Detection of bile reflux: in vivo validation of the Bilitec fibreoptic system

The Bilitec ambulatory bile reflux monitor is advocated for the assessment of bile reflux. However, it has only been validated by bench top (in vitro) studies. This in vivo study used controlled 'iatrogenic' episodes of bile reflux to determine the ability of the Bilitec system to detect bile acids in the oesophageal lumen. At least 10 ml of fluid was aspirated from the stomach of each of 32 patien ts, and analysed quantitatively for total bile acids (TBA) and total biliru bin (TB). An in vitro assessment of this fluid with the Bilitec probe was p erformed initially. The Bilitec probe was then sited in the distal oesophag us, and the gastric fluid was introduced into the mid-oesophagus, through a naso-oesophageal tube, as a series of 10-ml volumes of varying dilutions ( 1:8, 1:4, 1:2, undiluted) to simulate gastro-oesophageal reflux events. Int raoesophageal bilirubin absorbance was compared with TBA and TB for each di lution. Both TBA and TB values correlated strongly with bilirubin absorbance in vit ro (r=0.83, p < 0.0001 and r=0.82, p < 0.0001 respectively). However, this relationship was weaker in vivo (r=0.64, p=0.0001 and r=0.68, p < 0.0001 re spectively). Of the dilutions which contained potentially injurious concent rations of bile acid (> 1 mmol/L), 77% were associated with absorbance > 0. 14 (currently accepted absorbance threshold for oesophageal bile reflux). Although in vitro sensitivity of the Bilitec probe to bile acids was excell ent, in vivo sensitivity was less reliable, and 23% of significant 'bile re flux' episodes did not exceed the currently recommended Bilitec absorbance threshold for bile reflux. In contrast, false-positive results were unlikel y. These findings suggest that the overall accuracy of the Bilitec system i s probably sufficient for clinical use, although this method is not suffici ently reliable to be regarded as the gold standard for the evaluation of du odeno-oesophageal reflux.