374 Gut 1996; 39: 374-381 Evaluation of differential disaccharide excretion in
urine for non-invasive investigation of altered Gut: first published as 10.1136/gut.39.3.374 on 1 September 1996. Downloaded from intestinal disaccharidase activity caused by ox-glucosidase inhibition, primary hypolactasia, and coeliac disease
I Bjarnason, R Batt, S Catt, A Macpherson, D Maxton, I S Menzies
Abstract Intestinal disaccharidase deficiency, either Background/Aim-The reliability of a primary or secondary, is a frequent cause of quantitative method for the non-invasive gastrointestinal symptoms and it is reported assessment of intestinal disaccharide that as many as 50% to 90% of adult non-white hydrolysis was assessed. subjects may be lactase deficient.' Con- Methods-Differential excretion of intact ventional non-invasive assessment of intestinal disaccharide, expressed as ratios of lactu- disaccharidase deficiency is open to some lose to appropriate hydrolysable disacchar- criticism. The oral lactose tolerance test entails ides in urine collected following combined ingestion of non-physiological quantities of ingestion, has been investigated in healthy lactose (50-100 g, equivalent to 1-2 litres of volunteers with drug induced a-glucosidase milk) or other disaccharide, often accompanied inhibition, in subjects with primary hypo- by unpleasant symptoms.4 The test has a high lactasia, and patients with coeliac disease. false positive rate, thus a rise in blood glucose Results-Oral administration of the below 20 mg/100 ml (1.1 mmol/l), which is (K-glucosidase inhibitor 'Acarbose' (BAY taken to indicate disaccharidase deficiency4 g 5421, 200 mg) together with sucrose and may have alternative explanations. The high lactulose increased the urinary sucrose/ false positive rate can be reduced somewhat by lactulose excretion ratios (°/0 dose/10 h) intra-duodenal instillation of disaccharide but fivefold. The effect was quantitatively at the cost of losing the non-invasive nature of reproducible, a higher dose of 'Acarbose' the procedure.5 Alternatively, with respect to (500 mg) increasing the excretion ratio to hydrolysis of lactose, it has been suggested that http://gut.bmj.com/ about 1.0 indicating complete inhibition of measurement of the rise of blood galactose intestinal sucrase activity. The suitability might be preferable to that of glucose.6 ofthe method for measuring differences in Furthermore, as renal excretion of galactose is dose/response and duration of action was efficient, some authors have advocated the assessed by comparing three different measurement of galactose, or of a lactose/ a-glucosidase inhibitors (BAY g 5421, galactose ratio, in urine after ingestion of BAY m 1099, and BAY o 1248) and found lactose.7 8 Methods involving breath hydrogen on September 30, 2021 by guest. Protected copyright. to be satisfactory. Subjects with primary or `C-labelled carbon dioxide after ingestion Department of Clinical adult had urine lactose/ of appropriate disaccharide test doses5 9 10 are Biochemistry and hypolactasia Medicine, lactulose excretion ratios raised to values beset with somewhat similar problems. Diffi- King's College School indicating reduced rather than complete culties with these techniques are in part ofMedicine, absence of lactase activity whereas because the uptake and metabolism of mono- London I Bjarnason sucrose/lactulose ratios were not signifi- saccharide products, or generation ofhydrogen A Macpherson cantly affected. 'Whole' intestinal disac- in the bowel depend upon several other I S Menzies charidase activity assessed by this method variable factors beside the rate of hydrolysis by Department ofSmall demonstrated impairment of lactase, brush border disaccharidases.' 11-13 Animal Medicine and sucrase, and isomaltase in eight, one, and Diagnosis of disaccharidase deficiency can Surgery, The Royal seven, respectively, of 20 patients with also be made by in vitro assay of homogenised Veterinary College, University ofLondon coeliac disease. By contrast in vitro assay jejunal tissue. It is necessary to consider, R Batt of jejunal biopsy tissue indicated pan- however, how well these assays represent the in disaccharidase deficiency in all but five of vivo performance of the whole intestine, Department of Chemical Pathology, these patients. This shows the importance especially in coeliac disease where the severity St Thomas's Hospital of distinguishing between 'local' and of the disease is usually maximal in the proxi- Medical School, 'whole' intestinal performance. mal jejunum,4 ileal functions being com- London Conclusions-Differential ex- unaffected. S Catt urinary paratively D Maxton cretion of ingested disaccharides provides A further non-invasive technique has been Correspondence to: a reliable, quantitative, and non-invasive proposed that avoids error due to variations in Dr I Bjamason, technique for assessing profiles of monosaccharide transport and metabolism, Department of Clinical Biochemistry, King's College intestinal disaccharidase activity. and may therefore permit a more specific and School of Medicine and (Gut 1996; 39: 374-381) quantitative assessment of intestinal disacchar- Dentistry, Bessemer Road, 1 12 London SE5 9PJ. idase activity. 15 The principle of this Accepted for publication Keywords: coeliac disease, intestinal function, method, which entails measurement of the 9 May 1996 intestinal disaccharidases, intestinal permeability. renal excretion of several disaccharides Differential disaccharide excretion and intestinal disaccharidase activity 375
