The Biomaterials Silver Jubilee C o m p e n d i u m 175

Biomaterials
ELSEVIER Biomaterials 21 (2000) 2529-2543

Scaffolds in tissue engineering bone and cartilage

Dietmar W. Hutmacher
Laboratory for Biomedical Engineering, Institute of Engineering Science, Department of Orthopedic Surgery,
National University of Singapore, 10 Kent Ridge Crescent, Singapore 119260, Singapore

Abstract

Musculoskeletal tissue, bone and cartilage are under extensive investigation in tissue engineering research. A number of
biodegradable and bioresorbable materials, as well as scaffold designs, have been experimentally and/or clinically studied. Ideally,
a scaffold should have the following characteristics: (i) three-dimensional and highly porous with an interconnected pore network for
cell growth and flow transport of nutrients and metabolic waste; (ii) biocompatible and bioresorbable with a controllable degradation
and resorption rate to match cell/tissue growth in vitro and/or in vivo; (iii) suitable surface chemistry for cell attachment, proliferation,
and differentation and (iv) mechanical properties to match those of the tissues at the site of implantation. This paper reviews research
on the tissue engineering of bone and cartilage from the polymeric scaffold point of view. 9 2000 Elsevier Science Ltd. All rights
reserved.

Keywords" Tissue engineering of bone and cartilage; Design and fabrication of 3-D scaffold; Biodegradable and bioresorbable polymers

1. Introduction The currently applied scaffold fabrication technologies,

with special emphasis on the so-called solid-free form
Bone and cartilage generation by autogenous cell/tis- fabrication technologies, will also be bench marked. Fi-
sue transplantation is one of the most promising tech- nally, the paper discusses the author's research on the
niques in orthopedic surgery and biomedical engineering design and fabrication of 3-D scaffolds for tissue engi-
[1]. Treatment concepts based on those techniques neering an osteochondral transplant.
would eliminate problems of donor site scarcity, immune
rejection and pathogen transfer [-2]. Osteoblasts, chon-
drocytes and mesenchymal stem cells obtained from the 2. Polymer-based scaffold materials
patient's hard and soft tissues can be expanded in culture
and seeded onto a scaffold that will slowly degrade and The meaning and definition of the words biodegrad-
resorb as the tissue structures grow in vitro and/or able, bioerodable, bioresorbable and bioabsorbable
in vivo [3]. The scaffold or three-dimensional (3-D) con- (Table 1)--which are often used misleadingly in the tissue
struct provides the necessary support for cells to prolifer- engineering literature--are of importance to discuss the
ate and maintain their differentiated function, and its rationale, function as well as chemical and physical proper-
architecture defines the ultimate shape of the new bone ties of polymer-based scaffolds. In this paper, the polymer
and cartilage. Several scaffold materials have been inves- properties are based on the definitions given by Vert [13].
tigated for tissue engineering bone and cartilage includ- The tissue engineering program for bone and cartilage
ing hydroxyapatite (HA), poly(~-hydroxyesters), and in the author's multidisciplinary research curriculum has
natural polymers such as collagen and chitin. Several been classified into six phases (Table 2). Each tissue
reviews have been published on the general properties engineering phase must be understood in an integrated
and design features of biodegradable and bioresorbable manner across the research program--from the polymer
polymers and scaffolds [4-12]. The aim of this paper is to material properties, to the scaffold micro- and macro-
complete the information collected so far, with special architecture, to the cell, to the tissue-engineered trans-
emphasis on the evaluation of the material and design plant, to the host tissue. Hence, the research objectives in
characteristics which are of specific interest in tissue each phase are cross-disciplinary and the sub-projects are
engineering the mesenchymal tissues bone and cartilage. linked horizontally as well as vertically.

