Gut: first published as 10.1136/gut.48.4.571 on 1 April 2001. Downloaded from
`
`http://gut.bmj.com/
`
` on November 25, 2022 by guest. Protected by copyright.
`
`Gut 2001;48:571–577
`
`Review
`
`571
`
`Intestinal luminal pH in inflammatory bowel disease: possible
`determinants and implications for therapy with
`aminosalicylates and other drugs
`
`Summary
`luminal pH in the normal gastro-
`Measurements of
`intestinal tract have shown a progressive increase in pH
`from the duodenum to the terminal ileum, a decrease in
`the caecum, and then a slow rise along the colon to the
`rectum. Some data in patients with ulcerative colitis
`suggest a substantial reduction below normal values in the
`right colon, while limited results in Crohn’s disease have
`been contradictory. Determinants of luminal pH in the
`colon include mucosal bicarbonate and lactate production,
`bacterial
`fermentation of carbohydrates and mucosal
`absorption of short chain fatty acids, and possibly intestinal
`transit. Alterations in these factors, as a result of mucosal
`disease and changes in diet, are likely to explain abnormal
`pH measurements in inflammatory bowel disease (IBD). It
`is conceivable that reduced intracolonic pH in active
`ulcerative colitis impairs bioavailability of 5-aminosalicylic
`acid from pH dependent release formulations (Asacol,
`Salofalk) and those requiring cleavage by bacterial azo
`reductase (sulphasalazine, olsalazine, balsalazide), but fur-
`ther pharmacokinetic studies are needed to confirm this
`possibility. Reports that balsalazide and olsalazine may be
`more eYcacious in active and quiescent ulcerative colitis,
`respectively, than Asacol suggest that low pH may be a
`more critical factor in patients taking directly pH depend-
`ent
`release than azo bonded preparations. Reduced
`intracolonic pH also needs to be considered in the
`development of pH dependent colonic release formula-
`tions of budesonide and azathioprine for use in ulcerative
`and Crohn’s colitis. This paper reviews methods for meas-
`uring gut pH, its changes in IBD, and how these may influ-
`ence current and future therapies.
`
`Introduction
`Over the past 15 years, the development of radiotelemetric
`technology has made possible the measurement in vivo of
`the luminal pH of the entire human gastrointestinal tract
`using orally ingested free fall pH sensitive capsules.1 In this
`review, we compare methods available for investigating gut
`pH distal to the stomach, describe the pH profiles obtained
`in normal controls2–11 and in patients with inflammatory
`bowel disease (IBD),6–12 and discuss the mucosal and lumi-
`nal factors likely to account for diVerences in health and
`disease. Lastly, we consider the therapeutic implications of
`altered gut pH in IBD and, in particular, the potential
`influence of reduced colonic pH on the bioavailability of
`drugs such as 5-aminosalicylic acid (5-ASA), which are
`formulated in a pH dependent release system.
`
`Measurement of intestinal luminal pH
`Luminal gut pH can be measured directly in vivo using
`either radiotelemetric capsules13 (RTC) or tube mounted
`pH sensitive electrodes passed orally. Peri-mucosal colonic
`pH can be recorded in vivo by electrodes inserted
`endoscopically14 as well as applied directly in vitro to biop-
`sies or operative specimens.15
`
`RADIOTELEMETRIC MEASUREMENT OF INTRALUMINAL GUT pH
`RTC13 consist of a reference and pH sensitive electrode
`which samples and transmits the pH of the gut lumen.
`They are battery powered, approximately 20·7 mm in size,
`and contain a radiofrequency transmitter. Signals can be
`transmitted at frequencies of 6–60/second and are received
`by an aerial and stored on a data logger. The orally ingested
`RTC take 1–5 days to pass through the gastrointestinal
`tract by free fall.
`The approximate location of the capsule in relation to
`surface abdominal landmarks can be determined either by
`fluoroscopy or by identification of the maximal radio signal
`with the help of a radio receiving probe.4 Although this
`method of identifying the site of the capsule does not allow
`its precise location in relation to sphincters and other
`intestinal anatomical sites, the pH changes themselves
`indicate the location of the electrode. For example, a sud-
`den fall in pH when the probe is in the right iliac fossa
`indicates its arrival in the caecum.
