Phannaceutical Research, Vol. 15, No. 1, 1998
`
`Review
`
`Dissolution Testing as a Prognostic Tool for Oral Drug
`Absorption: Immediate Release Dosage Forms
`
`Jennifer B. Dressman,1•6 Gordon L. Amidon,2 Christos Reppas,3 and Vinod P. Shah4•5
`
`Received August 20, 1997; accepted September 27, 1997
`
`Dissolution tests are used for many purposes in the pharmaceutical industry: in the development of new
`products, for quality control and, to assist with the determination of bioequivalence. Recent regulatory
`developments such as the Biopharmaceutics Classification Scheme have highlighted the importance of
`dissolution in the regulation of post-approval changes and introduced the possibility of substituting
`dissolution tests for clinical studies in some cases. Therefore, there is a need to develop dissolution tests
`that better predict the in vivo performance of drug products. This could be achieved if the conditions in
`the gastrointestinal tract were successfully reconstructed in vitro. The aims of this article are, first, to
`clarify under which circumstances dissolution testing can be prognostic for in vivo performance, and
`second, to present physiological data relevant to the design of dissolution tests, particularly with respect
`to the composition, volume, flow rates and mixing patterns of the fluids in the gastrointestinal tract.
`Finally, brief comments are made in regard to the composition of in vitro dissolution media as well as
`the hydrodynamics and duration of the test.
`KEY WORDS: dissolution tests; prediction of in vivo performance; dissolution test conditions; composi(cid:173)
`tion of dissolution media.
`
`INTRODUCTION
`
`An important aspect of the development of a pharmaceuti(cid:173)
`cal product is to find an in vitro characteristic of potential
`formulations that reflects their in vivo performance. Although
`immediate release solid dosage forms are routinely subjected
`to tests such as content uniformity, weight, hardness, friability
`and disintegration, the test that is most often associated with
`the assessment of in vivo performance is the dissolution test.
`Currently there are about 500 tablet and capsule monographs
`in the USP which have dissolution requirements (1), and disso(cid:173)
`lution testing is an integral component of new drug applications
`to regulatory bodies worldwide.
`In vitro dissolution testing provides useful information at
`several stages of the drug development process. Formulation
`scientists use dissolution to assess the dissolution properties of
`the drug itself and thereby select appropriate excipients for the
`
`1 Johann Wolfgang Goethe-Universitat, lnstitut fiir Pharmazeutische
`Technologie, Marie-Curie-Str. 9, 60439 Frankfurt am Main, Germany.
`2 College of Pharmacy, The University of Michigan, Ann Arbor, Michi(cid:173)
`gan 48109- !065.
`3 Department of Pharmacy, University of Athens, Panepistimiopolis,
`157 71 Zografou, Greece.
`4 Office of Pharmaceutical Science, Center for Drug Evaluation and
`Research, Food and Drug Administration, 5600 Fishers Lane, HFD-
`350, Rockville, Maryland 20857.
`5 The views and suggestions presented in this article are those of the
`authors and not of the FDA.
`6 To whom correspondence should be addressed.
`
`formulation. Dissolution testing is also employed to assist in
`choosing among candidate formulations, with the aim of select(cid:173)
`ing the dosage form with the most suitable and reproducible
`release profile. However, if these tests are not performed under
`appropriate conditions, the prediction of which drugs and which
`dosage forms will exhibit the desired release profiles in vivo
`may be completely erroneous.
`Clinical scientists rely on dissolution tests to establish in
`vitro/in vivo correlations between release of drug from the
`dosage form and drug absorption. When in vitro results fail to
`adequately predict the in vivo performance of a drug product,
`more and larger clinical studies are needed to assess product
`bioavailability, thus adding substantially to the cost of product
`development. For drugs and formulations that have release rate
`limited absorption, it is also of particular interest to know
`whether the drug will be better absorbed when the product is
`given with food. Current pharmacopeial tests do not address
`this need.
