3

Dissolution Testing1
Sau Lawrence Lee, Andre S. Raw, and Lawrence Yu

3.1

Introduction

Ever since dissolution was known to have a signicant effect on bioavailability and
clinical performance, dissolution analysis of pharmaceutical solids has become
one of the most important tests in drug product development and manufacturing,
as well as in regulatory assessment of drug product quality. Not only can dissolution testing provide information regarding the rate and extent of drug absorption
in the body, it can also assess the effects of drug substance biopharmaceutical
properties and formulation principles on the release properties of a drug product. Nevertheless, despite the wide use of dissolution testing by the pharmaceutical industry and regulatory agencies, the fundamentals and utilities of dissolution
testing are still not fully understood. The objective of this chapter is to provide
a concise review of dissolution methods that are used for quality control (QC)
and bioavailability assessment, highlight issues regarding their utilities and limitations, and review challenges of improving some of these current dissolution methods, particularly those used for assessing in vivo drug product performance. In
this chapter, we rst provide some background information on dissolution, including the signicance of dissolution in drug absorption, theories of dissolution, and
factors affecting dissolution testing. Second, we examine the current roles of dissolution testing. Third, we evaluate the utilities and limitations of dissolution as
a QC tool under the current industry setting. Finally, we conclude this chapter
by discussing the biopharmaceutics classication system (BCS) and biorelevant
dissolution methods.

3.2

Signicance of Dissolution in Drug Absorption

Oral administration of solid formulations has been the most common route of
administration for almost a century. However, the importance of dissolution
processes in the oral drug absorption was only recognized about 50 years ago
when Nelson published his nding that showed a relationship between the blood
1 The opinions expressed in this chapter by the authors do not necessarily reect the views

or policies of the Food and Drug Administration (FDA).

47

48

S. L. Lee et al.

Gastric
Emptying
Metabolism
Dissolution

Transit

Dissolution

Permeation

F IGURE 3.1. Schematic representation of the simplied oral drug absorption process that
consists of transit (gastric emptying), dissolution, permeation, and rst pass metabolism

levels of orally administered theophylline salts and their in vitro dissolution

rate (Nelson, 1957). The need for dissolution testing can be understood easily
by considering the importance of dissolution on oral drug absorption, which is
described below.
When a systemically acting drug is administered in solid dosage forms, such
as a tablet or capsule, its absorption into the systemic circulation can be generally described by four consecutive steps (Fig. 3.1). The rst step involves delivery
of the drug into its absorption site through gastric emptying and intestinal transit ow. It is followed by the second step in which dissolution takes place in the
stomach and/or in the small intestine. It should be noted that the rst two steps
need not to be sequential and that lymphatic absorption is not considered. The
third step is characterized by the permeation of the dissolved drug across the gastrointestinal (GI) membrane. Finally, the absorbed drug passes through the liver
(rst pass metabolism) and reaches the systemic circulation. Although this is a
simplied description of the drug absorption process, it shows that transit (gastric
emptying), dissolution, absorption across intestinal membrane, and metabolism
constitute the fundamental processes of oral drug absorption. If the dissolution
process is slow relative to the other three processes, which is usually the case for
most poorly soluble drugs formulated in a conventional dosage form, dissolution
will be the rate limiting step. As a result, the dissolution rate will determine the
overall rate and extent of drug absorption into a systemic circulation, and hence
bioavailability.

3. Dissolution Testing

3.3

49

Theories of Dissolution

Dissolution is generally dened as a process by which a solid substance is solubilized into the solvent to yield a solution. This process is fundamentally controlled by the afnity between the solid substance and the solvent and consists
of two consecutive steps. The rst step involves the liberation of molecules from
the solid phase to the liquid layer near the solid surface (an interfacial reaction
between the solid surface and the solvent). It is followed by the transport of solutes
from the solidliquid interface into the bulk solution. The dissolution of solid substance is generally modeled based upon the relative signicance of these two transport steps.
The diffusion layer model proposed originally by Nernst and Brunner (Brunner,
1904; Nernst, 1904) is widely used to describe the dissolution of pure solid substances. In this model, it is assumed that a diffusion layer (or a stagnant liquid lm
layer) of the thickness h is surrounding the surface of a dissolving particle. The
reaction at the solidliquid interface is assumed to be instantaneous. Thus, equilibrium exists at the interface, and hence the concentration of the surface is the saturated solubility of the substance (Cs ). Once the solute molecules diffuse through
the lm layer and reach the liquid lmsolvent interface, rapid mixing takes place,
resulting in a uniform bulk concentration (C). Based upon this description (see
Fig. 3.2a), the dissolution rate is determined entirely by Brownian motion diffusion of the molecules in the diffusion layer.
To model the diffusion process through the liquid lm, Ficks rst law, which
relates ux of a solute to its concentration gradient, can be applied:
J = D

dC
,
dx

(3.1)

where J is the amount of solute passing through a unit area perpendicular to the
surface per unit time. D is the diffusion coefcient, and dC/dx is the concentration gradient, which represents a driving force for diffusion. At steady state, (3.1)
becomes
C Cs
,
(3.2)
J = D
h
where C is the bulk concentration, Cs is the saturation concentration, and h is
the thickness of the stagnant diffusion layer. Based on (3.2), the dissolution rate,
which is proportional to the ux of solutes across the diffusion layer, can be
described by
V

dC
= SJ
dt

(3.3)

50

S. L. Lee et al.

Diffusion Layer

a. Diffusion Layer Model

Cs

h
Film
Boundary

Particle
surface
C

b. Interfacial Barrier Model

h
Film
Boundary

Cs
Particle
surface
C

Cs

c. Danckwerts Model
C
Particle
surface

F IGURE 3.2. Schematic illustration of (a) the diffusion layer model, (b) the interfacial barrier model, and (c) the Danckwerts model

or

SD
dC
=
(3.4)
(Cs C) ,
dt
Vh
where S is the total surface area of particles, and V is the volume of dissolution
medium. The term Cs C represents the concentration gradient within the stagnant
diffusion layer with thickness h. This equation is known as the NernstBrunner
equation (Brunner and Tolloczko, 1900; Nernst, 1904).
In addition to lm theory, two other theories were also used to describe the dissolution process. These theories include the interfacial barrier model (Higuchi,
1961) and the Danckwerts model (Danckwerts, 1951). In contrast to the lm
model, the interfacial barrier model assumes that the reaction at the solid surface is signicantly slower than the diffusion across the interface. Therefore, no

3. Dissolution Testing

51

equilibrium exists at the surface, and the liberation of solutes at the solidliquid
interface controls the overall rate of the transport process. This model is illustrated
in Fig. 3.2b. Based on this model, the dissolution rate is given by
G = ki (Cs C),

(3.5)

where G is the dissolution rate per unit area, and ki is the interfacial transport
coefcient.
Under the assumption that the solid surface reaction is instantaneous, the
Danckwerts model suggests that the transport of solute is achieved by the macroscopic packets that reach the solid surface, absorb solutes at the surface, and
deliver them to the bulk solution. This transport phenomenon is depicted in
Fig. 3.2c. The dissolution rate is expressed as
dm
= S( D)1/2 (Cs C),
dt

(3.6)

where m is the mass of dissolved substances and is the interfacial tension.

These three models have been employed alone or in combination to describe the
mechanism of dissolution. Nevertheless, the diffusion layer model is the simplest
and most commonly used to describe the dissolution process of a pure substance
among these three models.

3.4

Factors Affecting Dissolution

Several physicochemical processes need to be considered along with the drug

substance dissolution process to determine the overall dissolution rate of drugs
from solid dosage forms under standardized conditions. The dissolution process
for a solid dosage form (or a drug product) in solution starts with the wetting and
the penetration of the dissolution medium into the solid formulation. It is generally followed by disintegration and/or deaggregation into granules or ne particles. However, this step is not a prerequisite for dissolution. The nal step involves
solubilization (or dissolution) of the drug substance into the dissolution medium.
A schematic diagram illustrating the processes involved in the dissolution of solid
dosage forms is shown in Fig. 3.3. It should be noted that these steps can also occur
simultaneously during the dissolution process. For most poorly soluble drugs, dissolution is considered to be dissolution controlled, since solubilization of drug
particles is slow relative to disintegration or deaggregation of the dosage form.
If the step of disintegration or deaggregation is rate-limiting, dissolution is considered to be disintegration controlled. The factors that affect the dissolution rate
of solid dosage forms can be classied under four main categories: (1) factors
related to the physicochemical properties of the drug substance, (2) factors related
to drug product formulations, (3) factors related to manufacturing processes, and
(4) factors related to dissolution testing conditions.

