Can Computers Improve The Accuracy Of Ultrasonographic Gastric Emptying Assessment?

Ultrasound can be used to obtain a reliable determination of gastric emptying by measuring the antral sectional area which increases after a meal and then decreases progressively in a linear manner.This technique is non-invasive, inexpensive and safe.Some parameters, such as final emptying time are relatively easy to achieve even when performing evaluations every 30' (solidliquid meals).On the contrary, carrying out the procedure at these intervals, it can be difficult to achieve precise results for maximum antral distension (MAD) which is also very important because of its possible correlation with dyspeptic problems.An accurate evaluation of this parameter at shorter intervals would be very time-consuming.The aim of the study was to verify whether 30' interval determinations were sufficient to provide an accurate assessment of the MAD time, when a computer analysis, with suitable software, was performed.Twenty healthy volunteers (9f, 1 Ira; mean age ± SD: 31.4 ± 4.12) were enrolled.Ultrasonographic evaluation of the antral area was performed by measuring the longitudinal and the anteroposterior diameters corresponding to the sagittal plane passing through the superior mesenteric vein and using the formula r~AB/4, in fasting state, at the end of a standardized solid-liquid meal (900 Kcal) (To) and at 5' intervals for 90'.Statistical analysis was performed using t test.The values obtained in fasting state, at T 0, at 30', 60', 90' were analyzed by a software program (Maple V 2.0) using cubic-spline interpolation of the data of each patient; the curve was made by pieces of polynomials of degree 3 on each time step and was the smoothest in a natural geometric sense (1).Performing evaluations every 5' MAD was 20', whereas if determinations were only carried out at 30' intervals MAD would be reached at 30' (p<0.0001).The interpolation analysis, carried out by Maple V 2.0 program on these last data, showed that MAD was reached at 19.72'.Maple V 2.0 analysis performed over results obtained at 30'-intervals and the evaluations carried out every 5' provided the same results.Using a suitable software program achieves an accurate evaluation of MAD avoiding too frequent determinations of the antral area which are highly time-consuming for the operator; it provides more accurate information from the ultrasonographic gastric emptying assessment without further loss of time.