Intragastric pH monitoring is unsuitable for diagnosis of duodenogastric reflux

Gastroesophageal Reflux Poses a Potential Risk for Late Complications of Bronchopulmonary Dysplasia: A Prospective Cohort Study

BACKGROUND

Bronchopulmonary dysplasia (BPD) is the most common respiratory disorder in extremely low birth weight infants. Although most symptoms of BPD improve, some late complications exist, even with regular treatment. Gastroesophageal reflux (GER), also common in extremely premature infants, may be related to many cardiorespiratory symptoms. However, the potential of GER as a risk factor for late complications associated with BPD is still unclear.

RESEARCH QUESTION

The goal of this study was to determine if GER increases the risk of late complications of BPD in infants.

STUDY DESIGN AND METHODS

A multicenter prospective cohort of 131 infants (79 male subjects, 52 female subjects) with BPD was enrolled. The development of late complications was assessed over an 18-month follow-up period. Twenty-four-hour pH-multichannel intraluminal impedance and gastric sodium concentrations were analyzed in all infants at 36 weeks' postmenstrual age and at the last interview. Prevalence and risk factors of late complications of BPD were analyzed by using forward logistic regression.

RESULTS

The prevalence of late complications in BPD infants was 63.79% and included respiratory symptoms (49.14%), vomiting (38.79%), retinopathy of prematurity (25.86%), hypoxic-ischemic injury (3.45%), rehospitalization (26.72%), and sudden death (0.86%). Respiratory diseases constituted the most frequent complication. The prevalence of GER in BPD was 42.24% and included acid GER (18.10%) and duodenogastroesophageal reflux (DGER; 24.14%). Risk factors for respiratory symptoms were gestational age ≤ 30 weeks (OR, 3.213; 95% CI, 1.221-8.460), birth weight < 1,500 g (OR, 2.803; 95% CI, 1.014-7.749), invasive ventilation > 7 days (OR, 4.952; 95% CI, 1.508-16.267), acid GER (OR, 4.630; 95% CI, 1.305-16.420), and DGER (OR, 5.588; 95% CI, 1.770-17.648). Infants with BPD and DGER were more prone to late complications than those with acid GER or no reflux.

INTERPRETATION

The prevalence of late complications is high in infants with BPD. GER (and in particular, DGER) poses a tentative risk for these late complications.

TRIAL REGISTRY

ClinicalTrials.gov; No.: NCT03014453; URL: www.clinicaltrials.gov.

Time pattern of gastric acidity in Barrett's esophagus

Increased gastroesophageal acid reflux is frequently found in patients with Barrett's esophagus, and it has been hypothesized that gastric acid hypersecretion could be an important factor aggravating the exposure of esophageal mucosa to acid and then contributing to the development of this disorder. The aim of the present study was to assess whether the circadian pattern of gastric acidity differs between refluxer patients with and without Barrett's esophagus and normal subjects. Continuous 24-hr gastric pH monitoring was performed in 119 healthy volunteers, 20 patients with Barrett's esophagus, 37 patients with moderate and 10 patients with severe reflux esophagitis without Barrett's esophagus. In all these diseases the final diagnosis was ascertained by means of endoscopy plus biopsy. There was no difference in the 24-hr and daytime patterns of gastric pH between healthy subjects and patients with Barrett's esophagus, while nocturnal acidity was significantly lower (P < 0.05) in the latter population. Gastric acidity, in contrast, was higher (P < 0.05) in controls than in patients with both moderate and severe reflux esophagitis without Barrett's esophagus during the whole 24-hr period. There was no difference between refluxer patients with and without Barrett's esophagus in any of the three time intervals we analyzed. Because normal subjects had lower gastric pH than patients with Barrett's esophagus during the night and than patients with reflux esophagitis during the whole 24-hr period, gastric hyperacidity is not a relevant factor in the development of both metaplastic columnar epithelium and inflammatory changes in the distal esophagus, and other pathophysiological mechanisms are involved in these histological alterations.