52

S. L. Lee et al.

Solid
Dosage
Form

Disintegration

Dissolution
(minor)

Granules
or
Aggregates

Disaggregation

Dissolution
(major)

Fine
Particles

Dissolution
(major)

Drug in vitro or in vivo

Absorption (in vivo)
Drug in Blood, Other Fluids and Tissues

F IGURE 3.3. Schematic illustration of a dissolution process of a solid dosage form (modied
from Wagner 1970)

3.4.1 Factors Related to the Physicochemical Properties

of the Drug Substance
The importance of the physicochemical properties on the dissolution of a drug
substance into the dissolution medium is best illustrated by (3.4)(3.6). Despite
the fact that these three equations are derived from different diffusion mechanisms,
they clearly show that the dissolution rate depends on the solubility and surface
area of a drug substance.
3.4.1.1 Solubility
From (3.4)(3.6), it is evident that compounds with high solubility generally
exhibit higher dissolution rates. The solubility of ionizable drugs, such as weak
acids and bases, depends upon both the pH of the medium and the pKa of the
compound. Therefore, it is important to ascertain the aqueous solubility of the
drug substance over the physiologically relevant pH range of 17.5 in order to
predict the effect of solubility on dissolution. The study of Yu et al. shows that
there is a good relationship between solubility and disk intrinsic dissolution rate
unless an extremely high or low dose is used. Solubility data may also be used as
a rough predictor for indicating any potential problems with oral absorption. For
example, when the dose/solubility of the drug, which provides an estimate of the
uid volume required to dissolve an individual dose, exceeds about 1 L, in vivo
dissolution is often considered problematic (Yu and Amidon, 1999).
3.4.1.2 Particle Size
According to (3.4) and (3.6), the dissolution rate is directly proportional to the
surface area of the drug. Reducing particle size leads to an increase in the surface

3. Dissolution Testing

53

area exposed to the dissolution medium, resulting in a greater dissolution rate.

Thus, the dissolution rate of poorly soluble drugs can often be enhanced markedly
by undergoing size reduction (e.g., through micronization). This is evidenced in
the case of glyburide tablets (Stavchansky and McGinity, 1989). However, particle size reduction does not always improve the dissolution rate. This is in part
attributed to adsorption of air on the surface of hydrophobic drugs, which inhibits
the wetting and hence reduces the effective surface area. In addition, ne particles
tend to agglomerate in order to minimize the surface energy, which also leads to a
decrease in the effective surface area for dissolution.
3.4.1.3 Solid Phase Characteristics
A drug substance may exist in different solid-state forms (polymorphism). These
different forms can be generally classied into three distinct classes including
(1) crystalline phases that have different arrangements and/or conformations of
the molecules in the crystal lattice, (2) solvates that contain either stoichiometric
or nonstoichiometric amount of a solvent, and (3) amorphous phases that do not
possess a distinguishable crystal lattice. Differences in the lattice energies among
various polymorphic forms can result in differences in the solubilities. Sometimes,
the solubility of different drug substance polymorphs can vary signicantly. For
example, the solubility of amorphous forms can be several hundred times greater
than that of the corresponding crystalline state. These solubility differences may
alter drug product in vivo dissolution, hence affecting oral drug absorption.
3.4.1.4 Salt Effects
Salt formation is frequently used to increase the solubility of a weak acid and
base. The solubility enhancement of a drug substance by salt formation is related
to several factors including the thermodynamically favored aqueous solvation of
cations or anions used to create the salt of the active moiety, the differing energies
of the salt crystal lattice, and the ability of the salt to alter the resultant pH. In
addition, even if the salt formation has no impact on the solubility of the drug, the
dissolution rate of the salt will often be enhanced due to the difference in the pH of
the thin diffusion layer surrounding the drug particles (Stavchansky and McGinity,
1989).

3.4.2

Factors Related to Drug Product Formulation

In addition to the physicochemical properties of a drug substance, inactive ingredients (or excipients) may inuence the dissolution of a drug product. The effect of
these excipients on the drug product dissolution rate depends on the dosage form.
For immediate-release dosage forms, excipients are often used to improve the drug
release from the formulation or the solubilization of a drug substance. For instance,
disintegrants such as starch are often used to facilitate the break up of a tablet and
promote deaggregation into granules or particles after administration (Peck et al.,
1989). For poorly soluble drugs, incorporation of surfactants (e.g., polysorbate)

54

S. L. Lee et al.

into the formulation may increase the dissolution rate of these products. The mechanism by which surfactants enhance the dissolution rate is to improve the solubility of the drug substance by promoting drug wetting, by forming micelles, and by
decreasing the surface tension of hydrophobic drug particles with the dissolution
medium (Banaker, 1991). Furthermore, coprecipitation with polyvinylpyrrolidine
(PVP) has been shown to signicantly inuence the dissolution (Corrigan, 1985).
This enhancing effect on the dissolution rate can be attributed to the formation of
an energetic amorphous phase or molecular dispersion. However, some excipients
may have an adverse effect on the dissolution rate. For example, lubricants such as
stearates, which are used to reduce friction between the granulation and die wall
during compression and ejection, are often hydrophobic in nature. Thus, these
hydrophobic lubricants may affect the wettability of a drug product (Pinnamaneni
et al., 2002).
For modied-release drug products, specic excipients are selected to control
the rate and extent of drug release from the formulation matrix, and/or to target
the delivery to selective sites in the GI tract. For instance, in matrix-based formulations, the active ingredient is embedded in a polymer matrix, which controls
drug release through using mechanisms such as swelling, diffusion, erosion, or
combinations (Gandhi et al., 1999). In designing these complex formulations, in
addition to the characteristics of modifying release excipients, the physicochemical properties of a drug substance, the interactions between the drug substance and
excipients, the type of the release mechanism and the target release prole must
be taken into consideration.

3.4.3

Factors Related to Manufacturing Processes

Many manufacturing process factors can have an impact on the dissolution characteristics of solid dosage forms. Very often, an appropriate unit operation is selected
to enhance the dissolution rates of a drug product. Wet granulation, in general, has
been shown to improve the wettability of poorly soluble drugs by incorporating
hydrophilic properties into the surface of granules, hence resulting in a greater
dissolution rate (Bandelin, 1990). Based upon the propensity for directly compressed tablets to deaggregate into ner drug particles, direct compression may be
chosen over granulation for improving dissolution (Shangraw, 1990). Manufacturing variables may also have both positive and negative effects upon drug product
dissolution. In tablet compression, there are always two competing factors: the
positive effect due to the increase in the surface area by breaking into smaller
particles, and the negative effect due to the enhancement in particle bonding that
inhibits solvent penetration. For instance, high compression may reduce the wettability of the tablet, since the formation of rmer and effective sealing layer by the
lubricant is likely to occur under the high pressure that is usually accompanied by
high temperature. The possible inuences of the force used to compress a mixture
of the drug and excipients into a tablet on the dissolution rate are summarized in
Fig. 3.4.

55

Dissolution rate

3. Dissolution Testing

Pressure

F IGURE 3.4. Possible effects of compression force on the dissolution rate (modied from
Finholt, 1974)

3.4.4

Factors Related to Dissolution Testing Conditions

External factors, such as temperature and viscosity of the dissolution medium can
inuence the dissolution rate of a drug substance or a drug product. This is in part
due to their effect on the diffusivity of a drug molecule. According to the Stokes
Einstein equation, the diffusion coefcient of a spherical molecule in solution is
given by
kT
,
(3.7)
D=
6r
where T is the temperature, r is the radius of a molecule in solution, is the viscosity of the solution, and k is the Boltzmann constant. This equation indicates that
diffusion is enhanced with increasing temperature but is reduced with increasing
viscosity.
Solution hydrodynamics also play an important role in determining the dissolution rate. One possible mechanism by which solution hydrodynamics inuences
the dissolution rate is through their effect on the stationary diffusion layer around
the drug molecule, as shown in (3.7). Since the thickness of this layer at the surface of the drug is determined by the shear force exerted by the uid, an increase
in the agitation (or stirring) rate may cause h to decrease, resulting in the improvement of drug dissolution. This hydrodynamic effect is demonstrated in the dissolution study of aspirin tablets, which shows that the dissolution half life of an
aspirin tablet decreases with increasing agitation intensity (Levy et al., 1965). In
addition, Armenante and Muzzio studied the velocity and shear stress/strain distribution in the USP Apparatus II (paddle). Their result shows that the ow rate
and shear rate vary signicantly at different locations near the vessel bottom of the
Apparatus II, thus resulting in different dissolution rates (Armenante and Muzzio,
2005).