`From the regulatory scientist's point of view, the evaluation
`of preclinical and clinical data would be greatly facilitated by
`the availability of validated, prognostic dissolution methodol(cid:173)
`ogy for the product. In certain cases, it may be appropriate to
`use dissolution test results to evaluate the biopharmaceutical
`implications of a product change, rather than to automatically
`require a bioequivalence study (2).
`Important aspects of the quality assurance of a drug product
`include the ability to confirm that the correct manufacturing
`procedures have been followed for a given batch, that batch(cid:173)
`to-batch reproducibility of the product meets regulatory require(cid:173)
`ments, and that the product performs adequately throughout its
`
`11
`
`0724-8741/98/0100-0011$15.00/0 © 1998 Plenum Publishing Corporation
`
`

`

`12
`
`Dressman, Amidon, Reppas, and Shah
`
`shelf life. Insofar as possible, the in vitro test conditions should
`bear a meaningful relationship to the conditions in the gastroin(cid:173)
`testinal tract (3). In the case of very poorly soluble drugs,
`however, the ability to test whether the product is able to release
`all of the active drug within the desired time-frame under physi(cid:173)
`ologically relevant test conditions may be difficult to achieve
`with current apparatus.
`In summary, there is a real need to develop dissolution
`tests that better predict in vivo performance of drug products.
`This could be achieved if the conditions in the gastrointestinal
`tract were successfully reconstructed in in vitro test systems.
`The development of prognostic in vitro tests should lead not
`only to a reduction in the work needed for formulation develop(cid:173)
`ment, but also in the number and size of clinical studies required,
`and to more meaningful quality assurance tests.
`In this article, we seek first to clarify under which circum(cid:173)
`stances dissolution testing can be prognostic for in vivo perfor(cid:173)
`mance, then to present physiological data relevant to the design
`of dissolution tests, particularly with respect to the composition
`of the medium, the hydrodynamics employed and the duration
`of the test. In a companion article, examples of drugs for which
`dissolution is highly dependent on test conditions will be used
`to illustrate the importance of selecting physiologically relevant
`test conditions for in vitro performance tests.
`
`RATE LIMITING FACTORS TO DRUG
`ABSORPTION
`
`Essentially, there are four possible sources of incomplete
`drug absorption following the oral administration of a solid
`dosage form (4):
`
`1) The drug is not delivered from its formulation over an
`appropriate time frame in solution form to those sites in the
`GI tract where it is well absorbed,
`2) The drug is decomposed in the gastrointestinal tract or
`forms a nonabsorbable complex,
`3) The drug is not transported efficiently across the gut
`wall in the apical to basal direction, and/or
`4) The drug is metabolized and/or eliminated en route to
`the systemic circulation. These possibilities are illustrated in
`Figure 1.
`
`Since the gastrointestinal tract is not a static system, the
`rate at which release, decomposition, complexation and gut
`
`Drug in systemic circulation
`
`Decompos1t1on
`Adsorption
`
`Drug in solution
`at uptake sites
`+ + +
`l'lele!'S!' -CGomp~eoxatio: _ - ~- ~- ~
`
`Decomposition
`Adsorption
`Complexation
`
`I
`I
`I
`1
`1
`1
`'
`'
`'
`Drug in solution
`at uptake sites
`
`- - ...
`
`-
`Transit
`
`-----,.....::,,_► Gut/Liver metabolism
`Biliary excretion
`
`Drug in systemic circulation
`Fig.1. Steps in drug absorption and sources of incomplete bioavailabil(cid:173)
`ity following oral administration of a solid dosage form.
`
`wall transport occur must additionally be weighed against the
`transit rate of the dosage form/drug through the gastrointestinal
`tract. In order for a drug to be well absorbed, release and uptake
`must be completed within the time taken for the dosage form/
`drug to traverse that part of the gastrointestinal tract up to and
`including the sites at which the drug is absorbed, whereas
`decomposition and complexation must occur more slowly than
`either release/uptake or transit.
`
`THE BIOPHARMACEUTICS CLASSIFICATION
`SCHEME
`
`Recently, aBiopharmaceutics Classification Scheme (BCS)
`has been proposed (5). Under this scheme (Table 1), drugs
`can be categorized into four basic groups according to their
`solubility properties and their ability to penetrate the gastroin(cid:173)
`testinal mucosa.
