Evaluation of an isotope ratio method for measurement of cholesterol absorption in man

Recently an isotope ratio method (IRM) was developed for measuring cholesterol absorption in rats by analysis of radioactivity in peripheral blood (Zilversmit, D. B. 1972. Proc. Soc. Exp. Biol. Med. 140: 862-865). To validate it in man we have compared cholesterol absorption by a fecal radioactivity method (FRM) with that simultaneously measured by IRM in 14 patients (15 experiments) hospitalized on a metabolic ward. Cholesterol absorption by FRM was assayed as fecal recovery of orally administered [(14)C]cholesterol, after correction with markers for fecal flow (chromic oxide) and cholesterol degradation (beta-sitosterol). Simultaneously, [(3)H]cholesterol was administered intravenously, and the dose-normalized ratio of [(14)C]- to [(3)H]cholesterol was repeatedly assayed in plasma. After 72 hours the ratio became constant in each patient and remained so for as long as 63 weeks (five additional outpatient studies). In three patients the fecal data were unsatisfactory because of poor recoveries of chromic oxide and radioactive cholesterol. In the remaining 11 patients (12 experiments) the mean cholesterol absorption by IRM was 42.1% (range 15.7-62.9%) and by FRM 36.6% (range 13.8-58.8%). There was good to excellent agreement between the two methods in the same patient, except in one experiment. Statistical analysis of these 12 comparisons by estimating confidence intervals showed that we can be 95% confident that the two absorption methods will produce results within 5 percentage points, and 99% confident that the differences are less than 7 percentage points. Although we conclude that IRM affords results that are concordant with those obtainable by earlier validated methods, we urge that its suitability for outpatient studies be further examined in more extensive trials.