3.5

Roles of Dissolution Testing

Dissolution testing plays many key roles in the development and production of
solid dosage forms. At the early stage of the drug research and development
(Phases 0 and 1), dissolution testing is used for active pharmaceutical ingredient

56

S. L. Lee et al.

(API) characterization and formulation screening. It is also employed to develop

and evaluate the performance of new formulations by examining drug release
from dosage forms, evaluating the stability of these formulations, monitoring and
assessing the formulation consistency and changes.
In addition to the use of dissolution testing in formulation optimization, process
development and scale up during Phases II and III, appropriate dissolution methods are developed to obtain an in vitroin vivo correlation (IVIVC) and other
biorelevant information that will guide bioavailability and/or bioequivalence
assessment of drug products.
For the release of drug products, dissolution testing serves as an important QC
tool which is used to verify manufacturing and product consistency. It is also
employed to evaluate the quality of the product during its shelf life, as well as
to assess postapproval changes and examine the need for bioequivalence studies
(FDA, 1997c).
Because of the diverse roles of dissolution testing in drug development and
manufacturing, it is often preferable to develop a single dissolution test that can
evaluate product quality and consistency, as well as predict in vivo performance.
However, developing such a dissolution method remains a signicant challenge.
Under most circumstances, this goal is not achievable since dissolution tests used
for QC and in vivo drug product performance assessment have very contrasting
characteristics, which is discussed below. Under the current industry setting, the
design of dissolution testing used for QC is primarily based upon the selection
of discriminatory media, apparatus, and conditions that can be used routinely for
QC purposes. Nevertheless, there is an increasing demand for the development of
biorelevant dissolution methods that can provide some predictive estimates of the
drug release with respect to the in vivo drug product performance. The remaining
chapter will be devoted to the review of dissolution methods that are currently
employed for QC and bioavailability assessment, as well as the discussion of scientic and regulatory issues associated with these two kinds of dissolution methods. Meanwhile, the signicance of the BCS is also emphasized in relation to its
use in the design of biorelevant testing.

3.6

In Vitro Dissolution Testing as a Quality Control Tool

The purpose of the dissolution test often dictates the choice of dissolution media.
In principle, dissolution testing should be carried out under physiological conditions if possible, allowing interpretation of dissolution data with respect to the in
vivo performance of a drug product. However, strict adherence to the GI environment is not necessary for routine dissolution testing. In fact, as mentioned previously, under the current setting of the pharmaceutical industry, the development
of dissolution methods for QC focuses more on discriminatory capability, ruggedness, and stability. Particularly, as a QC and testing tool, it is critical to develop a
dissolution method, which can consistently deliver a reliable test result and also

3. Dissolution Testing

57

assess drug product quality attributes (e.g., particle size, polymorphic form, or
excipients) that are sensitive to formulation and manufacturing changes.
For QC, dissolution tests are developed and optimized to target and assess product attributes by monitoring their effect on the rate and extent to which the drug
is released from the formulation. The design of a dissolution test used for QC is
therefore often dictated by the physicochemical properties (particularly solubility) of a drug substance and its formulation. The details regarding QC dissolution
testing of two solid dosage forms, immediate-release dosage forms and modiedrelease dosage forms are discussed below. In general, for QC purposes, the use of
the simplest dissolution medium is preferred whenever possible, regardless of the
dosage form.

3.6.1 Dissolution Method for Quality Control

of Immediate-Release Dosage Forms
3.6.1.1 Dissolution Media
Aqueous test media are generally preferred (USP, 2004; FDA, 1997b). Although
the design of a dissolution test used for QC is mainly based upon the physicochemical properties of the drug substance and the characteristics of the dosage
form, it is important to select dissolution media to at least reect the pH effect in
the GI environment. For this reason, the pH of these media should be within the
physiologic pH range of 1.26.8, where pH 1.2 and pH 6.8 represent the pH values
under the gastric and intestinal conditions, respectively (FDA, 1997a). Hydrochloric acid, acetate, or phosphate buffer in the physiological pH range are commonly
used and accepted as a dissolution medium for QC. The use of pure water in dissolution testing is usually not recommended primarily due to its limited buffering
capacity. The volume of these dissolution media should be based upon the drug
solubility, but is generally 500, 900, or 1,000 mL (FDA, 1997a). Sink conditions2
are often recommended, since the dissolution tests used for QC are intended to
provide conditions under which the majority of the drug (90%) can be released.
For some poorly soluble drugs that cannot dissolve adequately in aqueous solutions within the physiologic pH range, surfactants may be required to provide
sink conditions and achieve a complete drug dissolution within reasonable time.
The surfactants, such as sodium lauryl sulfate (SLS) and Tween, can be used to
improve the dissolution rate by acting as a wetting agent and/or increasing the
solubility of poorly soluble compounds through reduction of the interfacial tension and induction of micellar formation (Shah et al., 1989; Sievert and Siewert,
1998). They may also be used to improve the correlation between the in vitro dissolution data and in vivo drug product performance (Brown et al., 2004), as will
be explained below. The level and solubilizing capacity of a surfactant are critical
2 The term sink conditions is generally referred to as the condition where the medium

volume is at least greater than three times that needed to form a saturated solution of a drug
substance.

58

S. L. Lee et al.

to QC. When the level and/or the solubilizing capacity of the surfactant is too
high, the dissolution media may not be able to adequately discriminate differences
among formulations, such as changes in the polymorphic form or particle size, as
suggested in ICH Q6A. For hard and soft gelatin capsules as well as gelatin-coated
tablets, a specic amount of enzyme(s) may be added to the dissolution medium
to prevent pellicle formation.
3.6.1.2 Apparatus and Test Conditions
The most commonly used dissolution apparatus for solid oral dosage forms are
the basket method (USP Apparatus I), the paddle method (USP Apparatus II), the
reciprocating cylinder (USP Apparatus III) and the ow-through cell system (USP
Apparatus IV). The rst two apparatus are commonly used for dissolution testing of immediate-release dosage forms. The major advantage of these two devices
is that they are simple, robust, and well standardized. The reciprocating cylinder apparatus has also been used for the dissolution testing of immediate-release
products of highly soluble drugs, such as metoprolol and ranitidine, and some
immediate-release products of poorly soluble drugs, such as acyclovir (Yu et al.,
2002). However, this apparatus should be considered only when the basket and
paddle method are shown to be unsatisfactory. Due to the potential need for the
large volume of medium, the ow-through cell system is not suitable for a dissolution test that is used routinely for the QC purpose. Nevertheless, the reciprocating
cylinder device and the ow-through cell system may offer some advantages for
their use in a biorelevant method, as will be discussed below.
For QC or drug product release testing, mild agitation conditions should be
maintained during dissolution testing using the basket and paddle methods to
allow maximum discriminatory power. If the rotational speed is too low, coning
may occur, which leads to a low dissolution rate. However, if the rate of rotation is
too fast, the test will not be able to discriminate the differences between the acceptable and not acceptable formulations or batches. The common stirring speed used
for Apparatus I is 50100 rpm, while with Apparatus II the common stirring speed
is 5075 rpm (FDA, 1997a). All dissolution tests should be performed at physiological temperature (37 0.5 C). The test duration ranges from 15 min to 1 h.

3.6.2 Dissolution Method for Quality Control

of Modified-Release Dosage Forms
3.6.2.1 Dissolution Media
The media used for modied-release dosage forms are generally the same as those
used for immediate-release dosage forms. However, as opposed to the dissolution test used for immediate-release products that always uses one pH, more than
one dissolution media with different pH values may be employed for testing of
extended-release dosage forms to simulate the change in pH along the GI tract.

3. Dissolution Testing

59

3.6.2.2 Apparatus and Test Conditions

The most common types of apparatus used for routine quality testing of extendedrelease products are the basket and paddle methods. The reciprocating cylinder
may be used particularly for enteric-coated or extended-release dosage forms,
when the pH of the medium needs to be changed in order to mimic the pH changes
in the GI tract. The operating conditions for the basket and paddle methods are
very similar to those used for immediate-release dosage forms, with an exception
of the test duration, which can be as long as 12 h for extended-release products.