following their combined ingestion, is based on non-hydrolysable disaccharide should relate the finding that recovery of intact disaccharide inversely to the activity of intestinal disacchar- in urine after ingestion is inversely related to ides (lactase, sucrase and, in the case of pala- the rate of their intestinal hydrolysis. The test tinose, isomaltase).11 12 15 This technique has solution contains a non-hydrolysable 'refer- been used to demonstrate lactase deficiency Gut: first published as 10.1136/gut.39.3.374 on 1 September 1996. Downloaded from ence' disaccharide (for example, lactulose) associated with rotaviral infection and com- together with hydrolysable disaccharides (lac- bined sucrase and isomaltase deficiency in tose, sucrose, palatinose). As renal excretion of asucrasia. ' 12 15 The main purpose of this these disaccharides after entering the circu- study was further evaluation of the reliability of lation is known to be complete, without meta- the urine disaccharide ratio technique for both bolic loss, the excretion ratios of hydrolysable/ clinical and research purposes. This has been undertaken by assessing the ability of the differential disaccharide excretion technique to Variables Behaviour of markers distinguish the differences in dose/response and duration of action of three orally admin- Delivery istered ox-glucosidase inhibitors and, secondly, Lactose log to quantitate intestinal lactase activity in 11Coent of test solution [L Sucrose 10g Lactulose L-Rhamnose Palatinose 10g 6.7 g 1.0 g subjects with confirmed genetic hypolactasia (primary adult hypolactasia). Intestinal disac- 2 Ingestion/regurgitation [> 2. charide hydrolysis assessed in vivo by the 3 Gastric emptying > [ combined disaccharide ratio test was also compared with in vitro assay of disaccharidase Digestion activity in homogenised jejunal biopsy tissue in ydrolysis by disaccharidases v patients with coeliac disease to study the effect > of a disorder that mainly involves the jejunum. Intestinal permeation 5 Dilution by secretions > <> Methods 6 Rate of transit > <> |7 Mucosal area >
> [ of a non-hydrolysable disaccharide and hy-
11 Urine collection http://gut.bmj.com/ 2 drolysable disaccharides can give quantitative Interpretation of urinary ratios information of the rate of intestinal hydrolysis Lactase activity LAC/LACL of some disaccharides. Sucrase activity SUC/LACL Isomaltase activity PAL/LACL STUDIES WITH Ca-GLUCOSIDASE INHIBITORS Fourteen healthy adult white volunteers were Mucosal permeability LACL/RHAM on September 30, 2021 by guest. Protected copyright. studied. A strict sucrose free diet was imple- mented 18 hours before and throughout the test period. After an overnight fast each subject Nil [ | | = voided urine directly before the test to confirm Involvement -= + absence of sucrose. At 8 am the test solution containing 20 g sucrose (a-D-glucopyranosyl ,B-D-fructofuranoside) and 6-7 g lactulose (, 1-4 galactopyranosyl-fructose= 10 ml Figure 1: Principle ofthe differential urine excretion oforally administered test substances. 'Duphalac' lactulose syrup, Duphar Labora- The precise quantity ofintact disaccharide passing by unmediated diffusion across the tories, Southampton) dissolved in 300 ml intestinal mucosa into the circulation is determinted by thefactors shown in the Figure. The water (300 mmol/l) was within a inclusion oflactulose, which resists the action ofintestinal hydrolases in a test solution ingested containing hydrolysable disaccharides (lactose, sucrose orpalatinose either individually or in period of four minutes. Food and fluids were combination), enables a correctionfor most non-mucosal variables on urinary excretion to not permitted until 2-5 hours later. A complete be obtained. Thesefour disaccharides permeate the intestine as intact molecules by a urine collection was made from 8 am to 6 pm common mucosalpathway in quantities that depend upon their individual concentrations within the intestine. In turn, these concentrations are determined by the efficiency of (10 hours), the volume recorded, and an hydrolysis by appropriate disaccharidases. As lactulose will be affected in the same way as aliquot preserved with merthiolate (10 mg/100 the 'test' disaccharides by all thesefactors exceptfor the activity ofmucosal disaccharidase, ml minimum) for analysis of sugars by quan- which this disaccharide resists, the urinary excretion ratios oflactose, sucrose, orpalatinose, or all three, to lactulose can be regarded as specific indices ofthe efficiency with which the titative paper or thin layer chromatog- concentrations oflactose, sucrose, andpalatinose are reduced by hydrolysis within the small raphy." 17 18 Results are expressed as urinary intestine. In the absence ofhydrolysis oflactose thefraction ofthe ingested dose permeating sucrose/lactulose ratios of percentages of the the intestine will approach that oflactulose (which resists hydrolysis) so that the lactose! lactulose ratio ofpercentages ofthe oral dose excreted in urine would be 1.0. With normal oral doses excreted during 10 hours. (active) intestinal hydrolysis oflactose this ratio is much lower, usually between 0 1-0 3, and intermediate valus provide a quantitative measure ofthe degree ofimpaired hydrolysis. Nil=No effect. +=Plays a part in the overall urine excretion ofthe test substance but the effect on simultaneously ingested test substances is equal. ++=Plays a part in the overall Dose/response studies urine excretion ofthe test substance but the effect on simultaneously ingested test substances To evaluate the reliability of the method for differs. the quantitative differentiation of intestinal 376 Bjarnason, Batt, Catt, Macpherson, Maxton, Menzies
disaccharidase activity a study of the effect of ratios (of percentages excreted). Lactose/ different dose levels of a known a-glucosidase galactose urine excretion ratios were also inhibitory drug (BAY g 5421, 'Acarbose') was calculated to assess whether this ratio gave a