`Another problem with radiotelemetry pH recording is
`poor signal quality.10 11 EVective data transmission and
`retrieval
`is necessary to construct a pH profile for all
`segments of the gut. Low signal strength occurs when the
`capsule in the gut lumen and the aerial are not optimally
`aligned or when the capsule exceeds the optimal distance
`to the aerial for maximum reception of the transmitted sig-
`nal, a frequent problem in the colon. Some studies have
`reported up to 75% data loss in individual patients.10 11
`
`MEASUREMENT OF INTRALUMINAL GUT pH USING PER ORAL
`TUBE MOUNTED ELECTRODES
`Per oral tube mounted pH electrodes measure small bowel
`and right colonic luminal pH accurately and continuously.
`The pH catheter is passed into the stomach and the tip of
`the tube manoeuvred across the pylorus under fluoros-
`copy; a small balloon inflated at the tip assists passage
`through the small intestine into the colon. Luminal pH
`measurements are recorded and stored by a digitrapper
`from several electrodes positioned at specific intervals
`along the axis of the tube; their anatomical location can be
`identified fluoroscopically. This method avoids a potential
`hazard of the radiotelemetric capsule, namely impaction at
`the site of small bowel strictures in patients with Crohn’s
`disease with consequent intestinal obstruction.16
`
`MEASUREMENT OF PERI-MUCOSAL COLONIC pH
`Peri-mucosal pH can be measured by endoscopic place-
`ment of pH sensitive electrodes on to the luminal surface of
`the colonic mucosa.14 A surface layer of mucus approxi-
`mately 100–800 µm thick covers the mucosa. Beneath this
`layer and adjacent to the apical membrane lies an area
`apparently protected from the contents of the lumen and
`relatively unaVected by changes in the colonic lumen. The
`
`Abbreviations used in this paper: IBD, inflammatory bowel
`disease; 5-ASA, 5-aminosalicylic acid; SCFA, short chain fatty acids;
`RTC, radiotelemetric capsules.
`
`www.gutjnl.com
`
`Bausch Health Ireland Exhibit 2033, Page 1 of 7
`Mylan v. Bausch Health Ireland - IPR2022-00722
`
`

`

`Gut: first published as 10.1136/gut.48.4.571 on 1 April 2001. Downloaded from
`
`http://gut.bmj.com/
`
` on November 25, 2022 by guest. Protected by copyright.
`
`572
`
`Nugent, Rampton, Evans, et al
`
`Table 1 Intestinal luminal pH studies using radiotelemetry capsules in healthy volunteers
`
`Study
`
`Patients
`
`Watson, 19722
`
`Bown, 19743
`Evans, 19884
`Fallingborg, 19895
`Raimundo, 19926
`Fallingborg, 19988
`Sasaki, 19979
`Press, 199810
`Ewe, 199911
`
`N/A, data not available.
`
`2 normals+7 misc. GI
`disorders
`11 normals
`66 normals
`39 normals
`7 normals
`13 normals
`4 normals
`12 normals
`13 normals
`
`Small bowel pH
`
`Proximal
`
`5.5–7.0
`
`5.9
`6.6
`6.4
`6.6
`6.4
`6.8
`6.7
`6.5
`
`Distal
`
`6.5–7.5
`
`7.5
`7.5
`7.3
`7.4
`7.4
`7.7
`7.5
`7.6
`
`Caecum/right colon pH
`
`Left colon/ rectal pH
`
`5.5–7.5
`
`6.5–7.5
`
`6.0
`6.4
`5.7
`6.7
`5.8
`6.8
`6.1
`6.2
`
`7.0
`7.1
`6.6
`N/A
`N/A
`7.2
`6.1
`7.0
`
`Table 2 Colonic peri-mucosal pH in healthy volunteers and patients with
`ulcerative proctitis and neoplasia
`
`Study
`
`Method
`
`Patients
`
`Caecum/
`right colon
`pH
`
`McDougall,
`199314
`McNeil, 198715
`
`Colonoscopy
`probe
`Microelectrodes,
`human rectal
`biopsies
`
`21 normals
`37 neoplasia
`6 normals
`5 ulcerative
`proctitis
`
`7.1
`7.2
`N/A
`N/A
`
`N/A, data not available.
`
`Left
`colon/
`rectal pH
`
`7.2–7.5
`7.2–7.4
`6.3–6.8
`6.3–6.8
`
`pH probe is inserted down the biopsy channel to make
`contact with the mucosa and record the pH at its surface
`during the procedure. It is not possible with this method to
`record pH throughout the intestine or to record for long
`periods. Furthermore, in the large bowel, fasting and laxa-
`tives used for bowel preparation before colonoscopy may
`alter the luminal and surface properties of the colon and
`give pH recordings unrepresentative of those found in
`unprepared bowel.