3.6.3

Limitations of Quality Control Dissolution Tests

There are some issues regarding the current use of dissolution tests that were
developed for QC. Because these dissolution tests are developed to provide a
maximum discriminatory power to assess any formulation changes and manufacturing process deviations, they are often overly discriminating, meaning
that the differences detected by these dissolution tests may not have any clinical relevance. For instance, in the FDA-sponsored studies of metoprolol (Rekhi
et al., 1997), although the slow-dissolving tablets of metoprolol failed the USP
dissolution test, the in vivo pharmacokinetic studies showed that all metoprolol
tablets were bioequivalent with their corresponding formulations regardless of
their in vitro dissolution rates. Thus, these clinically insignicant differences
detected by the overly discriminating dissolution test often lead to the rejection
of batches that may have an acceptable clinical performance. In addition, dissolution specications, which are established based upon acceptable clinical, pivotal
bioavailability, and/or bioequivalence batches using such overly discriminating
dissolution tests, may not truly reect the in vivo performance of a drug product.
As a consequence, without a detailed knowledge on how dissolution affects the
bioavailability of the drug product, these specications are usually set to be very
tight to assure the product quality and consistency by identifying any possible
subtle changes in the product attributes before in vivo performance is affected.
These shortcomings further facilitate the need for the development of biorelevant
dissolution tests.
Dissolution tests used for QC can also be subjected to the limitation of being
nondiscriminating. This limitation becomes evident if testing conditions are
not selected appropriately (e.g., the agitation rate or surfactant level). This
situation is best illustrated by the case of mebendazole (Swanepoel et al.,
2003). Mebendazole, which is a broad-spectrum anthelmintic drug, exists in
three polymorphic forms (A, B, C) that display solubility and therapeutic differences. Among these three forms, polymorph C is therapeutically favored. Despite
these differences, the three polymorphs produced similar dissolution proles using
a dissolution method that employed 0.1N HCl with 1% SLS. Specically, all these
dissolution proles met the specication in which 75% of the drug dissolved
within 120 min. It has been understood that the use of a large amount of SLS

60

S. L. Lee et al.

in the dissolution medium eliminates the differences in the dissolution rates of

mebendazole polymorphs.
The precision and accuracy of dissolution testing are often very sensitive to
several subtle operational controls. These include, but are not limited to the eccentricity of the agitating element, vibration, stirring element alignment, stirring rate,
dosage form position, sampling probes, position, and lters. These factors may
have a signicant effect on the dissolution measurement if they are not controlled
properly. For instance, the study of nondisintegrating double layered tablets containing salicylic acid indicates that the stirring rate and basket placement inuence
the drug dissolution in the basket apparatus (Howard et al., 1979; Mauger et al.,
1979). In addition, the hydrodynamics in the paddle apparatus have been shown to
be very complex and vary with site in the vessel (McCarthy et al., 2004). Therefore, the exact location where the tablet lands after it is dropped into the vessel
may have a considerable inuence on the velocity prole around the tablet and
hence its dissolution behavior.

3.7

Biorelevant Dissolution Testing

In order to achieve an adequate estimate of in vivo release behavior for solid

dosage forms, the relevant physiological conditions, in addition to the physicochemical properties of a drug substance and its formulation, should be taken into
serious consideration during the development of a biorelevant dissolution testing
system. Specically, the biorelevant dissolution method should be able to simulate
the in vivo environment where the majority of the drug is released from the formulation. In principle, the design of such a system should, at minimum, account
for the following factors to reect the physiological conditions in the GI tract:
1. pH Conditions
2. Key aspects of the composition of the GI contents (e.g., osmolarity, ionic
strength, surface tension, bile salts, and phospholipids)
3. Volume of the GI contents
4. Transit times
5. Motility pattern
6. Dosing conditions (e.g., administered with food)
To reect the effect of these factors on the drug release, it is important to utilize the dissolution media that mimic the conditions in the GI tract and the apparatus that can simulate the dynamic environment that the dosage form experiences
in the GI tract. Thus, in comparison to dissolution methods used for QC, which at
best simulate pH effects and/or osmolality on the drug release under in vivo conditions, biorelevant dissolution media are generally more complex and are often
not suitable for the purpose of QC.
Prior to the discussion of biorelevant dissolution methods, it is important to
rst review the concept and application of the in vitroin vivo correlation and the
importance of the BCS on the biorelevant dissolution testing development. The

3. Dissolution Testing

61

biorelevant dissolution media, apparatus and test conditions will be discussed with
emphasis on their relevance to the physiological factors, including the pH, composition of the GI uids, volume, GI hydrodynamics/motility, and food effect. The
remaining challenges regarding the future development of biorelevant dissolution
testing will also be highlighted.

3.7.1

In VivoIn Vitro Correlations

The major objective of using biorelevant dissolution methods is to establish in

vivoin vitro correlation (IVIVC) so that in vitro dissolution data can be used
to predict bioavailability. The term in vivoin vitro correlation is dened as a
predictive mathematical model describing the relationship between an in vitro
property of a dosage form and a relevant in vivo response. In general, the physicochemical property or in vitro property of a dosage form is the in vitro dissolution
prole. The biological property or in vivo response is the plasma concentration
prole. Four correlation levels are dened in the FDA guidance (FDA, 1997b), as
described below:
1. Level A: a point-to-point relationship between in vitro dissolution rate and in
vivo input rate of the drug from the dosage form.
2. Level B: a comparison of the mean in vitro dissolution time to in vivo residence
time or the mean in vivo dissolution time.
3. Level C: a single point relationship between a dissolution parameter (t50% , t90% ,
etc.).
4. Multiple-level C: a correlation that relates one or several pharmacokinetic parameters of interest to the amount of drug dissolved at several time points of the
dissolution prole.
Among all levels of correlation, Level A is the most meaningful for predicting purposes, since it provides a relationship that directly links in vivo drug absorption to
in vitro dissolution. This level of correlation should be valid for a reasonably wide
range of values of formulation and manufacturing parameters that are essential for
the drug release characteristics. This level can be used as a surrogate for in vivo
performance of a drug product. Therefore, in vitro dissolution data, without any
additional in vivo data, can be employed to justify a change made in manufacturing
sites, raw material supplies, minor formulation modications, strength of a dosage
form, etc. However, the lower levels of correlation (B and C) are usually not very
useful for regulatory purposes, and are used primarily for the development of formulation or processing procedures.

3.7.2 The Importance of BCS on Biorelevant

Dissolution Testing
The BCS, which was proposed by Amidon et al. (1995), emphasizes the contribution of three fundamental factors, dissolution, solubility, and intestinal
permeability, to the rate and extent of drug absorption for solid oral dosage

62

S. L. Lee et al.

forms. The BCS identies three dimensionless numbers as key parameters including absorption number (An), dissolution number (Dn), and dose number (Do) to
represent the effects of dissolution, solubility and intestinal permeability on the
absorption process. These three dimensionless numbers are dened as:
An =

Peff
tres ,
R

(3.8)

Dn =

DCs 4r02
tres ,
r0 43 r03

(3.9)

Do =

M0 /V0
,
Cs

(3.10)

where Peff is the effective permeability, r0 is the initial particle radius, tres is the
mean residence time for the drug in the intestinal segment ( R 2 L/Q ow , where
R is the radius, L is the length of the segment, and Q ow is the ow rate of uid in
the small intestine), D is the diffusion coefcient, is the density, V0 is the initial
gastric volume, and M0 is the amount of drug that is administered, and Cs is the
saturated solubility.
According to the BCS, drug compounds are classied based upon their solubility and permeability described as follows:
Class I: High Permeability, High Solubility
Class II: High Permeability, Low Solubility
Class III: Low Permeability, High Solubility
Class IV: Low Permeability, Low Solubility
In this system, a compound is considered highly soluble when the highest dose
strength is soluble in 250 mL3 water over a range of pH from 1.0 to 7.5. For
a highly permeable drug substance, the extent of absorption in humans is 90%
of an administrated dose, based on mass-balance or in comparison to an intravenous reference dose. When 85% of the label amount of drug substance dissolves within 30 min using USP apparatus I (100 rpm) or II (50 rpm) in a volume
of 900 mL in each of the following media: (1) 0.1N HCl or USP simulated gastric uid (SGF) without enzymes, (2) a pH 4.5 buffer, and (3) a pH 6.8 buffer
or USP simulated intestinal uid (SIF) without enzymes, a corresponding drug
product is considered to be rapidly dissolving.
Although the BCS has been developed primarily for regulatory applicants and
particularly for oral immediate-release drug products, it has important implications in governing the dissolution test design during the drug development. Most
importantly, it provides a general guideline for determining the conditions under
which IVIVC is expected, as summarized in Table 3.1 (Amidon et al., 1995).
In other words, the BCS can provide an early insight into whether it is possible
3 This volume is derived based on typical bioequivalence study protocols that prescribe

administration of a drug product to fasting human volunteers with a glass of water (about
8 oz).