undertaken. Subjects underwent a control test better discrimination of hypolactasic from Gut: first published as 10.1136/gut.39.3.374 on 1 September 1996. Downloaded from without a-glucosidase inhibition (n= 14), which normolactasic subjects than lactose or was then repeated (with intervals of at least one galactose alone as has previously been week) adding either BAY g 5421 (Acarbose: suggested.7 8 dose range 25-500 mg, n=4-8), BAY m 1099, (dose 100 mg, n=7) or BAY o 1248 (dose 10 mg, n=6) to the test solution. One subject INVESTIGATION OF PATIENTS WITH COELIAC underwent 13 tests with BAY g 5421 in the DISEASE dose range 0-500 mg added to a test solution Nineteen patients undergoing jejunal biopsy as that contained lactose (,-1-4-D-galacto- a part of routine gastrointestinal investigation, pyranosyl-oL-D-glucose) 20 g in addition to in whom jejunal histology was normal and sucrose 20 g and lactulose 6.7 g dissolved in significant intestinal pathology not discovered, 300 ml drinking water, to assess response in served as a control group. The mean age ofthis relation to both sucrose and lactose hydrolysis. group was 41 years (range 21-67 years). Eleven patients with symptomatic, un- treated, and nine with treated coeliac diseases Duration ofaction were studied, the latter having achieved a full To demonstrate the ability of the method to clinical recovery with a gluten free diet of mean resolve differences in the duration ofdisacchar- duration six years (range nine months-12 idase inhibition the action of the three years). The mean age of the whole group was ox-glucosidase inhibitors on intestinal sucrose 54 years (range 28-74). Each patient was hydrolysis over a timed period was also admitted to a metabolic research ward for assessed in a single individual as follows: (a) investigation, in vivo studies and jejunal BAY g 5421 ('Acarbose' 500 mg) with, and biopsies being performed within a week of each then 30, 60, 120 minutes and 48 hours before other. oral administration of the test solution. (b) BAY m 1099 (200 mg) with, and then 30, 60, 120 minutes and 48 hours before the test Assessment ofintestinal disaccharide hydrolysis by solution. (c) BAY o 1248 (20 mg) with, and urinary excretion ratio then 60, 120, 180 minutes and 48 hours before Dietary sources of sucrose, lactose, and lactu- the test solution. lose were excluded for 18 hours before and throughout the test period. Each subject fasted overnight and voided urine directly before the http://gut.bmj.com/ INVESTIGATION OF SUBJECTS WITH PRIMARY test, a 'baseline' urine being retained to (GENETIC) ADULT HYPOLACTASIA exclude the presence of disacchariduria. At 8 Twenty three adult white volunteers acted as am each subject ingested a 300 ml test solution normolactasic control subjects, 16 Asian or containing: Lactose: 10 g (to assess lactase Afro-Caribbean together with two white activity), Sucrose: 10 g (to assess sucrase subjects comprised the hypolactasic group. activity), Palatinose: 10 g (to assess palatinase Hypolactasia was independently established by (isomaltase) activity (cx-1-6 D glucopyranosyl- demonstrating a positive conventional lactose D-fructofuranose, isomaltulose), Lactulose: on September 30, 2021 by guest. Protected copyright. tolerance test (blood glucose rise <20 mg/100 6.7 g (as non-hydrolysable reference for disac- ml and a diarrhoeal response after ingestion of charide permeation, also to assess 'large pore' 50 g lactose) and absence of intestinal disease permeation), L-rhamnose: 1 g (to assess 'small apart from milk intolerance. A lactose and pore' permeability). sucrose free diet was implemented 18 hours Incorporation of L-rhamnose permits simul- before, and throughout, each test. After an taneous assessment of intestinal permeability overnight fast each subject collected a baseline by measurement of lactulose-/L-rhamnose urine sample, which ensures that errors of urine excretion ratios, but as the data from this interpretation due to inadequate dietary study did not expand on the already extensive restriction, unsuspected lactulose medication, published literature these results are not endogenous lactosuria of pregnancy, etc, are presented.'3 avoided. Patients then ingested a test solution Foods and fluids were permitted 2.5 hours containing lactose (20 g) sucrose (20 g) and after ingestion of the test solution. lactulose (6-7 g) dissolved in 300 ml water. A Urine was collected for 0-5 and 5-10 hours, complete urine collection was then made in with merthiolate (10 mg/100 ml) as a preserva- two consecutive five hour periods (that is, 0-5 tive, for marker analysis as described.'7 18 hours and 5-10 hours). The urine volumes were recorded and aliquots preserved (10 mg merthiolate/100 ml minimum) for quantitative Assessment ofintestinal disaccharide activity by in analysis of lactose, sucrose, lactulose, and vitro assay galactose by quantitative paper chromatog- Intestinal biopsy specimens were obtained by raphy or an adaptation of this method to thin intubation technique with an adult Watson layer chromatography."` 17 18 Results were biopsy capsule, under radiological guidance, expressed as percentage of the dose of each just distal to the ligament of Trietz. A portion sugar excreted in urine during 0-5 or 0-10 of the biopsy specimen was allocated for hour period and as lactose-sucrose/lactulose histological examination, the remainder being Differential disaccharide excretion and intestinal disaccharidase activity 377
Cl) 1.0 0
-00C, 0.8 0 Gut: first published as 10.1136/gut.39.3.374 on 1 September 1996. Downloaded from
0 0.6-
0. 0
C.)