`
`MEASUREMENT OF COLONIC MUCOSAL pH IN VITRO
`Mucosal pH can also be measured at the epithelial cell sur-
`face in surgically resected colonic specimens and biopsies,
`using glass pH microelectrodes.15 Results obtained in organ
`bath preparations should be extrapolated to the intact
`human digestive tract with extreme care because factors
`such as buVers, trauma, nutrients, and absence of luminal
`contents may influence the pH significantly.
`
`Luminal pH in normal small bowel and colon
`Gastrointestinal luminal pH data recorded by RTC in nor-
`mal volunteers are shown in table 1. Luminal pH in the
`proximal small bowel ranges from 5.5 to 7.0 and gradually
`rises to 6.5–7.5 in the distal ileum. In almost every record-
`ing published there has been a fall in luminal pH from the
`terminal ileum to the caecum (range 5.5–7.5); pH then
`rises in the left colon and rectum to 6.1–7.5.
`
`COLONIC MUCOSAL SURFACE pH
`Colonic mucosal pH in healthy subjects is shown in table 2.
`In vitro, a mean perimucosal surface pH of 6.6 was
`recorded in rat colonic mucosa and human rectal biopsy
`specimens.15 However, the in vivo surface pH of human
`colonic mucosa ranged between 7.1 and 7.5 and was con-
`sistently higher at all anatomical segments than luminal
`pH.14 Although the eVect of bowel preparation prior to
`colonoscopy is uncertain, these findings suggest loss of the
`acidifying action of the luminal contents under the mucous
`barrier and the predominant eVect of submucous epithelial
`bicarbonate secretion.
`
`foregut luminal pH, other mechanisms play a role in the
`small bowel and colon. The acidic gastric contents are
`buVered by alkaline pancreatic secretions as they enter the
`proximal small bowel, resulting in a rise in luminal pH here
`by several units. Additionally, small bowel mucosal
`bicarbonate secretion results in a further gradual rise in
`luminal pH (7.5) in the terminal ileum.4
`The almost neutral small bowel contents then empty
`into the caecum where the luminal pH (6.4) is relatively
`acidic.4 This fall in luminal pH is in part attributable to the
`action of colonic bacteria which ferment carbohydrates
`entering the caecum from the ileum generating the short
`chain fatty acids (SCFA) acetic, propionic, and butyric
`acid, and hydrogen ions.17 The SCFAs are weak acids, pKa
`4.8, and are present as organic anions in the normal colonic
`lumen. The faecal concentration of these organic anions is
`negatively correlated with faecal pH.18 SCFAs, especially
`butyrate, are absorbed and metabolised by the colonic epi-
`thelium for which they are a principal energy source.19 A
`falling intraluminal concentration of SCFAs may contrib-
`ute,
`in common with colonic mucosal bicarbonate
`secretion, to a pH rise along the distal colon. A slight drop
`in pH may occur in the rectum due to faecal stasis and the
`subsequent action of colonic bacteria fermenting any
`remaining carbohydrates.17
`Ammonia is formed in the colonic lumen from the bac-
`terial metabolism of proteins, amino acids, and particularly
`urea. While,
`theoretically, a high protein diet20 may
`therefore raise colonic pH, the influence of ammonia on
`colonic pH is thought
`to be smaller
`than that of
`bicarbonate and organic acids.21
`Dietary intake can influence intracolonic pH through its
`eVects on SCFA production. Thus increased dietary
`fibre,22 as well as non-absorbable sugars such as lactulose,3
`increase caecal acidity by providing a carbohydrate meal to
`colonic flora.23
`The eVects of lactulose on gut pH may also be modified
`by its eVects on intestinal transit. However, the eVects of
`changes in colonic transit time on intraluminal pH are dif-
`ficult to predict. Theoretically, a shortened transit time
`could either increase pH by reducing the time available for
`bacterial
`fermentation of carbohydrates to SCFAs or
`decrease it by causing carbohydrate starved bacteria to
`produce more lactate.17 In fact, a mixture of magnesium
`sulphate and carbonate given to healthy volunteers in suf-
`ficient doses to increase stool weight threefold produced no
`change in pH in the colon itself, and a small rise in the rec-
`tum.3 Conversely, in a study of gall stone patients with slow
`transit constipation, there was a higher proximal colonic
`pH (6.8) than in controls (pH 6.4).24
`
`Determinants of normal intestinal luminal pH
`While hydrogen and bicarbonate ion secretion by the gas-
`tric and intestinal mucosa are major determinants of
`
`Intestinal luminal pH in ulcerative colitis
`The published reports of intraluminal pH in patients with
`ulcerative colitis6 7 10–12 indicate a wide range of pH values in
`
`www.gutjnl.com
`
`Bausch Health Ireland Exhibit 2033, Page 2 of 7
`Mylan v. Bausch Health Ireland - IPR2022-00722
`
`

`

`Gut: first published as 10.1136/gut.48.4.571 on 1 April 2001. Downloaded from
`
`http://gut.bmj.com/
`
` on November 25, 2022 by guest. Protected by copyright.