3. Dissolution Testing

63

TABLE 3.1. In Vitroin vivo correlation expectations for immediate-release products (Amidon et al. 1995)
Class

Solubility

Permeability

IVIVC expectation

High

High

II

Low

High

III

High

Low

IV

Low

Low

IVIVC is expected if dissolution rate is

slower than gastric emptying rate.
Otherwise limited or no correlation is
expected
IVIVC is expected if in vitro dissolution rate
is similar to in vivo dissolution rate, unless
dose is very high
Limited or no IVIVC is expected since
absorption (permeability) is rate
determining
Limited or no IVIVC is expected

to develop a dissolution method capable of predicting in vivo drug absorption for

immediate-release products, based primarily upon the solubility, permeability, and
dissolution data.
BCS Class I compounds (e.g., metoprolol) have a high absorption number (An)
and a high dissolution number (Dn), indicating that the rate determining step for
drug absorption is likely to be dissolution or gastric emptying. This class of drugs
is generally well absorbed if the drug is stable or does not undergo rst pass
metabolism. For immediate-release products of Class I compounds, the absorption
rate is likely dominated by the gastric emptying time, and no direct correlation
between in vivo data and in vitro dissolution data is expected. Thus, dissolution
tests for such IR drug products should be designed mainly to conrm that the drug
is released rapidly from the dosage form under the test conditions described above.
A dissolution specication for which 85% of drug contained in the IR dosage form
is dissolved in less than 15 min may be sufcient to ensure bioavailability, since the
mean gastric half emptying time is 1520 min (Amidon et al., 1995; CDER/FDA,
1997). For BCS Class I drugs, which are formulated in extended-release dosage
forms and have permeability that is site independent, dissolution becomes more
important and IVIVC (e.g., level A) may be expected.
Class II drugs (e.g., phenytoin) have a high absorption number (An) and a low
dissolution number (Dn). Dissolution is the rate limiting step for drug absorption.
The inuence of dissolution on absorption of BCS Class II drugs can be classied
into two scenarios: solubility-limited absorption or dissolution-limited absorption
(Yu, 1999). These two scenarios are best illustrated by grisefulvin and digoxin.
In the case of solubility-limited absorption, grisefulvin exhibits a high dose number (Do) and a low dissolution number (Dn). Although in theory, absorption of
grisefulvin can be improved by taking more water with the administered dose
(decreasing Do), this approach is impractical due to the limitation in the physiological and anatomical capacity of the stomach for water. Thus, the only practical way to improve the absorption of grisefulvin is to decrease Do and increase
Dn by enhancing its solubility through appropriate formulation approaches such

64

S. L. Lee et al.

as solid dispersion. On the other hand, in the case of dissolution-limited absorption, digoxin has a low dose number (Do) and a low dissolution number (Dn).
Despite the small volume (21 mL) of uids required to dissolve a typical dose of
digoxin (0.5 mg), this drug dissolves too slowly for the absorption to take place
at the site(s) of uptake. However, its dissolution rate can be improved simply by
increasing Dn through the reduction in particle size. Thus, for BCS Class II drugs,
a strong correlation between in vitro dissolution data and in vivo performance
(e.g., Level A) is likely to be established. When a BCS Class II drug is formulated
as an extended-release product, an IVIVC may also be expected.
For BCS Class III drugs (e.g., cimetidine), permeability is likely to be a
dominant factor in determining the rate and extent of drug absorption. Hence,
developing a dissolution test that can predict the in vivo performance of products
containing these compounds is generally not possible. Since BCS Class IV drugs,
which are low in both solubility and permeability, present signicant problems for
effective oral delivery, this class of drugs is generally more difcult to develop in
comparison to BCS Class I, II, and III drugs.
In spite of its usefulness in the drug product development and regulatory recommendations regarding biowaivers for in vivo bioequivalence studies, the BCS also
has its limitations. Drug instability in the GI tract, rst pass metabolism, and complexation phenomena of drugs with the GI contents may have signicant inuence
upon bioavailability, but are not addressed by the BCS. Furthermore, the BCS is
often considered to be a conservative measure with regard to highly soluble drugs,
since they are required to show high solubility across the range of pH from 1.2
to 7.5. It is important to note that the solubility of a weak acid and weak base
depends on pH. The solubility of weak bases is generally higher in the stomach
than in the small intestine. Therefore, a low solubility at high pH may not inhibit
absorption of weak bases as the absorption may already be complete prior to entering the low solubility, high pH GI region. In contrast, low solubility at low pH may
not present a problem for the absorption of weak acids since high solubility and
high permeability in the small intestine are sufcient for their complete absorption.

3.7.3

Biorelevant Dissolution Methods

Unlike dissolution methods used for QC in which their design is primarily based
upon drug substance physicochemical properties and formulation principles,
biorelevant dissolution methods are designed to closely simulate physiological conditions in the GI tract. However, it should be noted that the physicochemical properties of the drug substance (e.g., solubility) and its formulation
(e.g., immediate- or extended-release dosage forms) play a key role in selecting
an appropriate type of biorelevant dissolution medium (e.g., gastric or intestinal
medium), apparatus (e.g., a single vessel or multiple vessels), and test conditions (e.g., agitation speed and duration of a dissolution test), since these drug
substance and formulation characteristics impact the location where the drug dissolution takes place in the GI tract. For instance, weak acids that are not soluble
in the stomach (low pH) are usually very soluble in the small intestine (high pH).

3. Dissolution Testing

65

In addition, for drugs that are unstable in an acidic method, a delayed-release

formulation can be employed to ensure that the drug release occurs only in the
high pH GI region (the small intestine).
3.7.3.1 Biorelevant Dissolution Media for Gastric Conditions
For BCS Class I drugs that are formulated in immediate-release dosage forms or
any products that dissolve rapidly and completely in an acidic medium, it is logical
to use a dissolution medium that reects the gastric conditions. The minimum
physiological parameters that need to be considered here include pH, surfactants,
and enzymes. Food effects may also be considered if signicant food effects are
observed in vivo. It should be noted that biorelevant dissolution media or methods
are designed primarily to mimic GI conditions in healthy subjects under the fasted
and fed state, since in vivo bioequivalence studies are generally performed using
these healthy subjects.
The pH in the stomach has a signicant inuence on the dissolution rate due to
its effect on the solubility of a drug substance. In the fasted state of young healthy
subjects, values of gastric pH are generally between 1.4 and 2.1 (Dressman et al.,
1990). However, the fasted state gastric pH values are found to be higher in subjects who are either over 65 years old or receiving gastric acid blocker therapy
(Russell et al., 1993; Christiansen, 1968). The gastric pH values also increase
immediately following meal ingestion (pH 37) (Dressman et al., 1998). The gastric pH resumes the fasted state values in approximately 23 h depending on the
size and content of the meal (Dressman et al., 1998).
The surface tension of gastric uid is lower than that of water, and it was measured in the 3550 mN m1 range (Finholt and Solvang, 1968; Finholt et al., 1978;
Efentakis and Dressman, 1998). Although the decrease in surface tension suggests
the presence of surfactants in the stomach, substances that lower the surface tension in vivo have not been identied unequivocally. The enzyme, pepsin, is also
found to be in gastric uid. The presence of this enzyme in the stomach causes a
major problem for protein and polypeptide stability in addition to the acidity of
the gastric environment.
Based upon the physiological factors described above, to simulate gastric conditions in the fasted state, the pH values of a gastric dissolution medium should be
in the pH range of 1.52.5. In addition, surfactants, such as SLS, should be added
into the medium to lower its surface tension close to the in vivo values. As mentioned earlier, for some capsules, an enzyme (pepsin) can be added to the medium
to ensure timely dissolution of the shell by preventing pellicle formation. A sample
composition for SGF in the fasted state is shown in Table 3.2 (Dressman, 2000).
Due to its simplicity, this medium can also be used for QC dissolution testing.
To simulate the fed state in the stomach, the use of milk (Macheras et al., 1987)
R
(Ashby et al., 1989) may be appropriate, since these media offer
and Ensure
appropriate ratios of fat to protein and fat to carbohydrate. However, these two
media are not suitable for routine quality assurance testing due to the difculties
in ltering and separating the drug substance from the medium for analysis.

