Objective: The accuracy of the C-13-octanoic acid breath test is enhanced bk breath sampling over 6 h rather than 4 h, but this increases the cost of the test. Our ain was to validate a less costly but accurate sequence of breath sampling for measuring gastric emptying of solids. Methods: We performance the C-13-octanoic acid breath test and tested its reproducibility relative to simultaneous scintigraphy in 30 healthy volunteers. Hit stilts: There was a significant but weak correlation between t(1/2) measured by the two tests (r(s) = 0.54, p < 0.005), but not between the duration of the lag phase. The differences in the t(1/2) measurements between the tests were different between subjects but were highly reproducible within subjects, Within- and between-subject variations of measurements of gastric emptying with the C-13-octanoic acid breath test were not significantly different from the variations observed with scintigraphy. A subset of 11 breath samples collected over 6 h (24 samples) predicted (r(2) > 0.95) the variables characterizing the cumulative appearance of (CO2)-C-13 in breath; these samples were at 35, 30, 95, 110, 140, 155, 215, 245, 260, 290, and 335 min, The accuracy of this subset of sampling times was confirmed in a separate set of breath test samples over 6 h from the same 30 subjects, Conclusions: The C-13-octanoic acid breath test for gastric emptying of solids is as reproducible as scintigraphy. A subset of 11 sampling times provides sufficient information to characterize the whole breath-test curve, but the sampling period should be extended to 6 h after dosing. (C) 1998 by Am. Coll. of Gastroenterology.