1. The relative effects of changes in mucosal surface area and mucosal permeability on the passive uptakes of mannitol and raffinose have been studied in vitro using jejunal biopsies from 48 controls, 32 patients with coeliac disease and 11 patients with Crohn's disease. Total mucosal permeation was corrected for surface area measured morphometrically to provide an index of mucosal permeability.
2. In untreated coeliac disease, permeation of mannitol was reduced by 35% (P = 0.006) and that of raffinose was increased by 66% (P = 0.0095) compared with controls, whereas mucosal permeability to mannitol was increased twofold (P = 0.009) and to raffinose fivefold (P = 0.0001). Mucosal permeability was similar for each sugar.
3. In treated coeliac disease, permeation and permeability for mannitol were normal, but remained elevated for raffinose by 23% (P = 0.036) and 41% (P = 0.024), respectively.
4. In Crohn's disease, permeation of mannitol was reduced by 21%, but that of raffinose and mucosal permeability to both sugars were normal.
5. These findings suggest that surface area is quantitatively more important than mucosal permeability in determining the total permeation of mannitol, while the converse is true for raffinose. The findings are compatible with paracellular uptake of raffinose, but with both paracellular and transcellular uptake of mannitol. Both pathways are affected in coeliac disease, whereas only transcellular uptake is affected in Crohn's disease.