66

S. L. Lee et al.
TABLE 3.2. Sample composition for simulating gastric conditions in the fasted state
(Klein 2005)
SGF composition
Sodium chloride
Hydrochloric acid
Triton 100
Deionized water

0.6 g
2.1 g
0.3 g
qs ad 300 mL

3.7.3.2 Apparatus and Test Conditions for Simulating the Stomach

The basket and paddle methods are frequently used, in conjunction with biorelevant media for gastric conditions, to simulate the drug release in the stomach
under fasted and fed conditions. Since these two devices consist of a single vessel
for each dosage form and are operated with a xed volume of a single medium,
they are best suited for drug products in which the majority of drug release occurs
in the same section of the GI tract. Although the relationship between in vivo
hydrodynamics or motility and rotational speed is still not well understood, the
range of 50100 rpm, which is established empirically, appears to give data that
can be used to establish IVIVC.
In comparison to the basket and paddle methods, the reciprocating cylinder
(USP Apparatus 3) and the ow-through cell (USP Apparatus 4) may offer some
advantages regarding simulating gastric conditions. Apparatus 3, which originates
from the ofcial disintegration tester (Borst et al., 1997), can be used to improve
the study of food effects in the stomach by simulating changes in the composition and motility with time due to gastric secretion and digestion using a series
of different media and agitation rates in the vessels. Similarly, Apparatus 4 also
provides the possibility of changing the composition of a medium and ow rate
during the test.
Regarding the volume of these gastric media, it depends on the volume of
administered uids and endogenous secretion. For instance, in the fasted state,
gastric juice secretion is usually low. Therefore, by considering the quantity of
uid that is ingested with the dosage form, the medium volume should be in the
order of 200300 mL. The duration of a dissolution test should reect the time
available for dissolution in the stomach that is a function of the emptying pattern,
which can vary considerably depending on the size of the solid particles as well
as the size and the content of the meal (Meyer et al., 1988; Moore et al., 1981).
If a drug formulated in an immediate-release dosage form is administered in the
fasted state and is well absorbed from the upper small intestine, it is appropriate to
run the dissolution test with SGF for 1530 min. Nevertheless, there are still discrepancies in various pharmacopoeia regarding the duration of a dissolution test
for immediate-release drug products (e.g., 30120 min).

3. Dissolution Testing

67

3.7.3.3 Biorelevant Dissolution Media for Intestinal Conditions

For poorly soluble drugs (e.g., BCS Class II drugs that are neutral or weak acids),
it may be more appropriate to use a dissolution media that mimics the intestinal
conditions. The dissolution of drug products in the small intestine is inuenced by
physiological factors including but not limited to pH, endogenous secretions from
the pancreas and gall bladder (e.g., bile salts, lecithin, and digestion enzymes), and
food effects. The physiological aspects related to drug absorption in the colon will
not be addressed here and can be found elsewhere (Dressman et al., 1997).
The pH values of intestinal conditions are considerably higher than those of gastric conditions, and were measured to be in the range of 5.56.0 for the duodenum,
6.5 in the jejunum, 7 in the proximal, and 7.5 in the distal ileum (Dressman et al.,
1998). The high pH values in the small intestine are attributed to the neutralization
effect of bicarbonate ion secreted by the pancreas. It should be noted that pH values gradually increase from the duodenum to the ileum, resulting in a pH gradient
in the small intestine. In the fed state, the pH values in the duodenum (4.26.1),
jejunum (5.26.2), and ileum (6.87.8) are generally lower than those in the fasted
state (Dressman et al., 1990; Ovesen et al., 1986; Fordtran and Locklear, 1966).
In the small intestine, secretion of bile from the gallbladder in the duodenum
leads to a high concentration of bile salts and phospholipids (lecithin), resulting in the formation of mixed micelles even in the fasted state. These bile salts
and lecithin may have a signicant enhancing effect upon the dissolution rate of
poorly soluble drugs by improving the wettability of solids and by increasing the
solubility of a drug substance into mixed micelles (Mithani et al., 1996). As for
gastric secretion, the rate of bile secretion also depends strongly on the prandial
state, in which the concentration of bile salts and lecithin further increases in the
presence of food. Since the majority of bile salts (>90%) is reabsorbed by the
active transport mechanism (Davenport, 1982), a decreasing gradient of bile salts
is observed along the small intestine. The digestion enzymes, such as lipases, peptidases, amylases, and proteases, are also secreted by the pancreas in the small
intestine in response to food ingestion. Lipases and peptidases present a stability
problem for some drugs and hence may inuence the dissolution process.
A commonly used medium for simulating fasting conditions in the proximal
small intestine is fasted state simulated intestinal uid (FaSSIF). As evidenced
in the above discussion, one apparent difference between the SGF and FaSSIF is
that this simulated intestinal medium contains bile salts and lecithin. Thus, the
dissolution rate of poorly soluble, lipophilic drugs may be improved greatly in
this medium in comparison to the dissolution rate observed in simple aqueous
solutions. The composition of this medium is given in Table 3.3 (Dressman, 2000)
and it was based on experimental data in dogs and humans for the concentration
of bile components, pH value, buffer capacity, and osmolality (Greenwood, 1994).
The pH value was chosen to be 6.5, which closely resembles the values measured
from the midduodenum to the proximal ileum. Sodium taurocholate was often
used as a representative bile salt since cholic acid is one of the more common
bile salts in human bile (Carey and Small, 1972). In addition, because the pKa of

68

S. L. Lee et al.
TABLE 3.3. Sample composition for simulating the fasted state conditions in the
small intestine (note that the recommended
volume for dissolution studies is 1 L) (Klein
2005)
FaSSIF composition
Sodium taurocholate
Lecithin
NaH2 PO4
KCl
NaOH
Deionized water

3 mM
0.75 mM
3.9 g
7.7 g
qs ad pH 6.5
qs ad 1 L

TABLE 3.4. Sample composition for simulating the fed state conditions in the small
intestine (note that the recommended volume for dissolution studies is 1 L) (Klein
2005)
FeSSIF composition
Sodium taurocholate
Lecithin
Acetic acid
KCl
NaOH
Deionized water

15 mM
3.75 mM
8.65 g
15.2 g
qs ad pH 5.0
qs ad 1 L

taurine conjugate is very low, precipitation and an alteration in the micellar size
with small variations in pH values are unlikely to occur within the pH range in
the proximal small intestine (pH 4.27). The ratio of phospholipids to bile salts
employed in these media is approximately 1:3, which reects the in vivo ratio that
is generally found to be between 1:2 and 1:5 (Dressman et al., 1998).
In comparison to the fasted state, a dissolution medium simulating intestinal
conditions in the fed state should assume a lower pH value, higher buffer capacity,
and osmolarity (Greenwood, 1994). In addition, as described earlier, lipids in food
further simulate the release of bile salts and phospholipids, which certainly have
major effects on the dissolution rate of the drug. Most of these factors should be
taken into consideration during the development of a dissolution medium for simulating the proximal small intestinal conditions in the fed state. The sample composition of the fed state simulating intestinal uid (FeSSIF) is given in Table 3.4
(Dressman, 2000). It should be noted in Table 3.4 that an acetic buffer is used here
instead of the phosphate buffer to achieve the higher capacity and osmolarity while
maintaining the lower pH value, and that taurocholate and lecithin are present in
considerably higher concentrations than those in the fasted state medium.