0142-9612/00/$- see front matter 9 2000 Elsevier Science Ltd. All rights reserved.
PII: S0 142-96 12(00)001 2 1-6
 176 The Biomaterials Silver Jubilee Compendium
2530 D. W. Hutmacher / Biomaterials 21 (2000) 2529-2543

Table 1 polylactides (PLLA, PDLA), polycaprolactone (PCL),

Definitions given by Vert etc. II. A number of non-approved polymers, such as
Biodegradable are solid polymeric materials and devices which break
polyorthoester (POE), polyanhydrides, etc. which are
down due to macromolecular degradation with dispersion in vivo but also under investigation. III. The synthesis of entrepre-
no proof for the elimination from the body (this definition excludes neurial polymeric biomaterials, such as poly (lactic acid-
environmental, fungi or bacterial degradation). Biodegradable poly- co-lysine), etc., which can selectively shepherd specific cell
meric systems or devices can be attacked by biological elements so that
phenotypes and guide the differentiation and prolifer-
the integrity of the system, and in some cases but not necessarily, of the
macromolecules themselves, is affected and gives fragments or other ation into the targeted functional premature and/or ma-
degradation by-products. Such fragments can move away from their ture tissue.
site of action but not necessarily from the body. In general, polymers of the poly(~-hydroxy acids)
Bioresorbable are solid polymeric materials and devices which show group undergo bulk degradation. The molecular weight
bulk degradation and further resorb in vivo; i.e. polymers which are of the polymer commences to decrease on day one (PGA,
eliminated through natural pathways either because of simple filtration PDLA) or after a few weeks (PLLA) upon placement in
of degradation by-products or after their metabolization. Bioresorption
an aqueous media [12]. However, the mass loss does not
is thus a concept which reflects total elimination of the initial foreign
material and of bulk degradation by-products (low molecular weight start until the molecular chains are reduced to a size
compounds) with no residual side effects. The use of the word 'bio- which allows them to freely diffuse out of the polymer
resorbable' assumes that elimination is shown conclusively. matrix [ 14]. This phenomenon described and analyzed in
Bioerodible are solid polymeric materials or devices, which show sur- detail by a number of research groups [15-18], results in
face degradation and further, resorb in vivo. Bioerosion is thus a con- accelerated degradation and resorption kinetics until the
cept, too, which reflects total elimination of the initial foreign material physical integrity of polymer matrix is compromised. The
and of surface degradation by-products (low molecular weight com-
mass loss is accompanied by a release gradient of acidic
pounds) with no residual side effects.
by-products.
Bioabsorbable are solid polymeric materials or devices, which can In vivo, massive release of acidic degradation and
dissolve in body fluids without any polymer chain cleavage or molecu-
lar mass decrease. For example, it is the case of slow dissolution of
resorption by-products results in inflammatory reac-
water-soluble implants in body fluids. A bioabsorbable polymer can be tions, as reported in the bioresorbable device literature
bioresorbable if the dispersed macromolecules are excreted. [19-22]. If the capacity of the surrounding tissue to
eliminate the by-products is low, due to the poor vas-
cularization or low metabolic activity, the chemical com-
Table 2 position of the by-products may lead to local temporary
The research program for tissue engineering bone and cartilage classi- disturbances. One example of this is the increase of osmo-
fied into six phases tic pressure or pH manifested through local fluid
accumulation or transient sinus formation from fiber
I--Fabrication of bioresorbable scaffold
II--Seeding of the osteoblasts/chondrocytes populations into the
reinforced polyglycolide pins applied in orthopedic sur-
polymeric scaffold in a static culture (petri dish) gery [21]. Potential problems of biocompatibility in
III--Growth of premature tissue in a dynamic environment tissue engineering bone and cartilage, by applying degra-
(spinner flask) dable, erodable, and resorbable polymer scaffolds, may
IV--Growth of mature tissue in a physiologic environment (bioreactor)
also be related to biodegradability and bioresorbability.
V--Surgical transplantation
VI--Tissue-engineered transplant assimilation/remodeling Therefore, it is important that the 3-D scaffold/cell con-
struct is exposed at all times to sufficient quantities of
neutral culture media, especially during the period where
The first stage of tissue engineering bone or cartilage the mass loss of the polymer matrix occurs.
begins with the design and fabrication of a porous 3-D The incorporation of a tricalciumphosphate (TCP)
scaffold, the main topic of this review paper. In general, [23], hydroxyapatite (HA) [24] and basic salts [15] into
the scaffold should be fabricated from a highly biocom- a polymer matrix produces a hybrid/composite material.
patible material which does not have the potential to These inorganic fillers allow to tailor the desired degra-
elicit an immunological or clinically detectable primary dation and resorption kinetics of the polymer matrix.
or secondary foreign body reaction [9]. Furthermore, A composite material would also improve biocompatibil-
a polymer scaffold material has to be chosen that will ity and hard tissue integration in a way that ceramic
degrade and resorb at a controlled rate at the same time particles, which are embedded into the polymer matrix,
as the specific tissue cells seeded into the 3-D construct allow for increased initial flash spread of serum proteins
attach, spread and increase in quantity (number of compared to the more hydrophobic polymer surface [9].
cells/per void volume) as well as in quality. Currently, the In addition, the basic resorption products of HA or TCP
design and fabrication of scaffolds in tissue engineering would buffer the acidic resorption by-products of the
research is driven by three material categories: I. Regula- aliphatic polyester and may thereby help to avoid the
tory approved biodegradable and bioresorbable poly- formation of an unfavorable environment for the cells
mers (Table 3), such as collagen, polyglycolide (PGA), due to a decreased pH [15,23,24].
 t~