`Thus, the BCS addresses two of the potential four limita(cid:173)
`tions to oral drug bioavailability. Of these two, drug solubility
`recognizes the physicochemical limitations of the drug as a
`potential source of incomplete release from the dosage form.
`It is important to understand that the classification is based on
`the solubility properties of the drug substance throughout the
`upper GI tract. In the Commentary Section, appropriate media
`for such studies are suggested. The results of dissolution studies
`with the dtug in the same media can be subsequently used in
`the development process to assess the influence of formulation
`on the release rate. Permeability studies are needed to locate
`the main sites of drug absorption in the gastrointestinal tract,
`as well as assessing the efficiency of drug transport across the
`gut wall. A variety of cell culture, tissue and animal models
`are available for assessing permeability; currently there are
`also several groups studying permeability of drugs directly in
`humans (6,7). In principle, studies addressing drug stability
`problems in the lumen of the gastrointestinal tract should be
`run in media that reproduce the conditions to which the drug
`is likely to be subjected. In this respect, considerations for
`design of stability test media parallel those applicable to release
`and dissolution studies. In the case of drugs that undergo metab(cid:173)
`olism in the gut wall and/or liver, the rate and extent of the
`effect must be assessed using tissue preparations or from phar(cid:173)
`macokinetic analysis.
`Although the BCS is limited to two of the four important
`factors, it nonetheless provides a useful starting point for recog(cid:173)
`nizing when and how dissolution tests can aid in the design
`and evaluation of oral dosage forms. Compounds belonging to
`Class I, i.e. compounds with high solubility and permeability,
`should go into solution quickly when they are housed in immedi(cid:173)
`ate release dosage forms, and also be rapidly transported across
`
`Table 1. The Biopharmaceutics Classification Scheme
`
`Class I:
`HIGH SOLUBILITY
`HIGH PERMEABILITY
`
`Class III:
`HIGH SOLUBILITY
`LOW PERMEABILITY
`
`Note: From Ref. 5.
`
`Class II:
`LOW SOLUBILITY
`HIGH PERMEABILITY
`
`Class IV:
`LOW SOLUBILITY
`LOW PERMEABILITY
`
`

`

`Dissolution Testing as a Prognostic Tool for Oral Drug Absorption
`
`13
`
`the gut wall. Therefore, it is expected that they will be well
`absorbed unless they are unstable, form insoluble complexes,
`are secreted directly from the gut wall, or undergo first pass
`metabolism. Dissolution tests for immediate release formula(cid:173)
`tions of Class I drugs, therefore, need only to verify that the
`drug is indeed rapidly released from the dosage form under
`mild aqueous conditions.
`For Class II drugs, by contrast, the rate of dissolution of
`the drug is almost certain to be the principal limitation to its
`oral absorption. The limitation can be equilibrium or kinetic in
`nature. In the case of an 'equilibrium' problem there is not
`enough fluid available in the gastrointestinal tract to dissolve
`the dose. This can be checked by calculating the dose:solubility
`ratio (8). For example, at a dose of 500 mg and an aqueous
`solubility of 15 µ,g/ml at 37° C, 33 liters of fluid are required
`to dissolve one dose of griseofulvin. Since the total volume of
`fluid entering the gastrointestinal tract in a twenty-four hour
`period is only about five to ten liters (9), there is clearly insuffi(cid:173)
`cient volume present at any given time for the entire dose of
`griseofulvin to be dissolved. In the case of a 'kinetic' problem,
`the drug dissolves too slowly for the entire dose to become
`dissolved before the drug has passed by its sites of uptake.