`
`Intestinal luminal pH in IBD
`
`573
`
`Table 3 Intestinal luminal pH, measured using radiotelemetry capsules, in patients with ulcerative colitis (UC)
`
`Small bowel pH
`
`Study
`
`Patients with UC
`
`Proximal
`
`Distal
`
`Caecum/right colon pH
`
`Left colon/ rectal pH
`
`Raimundo, 19926
`
`Fallingborg, 19937
`
`Press, 199810
`
`Ewe, 199911
`Nugent, 200012
`
`N/A, data not available.
`
`7 active
`6 inactive
`3 active
`3 very active
`7 active
`4 inactive
`4 active
`6 active
`
`6.1
`5.9–6.6
`Normal range
`Normal range
`6.8
`6.6
`6.5
`7.3
`
`7.2
`6.9–7.4
`Normal range
`Normal range
`8.2
`7.9
`6.8
`8.3
`
`4.7
`4.9–5.5
`Normal range
`2.3–3.4
`7.2
`6.5
`5.5
`6.7
`(5.8–7.3)
`
`N/A
`N/A
`N/A
`N/A
`6.8
`6.5
`7.5
`6.7
`(4.8–7.3)
`
`the caecum and right colon with a shift towards much
`lower pH values in some patients with active disease (table
`3).
`In Fallingborg et al’s study of six patients with active
`ulcerative colitis,5 the three patients with the most severe
`disease (of whom two required urgent surgery) showed
`extremely acidic proximal colonic pH (ranging between
`pH 2.3 and 3.4). The remaining three patients had luminal
`pH profiles within the normal range. Raimundo et al
`reported similar findings in an abstract (right colonic lumi-
`nal pH as low as 4.7) in patients with both active and inac-
`tive ulcerative colitis.6 Nugent et al also reported, in an
`abstract, falls in colonic luminal pH to less than 5.5 in two
`of six patients with active ulcerative colitis.12 In contrast,
`Press et al reported slightly higher right colonic luminal pH
`in 11 patients with ulcerative colitis compared with normal
`controls.10 In a further recent study, four patients with
`mild-moderately active ulcerative colitis had no decrease in
`colonic luminal pH; pH was again higher than in normal
`controls.11
`Regrettably, these five studies are all small. Drawing firm
`conclusions is diYcult because of diVerences in the extent
`and severity of colitis and in dietary intake of the patients
`investigated. It has been suggested that recorded pH may
`sometimes be artefactually low as a result of signal loss,11
`but our own studies show transient reductions in colonic
`pH at
`times when simultaneously monitored signal
`strength is well maintained.12 On balance, it seems likely
`that right colonic pH is reduced in at least a proportion of
`patients with ulcerative colitis, but further studies of larger
`numbers of patients with well defined disease, and under
`strictly controlled conditions, are needed.
`
`Intestinal luminal pH in Crohn’s disease
`Existing data on luminal pH in Crohn’s disease are also
`limited by small numbers of patients recruited and diVer-
`ences in disease site, activity, and treatment (table 4).8–11
`In one study, a low colonic luminal pH, similar to that
`reported in patients with active ulcerative colitis, was found
`in patients with Crohn’s disease.9 Four patients with
`Crohn’s colitis, three active, had lower right (pH 5.3) and
`left (pH 5.3) colonic luminal pH values than normal con-
`trols (pH 6.8). The reported tendency for pH to rise from
`the right to the left colon was lost in two of the four patients
`
`but there was no obvious relation between gut luminal pH
`and mucosal disease activity or site. Press et al and Ewe et
`al failed to confirm these findings.10 11 In a total of 24
`patients with Crohn’s disease, small bowel and colonic
`luminal pH was similar to that recorded in healthy control
`subjects, irrespective of disease activity or site. In a fourth
`report,8 right colonic pH (mean 6.7) was higher in nine
`patients with an ileocaecal resection for Crohn’s disease
`than in 13 normal controls (mean pH 5.7) but was still
`within the normal range; neo-terminal ileal pH (7.3) was
`normal.