Table 3 t,,,,~ o
Properties of bioresorbable and bioerodable polymers

Polymer Comparison of Degradation and Molecular weight Mass loss References References Area of Products with
mechanical properties resorption process loss/loss of (in month) a (scaffolds) (medical device) application regulatory
of bioerodable and via hydrolysis mechanical properties approval
bioresorbable polymers (in month) a
,~,~o

Poly(L-lactide) + + + Bulk erosion 9-15 36-48 46, 48, 49, 19, 20, 24 Orthopedic FixSorb System
60-64, 70-72 Surgery, Oral and (screws, nails, pins)
Maxillofacial Neofix (screws,
Surgery nails, pins)
Poly(L-lactide-co-D, + + Bulk erosion 5-6 12-18 22, 23 Oral and ResorPin, Leadfix
L-lactide) 70/30 Maxillofacial MacroSorb System
Surgery, Orthopedic (screws and plates,
Surgery mesh, nails, pins)
PolyPin
Poly(L-lactide-co- + + Bulk erosion 1-2 3-4 28, 63 Suture Periodontal Vicryl Suture,
glycolide) 10/90 Surgery, Surgery, Vicryl Mesh
Polyglycolide + + + Bulk erosion 0.5-1 3-4 6, 30-35, 41, 65 Orthopedic Surgery Biofix
Poly(D,L-lactide) + Bulk erosion 1-2 5-6 6, 31, 34, 49, 60 t..j

Poly(D,L-lactide- + Bulk erosion 1-2 4-5 53-56, 60, 70

co-glycolide) 85/15
Poly(D,L-lactide- + Bulk erosion 1-2 4-5 49, 61
co-glycolide) 75/25 t~
t-.a
Poly(D,L-lactide- + + Bulk erosion 1-2 3-4 28 I
co-glycolide) 50/50
4~
Polycaprolactone + Bulk and surface 9-12 24-36 29 Drug delivery Capranor
erosion
Polyorthoester + + Surface erosion 4-6 12-18
Polyanhydrides + + Surface erosion 4-6 12-18 59 Animal experiments

a Molecular weight and mass loss vary depending on factors such as chemical structure and composition; presence of ionic groups and of side group defects; configuration of the structure molecular

weight and molecular weight distribution (polydispersity); presence of low molecular weight components (monomers, oligomers, solvents, softeners, drugs, growth factors, etc.); production and
manufacturing procedures and their process parameters, implant design, sterilization method, morphology (amorphous versus semi-crystalline, presence of microstructures and stress within the
components), tempering, storage, implant site. + + + , good; + + , average; +, poor.

-.q

t.J "q
t.a.a
 178 The Biomaterials Silver Jubilee Compendium
2532 D. W. Hutmacher / Biomaterials 21 (2000) 2529-2543