`Digoxin, for example, with a dose of 0.25 mg and a solubility
`of 20 µ,g/ml, has a dose:solubility ratio of just 12.5 ml. Despite
`the small volume of fluids required to dissolve the drug, digoxin
`exhibits dissolution rate limited absorption at particle sizes of
`greater than 10µ, in diameter (8) because the poor driving force
`for dissolution supplied by the solubility, combined with the
`low surface area of drug at larger particle sizes, is insufficient
`to ensure timely dissolution. For Class II drugs, it should there(cid:173)
`fore be possible to establish a strong correlation between the
`results of dissolution tests and the in vivo absorption rate. Estab(cid:173)
`lishment of an in vitro/in vivo correlation and the resultant
`ability to discriminate between formulations with different bio(cid:173)
`availabilities will be dependent on how well the in vitro tests
`are designed. In order to be successful, it is necessary to repro(cid:173)
`duce the conditions extant in the gastrointestinal tract following
`administration of the dosage form as closely as possible. Ade(cid:173)
`quate comparison of formulations for Class II drugs requires
`dissolution tests with multiple sampling times in order to charac(cid:173)
`terize the release profile (2), and in some cases the use of more
`than one dissolution medium may also be worth considering.
`Like compounds belonging to Class I, Class III drugs are
`rapidly dissolving and the test criterion should be that the
`formulation can release the drug under mild aqueous conditions
`within a predetermined time. Rapid dissolution is particularly
`desirable for Class III drugs, in order to maximize the contact
`time between the dissolved drug and the absorbing mucosa,
`and consequently the bioavailability of the compound. There(cid:173)
`fore, the duration of the dissolution test should be at least
`as stringent for Class III drugs as for Class I drugs. Class
`IV drugs are expected to have poor absorption in general,
`but it is anticipated that, as in the case of Class II drugs,
`poor formulation could have an additional, negative influence
`on both the rate and extent of drug absorption. Thus, for all
`four categories, it is anticipated that well-designed dissolution
`tests can be a key prognostic tool in the assessment of both
`the drug's potential for oral absorption and of the bioequi valence
`of its formulations.
`
`IMPORTANT CONSIDERATIONS IN DISSOLUTION
`AND THEIR CORRESPONDING
`PHYSIOLOGICAL PARAMETERS
`
`From the following equation, based on the Nemst-Brunner
`and Levich modifications of the Noyes-Whitney model ( l 0-12),
`the factors important to the kinetics of drug dissolution can
`be identified:
`
`dXct A*D
`-= - - * (C -XctN)
`dt
`8
`s
`
`where A is the effective surface area of the solid drug, D is
`the diffusion coefficient of the drug, 8 is the effective diffusion
`boundary layer thickness adjacent to the dissolving surface, C,
`is the saturation solubility of the drug under lumenal conditions,
`Xct is the amount of drug already in solution and V is the volume
`of the dissolution medium. Some of these factors are primarily
`influenced by physicochemical properties of the drugs, but most
`are also influenced by the conditions in the gastrointestinal tract.
`A summary of the relevant physicochemical and physiological
`parameters is given in Table 2.
`The key factors in the dissolution of drugs in the gastroin(cid:173)
`testinal tract are thus the composition, volume and hydrodynam(cid:173)
`ics of the contents in the lumen following administration of the
`dosage form. Only when these factors are adequately repro(cid:173)
`duced in vitro can we expect to accurately predict dissolution
`limitations to absorption.
`In addition to these factors, the permeability of the gut
`wall to the compound plays a role in the maintenance of sink
`(less than 20% of saturation concentration) conditions for disso(cid:173)
`lution, which are required for the fastest possible dissolution
`rate. For highly permeable drugs sink conditions are likely to
`be maintained, in which case the dissolution rate per unit surface
`area will be constant and close to the initial dissolution rate.
`For less permeable drugs, the dissolution rate per unit surface
`area will decrease with time, due to the gradual buildup of drug
`in solution in the lumen.
