Assessment of oesophagitis by histology and morphometry

Histopathology of reflux-induced esophageal and supraesophageal injuries

As many as half of patients who have symptoms and objective evidence of gastroesophageal reflux disease (GERD) will have normal mucosa or only hyperemia at endoscopy. Because inflamed esophageal mucosa may appear normal endoscopically, and because hyperemia may or may not reflect histologic espophagitis, biopsy to document tissue injury in symptomatic patients with these minimal endoscopic findings may be helpful. Reflux may induce inflammation in the squamous mucosa of the esophagus, but in many patients only hyperplasia of the epithelium is seen. This hyperplasia is defined by a basal zone that exceeds 15% of the thickness of the mucosa and subepithelial papillae that exceed 67% of the thickness of the mucosa. Because these changes may be present normally in the distal 2.5 cm of the esophagus, and because they may be distributed over the distal 8 cm in a patchy fashion, multiple biopsies taken more than 2.5 cm above the esophagogastric junction are necessary to detect them reliably. Supraesophageal complications of GERD include posterior laryngitis, inflammatory polyp of the larynx (contact ulcer or laryngeal granuloma), subglottic stenosis and laryngeal squamous cell carcinoma.

Diagnostic value of histology in non-erosive gastro-oesophageal reflux disease

BACKGROUND

In the absence of oesophageal erosions longterm pH monitoring is the present gold standard for diagnosing gastro-oesophageal reflux disease (GORD). This method, however, is invasive, time consuming, expensive, and not generally available.

AIMS

As histological changes have been described in GORD, this study looked at the possibility of whether the diagnosis of non-erosive reflux disease could be made by histological examination routinely during endoscopy.

SUBJECTS

A total of 24 prospectively selected patients with symptoms suggestive of GORD and seven healthy volunteers.

METHODS

Oesophageal erosions and other peptic lesions were excluded by oesophago-gastroduodenoscopy. Oesophageal pinch biopsy specimens were taken 2 cm and 5 cm above the oesophagogastric junction and evaluated blindly for the histological parameters cellular infiltration, basal zone hyperplasia, and papillary length. Twenty four hour pH monitoring was used as gold standard for the definition of reflux disease. It was abnormal in 13 patients (reflux patients) and normal in 11 patients (symptomatic controls) and in seven healthy volunteers.

RESULTS

Sparse infiltration of the epithelium with lymphocytes in at least one biopsy specimen was found in all patients and volunteers, with neutrophils in three reflux patients, and with eosinophils in two reflux patients and in two healthy volunteers. The basal zone thickness was increased in three reflux patients, in one symptomatic control, and in one healthy volunteer. The papillary length was greater than two thirds of total epithelium in six of 13 reflux patients in contrast with none in 11 symptomatic controls (p < 0.05) and to one healthy volunteer. The sensitivity of the parameter papillary length hence was only 46%.

CONCLUSIONS

Although gastro-oesophageal reflux produces slight histological changes apart from oesophageal erosions in a few subjects, none of the established histological parameters can fulfil the for the diagnosis of GORD in patients without visible oesophageal erosions.

The application of microridge analysis in the diagnosis of gastro-oesophageal reflux disease

BACKGROUND

The percentage of epithelial surface area covered by microridges (%MR) seen during scanning electron microscopy of oesophageal biopsy specimens has previously been shown to correlate with symptomatic reflux disease, a result < or = 35% being abnormal. The aim of this study was to compare %MR with endoscopy, light microscopy, and pH monitoring results.

METHODS

Sixty-seven patients with heartburn were divided into oesophagitis or none on the basis of endoscopy and light microscopy findings and into those with and without abnormal acid reflux on the basis of pH monitoring.

RESULTS

The endoscopic and light microscopic oesophagitis groups had significantly greater degrees of acid reflux than those without oesophagitis (p < 0.05), even though neither the specific %MR nor the number of patients below the 35% cutoff showed any difference between those with and without endoscopic oesophagitis, light microscopic oesophagitis or those with normal and abnormal acid reflux on pH monitoring.

CONCLUSION

Despite the significant relationship between endoscopic and light microscopic oesophagitis and abnormal pH monitoring microridge analysis did not correlate with any of these variables

Effect of general anaesthesia on prolonged intraoesophageal pH monitoring

Investigation of gastro-oesophageal reflux often includes endoscopy, usually under general anaesthesia, and pH monitoring. In most cases, the pH probe is passed when the child is awake and is poorly tolerated. The effect of general anaesthesia on pH monitoring is unknown. The aim of the study was to determine if placing the probe in the anaesthetised child gives a representative pH study. Twenty children aged 4 months to 13 years underwent oesophago-gastroduodenoscopy under general anaesthesia. A pH electrode was placed under direct vision in the distal oesophagus. pH monitoring was begun after completion of anaesthesia and continued for 18-24 hours. The study was repeated within 14 days without anaesthetic. The reproducibility of values of percent pH < 4, number of reflux episodes/hour, reflux episodes lasting > 5 min, and longest reflux episode was 85%, 90%, 75%, and 75% respectively. These results are comparable with those in adults and children in whom pH studies were performed on consecutive days (without anaesthetic) keeping all variables constant. Therefore pH data collected in a child within 24 hours of endoscopy under general anaesthesia are representative.

Oesophageal mucosa in a population at risk of oesophageal cancer: post-mortem studies

We performed post-mortem studies of oesophageal mucosa from 513 consecutive autopsies of cases who died from unnatural causes or from diseases not related to the oesophagus. Of these, 170 cases were rural blacks from endemic high-risk areas, 98 were urban blacks at high risk, 158 were coloureds at moderate risk and 87 whites at low risk for oesophageal carcinoma. Oesophagi were studied macroscopically and histologically to determine malignant and precursor lesions. A prevalence of 1.0 to 1.8% of squamous carcinoma and 7 to 7.5% of mucosal dysplasia was detected in cases older than 30 years from high-risk urban and rural population groups. These results are in accord with our previous cytological and endoscopical screening studies. No significant differences in the prevalence of oesophagitis, glycogenic acanthosis and atrophy were found in the groups studied. A study of oesophageal melanocytes showed a 3% prevalence, unrelated to dysplastic and malignant lesions, and suggested that these cells were of little if any practical significance as precursor lesions.