Control of the hydrodynamic and biochemical envi- the scaffold matrix must serve an additional function; it
ronment is essential for the successful in vitro engineering must provide sufficient temporary mechanical support to
of 3-D scaffold/tissue constructs for potential clinical use withstand in vivo stresses and loading. In Strategy I re-
[25]. Computer-controlled bioreactors that continuously search programs, the material must be selected and/or
supply physiological nutrients and gases, serve to regu- designed with a degradation and resorption rate such
late the required cell/tissue culture conditions for a long that the strength of the scaffold is retained until the tissue
period of time. After the in vitro culturing of the 3-D engineered transplant is fully remodeled by the host
scaffold/tissue construct, the degree of remodeling and tissue and can assume its structural role.
cell/tissue replacement of the bone/cartilage transplant Bone is able to remodel in vivo under so-called physio-
by the host tissue has to been taken into consideration logical loading [27]. It is a requirement that the degrada-
[26]. Cell and tissue remodeling is important for achiev- tion and resorption kinetics have to be controlled in such
ing stable mechanical conditions and vascularization at a way that the bioresorbable scaffold retains its physical
the host site. Hence, the 3-D scaffold/tissue construct properties for at least 6 months (4 months for cell cultur-
should maintain sufficient structural integrity during the ing and 2 months in situ). Thereafter, it will start losing its
in vitro and/or in vivo growth and remodeling process. mechanical properties and should be metabolized by the
The degree of remodeling depends on the host anatomy body without a foreign body reaction after 12-18 months
and physiology [26]. The polymer selection from a ma- (Fig. 1). The mechanical properties of the bioresorbable
terial science point of view is based on two different 3-D scaffold/tissue construct at the time of implantation
strategies in regard to the overall function of the scaffold. should match that of the host tissue as closely as possible
[7]. It should posses sufficient strength and stiffness to
2.1. Strategy I function for a period until in vivo tissue ingrowth has
replaced the slowly vanishing scaffold matrix.
In the first strategy (Fig. 1), the physical scaffold struc- Thompson et al. [28] studied a poly(D,L-lactide-co-
ture supports the polymer/cell/tissue construct from the glycolide) matrix under cyclic compressive loading.
time of cell seeding up to the point where the hard tissue They concluded that changes in surface deformation and
transplant is remodeled by the host tissue. In the case of morphology suggest that the compressive loading ini-
load-bearing tissue such as articular cartilage and bone, tially collapses and stiffens the polymer matrix. The de-

I Hydration
II Hydration and Degradation
III Degradation and Mass Loss
IV Resorption and Metabolisation
I I I V Metabolisation
I II 1 III I IV 1 V
100% i
i
I i
i
I
!
l i
,' Mdle~zular w~ight ios~ ,
i , I i
, ' ~D Scat~fold t
=.-.K,.r-.~,
, \ ,
~, .
, \/
. , . ~ .
,
- . ~ . - , . - , olOSS Ol
lvlass ~ ,, - .
' i\
. . . .
i [ x~ I
i i '1 i

b,oresor~able
,
I
/\.
'

3I~,, scaffol
~ ,

Remodelin", g of tissue
'
.. . .. . . . \]......I- ,,
", ,' / henn g~
~ mt ee;s:ed
e r e d ~trans
; ; ; P ~P lant b
;intn bo
50% l , I i ~ 11 I i / / i _

', I I I "~[', ] ~ k e l e t ~ l tissue in vi

i , ' , ~ I / /

, i ' i ,'.\ I f,
',I !, ',i i! I X".\
I ~ ~I I / 'i
| ! ! I i i

i i 9 i ~ i

0%
! i i I !

i i I i

i0 1 2 ~ 6 9 12 15 24
, ,,
,' ', Numb, er of weeks
I ! !
i
, B C D ', F 'o
I I I
iT M f-~, ~ -,q

A E

Fig. 1. Graphical illustration of the complex interdependence of molecular weight loss and mass loss of a strategy 1 3D scaffold matrix plotted against
the time frame for tissue engineering a cartilage/bone transplant. (A) Fabrication of bioresorbable scaffold; (B) seeding of the osteoblast/cartilage
populations into the polymeric scaffold in a static culture (petri dish); (C) growth of premature tissue in a dynamic environment (spinner flask);
(D) growth of mature tissue in a physiologic environment (bioreactor); (E) surgical transplantation; (F) tissue-engineered transplant assimila-
tion/remodeling.
 The Biomaterials Silver Jubilee Compendium 179
D.W. Hutmacher / Biomaterials 21 (2000) 2529-2543 2533