`The lumenal conditions in the gastrointestinal tract vary
`widely both within and between subjects. Intersubject variabil-
`
`Table 2. Physicochemical and Physiological Parameters Important to
`Drug Dissolution in the Gastrointestinal Tract
`
`Factor
`
`Surface area of
`drug
`Diffusivity of
`drug
`boundary layer
`thickness
`Solubility
`
`Amount of drug
`already
`dissolved
`Volume of sol(cid:173)
`vent available
`
`Physicochemical
`parameter
`
`Physiological
`parameter
`
`particle size,
`wettability
`molecular size
`
`hydrophilicity,
`crystal structure,
`solubilization
`
`surfactants in gastric
`juice and bile
`viscosity of lumenal
`contents
`motility patterns &
`flow rate
`pH, buffer capacity,
`bile, food components
`
`permeability
`
`secretions,
`coadministered
`fluids
`
`

`

`14
`
`Dressman, Amidon, Reppas, and Shah
`
`ity arises from normal genetic variation in the population (as
`in the case of heart rate, liver function and other physiological
`parameters) as well as from disease states implicating the gastro(cid:173)
`intestinal tract. Intrasubject variability may occur as the result
`of circadian rhythm, food ingestion, physical activity level and
`stress level, among others. This variability notwithstanding, the
`remainder of this Section is devoted to a summary of representa(cid:173)
`tive values for key parameters in the fed and fasted states in
`different segments of the gastrointestinal tract.
`
`Lumenal composition in the GI tract
`In addition to food and beverages ingested with the dosage
`form, various fluids are secreted by the gastrointestinal tract,
`including hydrochloric acid, bicarbonate, enzymes, surfactants,
`electrolytes, mucus and, of course, water. Thus, parameters that
`can profoundly influence the solubility and dissolution rate of
`a drug, e.g. pH, buffer capacity, presence of surfactant concen(cid:173)
`tration and volume of lumenal contents, may vary widely with
`position in the gastrointestinal tract and with timing of adminis(cid:173)
`tration of the drug in relation to meal intake.
`
`pH
`
`Values of gastric pH in the fasted state can fluctuate on a
`minute-to-minute basis over the range pH 1 to pH 7, but in
`healthy, young Caucasians gastric pH lies below pH 3 during
`90% of the fasted state (13), with an interquartile range of pH
`1.4 to pH 2.1. Suitable dissolution media for simulating the
`fasted state gastric conditions will therefore have pH values
`between pH 1.5 and pH 2. Fasted state gastric pH values of
`pH 6 and higher are found in two significant subpopulations:
`those receiving gastric acid blocker therapy and those over the
`age of 65 years (about 10--20 % of North Americans (14)
`and Europeans (15) acquire hypo/achlorhydria; the incidence
`appears to be much higher in Japan (16)). With ingestion of a
`meal, the gastric juice is initially buffered to a less acidic pH,
`which is dependent on the meal composition. Typical gastric
`pH values immediately following meal ingestion are in the
`range pH 3 to pH 7. Depending on meal size, the gastric pH
`returns to fasted state values within two to three hours. Thus,
`only dosage forms ingested with or soon after meal intake will
`encounter elevated gastric pH under normal physiological
`circumstances.
`Intestinal pH values (Table 3) are considerably higher than
`gastric pH values due to the neutralization of incoming acid
`
`Table 3. pH in the Small Intestine in Healthy Humans in the Fasted
`and Fed States"
`
`Location
`
`fasted state pH
`
`fed state pH
`
`mid-distal duodenum
`
`jejunum
`
`ileum
`
`4.9
`6.1
`6.3
`6.4
`4.4--6.5
`6.6
`6.5
`6.8-8.0 (range)
`7.4
`
`5.2
`5.4
`5.1
`
`5.2-6.0
`6.2
`6.8-7.8
`6.8-8.0
`7.5
`
`a Reproduced from Ref. 17, which summarized results from several
`studies in the literature.
`
`Table 4. Comparison of Average pH and Buffer Capacity of Chyme
`Recovered at Midgut From Fistulated Dogs, After Administration of
`Nonnutrient and Nutrient 'Meals', with Those of Simulated Intestinal
`Fluid USP, Without Pancreatin
`
`Sample
`
`Nonnutrient 'meal' (water)
`Nutrient meal ( cheeseburger,
`fries, water)
`SIFsp (USP)*
`
`pH
`
`6.0"
`
`5.2a
`7.5
`
`Buffer capacity
`(mEq/L/pH
`unit)
`
`0.16 ± 0.16/J
`
`76 :!: 25h
`25.8 ± 0.8
`
`Note: Excerpted from Ref. 20.