`
`Determinants of colonic luminal pH in IBD
`Reduced mucosal bicarbonate secretion, increased mu-
`cosal and bacterial lactate production, and impaired SCFA
`absorption and metabolism may each contribute to a
`reduction in colonic luminal pH in patients with inflamed
`colonic mucosa.17 Changes in intestinal transit and dietary
`fibre intake during an acute flare up may also influence
`colonic pH.10
`Decreased faecal bicarbonate concentration and re-
`duced rectal mucosal bicarbonate secretion are found in
`patients with active ulcerative colitis,25 26 and could account
`for the acidic colonic lumen. However, bicarbonate secre-
`tion appears to be unaltered in Crohn’s disease.26
`Elevated colonic luminal concentrations of SCFAs have
`been found in active ulcerative colitis,19 decreasing colonic
`pH,18 and this could be explained by impaired SCFA
`absorption and utilisation reported in some26–28 but not all
`studies.29–32
`In contrast, it has been suggested that a reduced intake of
`dietary fibre in patients with active colitis could limit the
`amount of carbohydrate available for utilisation as an
`energy source by colonic bacteria,10 resulting in the prefer-
`ential production of lactate instead of SCFAs. Indeed,
`elevated concentrations of luminal lactic acid have been
`reported in active colitis.17 31
`The eVects of increased SCFAs or lactate concentrations
`on colonic luminal pH are likely to be buVered in active
`colitis by the presence of blood and mucus, although the
`quantitative importance of these mechanisms is uncer-
`tain.10 Furthermore, bacterial generation of ammonia from
`urea and other nitrogenous blood constituents may also
`
`Table 4 Intestinal luminal pH, measured using radiotelemetry capsules, in patients with Crohn’s disease
`
`Small bowel pH
`
`Study
`
`Patients with CD
`
`Proximal
`
`Distal
`
`Caecum/right colon pH
`
`Left colon/ rectal pH
`
`Fallingborg, 19988
`
`Sasaki, 19979
`Press, 199810
`
`Ewe, 199911
`
`N/A, data not available.
`
`9 with ileocaecal
`resections
`3 active+1 inactive
`5 active
`7 inactive
`12 active
`
`6.3
`
`7.2
`6.5
`6.8
`6.5
`
`7.3
`
`7.8
`7.9
`8.2
`7.5
`
`6.7
`
`5.3
`7.2
`6.5
`6.2
`
`N/A
`
`5.3
`6.8
`6.5
`6.5
`
`www.gutjnl.com
`
`Bausch Health Ireland Exhibit 2033, Page 3 of 7
`Mylan v. Bausch Health Ireland - IPR2022-00722
`
`

`

`Gut: first published as 10.1136/gut.48.4.571 on 1 April 2001. Downloaded from
`
`http://gut.bmj.com/
`
` on November 25, 2022 by guest. Protected by copyright.
`
`574
`
`Nugent, Rampton, Evans, et al
`
`antagonise any tendency of colonic pH to fall in patients
`with active colitis.21
`Contrary to widespread assumption, mouth to anus
`intestinal transit times in ulcerative colitis are not reduced;
`indeed, small bowel transit time is prolonged.33–35 Further-
`more, transit time through the whole colon is similar to that
`of healthy controls.33 Several studies, however, show
`regional diVerences in transit within the colon in ulcerative
`colitis.36–38 Passage of
`luminal contents
`through the
`proximal colon is delayed while that through the left colon
`is accelerated.33 These changes tend to be more marked in,
`but are not restricted to, patients with distal disease, but
`their eVects on intracolonic pH, as indicated earlier, are
`diYcult to interpret.
`
`Therapeutic implications of low colonic luminal pH
`in IBD
`Several drugs used for the treatment of ileal and colonic
`IBD have been formulated so as to deliver the active agent
`directly to the site of inflammation, thereby reducing their
`absorption in the proximal gastrointestinal
`tract and
`reducing systemic side eVects. Some of these agents utilise
`pH dependent release systems (for example, Asacol, Salo-
`falk, and budesonide) while others depend on bacterial
`enzymatic metabolism (sulphasalazine, olsalazine, bal-
`salazide) which may also be aVected by changes in colonic
`luminal pH.
`
`5-ASA drug delivery to the colon
`Sulphasalazine was the first 5-ASA containing drug to
`show therapeutic benefit in ulcerative colitis. The active
`component, 5-ASA, is bound to an inert carrier, sulphapy-
`ridine,39 and is released in the colon by the action of colonic
`bacterial azo reductase. Newer preparations dependent on
`bacterial azo reduction are olsalazine (two 5-ASA mole-
`cules azo bonded together), and balsalazide (5-ASA azo
`bonded to an inert carrier, 4-amino-benzoyl-alanine).