crease in molecular weight is slowed down due to the 7 month in a bioreactor reached 40% of the mechanical
reduction of surface area from hydrolysis, until the properties of natural cartilage. The next phase of those
matrix architecture no longer accommodates the mech- research programs will be to evaluate how these tissue-
anical loading and begins to lose its integrity. Conclus- engineered cartilage transplants assimilate and remodel
ively, in Strategy I the scaffold architecture has to in vivo [35]. In an in vivo model, one of the major
withstand mechanical loading in vitro and in vivo. problems from a biomechanical and clinical view point, is
the primary mechanical stabilization of cartilage trans-
2.2. Strategy H plants [28]. This aspect will be discussed below, under
scaffold design, in more detail.
For Strategy II (Fig. 2), the intrinsic mechanical prop-
erties of the scaffold architecture templates the cell prolif-
eration and differentiation only up to the phases where 3. Scaffold design
the premature bone or cartilage is placed in a bio-
reactor. The degradation and resorption kinetics of the Skeletal tissue, such as bone and cartilage, is usually
scaffold are designed to allow the seeded cells to prolifer- organized into 3-D structures in the body [36]. For the
ate and secrete their own extracellular matrix in the static repair and generation of hard and ductile tissue, such as
and dynamic cell seeding phase (weeks 1-12), while the bone, scaffolds need to have a high elastic modulus in
polymer scaffold gradually vanishes leaving sufficient order to be retained in the space they were designated for;
space for new cell and tissue growth. The physical sup- and also provide the tissue with adequate space for
port by the 3-D scaffold is maintained until the engineer- growth [37]. If the 3-D scaffold is used as a temporary
ed tissue has sufficient mechanical integrity to support load-bearing device (Strategy II), the mechanical proper-
itself. ties would maintain that load for the required time with-
Different research groups [29-33] have shown in out showing symptoms of fatigue or failure. Therefore,
a number of studies that a nonwoven mesh made of one of the basic problems from a scaffold design point of
polyglycolide fibers offers degradation and resorption view, is that to achieve significant strength the scaffold
kinetics for Strategy II. However, the challenge for the material must have sufficiently high interatomic and in-
grown cell/tissue construct is to have similar mechanical termolecular bonding, but must have at the same time
properties to the host bone and cartilage. Ma and Langer a physical and chemical structure which allows for hy-
[34] showed that cartilage which was cultured for drolytic attack and breakdown.

[ Hydration Hydration .] Bulk Erosion .] Bulk Erosion .] Metabolisation

[ i Degradation [ Mass loss i Metabolisation
.
9 | .

I I ',
i
I I I
9 . i 9 9

00% i_ . . . . . . ,:

: Molelzula \ 9construct r e n i o d e l i n g a n d
I w;igl~t los; ! /ii ~ //;\ ~ell/tissue repla[~ement by the
: ] ' i / I i ~ / 9i host tissue
I9 ,, !9/ 1 ! ~ \',: ":

50~176i i / Mass l~ ~f'~3 D / s c a f f ~ I

9 , 9 construct in:vitro : :
I)% :i
I "i I ', '1 I Ii
I~ i
v , 3 , 6 12 ',15
i !
'
I
,
I
Number of mo~hs
IB C', D ', F ', i

! w"i'~

'E

Fig. 2. Graphical illustration of the complex interdependence of molecular weight loss and mass loss of a strategy II 3D scaffold matrix plotted against
the time frame for tissue engineering a cartilage/bone transplant. (A) Fabrication of bioresorbable scaffold; (B) seeding of the osteoblast/cartilage
populations into the polymeric scaffold in a static culture (petri dish); (C) growth of premature tissue in a dynamic environment (spinner flask);
(D) growth of mature tissue in a physiologic environment (bioreactor); (E) surgical transplantation; (F) tissue-engineered transplant assimila-
tion/remodeling.
 180 The Biomaterials Silver Jubilee Compendium
2534 D. W. Hutmacher / Biomaterials 21 (2000) 2529-2543

Ingber and his group [38,39] design their scaffolds For tissue engineering a bone transplant, the creation
based on a concept which they named tensegrity. of a vascularized bed ensures the survival and function of
Geodesical 3-D constructs were designed by applying seeded cells, which have access to the vascular system for
tensegrity so that the entire scaffold structure evenly nutrition, gas exchange, and elimination of by-products
distributes and balances mechanical stresses. The walls, [41]. The vascularization of a scaffold may be compro-
layers or struts that make up the interconnecting scaffold mised by purely relying on capillary ingrowth into the
framework are connected into triangles, pentagons or interconnecting pore network from the host tissue.
hexagons, each of which can bear tension or compres- In situ, the distance between blood vessels and mesen-
sion. However, the mechanical rational of the tensegrity chymal cells are not larger then 100 gm [42]. Therefore,
design concept has been known for centuries in the area the time frame has to be taken into account for the
of civil engineering. Gothic architects used a stone skel- capillary system to distribute through larger scaffold
eton structure of stone columns and ribs supported by volume. It may also be possible to control the degree and
arches and buttresses to build cathedrals (Fig. 3). rate of vascularization by incorporating angiogenic and
Another point, which has to be focused on is the anti-angiogenic factors in the degrading matrix of the
diffusion of nutrients into the 3-D scaffold. Although, an scaffold.
interconnected macropore-structure of 300-500 ~tm en- From a biomechanical and clinical point of view, the
hances the diffusion rates to and from the center of tissue-engineered bone or cartilage transplant should
a scaffold, transportation of the nutrients and by-prod- allow for a mechanically secure and stable fixation on or
ucts is not sufficient for large scaffold volumes. A fluid- to the host tissue [29]. For bone, the currently available
dynamic microenvironment provided by a bioreactor can medical devices, such as pins, screws, and plates might be
mimic the interstitial fluid conditions present in natural used. However, the integration of a device-like part into
bone and cartilage in a macroporous scaffold architec- the 3-D scaffold design can be advantageous. The ration-
ture. Bioreactors permit in vitro culture of larger and ale for such an innovative design concept will be for the
better organized 3-D cell communities than can be tissue engineering of an osteochondral bone transplant is
achieved using standard tissue culture techniques [40]. described below.

