Journal of Chromatography B, 834 (2006) 183–187 Lactulose and mannitol intestinal permeability detected by capillary electrophoresis Rita Paroni a,∗ , Isabella Fermo b , Laura Molteni c , Laura Folini c , Matteo Rocco Pastore c , Andrea Mosca d , Emanuele Bosi c a Department of Medicine, Surgery and Dental Science, University of Milan, H San Paolo, via Di Rudini’ 8, 20142 Milan, Italy b Separative Techniques Unit, Vita-Salute San Raffaele University Hospital and Scientific Institute, Milan, Italy c General Medicine, Diabetes and Endocrinology Unit, Vita-Salute San Raffaele University Hospital and Scientific Institute, Milan, Italy d Department of Science and Biomedical Technology, University of Milan, Milan, Italy Received 5 October 2005; accepted 23 February 2006 Available online 10 March 2006 Abstract Aim of this study was to set up a method by capillary electrophoresis to detect lactulose and mannitol in urine after an oral load, and to estimate the intestinal permeability in controls and in type I diabetes patients. The underivatized carbohydrates were monitored by indirect UV detection using sorbate, cetyltrimethylammonium bromide and LiOH as background electrolyte. Urines were purified by solid phase extraction, shaken with cation exchange resin, filtered and analysed. Carbohydrates migrated in <10 min in relation to their pK a and M r . Controls (n = 33) and patients (n = 23) had an excretion ratio lactulose/mannitol 0.025 (0.018–0.051) and 0.067 (0.050–0.127), respectively (p < 0.01, median, interquartile range). © 2006 Elsevier B.V. All rights reserved. Keywords: Mannitol; Lactulose; Intestinal permeability; Urine; Capillary electrophoresis 1. Introduction The lactulose-mannitol (LAC-MAN) is a highly sensitive test for the screening of the diseases that affect intestinal permeabil- ity [1,2]. The test is based on the coupled oral administration of two sugar probes of different molecular weight and on the determination of their excretion rate in urine. These two small molecules are neither hydrolysed nor actively transported across the intestinal barrier. The sugar alco- hol MAN (∼5–7 ˚ A) permeates the intestinal mucosa via a tran- scellular pathway through the water-filled pores on the cell membrane, whereas the disaccharide LAC (∼10–12 ˚ A) uses a paracellular route through the intercellular junctional complexes between adjacent enterocytes and extrusion zones at the villous tip [1,3]. If the mucosal barrier is damaged as in celiac disease, Crohn’s disease, atopic dermatitis, cow’s milk proteins intoler- ance, cystic fibrosis, diarrhoea, HIV infection and diabetes, the gut is more permeable to intact sugars and proteins [3–8]. ∗ Corresponding author. Fax: +39 02 50323245. E-mail address: rita.paroni@unimi.it (R. Paroni). Therefore, an abnormal intestinal permeability with an unbal- ance in the di/mono saccharide (or sugar alcohol) excretion ratio in urine would indicate either an abnormal passive uptake of large molecules (sugars or antigens) through the paracellular route across the damaged mucosal barrier, or a reduction of the integral mucosa area with a reduced passage of the small molecules. So far, several techniques have been proposed to set up methods for quantitative determination of sugars in urine, but all of them present some limits. Thin-layer chromatogra- phy [9] and colorimetric/enzymatic procedures [10] are time- consuming and unspecific methods, HPLC must be coupled to a refractive index, pulsed amperometric, fluorescence or evap- orative light scattering detector [11–17], not always available in all analytical laboratories, while gas chromatography or gas chromatography–mass spectrometry needs a derivatization step [18–22]. Capillary electrophoresis (CE) is the technique of choice for the analysis of hydrophilic mono- and oligosaccharides carbohydrates, with an impressive number of different sepa- ration approaches, different detection modes (with or without pre-column derivatization) and different applications [23–33]. 1570-0232/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.jchromb.2006.02.050