`Shared letters indicate significant differences, "p < 0.05, hp < 0.005.
`* SIFsp was prepared according to the USP, but without pancreatin.
`At the time the studies were initiated, the official pH of the medium
`was 7.5.
`
`with bicarbonate ion secreted by the pancreas. Furthermore,
`there is a pH gradient in the small intestine, with values gradu(cid:173)
`ally rising between the duodenum and ileum. pH values in the
`colon are heavily influenced by products of bacterial exoenzyme
`reactions. Undigested carbohydrate that is passed into the colon
`is converted into short chain fatty acids (C2-C4) that lower
`the local pH value to around pH 5 ( 18). Thus, when suitable
`carbohydrate substrates are present, the pH in the proximal
`colon may be 2-3 pH units lower than in the terminal ileum.
`
`Buffer Capacity
`
`The microclimate pH in the diffusion boundary layer adja(cid:173)
`cent to the dissolving surface is an important determinant to
`the dissolution of ionizable drugs. In addition to the intrinsic
`solubility and ionization constant of the drug and the pH of
`the medium, the buffer capacity of the medium plays an
`important role in determining the microclimate pH (19). Data
`obtained in a fistulated dog model (20) suggest that the buffer
`capacity at midgut is far greater after a cheeseburger/fries/water
`meal than following administration of water (Table 4).
`
`Surfactants
`
`The surface tension of gastric fluid is considerably lower
`than that of water, suggesting the presence of surfactants in
`this region. Usual values in the fasted state lie between 35 and
`45 mN.m- 1 (21). In the small intestine, secretion of bile results
`in substantial concentrations of bile salts and lecithin, which
`form mixed micelles even at fasted state concentrations.
`Fasting bile salt concentrations of about 3-5 mM have been
`reported for the proximal small intestine (Table 5). Although
`
`Table 5. Fasting Bile Salt Concentrations in the Human Small
`Intestine"
`
`statistic
`
`mean ± s.d. (mM)
`
`duodenum
`
`6.4 ± 1.3
`4.3 ± 1.2
`
`median
`range
`
`" Data from Refs. 22-25.
`
`upper
`jejunum
`
`lower
`jejunum
`
`5
`
`3
`0-14
`
`6
`
`5
`0-17
`
`

`

`Dissolution Testing as a Prognostic Tool for Oral Drug Absorption
`
`15
`
`Table 6. Postprandial Bile Salt Concentrations in the Human Small
`Intestine
`
`Time
`
`Location
`
`Statistic
`
`Reference
`Number
`
`0-30 min
`
`duodenum
`
`upper jejunum
`
`30--60 min
`
`duodenum
`upper jejunum
`
`120--150 min
`
`upper jejunum
`
`mean 14.5 ± 9.4
`range 5.8-39.6
`mean 16.2 ± 1.5
`mean 15
`range 4-34
`mean 5.2 ± 2.3
`mean 9.7 ± 1
`mean 8
`range 3-17
`mean 6.5 ± 0.9
`
`29
`
`30
`23
`
`29
`30
`23
`
`30
`
`concentrations vary widely between individuals, average values
`are similar in the duodenum and jejunum. Levels fall rapidly
`in the ileum where bile salts are absorbed by an active transport
`mechanism, and are insignificant in the colon in healthy
`individuals.
`After eating, the bile output and lumenal concentration of
`bile components (Table 6) peak within thirty minutes (26).
`Thereafter levels gradually decline, mostly because of dilution
`with chyme. The peak level averages about 15 mM in the
`proximal small intestine. Since the gallbladder empties into the
`upper small intestine, duodenal levels tend to fluctuate more
`with meal ingestion than levels in the distal small intestine
`(27,28).
`
`Enzymes
`
`The primary enzyme found in gastric juice is pepsin, an
`exopeptidase. Lipases, amylases and proteases (Table 7) are
`secreted from the pancreas (31) into the small intestine in
`response to meal ingestion; these enzymes are responsible for
`the bulk of nutrient digestion. Pepsin and the pancreatic prote(cid:173)
`ases pose a particular threat to stability of proteins and peptides
`in the lumen, while lipases may affect release of drugs from
`fat/oil containing dosage forms.