3. Dissolution Testing

69

3.7.3.4 Apparatus and Test Conditions for Simulating Small Intestine

Using biorelevant media that mimic intestinal conditions (e.g., FaSSIF and FeSSIF), the basket and paddle methods can also be employed to study the drug
release in the small intestine. The advantages and disadvantages of these two
apparatus used for simulating intestinal conditions are similar to those used for
simulating gastric conditions. Since the relationship between in vivo hydrodynamics (or motility) and rotational speed is not known, the agitation rate (50100 rpm)
is once again determined empirically to give data that provide the best IVIVC.
With the possibility of varying the composition of media and the agitation rate
(or the ow rate), both the reciprocating cylinder and ow-through cell systems
can be used to simulate the pH and composition changes from the duodenum to
the ileum. Furthermore, the ow-through cell system can be operated as an open
system, allowing removal of dissolved drugs and hence providing sink conditions
for poorly soluble drugs to mimic conditions in the small intestine. However, the
open system mode requires a large volume of media. Therefore, its practical use
is severely limited to product development, especially when biorelevant media
are used.
With regard to the volume of the fasted state simulated intestinal medium, pharmacokinetic studies in the fasted state show that by ingesting 200250 mL of water
with the dosage form, a total volume of 300500 mL will become available in the
proximal small intestine. Based upon this evidence, a volume of 500 mL is recommended for the FaSSIF. The total volume of the fed state simulated intestinal
medium should take into consideration the volume of coadministered uid, the
volume of uid ingested meal, and the secretions of the stomach, pancreas, and
bile (Fordtran and Locklear, 1966). As a result, in comparison to FeSSIF, a larger
volume (up to 1 L) is generally required for dissolution testing using FeSSIF.
3.7.3.5 Biorelevant Methods for Extended-Release Dosage Forms
For extended-release drug products, the dissolution method must capture, at
minimum, the changes in composition, pH, and residence times along the GI tract,
since absorption of these dosage forms takes place throughout the entire intestine.
Thus, the reciprocating cylinder and ow-through cell systems can be used, in
conjunction with different biorelevant dissolution media, to assess the in vivo
release behavior of extended-release dosage forms.
3.7.3.6 Remaining Challenges
Currently, similar to the dissolution test used for QC, the biorelevant dissolution
method is generally drug product specic. In other words, no universal biorelevant
methods have been devised. If the same drug is formulated differently, even a subtle difference in the formulation may require the development of different in vitro
dissolution methodology in order to obtain an IVIVC. Although some progress
has been made in understanding the GI tract environment (e.g., pH, composition,
and volume), the establishment of IVIVC is still primarily based upon a trial and

70

S. L. Lee et al.

error approach (Zhang and Yu, 2004). Furthermore, none of the dissolution media,
apparatus, and test conditions described previously for gastric and intestinal conditions reects all physiological parameters that are important for determining the
effects of composition, food, motility patterns, and transit times on drug release
in the stomach and small intestine. In addition, transient changes in composition,
motility, and volume in both the fasted and fed states are not fully captured by the
current biorelevant dissolution methods.
Therefore, developing biorelevant methods that truly capture the drug release
behavior under in vivo conditions remains extremely challenging, since the physiological environment of the GI tract is still not fully understood. For instance,
despite the fact that the hydrodynamics in the GI tract are known to play an important role in dissolution, they have not been studied in detail. Thus, to devise such
dissolution methods, we must seek a complete understanding of how all the key
factors such as composition, hydrodynamics, volume, and transit times affect the
dissolution of drugs in the GI tract. We can then utilize this knowledge in the
design of biorelevant dissolution testing.

3.8

Conclusions

Dissolution testing is critical to the drug product development and production. It is

routinely used in QC as well as research and development. The objective of dissolution testing in QC is to assure batch to batch consistency and detect manufacturing deviations. In research and development, dissolution testing is used to evaluate
the performance of new formulations by measuring the rate of drug release from
dosage forms, examining the stability of these formulations, and assessing formulation changes. More importantly, dissolution testing is employed to provide some
predictive estimates of the drug release under physiological conditions by establishing IVIVCs. For QC, dissolution tests are developed and optimized to target
and assess specic product properties (e.g., particle size and excipient composition) by monitoring their effects on the rate and extent to which the drug is released
from the formulation. The design of a dissolution test used for QC is, therefore,
dependent on the drug substance physicochemical properties (e.g., solubility) and
formulation principles (e.g., extended-release dosage forms). On the other hand,
in addition to the physicochemical properties of the drug substance and formulation characteristics, physiological factors also play an important role in the design
of biorelevant dissolution methods, since these methods are developed mainly to
simulate relevant conditions where the drug is being released from the formulation
in the GI tract. Although progress has been made in developing dissolution media
that reect gastric (e.g., SGF) and intestinal conditions (e.g., FaSSIF), developing
dissolution methods, which reect all physiological parameters (e.g., motility pattern and transit times) that may inuence the drug release behavior in the GI tract,
remains as a signicant challenge.

3. Dissolution Testing

71

References
Amidon, G.L., Lennernas, H., Shah, V.P., and Crison, J.R. (1995). A theoretical basis for a
biopharmaceutics drug classication: the correlation of in vivo drug product dissolution
and in vivo bioavailability. Pharm. Res. 12:413420.
Armenante, P., and Muzzio, F. (2005). Inherent method variability in dissolution testing:
the effect of hydrodynamics in the USP II Apparatus. A Technical Report Submitted to
the Food and Drug Administration.
Ashby, L.J., Beezer, A.E., and Buckton, G. (1989). In vitro dissolution testing of oral controlled release preparations in the presence of articial foodstuffs. I. Exploration of alternative methodology: microcalorimetry. Int. J. Pharm. 51:245251.
Bandelin, F.J. (1990). Compressed Tablets by Wet Granulation. In: Lieberman, H.A.,
Lachman, L., and Schwartz, J.B. (eds.), Pharmaceutical Dosage Forms: Tablets, Volume 1. Marcel Dekker, Inc., New York, pp. 199302.
Brown, C.K., Chokshi, H.P., Nickerson, B., Reed, R.A., Rohrs, B.R., and Shah, P.A. (2004).
Acceptable analytical practices for dissolution testing of poorly soluble compounds.
Pharm. Tech. 5665.
Brunner, E. (1904). Reaktionsgeschwindigkeit in heterogenen Systemen. Z. Phys. Chem.
43:56102.

Brunner, L., and Tolloczko, S. (1900). Uber

die Auosungsgeschwindingkeit fester Korper.
Z. Physiol. Chem. 35:283290.
Borst, I., Ugwu, S., and Beckett, A.H. (1997). New and extended application for USP drug
release Apparatus 3. Dissolut. Technol. 16.
Carey, M.C., and Small, D.M. (1972). Micelle formation by bile salts. Physicalchemical
and thermodynamic considerations. Arch. Intern. Med. 130:506527.
Christiansen, P. (1968). The incidence of achlorhydria in healthy subjects and patients with
gastrointestinal diseases. Scan. J. Gastroenterol. 3:497508.
Corrigan, O.I. (1985). Mechanisms of dissolution of fast release solid dispersions. Drug
Dev. Ind. Pharm. 11:697724.
Danckwerts, P.V. (1951). Signicance of liquid-lm coefcients in gas absorption. Ind.
Eng. Chem. 43:14601467.
Davenport, H.W. (1982). Physiology of the Digestive Tract, 5th ed. Year Book Medical
Publishers, Inc., Chicago.
Dressman, J.B. (2000). Dissolution testing of immediate-release products and its application to forecasting in vivo performance. In: Dressman, J.B., and Lennernas, H. (eds.),
Oral Drug Absorption: Prediction and Assessment, Volume 106. Marcel Dekker, Inc.,
New York, pp. 155181.
Dressman, J.B., Berardi, R.R., Dermentzoglou, L.C., Russell, T.L., Schmaltz, S.P.,
Barnett, J.L., and Jarvenpaa, K.M. (1990). Upper gastrointestinal (GI) pH in young
healthy men and women. Pharm. Res. 7:756761.
Dressman, J.B., Amidon, G.L., Reppas, C., and Shah, V.P. (1998). Dissolution testing as a
prognostic tool for oral drug absorption: immediate release dosage forms. Pharm. Res.
15:1122.
Efentakis, M., and Dressman, J.B. (1998). Gastric juice as a dissolution medium: surface
tension and pH. Eur. J. Drug Metab. Pharmacokinet. 23:97102.
FDA (1997a). Center for Drug Evaluation and Research, Guidance for industry. Dissolution
testing of immediate release solid oral dosage forms.

