Evaluation of mannitol for use as a probe marker of gastrointestinal permeability in man.

The absorption, metabolism, space distribution, renal excretion and natural occurrence of mannitol was studied in healthy human volunteers to determine its suitability for use as a probe marker of gastrointestinal permeability. Urinary recovery following oral loading was 8.0-39.6% of the ingested amount. Net absorption during intubation experiments in which a 30 cm segment of small intestine was perfused was 3.6% of perfusate mannitol. The slow rate of removal from the jejunum suggests that mannitol has a very low affinity for facilitated transport systems. Urinary recovery of intravenously injected mannitol approached 100% and renal clearance averaged 131 ml/min. The space distribution, progression of renal excretion and falling plasma concentration simulated that of lactulose rather than 3-0-methyl-D-glucose suggesting a mainly extracellular distribution. This was supported by measurement of erythrocyte penetration. Stool cultures degraded mannitol to products which included acid and carbon dioxide, when incubated aerobically and anaerobically. Excretion of mannitol, which was found to be naturally present in urine, was reduced but not abolished by fasting. Since mannitol occurs naturally in human urine we conclude that measurement of urinary mannitol following oral administration to assess intestinal permeability may be subject to error.