`
`Bacteria, which mainly populate the distal ileum and the
`colon, also secrete diverse enzymes. Table 8 (32) illustrates
`some of the enzymes available, classified according to the
`reactions that they catalyze. The ability of bacterial exoenzymes
`to split certain chemical bonds has been used to design dosage
`forms intended for colonic delivery, such as azo polymers and
`some hydrogels (33-35).
`
`Volume
`The volume of fluids available in the gastrointestinal tract
`for drug dissolution is dependent upon the volume of coadminis(cid:173)
`tered fluids, secretions and water flux across the gut wall.
`About 2 liters per day are ingested orally, though this varies
`considerably with climate, body weight, activity and personal
`habit (9). The volume of the stomach in the fasted state may
`be as little as 20-30 mL, mostly present as wet mucus rather
`than as a fluid pool. At the other extreme, gastric pressure starts
`to rise when a volume of about 1.5 liters is exceeded (36).
`The secretions of the para-gastrointestinal organs (salivary
`glands, liver, pancreas) as well as the secretion of the stomach,
`are received by the first portion of the duodenum. These endoge(cid:173)
`nous secretions, totalling about 6 liters per day, are essential
`for the normal lumenal digestion of foodstuffs. Approximately
`1-2 L of pancreatic juice are secreted into the duodenum over
`a 24 hour period (37) while bile output in a 24 hour period
`totals about 600 mL. Most of the pancreatic and biliary output
`is secreted postprandially. In addition, the intestine secretes
`about 1 liter of water per day, mostly as a component of mucus.
`According to the perfusion studies of Dillard et al. (38),
`the volume of fluid in the jejunum and ileum varies from
`120-350 mL, depending on the perfusion rate. In a landmark
`study by Fordtran and Locklear (39) (Figure 2), electrolyte and
`volume measurements were compared at different sites within
`the small intestine after ingestion of hypertonic (milk/dough(cid:173)
`nuts) and hypotonic (steak and water) meals. Volumes were
`considerably higher following administration of a hypertonic
`meal than after administration of a hypotonic meal. In the case
`of a hypertonic meal, net water efflux across the mucosa into
`the lumen occurs due to the osmotic pressure difference, while
`in the case of a hypotonic meal, there is net water absorption
`from the meal.
`
`Table 7. Characteristics of Some Exocrine Pancreatic Enzymes
`
`(pro )enzyme
`
`trypsin( ogen)
`chymotrypsin( ogen)
`(pro)carboxypeptidase A
`(pro)carboxypeptidase B
`(pro )elastase
`ribonucleases
`lipase I
`
`lipase 2
`amylase
`
`%
`output
`
`33
`16
`12
`9
`8
`I
`8.5
`
`3.4
`3.6
`
`Note: Excerpted from Ref. 31.
`
`substrates
`
`products
`
`proteins/polypep
`proteins/polypep
`proteins/polypep
`proteins/polypep
`proteins/polypep
`nucleic acids
`triglycerides
`
`triglycerides
`polysaccharides
`
`peptides, amino acids
`peptides, aminoacids
`amino acids
`amino acids
`amino acids
`mono-nucleotides
`fatty acids
`monoglycerides
`monoglycerides
`disaccharides
`trisaccharides
`limit dextrins
`
`

`

`16
`
`Dressman, Amidon, Reppas, and Shah
`
`Table 8. Bacterial Flora in the Colon and Their Exoenzymes
`
`Bacteria
`
`Bacteroides
`Clostridia
`
`Enterobacteria
`
`Lactobacilli
`
`Reductive
`Reactions
`
`Hydrolytic
`Reactions
`
`nitroreductase
`azoreductase
`hydrogenase
`nitroreductase
`N-oxide-reductase
`sulfoxide-reductase
`azoreductase
`hydrogenase
`
`glucosidase
`sulfatase
`esterase, amidase
`glucuronidase
`sulfatase
`
`Note: Reproduced with permission from Ref. 32.
