How reproducible is cutaneous electrogastrography? An in-depth evidence-based study

To check on reproducibility of parameters of the cutaneous electrogastrogram registered at a close or a distant time span.

Twenty-two volunteers recruited by an advertisement (11 females and 11 males, median age 25 years, range: 18-35) underwent three surface electrogastrography examinations of which two were taken on consecutive days and the third one was accomplished at least 2 weeks before or after the two other sessions. The examination involved a 30-min fasted recording, followed by a 90-min postprandial registration after intake of a 394-kcal mixed solid-liquid test meal.

Parameters of the electrogastrogram pertaining to the frequency of the gastric slow waves exhibited good to moderate reproducibility, whereas fair reproducibility characterized parameters expected to describe the power of gastric slow waves. With the exception of the difference fed minus fasted power (DeltaDP), in no instance was the medium term reproducibility any worse than the short term one. Categorical data analysis revealed that the relative time share of normogastria postprandially exhibited a better reproducibility than in the fasted period. The Cohen's kappa-value of 0.459 for the DeltaDP for the medium term reproducibility placed this parameter within the range of moderate agreement between repeat examinations. Of the two two-parameter combinations considered, the alliance of the fasted and fed normogastria performed worse than any of those parameters considered alone, whereas a combination of the DeltaDP with the fed-state normogastria revealed a kappa-value amounting to 0.510 for the medium term reproducibility.

The feasibility of some electrogastrographic parameters to convey clinically useful information may be hampered by their fair reproducibility. Recoding of parameters of the cutaneous electrogastrogram from primary continuous to secondary categorical may help achieve a better agreement between repeat examinations.