Impact of hiatal hernia on histological pattern of non-erosive reflux disease.

BACKGROUND: Hiatus hernia (HH) has major pathophysiological effects favoring gastroesophageal reflux and hence contributing to esophageal mucosa injury, especially in patients with severe gastroesophageal disease. However, prospective studies investigating the impact of HH on the esophageal mucosa in non-erosive reflux disease (NERD) are lacking. This study evaluated the association between the presence of (HH) and the histological findings in symptomatic patients with NERD.
METHODS: Fifty consecutive patients with gastroesophageal reflux disease (GERD) were enrolled. After conventional endoscopy, Lugol solution was applied and biopsy specimens were obtained. Histological parameters including basal zone hyperplasia, papillary length and cellular infiltration were evaluated. The chi-square test with Yates' correlation was used for comparing discrete parameters between groups. However, Fisher's exact probability test was used where the expected frequencies were lower than 5. Wilcoxon's test for unpaired samples was preferred in cases of semi-quantitative parameters.
RESULTS: The presence of HH along with more severe findings (0.01 <P < 0.05) was confirmed in 18 patients. NERD was observed in 29 (58%) patients. Basal zone hyperplasia and loss of glycogen accompanied HH in all cases, and the correlation was significant in NERD (P < 0.001). The remaining histological patterns were similar between erosive reflux disease and NERD in the presence of HH.
CONCLUSION: The presence of HH is correlated with more severe endoscopy findings, and predisposes for severe histological abnormality in cases of NERD.

Background

Gastroesophageal reflux disease (GERD) is a common condition that affects 25–30% of the population [1]. It clearly involves multifactorial pathophysiology, yet the factors underlying why only some patients develop reflux esophagitis are unclear [2].

Symptoms and demographic data do not allow differentiation between the endoscopy-negative (non-erosive reflux disease; NERD) and endoscopy-positive (erosive reflux disease; ERD) forms of the disease. In fact most patients with typical symptoms of GERD have normal esophageal mucosa on upper endoscopy. Indeed, more than two-thirds of all patients with reflux symptoms never develop esophageal erosions, ulcers or strictures [3]. This group of NERD patients constitutes a significant clinical problem since they appear to be relatively resistant to proton-pump inhibitors (PPIs) [4,5].

Hiatal hernia (HH) has been considered to be one of the pathophysiological mechanisms that contributes to the development of GERD, promoting refluxate access and impaired acid clearance; however, the impact of this mechanism in NERD is unclear [2,6,7].

The aim of the present study was to clarify the possible association of HH with histological findings on a group of prospectively studied symptomatic patients with NERD.

Methods

Fifty patients (29 men, 21 women; aged 49.9 ± 6.6 years, mean ± SD) were evaluated prospectively in our endoscopy unit for symptoms compatible with GERD, namely heartburn, acid regurgitation and belching. A standardized questionnaire was completed for each patient during an interview with an experienced gastroenterologist. Demographic details of the GERD patients were recorded, including age, sex, smoking habits, tea, coffee and alcohol consumption, and concurrent medical conditions including hypertension and diabetes mellitus.

None of the patients included in this study had a current or past history of peptic ulcer disease, previous gastric surgery, antihelicobacter therapy, or use of PPIs, non-steroidal anti-inflammatory drugs, steroids or tetracycline during the previous 4 weeks. Ethics approval was obtained from the ethics committee of the University Hospital of Alexandroupolis, and patients provided signed, informed consent for their biopsy specimens to be taken.

Routine endoscopy was performed in all patients by the same endoscopist using a flexible endoscope (GIF-Q145, Olympus). The distance between the esophagogastric junction and the incisor teeth was recorded. Reflux esophagitis was graded in accordance with the Los Angeles classification [8]. HH was considered present if gastric folds were assessed as extending ≥2 cm above the diaphragmatic hiatus during quiet respiration [2].

At least four biopsy specimens were taken at 3 cm above the lower esophageal sphincter with biopsy forceps (Olympus) in a criss-cross manner. In order to improve endoscopic visualization and provide biopsy orientation, 20 ml of 2% potassium iodine solution (Lugol) was applied through a "spray" catheter [9-11]. To obtain sufficient material and to ensure an almost vertical pinch biopsy specimen, the opened forceps were withdrawn towards the tip of the endoscope, which was bent forwards maximally, and hence the forceps were pressed vertically against the esophageal wall. Specimens were fixed in 40 mg/L formaldehyde [12].

After all the sections had been obtained, they were assessed histologically in a blinded manner (i.e. without endoscopic or clinical information). Standardized reports completed by the histopathologist comprised an evaluation of the following histological parameters: basal zone hyperplasia, papillary length, dilatation of intraepithelial blood vessels, and semi-quantitative cellular infiltration by T-lymphocytes, neutrophils and eosinophils. Alterations in glycogen content, erosion, ulceration and chronic inflammation were also assessed as described previously [12-17].

The chi-square test with Yates' correlation was used to compare discrete parameters between groups. However, Fisher's exact probability test was used where expected frequencies were lower than 5. Wilcoxon's test for unpaired samples was preferred in cases of semi-quantitative parameters due to its greater power. Mean values and their 95% confidence limits were calculated. Statistical significance was set at P ≤ 0.05. All analyses were performed using the statistical software package "Statistica (version 6)".

Results

Endoscopy findings

Endoscopy revealed esophageal mucosa with a normal appearance in 29 patients. The remaining 21 patients had esophagitis of variable severity (Table 1).

Table 1

Endoscopy findings in patients with reflux disease. Endoscopy findings in patients with reflux disease, for HH+ and HH-.

	NERD	ERD grade A	ERD grade B	ERD grade C	ERD grade D	Total
HH+	7	5	4	2	0	18
HH-	22	8	2	0	0	32
Total	29	13	6	2	0	50

HH was observed in 18 patients. Its presence (HH+) was correlated not only with the presence of erosions (P = 0.0196) (Figure 1), but also with the severity of the endoscopy findings (Wilcoxon's T1 score for unpaired samples: 576 for N1 = 18 and N2 = 32, 0.01