`The pH dependent delayed release formulations of
`5-ASA release the active moiety when their Eudragit coat-
`ing dissolves as luminal pH rises above a critical value (for
`Asacol, Eudragit S dissolves when pH > 7.0; for Salofalk,
`Eudragit L dissolves when pH > 6.0).40 They are designed
`to release the maximum concentration of the drug in the
`terminal ileum and right colon. For Asacol, for example,
`optimal activity depends on a rise in distal small bowel
`luminal pH above pH 7.0 for suYcient duration to ensure
`complete release of 5-ASA from the polymer coating,
`before it enters the caecum where luminal pH is lower
`(table 1).
`The slow release formulation, Pentasa, releases 5-ASA
`from ethylcellulose microspheres in a time dependent
`manner throughout the small bowel and colon.41 Pentasa
`relies, like pH sensitive capsules, on normal intestinal tran-
`sit for optimal delivery of the drug but is not, in contrast,
`aVected by fluctuations in luminal pH.
`
`Pharmacokinetics of 5-ASA in healthy volunteers
`The proximal gastrointestinal tract rapidly absorbs orally
`ingested 5-ASA42 which is then metabolised in the gut
`mucosa to an inactive metabolite,43 44 N-acetyl-5-ASA, by
`epithelial acetyl coenzyme A.45 The activity of this is greater
`in the colonic than small bowel mucosa.46 As indicated
`above, the 5-ASA formulations incorporate various mecha-
`nisms to delay the release of 5-ASA in the proximal gastro-
`intestinal tract, minimise systemic absorption, and produce
`high luminal concentrations of 5-ASA at the site of inflam-
`mation.47
`After oral Asacol, approximately 10–40% of the ingested
`dose is absorbed and excreted in the urine of healthy
`volunteers as 5-ASA and its metabolite N-acetyl-5-ASA,
`
`Table 5 Total 5-aminosalicylic acid (5-ASA +N-acetyl-5-ASA) faecal
`and urinary recovery (as percentage of ingested dose) of Asacol, Pentasa,
`and olsalazine in volunteers with normal or rapid transit time (TT) and
`in patients with inactive or active ulcerative colitis
`
`Healthy volunteers
`
`Ulcerative colitis patients
`
`Drug
`
`Normal TT Rapid TT
`
`Inactive
`
`Active
`
`Total faecal 5-ASA
`Asacol
`23–40
`Pentasa
`16–47
`Olsalazine 47
`Total urinary 5-ASA
`Asacol
`13–36
`Pentasa
`26–56
`Olsalazine 19–25
`
`48
`29–52
`79 (53% unsplit)
`
`10–31
`14–28
`5
`
`44–53
`38–40
`39–53
`
`17–35
`25–36
`9–22
`
`60–90
`57
`65 (47% unsplit)
`
`16
`23
`5
`
`Data taken from references48 54
`
`Table 6 Comparison of 5-aminosalicylic acide (5-ASA) and
`N-acetyl-5-ASA excretion in faeces and urine of three diVerent 5-ASA
`preparations, each given in a dose of 2 g/day for >6 days, in healthy
`volunteers and in patients with ulcerative colitis
`
`Healthy volunteers
`
`Ulcerative colitis patients
`
`Drug
`
`5-ASA
`
`N-acetyl-5-ASA
`
`5-ASA
`
`N-acetyl-5-ASA
`
`Faecal 5-ASA and N-acetyl-5-ASA excretion
`Asacol
`6 [3–13]
`14 [3–19]
`Pentasa
`7 (1)
`20 (2)
`Olsalazine 12 (2)
`10 (2)
`Urinary 5-ASA and N-acetyl-5-ASA excretion
`Asacol
`N/A
`N/A
`Pentasa
`4 (3)
`27 (3)
`Olsalazine 2 (3)
`23 (3)
`
`28 (4)
`13 (2)
`36 (5)
`
`6 (2)
`6 (2)
`3 (1)
`
`15 (2)
`25(3)
`16 (3)
`
`24 (5)
`31 (8)
`19 (3)
`
`Values are percentage mean (SEM) or [range] of ingested dose.
`N/A, data not available.