U1) 0 0 0.1 0.2 0.3 0.4 0.5 0 0.1 0.2 0.3 0.4 0.5 0 10 20 30 40 50 Orsal Bayg 5421 (g) Bay m 1099 (g) Bayo 1248 (mg) Figure 2: Use ofthe disaccharide ratio method to compare the dose/response relation ofthree different a-glucosidase inhibitors on intestinal sucrose hydrolysis. This is assessed by measuring the excretion ratio ofsucrosellactulose (% dose in a 10 hour urine collection) foUlowing ingestion ofthe disaccharide test solution together with increasing dose ofBAYg 5421 (Acarbose), BAY m 1099, and BAYo 1248. Equivalent effects were produced by 0.2 g, 0 1 g, and 10 mg doses ofthese inhibitors, respectively.
'flash' frozen in liquid nitrogen and stored at Results -70°C. No more than two minutes lapsed between firing the capsule and freezing the STUDIES WITH 0L-GLUCOSIDASE INHIBITORS samples. Lactase and sucrase activities in Figure 2 shows the 10 hour urine sucrose/ homogenised biopsy tissue were measured by lactulose excretion ratios obtained when the a micro-assay as described previously.`9 The test solution had been ingested without, and same method was used for measurement of then together with progressively increasing isomaltase activities using palatinose as the doses ofthe three aL-glucosidase inhibitors. The substrate, dissolving 0056 mol/litre in 0.1 M normal control 10 hour sucrose/lactulose sodium maleate buffer, pH 6-0. Enzyme units excretion ratio (mean SD)) was 0*12 (0. 10), were expressed as units/mg protein. All range 0 05-0 30). subjects gave informed consent and the studies With increasing amounts of simultaneously were approved by the Harrow Health ingested BAY g 5421 (Acarbose 25-500 mg) Authority and St Thomas's Hospital Ethical there was stepwise and significant (p<0.01) http://gut.bmj.com/ Committee. increase in urine excretion ratio of sucrose/ lactulose with a mean of 0.61 (0.12) and 0.91 (0.09) for 200 and 500 mg BAY g 5421, STATISTICS respectively. Ingestion of the two other Statistical difference between sequential data a-glucosidase inhibitors also produced a signi- were assessed by the paired Student t test, ficant (p<0 01) rise in urinary sucrose/lactulose otherwise Wilcoxon's rank sum test was used. excretion ratios. The relation between dose on September 30, 2021 by guest. Protected copyright. and response was very different, approximately Surcrose/lactulose Lactose/lactulose similar effects being produced by 200 mg BAY g 5421, 100 mg BAY m 1099, and 10 mg BAY 1.0 r = No hydrolysis 0 1248, giving mean sucrose/lactulose ratios of 0.61 (0.12), 0.58 (0.13), and 0 55 (0.08), 0 0 respectively (see Fig 2). 0 Figure 3 shows sucrose/lactulose and Uo- 0 lactose/lactulose responses obtained from a ~ c.o single subject receiving nine dose levels ofBAY 0 g 5421 (range 10-500 mg). The progressive 0X 0 0.5 [- 0 rise in the sucrose/lactulose excretion ratio .o S X U) shows satisfactory resolution between different a) a) levels ofax-glucosidase inhibition while absence a) COm) ._ 0 of a significant change in lactose/lactulose D a) 4 excretion ratio provides evidence of specificity. a) 0 h- 0.0 0@ Figure 4 shows the rate at which the * 0 ct-glucosidase inhibitory action of the three 0 t drugs declines. Whereas the sucrose/lactulose 0 100 200 300 400 500 0 100 200 300 400 500 excretion ratio after ingestion of BAT g 5421 Dose (mg) of Bay g 5421 'Acarbose' (ax-glucosidase inhibitor) and BAY m 1099 fall offrapidly, reaching 30% ofthe initial value when the test dose is delayed Figure 3: Dose response to the oL-glucosidase inhibitor BAYg 5421 (Acarbose). Ten hour urinary disaccharide excretion recordedfrom a single volunteer in response to co-ingestion of by 30 minutes, the effect of BAY o 1248 is sucrose/lactose/lactulose test solution together with nine different dose levels ofBA Yg 5421. more prolonged, taking three hours to reach This shows a steady rise in the sucrose/lactulose ratio as the oral dose is increased, the 30% level. The resolution obtained seems confirming that the action ofthe inhibitor on intestinal sucrase activity is dose dependent. Absence of a significant change in the lactose/lactulose excretion ratio is evidence ofthe to be satisfactory for investigating the duration specificity of inhibitory action. of action of such drugs. 378 Bjarnason, Batt, Catt, Macpherson, Maxton, Menzies
