Letter: studies of salivary pepsin in patients with gastro-oesophageal reflux disease.

EDITORS,

We read with great interest the study by Race et al investigating the role of salivary pepsin in diagnosing gastro‐oesophageal reflux disease (GERD).1 Studies were performed on patients referred for consideration for anti‐reflux surgery and asymptomatic healthy volunteers.

All subjects were intubated for the duration of the study and no control was used to show that intubation was not a cause of reflux episodes.2, 3 For good experimental design, we would recommend having a separate control group without intubation.

The evidence of validation for the pepsin ELISA has a series of flaws. The authors have failed to understand the significance of using antibodies and recombinant proteins, made specifically as research tools used primarily in Western blot which are difficult to interpret due to lack of MW markers to identify pepsin and are not recommended by the manufacturer for diagnostic procedures. The manufacturer demonstrated that the antibody, although monoclonal, had specificity for proteins other than human pepsin, detecting a protein the wrong MW for pepsin A at 45 kDa. The MW of pepsin is 34 kDa and pepsinogen is 41 kDa.4

The data on the ELISA were not robust; the amount of pepsin present in gastric juice is approximately 0.9 mg/mL.5 Reflux into the airways would be diluted with a wide concentration range. The ELISA test results gave a dynamic range of 0‐100 ng/mL, with 90% less than 50 ng/mL and an average of 20 ng/mL. The expected range was not present with the ELISA, suggesting either pepsin was not being detected or a high‐dose hook effect had been exceeded. However, Peptest gave a range of 0 to >500 ng/mL with an appropriate standard deviation. This is more typical of a single analyte being measured.

Literature presented describing the use of salivary pepsin as a diagnostic marker for GERD was used to argue that there were no differences between patients and controls—this was misleading. Kim et al6 stated that all the healthy volunteers were negative for pepsin and Lannella et al7 stated that subjects belonging to the control group were pepsin negative, and likewise, Birtic et al8 came to similar conclusion. In another paper referenced by Race et al, they stated no difference between controls and GERD patients suggesting low reproducibility; this was not shown or stated in the paper by Du et al.9

Finally, in response to the comment on the expression of pepsin in the tongue used as an argument for the presence of pepsin in saliva was misleading. The evidence presented for pepsinogen expression referenced the Fantom 59 project. The Human Expression Atlas reports 9 studies on tissue expression of pepsinogen 3. Only the Fantom 5 project predicts any tongue expression. The other 8 had none. The level reported by Fantom 510 is four transcripts per kilo base millions and the ovaries is 14 141, a tissue not recognised as a major secretor of pepsin. Consequently, the level of expression reported in the tongue could not account for levels measured in saliva.

FUNDING INFORMATION

None.