4. Scaffold fabrication

A number of fabrication technologies have been

applied to process biodegradable and bioresorbable
materials into 3-D polymeric scaffolds of high porosity
and surface area. The conventional techniques for scaf-
fold fabrication include fiber bonding, solvent casting,
particulate leaching, membrane lamination and melt
molding (Table 4). Several papers have reviewed the past
and current research on scaffold fabrication techniques
[43-45]. However, none of those papers has directly
compared the 3-D scaffold-processing technologies for
the tissue engineering community. From a scaffold design
and function view point each processing methodology
has its pro and cons. It is the aim of this paper to
aggregate the compiled information and to present this
data in a comprehensive form. Table 4 summarizes the
key characteristics and parameters of the techniques cur-
rently used. The aim of this part of the review paper is to
assist research teams with their choice for a specific 3-D
scaffold-processing technology by providing the informa-
tion needed to determin the critical parameters. As dis-
cussed above, at present the challenge in tissue
engineering bone and cartilage is not only to design, but
also to fabricate reproducible bioresorbable 3-D scaf-
folds, which are able to function for a certain period of
Fig. 3. The Cologne cathedral was built in the 18th and 19th century by
time under load-bearing conditions.
gothic architects who used a stone skeleton structure of stone columns Solvent casting, in combination with particle leaching,
and ribs supported by arches and buttresses. works only for thin membranes or 3-D specimens with
 e~
t,,~~

Table 4
Currently applied 3D scaffold fabrication technologies
c,z
Fabrication technology Processing Material properties Scaffold design and Achievable pore Porosity Architecture Reference
required for reproducibility size in Jam in %
processing
t...~ o
Solvent casting in Casting Soluble User, material and 30-300 20-50 Spherical pores, salt particles remain 47
t~
combination with technique sensitive in matrix
particular leaching
Membrane lamination Solvent bonding Soluble User, material and 30-300 <85 Irregular pore structure 48
technique sensitive
Fabrication of non-woven Carding, Needling, Fibres Machine controlled 20-100 <95 Insufficient mechanical properties 30-35, 41,
Plate pressing 63-65
Melt moulding Moulding Thermoplastic Machine controlled 50-500 < 80 r~

Extrusion in combination Extrusion Thermoplastic Machine controlled < 100 < 84 Spherical pores, salt particles remain 49
with particular leaching through dies in matrix
Emulsion freeze drying Casting Soluble User, material and < 200 <97 High volume of inter-connected 54-56
technique sensitive micropore structure
Thermally induced Casting Soluble User, material and < 200 <97 High volume of inter-connected 58-61
phase separation technique sensitive micropore structure t..a