_ 100 lactase activity. Sucrose excretion did not differ 0) significantly between the groups and, although 00 lactulose excretion was less in hypolactasic sub- 80 - jects, this did not reach a statistical significance. Gut: first published as 10.1136/gut.39.3.374 on 1 September 1996. Downloaded from 2- C The 10 hour urinary excretion ratio of lactose/ 0 6 lactulose differed significantly between the ~CO groups and none of the 18 subjects defined as hypolactasic by other criteria had a ratio that was 40 below the mean (2 SD) (0.24) for the normo- lactasic group. Alternatively, although the dis- crimination ofhypolactasic subjects by excretion 20- of lactose or galactose is incomplete in 10 hour urine samples, combination of these measure- 0) ments, which respond to reduced lactase activity 0 in opposite directions, as lactose/galactose ex- 0 30 60min 2h 3h 4h 48 h cretion ratios provides almost complete discrimi- Time interval between ingestion of inhibitor nation. Sucrose/lactulose ratios in Figure 5 show and SUC/LAC test solution that eight of the hypolactasic subjects have Figure 4: Use ofthe differential disaccharide excretion method to compare short--time higher values than any of the normolactasic sub- duration ofdifferent a-glucosidase inhibitors on intestinal sucrase activity. Equivalent doses jects, which could suggest a coexistent impair- ofBAYg 5421 (open circles), BAYm 1099 (closed circles), and BAYo 1248 triangles) were ingested by a single volunteer at the same time as and, on separa(teoccasions, ment of sucrose hydrolysis. However, this seems 30, 60 minutes and two, three, and 48 hours before, oral administration oftheedisaccharide to be the outcome of a decrease in lactulose rather test solution. A 100% 'inhibition' score was allocated to the sucrose/lactulose exccretion ratio than an increase in sucrose excretion, and more obtainedfollowing simultaneous administration oftest solution and inhibitor, anid in drug action seen after the lapse ofdifferent periods oftime were expressed as tl changes likely to be the consequence ofosmotic accumula- ofthe initial dose that would be required to produce the observed sucrose/lactulose ratio ifit tion of fluid due to retention of unabsorbed had been given simultaneously with the test solution. The latter value was caleulated by lactose (due to hypolactasia) in the intestine, extrapolation from graph relating dose responsefor each inhibitorfrom the same subject. lactulose permeation being reduced by the result- ing dilution and hurry.20 Sucrose uptake would GENETICALLY DETERMINED HYPOLAC'TASIA not be affected to the same extent because, unlike Ten hour urine excretions of lactulos;e, lactose, lactulose, it is largely hydrolysed before reaching sucrose, and galactose and the calculated the lower part of the small intestine where these lactose-sucrose/lactulose and lactose/g.alactose of osmoticaly induced effects are most pronounced. two groups of healthy subjects, one wiith normal intestinal lactase and the other with,genetically determined hypolactasia, are shown ini Figure 5. DISACCHARIDE HYDROLYSIS IN PATIENTS WITH The five hour urinary data (not shown) showed COELIAC DISEASE less discrniination between normal cc)ntrols and http://gut.bmj.com/ hypolactasic subjects. Subjects with h)ypolactasia Non-invasive in vivo assessment excreted significantly (p<0 01) more 1actose and Figure 6 shows the 10 hour urinary excretion of less galactose than those with active intestinal the four disaccharides and the calculated lactose/
Lactulose Lactose Sucrose Galactose LAC/LACL SUC/LACL LAC/GAL on September 30, 2021 by guest. Protected copyright. 1.5 r 1.0 r 5 r-
0 0.8 F- 4 .
0 1.0 K 0 co 0 0.6 [- 3 _- 00 0 W 0 0.0 00 0 0 0@ - 00 #0L #0 Lo00 01 0.4 H 000 2 - 0% 00 0.5 K 0%0o *0 0 we 0 0
0 0.2 F- 1 00-0 0 00 0 0 0 -(.C-&-0 0g 0 -0000C6 O L 0 ' 0 _ Mean 0.68 0.47 0.09 0.21 0.08 0.08 0.21 0.09 0.14 0.48 0.12 0.19 0.48 0.27 (SD) (0.26) (0.23) (0.03) (0.07) (0.03) (0.03) (0.09) (0.03) (0.05) (0.14) (0.04) (0.07) (0.25) (2.07)
Figure 5: Ten hour urine excretion (% dose) oflactulose, lactose, sucrose, and galactose and the lactose-sucrose/lactulose and lactose/galactose urine excretion ratios in normolactasic (open circles) and hypolactasic (filled circles) subjects after an oral test solution containing lactose (20 g), sucrose (20 g), and lactulose (6-7 g). Differential disaccharide excretion and intestinal disaccharidase activity 379
1 a) ~00 1.8 Gut: first published as 10.1136/gut.39.3.374 on 1 September 1996. Downloaded from ,1.6 0.8 F- a) CO -_ 0 ,ccc 1.4 C.) 0 In 1.2 a1) 0.6 F- 0 x c 1.0 ._ 0 cJ
° 0.8 0.4 - CU 0.6 X 0
0 0. oc 0.2 0.2 H
0
0
Mean 0.499 0.819 0.66210.103 0.381 0.26310.091 0.150 0.121 0.078 0.159 0.175 0.227 0.366 0.38610.197 0.216 0.190 10.169 0.284 0 (SD) 0.200 0.529 0.38510.035 0.260 0.35610.045 0.139 0.109 0.049 0.175 0.249 0.065 0.201 0.29710.080 0.168 0.092 0.073 0.255 0 Figure 6: Ten hour urine excretion ofdisaccharides and ratios oflactose-sucrose-palatinose/lactulose from controls (open circles), untreated (filled boxes), and treated (filled circles) patients with coeliac disease.