`
`Fluid levels tend to be lower at more distal locations. Only
`about 1.5 liters are presented to the colon daily, of which about
`1.3 liters are absorbed, with the rest forming a component of
`the stool (40).
`
`Hydrodynamics in the GI Tract
`
`Mixing Patterns in the Gut
`
`The hydrodynamics in the gastrointestinal tract, that is,
`how well the lumenal contents are mixed, play an important role
`in dissolution through their influence on the effective boundary
`layer thickness, 6. In the upper GI tract, there are basically four
`motility patterns: no activity (quiescence), segmental move(cid:173)
`ments, propagative movements (short or long range) and tonic
`contractions ( 41 ).
`In the fasting stomach, long periods of little or no motor
`activity occur. About once every two hours, contractions start
`to occur which gradually become more frequent and more
`forceful, until they culminate in a burst of activity that clears
`the contents of the stomach into the intestine (so-called Phase
`III activity). The quiescent phase can, of course, be modelled
`by a stagnant system. To date, however, no useful quantitative
`model of the hydrodynamics of the stomach in the more active
`phases of the fasted state or in the fed state pattern has been
`developed.
`As far as the small intestine is concerned, some data is
`available regarding the ratio of segmental to propagative motil(cid:173)
`ity as a function of the fed and fasted states. Segmentation is
`
`STEAK MEAL
`
`Ml!..K S DOUGHNUTS j
`I
`MEAL
`
`a
`w
`~ 80
`
`Fig. 2. Water volumes and electrolyte concentrations in the small
`intestine following ingestion of a hypotonic steak/water meal (Panel
`A), and a hypertonic milk/doughnuts meal (Panel B). (Reproduced
`with permission from Ref. 39).
`
`Valve
`
`Meal Marker Ingested
`(hours before radiographs)
`12 e
`24
`0
`
`36 □
`
`Fig. 3. Distribution of markers in the colon after administration at 12
`hour intervals. (Reproduced with permission from Ref. 43).
`
`the predominant mixing pattern in the small intestine, and is
`characteristic of the fed state. Segmental contractions tend to
`occur over very short distances, typically less than 2 cm, and
`serve to mix the lumenal contents thoroughly. Short propulsive
`movements, on the other hand, provide the main mechanism
`by which the lumenal contents are moved down the intestine.
`These usually result in movement of the chyme over distances
`of 15 cm or less (41).
`With the change in the ratio of segmenting to propulsive
`activity as a function of the phase of the motor pattern and
`upon meal ingestion, the efficiency of absorption also changes.
`Absorption is least efficient during Phase III (fasted state long
`range contractions), is intermediate during Phase I and Phase
`II and is greatest during the fed state motor pattern (42). Simi(cid:173)
`larly, it is expected that because of better mixing, dissolution
`will also be most efficient in the fed state.
`In the stomach and small intestine, movement of lumenal
`contents is virtually always in the distal direction. In the proxi(cid:173)
`mal colon, however, mixing can occur longitudinally as well
`as laterally, because some contractions drive the contents in
`the proximal rather than the distal direction. Figure 3 shows
`representative results from a study in which subjects ingested
`radio opaque markers with meals 36, 24, and 12 hours prior
`to radiography. Some of the markers taken 36 hours earlier are
`still in the ascending colon while some of the markers taken
`only 12 hours before are already at the end of the transverse
`colon (43). Based on these data, one may assume that some
`degree of mixing of fluids with solids occurs in the colon,
`although this has not been quantitatively defined.
`
`Flow Rates in the GI Tract
`
`Flow rates out of the stomach and at various locations in
`the small intestine have been measured in both the fasted and
`fed states. Emptying rates can be as high as 40 mL/minute
`immediately after ingestion of a 400 mL volume of normal
`saline ( 44 ). On the other hand, when small volumes are ingested
`in the quiescent phase of gastric motor activity the emptying
`rate may be virtually negligible (Table 9) (45). Nutrient fluids
`empty from the stomach according to zero order kinetics, with
`em