72

S. L. Lee et al.

FDA (1997b). Center for Drug Evaluation and Research, Guidance for industry. Extended
release oral dosage forms: development, evaluation, and application of in vitro/in vivo
correlations.
FDA (2000). Center for Drug Evaluation and Research, Guidance for industry. Waiver of in
vivo bioavailability and bioequivalence studies for immediate-release solid oral dosage
forms based on a biopharmaceutics classication system.
Finholt, P. (1974). Inuence of formulation on dissolution rate. In Leeson, L.J., and
Carstensen, J.T. (eds.), Dissolution Technology, American Pharmaceutical Association,
Washington, DC, pp. 106146.
Finholt, P., and Solvang, S. (1968). Dissolution kinetics of drugs in human gastric juice
the role of surface tension. J. Pharm. Sci. 57:13221326.
Finholt, P., Gundersen, H., Smit, A., and Petersen, H. (1978). Surfactant tension of human
gastric juice. Medd. Norsk. Farm. Selsk. 41:114.
Fordtran, J.S., and Locklear, T.W. (1966). Ionic constituents and osmolality of gastric and
small-intestinal uids after eating. Am. J. Dig. Dis. 11:503521.
Fraser, E.J., Leach, R.H., and Poston, J.W. (1972). Bioavailability of digoxin. Lancet 2:541.
Gandhi, R., Lal Kaul, C., and Panchagnula, R. (1999). Extrusion and spheronization in
the development of oral controlled-release dosage forms. Pharm. Sci. Technol. Today
4:160170.
Gordon, M.S., and Rudraraju, V.S. (1993). Effect of the mode super disintegrant incorporation on dissolution in wet granulated tablets. J. Pharm. Sci. 82:220226.
Greenwood, D.E. (1994). Small intestinal pH and buffer capacity: implication for dissolution of ionizable compounds. Doctoral Dissertation. The University of Michigan,
Ann Arbor.
Hanson, W.A. (1982). Handbook of Dissolution Testing, Volume 49. Pharmaceutical Technology Publications, Oregon.
Higuchi, T. (1961). Rate of release of medicaments from ointment bases containing drugs
in suspension. J. Pharm. Sci. 50:874875.
Howard, S.A., Mauger, J.W., Khwangsopha, A., and Pasquerelli, D.A. (1979). Tablet
position and basket type effects in spin-lter dissolution device. J. Pharm. Sci. 68:
15421545.
Klein, S. (2005). Biorelevant dissolution test methods for modied release dosage forms.
Doctoral Thesis. Johann Wolfgang Goethe University Frankfurt, Shaker-Verlag.
Levy, G., Leonards, J.R., and Procknal, J.A. (1965). Development of in vitro dissolution
tests which correlate quantitatively with dissolution rate-limited drug absorption in man.
J. Pharm. Sci. 54:17191722.
Loftsson, T., Hreinsdottir, D., and Masson, M. (2005). Evaluation of cyclodextrin solubilization of drugs. Inter. J. Pharm. 302:1828.
Macheras, P., Koupparis, M., and Apostelelli, E. (1987). Dissolution of four controlledrelease theophylline formulations in milk. Int. J. Pharm. 36:7379.
Mauger, J.W., Howard, S.A., and Khwangsopha, A. (1979). Hydrodynamic characterization
of a spin-lter dissolution device. J. Pharm. Sci. 68: 10841087.
McCarthy, L.G., Bradley, G., Sexton, J.C., Corrigan, O.I., and Healy, A.M. (2004). Computational uid dynamics modeling of the paddle dissolution apparatus: agitation rate,
mixing patterns, and uid velocities. AAPS PharmSciTech. 5:110.
Meyer, J.H., Elashoff, J., Porter-Fink, V., Dressman, J., and Amidon, G.L. (1988).
Human postprandial gastric emptying of 13 millimeter spheres. Gastroenterology 94:
13151325.

3. Dissolution Testing

73

Miller, D.A., McConville, J.T., Yang, W., Williams III, R.O., and McGinity J.W. (2006).
Hot-melt extrusion for enhanced delivery of drug particles. J. Pharm. Sci. 96:361376.
Mithani, S.D., Bakatselou, V., TenHoor, C.N., and Dressman, J.B. (1996). Estimation of
the increase in solubility of drugs as a function of bile salt concentration. Pharm. Res.
13:163167.
Moore, J.G., Christian, P.E., and Coleman, R.E. (1981). Gastric emptying of varying meal
weight and composition in man. Evaluation by dual liquid- and solid-phase isotopic
method. Dig. Dis. Sci. 26:1622.
Nelson, E. (1957). Solution rate of theophylline salts and effects from oral administration.
J. Am. Pharm. Assoc. 46:607614.
Nernst, W. (1904). Theorie der reaktionsgeschwindigkeit in heterogenen systemen.
Z. Physiol. Chem. 47:5255.
Ovesen, L., Bendtsen, F., Tage-Jensen, U., Pedersen, N.T., Gram, B.R., and Rune, S.L.
(1986). Intraluminal pH in the stomach, duodenum, and proximal jejunum in normal subjects and patients with exocrine pancreatic insufciency. Gastroenterology 90:
958962.
Peck, G.E., Baley, G.J., McCurdy, V.E., and Banker, G.S. (1989). Tablet Formulation and
Design. In: Lieberman, H.A., Lachman, L., and Schwartz, J.B. (eds.), Pharmaceutical
Dosage Forms: Tablets, Volume 1. Marcel Dekker, Inc., New York, pp. 73130.
Pinnamaneni, S., Das, N.G., and Das, S.K. (2002). Formulation approaches for orally
administered poorly soluble drugs. Pharmazie 57:291300.
Rekhi, G.S., Eddington, N.D., Fossler, M.J., Schwartz, P., Lesko, L.J., and Augsburger, L.L.
(1997). Evaluation of in vitro release rate and in vivo absorption characteristics of
four metoprolol tartrate immediate-release tablet formulations. Pharm. Dev. Technol.
2:1124.
Russell, T.L., Berardi, R.R., Barnett, J.L., Dermentzoglou, L.C., Jarvenpaa, K.M.,
Schmaltz, S.P., and Dressman, J.B. (1993). Upper gastrointestinal pH in seventy-nine
healthy, elderly North American men and women. Pharm. Res. 10:187196.
Schott, H., Kwan, L.C., and Feldman, S. (1982). The role of surfactants in the release of
very slightly soluble drugs from tablets. J. Pharm. Sci. 71:10381045.
Shah, V.P., Konecny, J.J., Everett, R.L., McCullough, B., Noorizadeh, A.C., and Skelly, J.P.
(1989). In vitro dissolution prole of water-insoluble drug dosage forms in the presence
of surfactants. Pharm. Res. 6:1218.
Shangraw, R. F. (1990). Compressed Tablets by Direct Compression. In: Lieberman, H.A.,
Lachman, L., and Schwartz, J.B. (eds.), Pharmaceutical Dosage Forms: Tablets,
Volume 1, Marcel Dekker, Inc., New York, pp. 195246.
Sievert, B., and Siewert, M. (1998). Dissolution tests for ER products. Dissolut. Technol.
5:17.
Stavchansky, R., and McGinity, J. (1989). Bioavailability and Tablet Technology. In:
Lieberman, H.A., Lachman, L., and Schwartz, J.B. (eds.), Pharmaceutical Dosage
Forms: Tablets, Volume 2, Marcel Dekker, Inc., New York, pp. 349553.
Swanepoel, E., Liebenberg, W., and de Villiers, M.M. (2003). Quality evaluation of generic
drugs by dissolution test: changing the USP dissolution medium to distinguish between
active and non-active mebendazole polymorphs. Eur. J. Pharm. Biopharm. 55:345349.
Wagner, J.G. (1970). Rate of dissolution in vivo and in vitro, part II. Drug Intell. Clin.
Pharm. 4:32.
Yu, L.X. (1999). An integrated absorption model for determining causes of poor oral drug
absorption, Pharm. Res. 16:18831887.

74

S. L. Lee et al.

Yu, L.X., and Amidon, G.L. (1999). Analytical solutions to mass transfer. In Amidon, G.L.,
Lee, P.I., and Topp, E.M. (eds.), Transport Processes in Pharmaceutical Systems, Volume
102, Marcel Dekker, Inc., New York, pp. 2354.
Yu, L.X., Wang, J.T., and Hussain, A.S. (2002). Evaluation of USP Apparatus 3 for Dissolution Testing of Immediate-Release Products. AAPS PharmSci. 4:15.
Zhang, H., and Yu, L.X. (2004). Dissolution testing for solid oral drug products: theoretical
consideration. Am. Pharm. Rev. 2631.