lactulose, sucrose/lactulose, and palatinose/ levels were 76 (8), 107 (9), and 34 (2) units/mg lactulose excretion ratios from patients without protein, respectively. Corresponding values for gastrointestinal disease and patients with patients with coeliac disease were, 11 (3), 36 untreated and treated coeliac disease. Only the (5), and 11 (2), all of which differed signifi- differential urinary excretion of lactose/lactulose cantly (p<0.001) from controls. Two treated from patients with untreated coeliac disease patients had normal activity for all three disac- differed significantly from controls. charidases and three patients each had normal (p=0047) http://gut.bmj.com/ Eight, one and seven of the coeliac group were activity of a single disaccharidase. found to have varying degrees of defective There was no significant (p>O 1) correlation lactose, sucrose, and palatinose hydrolysis, between the lactose/lactulose urine excretion respectively. The five hour urinary data showed ratio (0-10 hours) and lactase activities by in less discrimination (data not shown) between vitro assay within the control group (r=-040) or controls and patients with coeliac disease. collectively for all the subjects studies (r=0 28). Similarly, there was no significant (p>0.5) corre- lation between sucrose/lactulose and palatinose/ on September 30, 2021 by guest. Protected copyright. In vitro assay ofjejunal biopsy tissue lactulose urine excretion values and in vitro Figure 7 shows the in vitro lactase, sucrase, and sucrase (r=0.00) and palatinase (r=0.08) palatinase (isomaltase) activities. The control activities in the whole group of patients.
200 r 250 r 60 0 0
c 50 200 0 0 0 150 h 0 0.m 0 E 40 0 0 0 :D 0 oO 150 0 00 Oo 00 E 0 CA 100 30 -0 a) 9 0 0 0 0 0 0 C.) 0 100 0 0 '._-Cu 0 0 ci a) 20 8o 0. 0
80 0 E(c 50;- 0 0 N o80 0 0 * 000 0r 0 50 C_ 0o a 10 wU :- 00 0 0 @00 a aEam. *- *E* 0 0 0 Lactase Sucrase Palatinase Figure 7: In vitro lactase, sucrase, andpalatinase activity injejunal biopsy samplesfrom controls (open circles) andpatients with untreated (filed boxes) and treated (filled circles) coeliac disease. 380 Bjarnason, Batt, Catt, Macpherson, Maxton, Menzies
Despite full clinical recovery with gluten verified by the conventional lactose tolerance restriction, jenunal biopsy specimens from the test, and also in context with temporary treated patients were histologically abnormal impairment of sucrase activity in healthy showing partial villous atrophy, in keeping with volunteers by drug induced a-glucosidase previous studies.' blockade where the test distinguishes between Gut: first published as 10.1136/gut.39.3.374 on 1 September 1996. Downloaded from the potency of the a-glucosidase inhibitors and the duration of drug action. The three Discussion oa-glucosidase inhibitors used for the latter pur- Construction of non-invasive systems for poses are complex oligosaccharides or, more assessing aspects of intestinal function by non- precisely, pseudo-tetrasaccharides, which have invasive methods such as recovery of orally been shown to inhibit intestinal sucrose in vitro administered probes in urine calls for attention and, by indirect methods, in vivo.25 26 The use to the detailed conduct of the test, especially and reliability of the disaccharide ratio method selection of test substances with suitable combining the use of lactose, sucrose, and properties.`3 An important problem affecting palatinose with lactulose reference for the such indirect procedures concerns the dis- investigation ofpatients with primary intestinal tortion of results by adventitious non-mucosal sucrase-isomaltase, has already been investi- factors such as the rate of gastric emptying and gated and found to be satisfactory." 12 intestinal transit, systemic distribution and Subjects with primary hypolactasia had signi- metabolism, and renal clearance. Use of the ficantly higher urinary lactose/lactulose ex- 'differential absorption' principle to overcome cretion ratios than those with normal intestinal such irrelevant variables is intrinsic to the lactase activity and the range of values (range strategy, illustrated in Figure 1, for simul- 0.25-0.78) indicating that most had partial taneous estimation of intestinal lactase, rather than complete lactase deficiency, which sucrase, isomaltase, and permeability based on would be suggested had the ratio approached the measurement of lactose, sucrose, pala- 1.0. At the same time the opportunity was tinose, lactulose, and L-rhamnose excreted in taken to compare urinary excretion of galactose urine following oral administration. Such test with that of lactose in healthy normolactasic systems should be based on sound theory, but and hypolactasic subjects. These results also also shown to be reliable and convenient for produced satisfactory discrimination provided clinical and research purposes. they were expressed as galactose/lactose ex- Evaluation of an original method is usually cretion ratios. However well urinary or blood undertaken by demonstrating ability to quanti- galactose estimations7 may discriminate, it tate a specific function, in this study the activity must nevertheless be emphasised that galactose of intestinal disaccharidases, with respect to intolerance due, for intance, to ethanol intake,27 the detection of a clinically important defect. galactose pathway defects or hepatic disease, in This is usually achieved by comparison with addition to impairment ofgalactose absorption, http://gut.bmj.com/ previously established techniques. A familiar will