Influence of urine volume in the assessment of intestinal permeability in affected
children by multiple sugar probes
Riccardo Addobbati1, Lorella Pascolo1,§,, Nicola Di Toro1, Giulia Barucca Sebastiani1, Stefano
Martellossi and Tarcisio Not1,2.
Supplemental Data
Figure S1
Figure S1. Calibration curves in gas chromatography-tandem mass spectrometry (GC/MS).
Calibration curves are generated for all the four sugars in a range from 0 to 400 µg/ml. In the
graphs, concentration rates are expressed as µg/ml, while “response ratio” is in arbitrary units from
the instrument. The concentrations of the patient sample are obtained comparing to these curves.
When sample concentrations give values above these calibrations (i.e. rhamnose concentration in
control subjects with a diuresis lower than 250 ml,), the initial samples were diluted 1.2 and
measured again.
Figure S2
Figure S2. Diuresis volume effect on sugar permeability in healthy subjects.
Panels A: ROC curve simulations assigning different cut-offs to the volume variable and testing
rhamnose recovery below and over the limits in the healthy population. As shown in the inset at the
cut-off of 240 ml the ROC area (AUC) is 0.855, similarly to that at 200 ml. Higher volume cut off
reduces the AUC values. B, C and D show the percentages of recovery vs the different urine
volumes of all the healthy subjects in the study collected at 5 hours, for lactulose, sucralose and
sucrose, respectively. A clear correlation for rhamnose recovery vs volume results in panel A. sis
the Spearman correlation parameter of the correlation; p shows the statistical significance. No
statistically significant correlation was found for the other sugars. Recovery has been calculated as
described in material and methods section.
Figure S3
Figure S3. Diuresis volume effect on sugar permeability in patients.
Panels A, B, C and D show the percentages of recovery of rhamnose, lactulose, sucralose and
sucrose (respectively) vs the different urine volumes collected at 5 hours from 86 patients.
Statistically significant correlation is calculated for rhamnose recovery vs volume results in panel A.
(r) is Spearman correlation parameter and p values are referred to correlation significance. Similar
correlation is found for sucralose (panel C) and sucrose (panel D) recoveries vs. volume. No
statistically significant correlation is founding the case of lactulose (panel B). Recovery has been
calculated as described in material and methods section.
Figure S4.
Figure S4. Rhamnose / Lactulose ratio in healthy subjects and patients.
The ratio between lactulose and rhamnose is calculated for the control subjects and patients, divided
accordingly to urine volume range (≤ 240 and >240). Two subgroups of patients are in evidence:
Chron and RCU diseased patients. Graph represent confidence value ranges (bars), median values
(line inside box), media (+) and statistical significance for differences among groups (star
symbols). Each volume subgroup of patients, Chron and RCU is compared to the respective volume
subgroup of healthy subjects. (**) corresponds to p<0.01; (***) to p<0.001. Statistical evaluations
for the variation among groups assign a very high significance comparing both Crohn’s disease and
UC affected patients (p<0.001) vs. healthy subjects, supporting a good clinical predictivity of the
test.