Fiberoptic technique for 24-hour bile reflux monitoring. Standards and normal values for gastric monitoring

Physiologic bile reflux was assessed in 27 in vivo test with healthy volunteers to define a standardized protocol and normal values for 24-hour enterogastric bile reflux monitoring (protocol with supine, upright, and meal phases and a free diet avoiding alcohol, smoking, and coffee, evaluation with different thresholds of absorbance units: 0.14, 0.25). In vitro tests with bile-sodium solutions demonstrated a linear dependence of absorbance for bilirubin up to 600 mumol/liter (range of the fiberoptic device: 0.0-1.0). Fluids and food might interfere with absorbances below 0.25 (exception: coffee). In vivo bile often remains in the stomach for more than 1 hr; these events were defined as reflux episodes. The upper limits for physiologic bile reflux are a percentage of total time of bile reflux of 28.2% and an average absorbance during a reflux episode of 0.62 (95th percentile with threshold 0.25). Comparing bile with pH monitoring (absorbance > 0.25 and/or pH > 4), an increase of bilirubin was found most frequently with constant pH (45%) or an increase of pH with constant bilirubin (36%). The hypothesis was drawn that bile and duodenal or pancreatic secretions may separately contribute to duodenogastric reflux.

Ambulatory oesophageal bile reflux monitoring in Barrett's oesophagus

Bile reflux has been implicated in the pathogenesis of Barrett's oesophagus but evaluation remains difficult. Bilitec 2000 is an ambulatory system that detects bilirubin based on its spectrophotometric properties. Oesophageal bile exposure was evaluated in three groups of patients. Group 1 (n = 11) were normal controls, group 2 (n = 13) were patients with uncomplicated gastro-oesophageal reflux and group 3 (n = 12) were patients with Barrett's oesophagus. Bile reflux was greater in patients with Barrett's mucosa than in controls or those with uncomplicated reflux. This difference was seen in the supine and interdigestive periods. The percentage of time at which gastric pH was greater than 4 and oesophageal pH was above 7 did not differ between the groups. Bilitec 2000 detects greater bile reflux in patients with Barrett's oesophagus. No corresponding gastric or oesophageal alkaline shift is found. This ambulatory bile reflux monitoring system may be a useful tool in clinical practice.

An ambulatory bile reflux monitoring system: an in vitro appraisal

Assessment of duodenogastric bile reflux has hitherto been unsatisfactory. An ambulatory system which utilizes the optical properties of bilirubin is examined. Test readings are correlated with the laboratory values for bilirubin in each of a number of physiological solutions. For dilutions of pure bile there was a linear correlation between absorbance and bilirubin concentration (r = 0.93, p < 0.001). In the more acidic environment of gastric juice there was also a linear correlation between absorbance and bilirubin concentration (r = 0.65, p < 0.001), but absorbance values were significantly higher than those of pure bile solutions (F ratio = 130, p < 0.0001). Normal gastric secretions and saliva give low absorbance values (0.02-0.04). Common foodstuffs such as soups, tea and coffee give higher readings and may interfere with bile reflux assessment in a clinical setting. Bilitec 2000 has potential for use as an ambulatory bile reflux monitoring system but consideration should be given to a standard diet or attention must focus on the fasting period.

Simultaneous gastroesophageal pH monitoring and the diagnosis of alkaline reflux

The diagnosis of alkaline gastroesophageal reflux (AGER) by esophageal pH-monitoring is elusive because several other causes may raise esophageal pH above 7. This study approaches the issue by simultaneously recording esophageal and gastric pH with a 2.1 mm assembly of two antimony electrodes in refluxing (n = 69) and nonrefluxing (n = 40) children. On the basis of the information gathered, patients were divided into four groups: acid refluxers (GER, n = 40 or 57%), alkaline refluxers (AGER, n = 8 or 11%), alkalacid refluxers (AAGER, n = 15 or 22%), and silent refluxers or false negatives (SGER, n = 6 or 9%). Children with AGER could not be differentiated from other refluxers nor from controls by esophageal pH information alone. Prolonged gastric buffering or alkalinization could be documented in both AGER and AAGER children, indicating extended duodenogastric reflux (DGR) in them. The incidence of esophagitis was not significantly different between the groups of refluxers: 75% for AGER and AAGER, 73% for GER, and 67% for SGER. AGER, a disorder involving both pyloric and gastroesophageal barriers, is as significant a problem in children as in adults and may lead to severe complications even in the absence of excessive acid exposure. Monitoring esophageal pH alone is insufficient for making the diagnosis of AGER. The diagnosis can be made with little more expense and similar patient discomfort using the two-probe test. We suggest the need for a critical reassessment of the current antacid and prokinetic GER treatments and of antireflux operations which leave DGR intact while correcting GER.