render interpretation unreliable while the problem, however, concerns the reliability of excretion ratios of lactose/lactulose would test procedures available for reference pur- remain unaffected.'3 16 Furthermore, abnormal poses. Existing methods that depend upon the intestinal permeability due, for instance, to rise in blood monosaccharide concentration or villous atrophy or to the effect of non-steroidal generation of breath "`C-carbon dioxide or anti-inflammatory drugs28 will interfere with hydrogen after oral administration of disac- interpretation by producing a false rise in the on September 30, 2021 by guest. Protected copyright. charide are not only affected by the rate of galactose/lactose excretion ratio. hydrolysis, but also by the state of intestinal The suitability of the disaccharide ratio absorption and systemic metabolism of the systems for assessing temporary changes in monosaccharide products. Newcomer and intestinal disaccharides activity is well illus- McGill22 have shown that as many as 25% of trated by the results of the study of the three subjects without lactase deficiency may have a-glucosidase inhibitors, BAY g 5421, BAY m misleadingly 'flat' blood glucose responses to 1099, and BAY o 1248, reported here. These lactose tolerance tests (a rise ofless than 20 mg/ are shown in Figures 5 and 6 and indicate that 100 ml following 50 g and 100 g oral lactose). the sensitivity and precision of the method is The lactose-hydrogen breath test may become adequate to resolve the quantitative differences unreliable in the presence of monosaccharide in dose response and also differences in the malabsorption, bacterial overgrowth in the duration of disaccharidase inhibition over a upper intestine, or of colonic bacteria in- short time interval. capable of generating hydrogen, and these Lactosuria and sucrosuria have long been possibilities require exclusion.23 24 In vitro known to occur in patients with coeliac assays performed on a homogenate ofintestinal disease29 but the extent to which this was due biopsy tissue are generally considered reliable to increased intestinal (paracellular) permea- evidence of disaccharidase status, but the bility or reduced disaccharide hydrolysis has measurements obtained relate to the region not been clear.'3 30 Blood glucose responses from which the biopsy tissue was taken and do following the conventional lactose tolerance not necessarily parallel the hydrolytic per- test generally suggest a high prevalence of formance ofthe whole intestine. lactase deficiency in untreated patients,3' but During this study performance of the disac- the outcome ofthis test depends as much upon charide ratio system was assessed by com- intestinal monosaccharide transport as upon paring responses obtained from healthy disaccharidase activity, both ofwhich are often 'normolactasic' volunteers, lactase status being impaired in coeliac disease.3' Changes in Differential disaccharide excretion and intestinal disaccharidase activity 381
monosaccharide absorption may, similarly, 5 Fairclough PD. Digestion and malabsorption and carbo- hydrate. In: Bouchier IAD, Allan RN, Hodgson HJF, interfere with tests that depend upon the rise Keighley MRB, eds. Textbook ofgastroenterology. London: in blood level or recovery in urine of galactose, Balliere Tindall, 1984: 361-7. 6 Isokoski H, Jussila J, Sama S. A simple screening method of breath '4CO2 derived from ingested for lactose malabsorption. Gastroenterology 1972; 62: Gut: first published as 10.1136/gut.39.3.374 on 1 September 1996. Downloaded from '4C-lactulose, and also breath hydrogen, which 28-32. 7 GrantJ, BezerrajA, Thompson SH, Lemen RJ, Koldovsky O, is generated when either unabsorbed monosac- Udall JN. Assessment of lactose absorption by measure- charide or disaccharide enters the colon.7 8 ment of urinary galactose. Gastroenterology 1989; 97: 895-9. This problem has usually been solved by 8 Arola H. The strip test for hypolactasia also works without repeating the test using an equivalent dose of ethanol. ScandJ7 Gastroenterol 1988; 23: 851-5. 9 Hepner GW. Breath analysis: gastroenterological appli- monosaccharide. In vitro assay of intestinal cations. Gastroenterology 1974; 63: 1250-6. biopsy tissue usually demonstrates the pres- 10 Newman A. Progress report: breath-analysis test in gastro- enterology. Gut 1974; 15: 308-23. ence of lactase deficiency in coeliac disease 11 Maxton DG, Catt SD, Menzies IS. Combined assessment regardless of treatment.32 The contract be- of intestinal disaccharidases in congenital asucrasia by differential urinary disaccharide excretion. J Clin Pathol tween estimates of intestinal disaccharidase 1990; 43: 406-9. activity in coeliac disease by in vitro assay of 12 Maxton DG, Catt SD, Menzies IS. Intestinal disac- charidases assessed in congenital asucrasia by differential jejunal tissue and disaccharide hydrolysis urinary disaccharide excretion. Dig Dis Sci 1989; 34: assessed in vivo by the disaccharide ratio test 129-31. 