`Data from references48 52 53
`
`0–15% is excreted in the faeces unchanged, and a further
`in the faeces as N-acetyl-5-ASA.48
`0–20% appears
`Depending on their release profile, the various 5-ASA for-
`mulations diVer in the proportions of 5-ASA:N-acetyl-5-
`ASA absorbed and excreted in the urine and faeces (tables
`5, 6).
`For each formulation, serum and urine concentrations of
`the metabolite N-acetyl-5-ASA are greater than those of
`5-ASA.49 A high urinary excretion of N-acetyl-5-ASA indi-
`cates early release of 5-ASA from the formulation in the
`proximal gastrointestinal tract.49 Recovery of N-acetyl-5-
`ASA in the faeces indicates timely release of 5-ASA in the
`colonic lumen with its subsequent mucosal absorption,
`metabolism to N-acetyl-5-ASA, and release of the latter
`back into the lumen. Any 5-ASA recovered in the faeces
`represents late or impaired release of 5-ASA from the for-
`mulation. Thus an ideal 5-ASA formulation should achieve
`a high faecal N-acetyl-5-ASA:5-ASA ratio and low urinary
`5-ASA and N-acetyl-5-ASA recoveries: this profile indi-
`cates maximised colonic delivery, minimal proximal
`absorption, and low systemic toxicity.49
`How might changes in intraluminal gut pH and transit
`time in IBD mitigate against optimal bioavailability of
`5-ASA from its presently available formulations?
`
`Potential eVects of altered colonic pH and transit on
`bioavailability of 5-ASA in IBD
`Theoretically, it is possible that reduced right colonic pH in
`ulcerative colitis could reduce bioavailability of 5-ASA
`from both Eudragit coated pH dependent and azo reduct-
`ase dependent formulations, without aVecting bioavailabil-
`ity of 5-ASA from the slow release preparation Pentasa.
`Thus intraluminal pH could inhibit release of 5-ASA
`from Asacol and Salofalk if it failed to exceed 7.0 and 6.0,
`respectively, for long enough to ensure complete coat dis-
`solution. Direct evidence on pH dependent release in
`ulcerative colitis is not yet available but preliminary data
`
`www.gutjnl.com
`
`Bausch Health Ireland Exhibit 2033, Page 4 of 7
`Mylan v. Bausch Health Ireland - IPR2022-00722
`
`

`

`Gut: first published as 10.1136/gut.48.4.571 on 1 April 2001. Downloaded from
`
`http://gut.bmj.com/
`
` on November 25, 2022 by guest. Protected by copyright.
`
`Intestinal luminal pH in IBD
`
`575
`
`suggest that in most patients small bowel pH, measured
`with a radiotelemetry capsule,12
`is high enough for
`suYcient time to allow capsule dissolution.47 In vitro stud-
`ies have shown that a low pH inhibits colonic bacterial
`metabolism of carbohydrate, urea, and other nitrogenous
`compounds21: it is possible that increased colonic acidity
`could also reduce azo reductase activity and release of
`5-ASA from sulphasalazine, olsalazine, and balsalazide.49a
`Rapid transit of luminal contents reduces the duration of
`contact of released 5-ASA with the mucosa as well as the
`time for this release to occur and for exposure of azo
`bonded 5-ASA formulations to bacterial azo reductase. In
`normal subjects, intestinal transit accelerated by Bisacodyl
`decreases systemic absorption, as indicated by reduced
`urinary excretion, and increases faecal excretion of 5-ASA
`from all formulations (table 5).50–53 This eVect is most pro-
`nounced with azo bound 5-ASA formulations as much of
`the 5-ASA remains bound to its carrier. Under conditions
`of accelerated intestinal transit the proportion of N-acetyl-
`5-ASA in faeces is reduced,50 indicating that although
`luminal 5-ASA concentrations are increased, 5-ASA is
`released more distally in the colon.
`The relevance of these points to what actually occurs in
`patients with IBD in relation to the bioavailability of 5-ASA
`is uncertain. As indicated above, low colonic pH has not
`been found universally and transit appears to be delayed in
`the small intestine and right colon, and accelerated only
`distally in patients with ulcerative colitis.
`
`Bioavailability of 5-ASA in IBD
`The eVect of ulcerative colitis on the distribution of 5-ASA
`derived from a representative of each of the main types of
`5-ASA formulations is summarised in tables 5 and 6.
