations were performed using SPSS version 10.0 for Windows software (SPSS Inc., Chicago, IL, USA).

There was no difference in alcohol drinking, smoking, body mass index, or the presence of hiatal hernia according to the grade of atrophic gastritis. The patients with the O-type were significantly older than those with the C-type (p<0.001). The O-type was significantly associated with a lower prevalence of reflux symptoms and reflux esophagitis (p=0.01, p=0.001, respectively) (Table 1).

All the variables showing gastroesophageal reflux in the distal probe were significantly lower in the O-type than in the C-type (p<0.05). Similarly for the proximal probe, all variables, except the supine time of pH < 4, were significantly lower in the O-type than in the C-type (p<0.05) (Table 2).

The frequency of gastroesophageal reflux disease, as defined by ambulatory pH monitoring, was significantly lower in the O-type than in the C-type (46/152, 30.3% vs. 253/475, 53.3%, p<0.001). The frequency of gastropharyngeal reflux disease, as defined by ambulatory pH monitoring, was also significantly lower in the O-type than in the C-type (61/152, 40.1% vs. 246/475, 51.8%, p=0.012) (Figure 2).