13 Menzies IS. Transmucosal passage of inert molecules in is therefore a matter of interest and import- health and disease. In: Skadhauge E, Heintze K, eds. ance. Ratios of urinary disaccharide excretion, Intestinal absorption and secretion. Lancaster: Falk Symposium 36 MTP Press, 1984: 527-43. which are not affected by overall variations in 14 Stewart JS, Pollock DJ, Hoffbrand AV, Hollin DL, monosaccharide transport capacity or permea- Booth CC. A study of proximal and distal intestinal structure and absorptive function in idiopathic bility to disaccharide, suggest that impairment steatorrhoea. QJfMed 1967; 26: 425-45. ofintestinal disaccharide hydrolysis may be less 15 Noone C, Menzies IS, Banatvala JE, Scopes JW. Intestinal permeability and lactulose hydrolysis in human rotaviral prevalent in coeliac disease than suggested by gastroenteritis assessed simultaneously by non-invasive the jejunal disaccharidase assays. differential sugar permeation. EurJ Clin Invest 1986; 16: 217-25. It is of note that although palatinose/lactu- 16 Bjarnason I, Macpherson A, Hollander D. Intestinal lose excretion ratios suggest that the in vivo permeability: an overview. Gastroenterology 1995; 108: 1566-81. intestinal hydrolysis ofpalatinose is as efficient 17 Menzies IS. Quantitative estimation of sugars in blood and as that of sucrose, in vitro assays show that urine by paper chromatography using direct densito- metry. J Chromatogr 1973; 81: 109-27. jejunal palatinase activity is much lower than 18 Menzies IS, Mount JN, Wheeler MJ. Quantitative that of sucrase (34 (2) compared with 107 (9) estimation of clinically important monosaccharides in plasma by rapid thin layer chromatography. Ann Clin mU/mg protein, respectively). Evidently Biochem 1978; 15: 65-76. palatinose is a satisfactory substrate for in vivo 19 Peters TJ, Batt R, Heath JR, Tilleray J. The microassay of intestinal disaccharidases. Biochem Med 1976; 15: 145-8. assessment ofintestinal isomaltase activity but, 20 Menzies IS, Jenkins AP, Heduan E, Catt SD, Segal MB, unlike isomaltose, it seems to exert severe Creamer B. The effect of poorly absorbed solute on of in intestinal absorption. Scand J Gastroenterol 1990; 25: substrate inhibition under the conditions 1257-64. http://gut.bmj.com/ vivo assay.33 21 Slavin G, Sowter C, Robertson S, MacDermott S, Paiton K. Measurement in jejunal biopsy by computer aided In summary, the evidence presented suggests microscopy. J Clin Pathol 1980; 33: 254-61. that measurement of urinary hydrolysable/non- 22 Newcomer AD, McGill DB. Lactose tolerance test in adults with normal lactase activity. Gastroenterology 1966; 50: hydrolysable disaccharide excretion ratios 340-6. following oral administration of appropriate 23 Calloway DH, Murphy EL, Bauer D. Determination of lactose tolerance by breath analysis. Am J Dig Dis 1969; disaccharides is a reliable non-invasive tech- 14: 811-5. 24 Metz Peters Newman Blendis LM. nique capable of providing simultaneous G, Jenkins JA, TJ, A, on September 30, 2021 by guest. Protected copyright. Breath hydrogen as a diagnostic method for hypolactasia. quantitative assessment of intestinal lactase, Lancet 1975; i: 1155-7. sucrase, and isomaltase activity. We have found 25 Walton RJ, Sherif IT, Noy GA, Alberti KGMM. Improved metabolic profiles in insulin-treated diabetics given an the method convenient for studying the dose/ alpha-glucosidase hydrolase inhibitor. BMJ 1979; i: response and temporary duration of action of 220-1. 26 Hillebrand I, Boehme K, Frank G, Fink H, Berchold F. The ao-glucosidase inhibitors on sucrose hydrolysis, effect of the cs-glucosidase inhibitor Bay g 5421 and for the investigation oflactose hydrolysis in (Acarbose) on meal stimulated elevations of circulating glucose, inulin and triglyceride levels in man. Res Exp Med subjects with primary (genetic) adult hypolac- 1979; 175: 81-6. tasia. A study of patients with coeliac disease 27 Salaspuro MP, Kesaniemi YA. Intravenous galactose elimination test with and without ethanol loading in shows that in most cases impairment of in vivo various clinical conditions. ScandJ7 Gastroenterol 1973; 8: disaccharide hydrolysis is only moderate. 274-80. 28 Bjamason I, Williams P, So A, et al. Intestinal permeability and inflammation in rheumatoid arthritis: effects of non- We thank Bayer UK for the gift of the ae-glucosidase inhibitors steroidal anti-inflammatory drugs. Lancet 1984; ii: and Tate and Lyle Industries Ltd, Reading, Berks, UK, for the 1171-4. gift of palatinose used in these studies. 29 Weser E, Sleisenger MH. Lactosuria and lactase deficiency in adult coeliac disease. Gastroenterology 1964; 48: 571-8. 30 Muller M, Walker-Smith J, Schmerling DH, Curtius HCH, 1 Menzies IS. Medical importance of sugars in the alimentary Prader A. Lactulose: a gas-liquid chromatographic tract. In: Grenby TH, Parker KJ, Lindley MG, eds. method of determination and evaluation of its use to Developments in sweeteners-2. London: Applied Science assess intestinal mucosal damage. Clin Chim Acta 1969; Publishers, 1983: 89-117. 24: 45-9. 2 Bedine MS, Bayless TM. Intolerance of small amounts of 31 Brode S, Gudman-Hoyer E. Incidence and general signifi- lactose by individuals with low lactase levels. Gastro- cance of lactose malabsorption in adult coeliac disease. enterology 1973; 65: 735-43. ScandJ Gastroenterol 1988; 23: 2848. 3 Gray GM. Intestinal disaccharidase deficiencies and glucose- 32 Peters TJ, Jones PE, Wells G. Analytical subcellular galactose malabsorption. In: Stainbury JB, Wyndgaarden fractionation of jejunal biopsy specimens. Enzyme JB, Fredrickson DS, eds. The metabolic bases of inherited activities, organelle pathology and response to gluten diseases. 4th ed. New York: McGraw-Hill, 1974: 1526-36. withdrawal in patients with coeliac disease. Clin Sci Mol 4 Newcomer AD, McGill DB. Disaccharidase activity in the Med 1987; 55: 582-92. small intestine. Prevalence of lactase deficiency in 100 33 Dahlquist A. Substrate inhibition of intestinal glucosidases. healthy subjects. Gastroenterology 1967; 52: 881-9. Acta Chem Scand 1960; 14: 1797-808.