`Rijk et al compared five diVerent formulations in 20 IBD
`patients with and without diarrhoea. The azo formulations
`sulphasalazine and olsalazine were less completely split in
`patients with diarrhoea than in those without diarrhoea.49
`Release of 5-ASA from Asacol in patients with diarrhoea
`was characterised by a high proportion of 5-ASA in stools
`but little in the acetylated form, indicating release primarily
`in the distal colon.49 In patients with diarrhoea, release of
`5-ASA from Salofalk and Pentasa was also impaired but
`the changes were less substantial and their bioavailability
`more favourable. However, in the absence of diarrhoea,
`faecal 5-ASA concentrations were highest with olsalazine
`and Asacol, consistent with predominantly colonic release
`of 5-ASA from these formulations.49
`In another study of Asacol bioavailability in ulcerative
`colitis, greater faecal excretion of 5-ASA was confirmed in
`patients with active compared with inactive disease.54
`Lastly, a comparative study of four 5-ASA formulations in
`quiescent ulcerative colitis showed urinary and faecal
`N-acetyl-5-ASA excretion to be greatest after ingestion of
`Pentasa and Salofalk.53
`These studies indicate that bioavailability of 5-ASA from
`all its formulations is reduced in patients with active IBD
`with results being least untoward for Pentasa and Salofalk.
`However, further comparative studies of the various 5-ASA
`formulations in patients with IBD are needed to clarify the
`eVect of disease severity and extent on the bioavailability of
`5-ASA and in particular its relation to changes in intralu-
`minal pH as well as transit time.
`
`Clinical eYcacy of 5-ASA formulations in IBD
`Although the pharmacokinetic data described above
`suggest that pH dependent or azo bonded formulations of
`5-ASA could be less eVective in active ulcerative colitis
`than slow release preparations,
`there are no direct
`comparative clinical trials of Pentasa with other 5-ASA
`formulations to confirm or refute this possibility.
`
`Most trials of 5-ASA formulations in mild-moderately
`active ulcerative colitis indicate that they all achieve similar
`remission rates (40–80%). A recent comparative study did,
`however, suggest that balsalazide may be more potent than
`Asacol in moderately active ulcerative colitis.55
`Since the early trials with sulphasalazine56 57 it has been
`clear that 5-ASA formulations are more eVective in main-
`taining remission than in treating active ulcerative colitis,58
`and this may be due at
`least
`in part
`to impaired
`bioavailability of 5-ASA in patients in relapse. While a
`meta-analysis published in 199359 suggested that the newer
`5-ASAs, including Pentasa, have similar eYcacy to each
`other and to sulphasalazine in maintenance of remission in
`quiescent ulcerative colitis, both olsalazine60 and bal-
`salazide61 have more recently been claimed to have advan-
`tages over Asacol, particularly in patients with left sided
`disease.60 The olsalazine study, however, has been criticised
`for its single blind design and insuYcient use of sigmoido-
`scopic review, and for the unusually high relapse rate found
`in the Asacol treated group.60 Furthermore, in the other
`trial, the delay in time to relapse in balsalazide treated
`patients was not accompanied by any diVerences in remis-
`sion rate at one year compared with Asacol.61 Nevertheless,
`if substantiated, these reports suggest that any eVects of pH
`in quiescent ulcerative colitis are more marked for the
`directly pH dependent than azo bonded preparations.
`Limited data, none of which are directly comparative,
`show no major diVerences in eYcacy between Pentasa,
`Asacol,
`and Salofalk in active
`ileocaecal Crohn’s
`disease.62–64 Similarly, a recent meta-analysis of trials of
`5-ASA formulations as maintenance therapy in Crohn’s
`disease65 showed clinically unimpressive benefits:
`the
`release formulation did not
`influence the success of
`therapy. Low peri-anastomotic mucosal concentrations of
`5-ASA in patients on postoperative maintenance therapy
`with Asacol were associated with local recurrence66 but the
`relation of such findings to gut pH or transit is not known.
`In summary, clinical
`trial data suggest
`that
`low
`intraluminal pH could have an adverse eVect on 5-ASA
`bioavailability in patients with ulcerative colitis, particu-
`larly if active, but probably does not in Crohn’s disease. A
`head to head comparison of Pentasa with a pH dependent
`formulation is needed to test this conclusion.
`
`New formulations of other drugs in IBD:
`budesonide and azathioprine
`Changes in intraluminal
`intestinal and colonic pH in
`aVected patients also require consideration in the assess-
`ment and design of other existing and novel drugs for the
`treatment of IBD.
`Controlled ileal release budesonide approaches pred-
`nisolone in eYcacy for the treatment of active ileocaecal
`Crohn’s disease.67 68 Two diVerent pH dependent prepara-
`tions of budesonide are now available. Budesonide CR
`(Entocort CR) gelatin capsules contain acid stab