Supercritical-fluid Casting Amorphous Material and < 100 10-30 High volume of non interconnected 53
technology technique sensitive micropore structure
Supercritical-fluid Casting Amorphous Material and Micropores < 50 <97 Low volume of non-interconnected 54
t-.a
technology in technique sensitive macropores < 400 micropore structure combined t~
t-,a
combination with with interconnected
t..a
particle leaching macropore structure t~
4~
3-D printing in and without Solid free form Soluble Machine and 45-150 < 60 100% interconnected macropore 69-72 t~

combination of particle fabrication computer controlled (triangles, pentagons, honey comb, etc.),
leaching design and fabrication layer by layer,
by use of water-based binder incorpo-
ration of biological agents into
matrix possible
Fused deposition modelling Solid free form Thermoplastic Machine and > 150 < 80 100% interconnected macropore structure 29
fabrication computer controlled (triangles, pentagons, honey comb, etc.),
design and fabrication layer by layer

c~
t,3
 182 The Biomaterials Silver Jubilee C o m p e n d i u m
2536 D.W. Hutmacher/ Biomaterials 21 (2000) 2529-2543

very thin wall sections: otherwise, it is not possible to fabrication of a truly interconnecting pore structure de-
remove the soluble particles from within the polymer pends on the processing method and parameters as well
matrix [46]. Mikos et al. [47], using the above-described as on the used equipment [57,58].
technology, fabricated porous sheets and laminated them Several groups [57-60] studied thermally induced
to 3-D structures. Chloroform was used on the attach- phase separation technology to process polymeric 3-D
ment interface for the lamination process. This fabrica- scaffolds. This technique has been used previously to
tion technology is time consuming because only thin fabricate synthetic membranes for non-medical applica-
membranes can be used. Another disadvantage is that the tions. The method has been extensively applied in the
layering of porous sheets allows only a limited number of field of drug delivery to fabricate microspheres, which
interconnected pore networks. Solvent-casted polymer- allows the incorporation of pharmaceutical and biolo-
salt composites have also been extruded into a tubular gical agents, such as bone morphogenetic proteins
geometry [48]. The disadvantages of the above technolo- (BMPs) into the polymer matrix. In general, the micro-
gies include the extensive use of highly toxic solvents, and macro-structure is controlled by varying the polymer
time required for solvent evaporation (days-to-weeks), material, polymer concentration, quenching temper-
the labor intensive fabrication process, the limitation to ature, and solvents. However, current research shows
thin structures, residual particles in the polymer matrix, that the method, similar then emulsion freeze-drying
irregularly shaped pores, and insufficient interconnectivity. technique, is user and technique sensitive and that the
The supercritical fluid-gassing process has been known processing parameters have to be well controlled. Nam
for many years in the non-medical polymer industry [49] and Park [57] as well as Zhang and Ma [58] fabricated
as well as in the pharmaceutical community [50]. This polymer and polymer/HA specimens with a porosity of
technology is used to produce foams and other highly
porous products. The polymers which can be used for
this technology have to have an high amorphous frac-
tion. The polymer granules are plasticized due to the
employment of a gas, such as nitrogen or carbon dioxide,
at high pressures. The diffusion and dissolution of the gas
into the polymer matrix results in a reduction of the
viscosity, which allows the processing of the amorphous
bioresorbable polyesters in a temperature range of
30-40~ [51]. The supercritical fluid-gassing technology
allows the incorporation of heat sensitive pharmaceut-
icals and biological agents. However, on average only
10-30% of the pores are interconnected [51,52]. Harris
et al. [53] combined this technology with particulate
leaching to gain a highly interconnected void network.
The researchers could control porosity and pore size by
varying the particle/polymer ratio and particle size.
Whang et al. [54,55] developed a protocol for the
fabrication of aliphatic polyester-based scaffolds by using Fig. 4. Polymerdisks with a diameter of 500 ~tm and 40 gm thickness
the emulsion freeze-drying method. Scaffolds with poros- allow to stack a 3D scaffold with a porosity of 98%.
ity greater than 90%, median pore sizes ranging from 15
to 35 gm with larger pores greater than 200 gm were
fabricated. The scaffold pore architecture was highly in-
terconnected which is necessary for tissue ingrowth and
regeneration [54]. Based on their results from an animal
experiment, the interdisciplinary group proposed a scaf-
fold design concept which results in in vivo bone regen-
eration based on hematoma stabilization [56]. The
authors compare their in vivo bone engineering concept
to the induction phase of fracture healing. The osteop-
rogenitor cells which are in the blood of the osseous
wound are embedded in the scaffold microarchitecture
via the hematoma. The multipotent cells differentiate to
osteoblasts due to the presence of growth factors which
are released by the host bone. However, the emulsion Fig. 5. Schematicillustration of the fused deposition modeling(FDM)
freeze-drying method is user and technique sensitive. The process.
 The B i o m a t e r i a l s Silver J u b i l e e C o m p e n d i u m 183
D. W. Hutmacher/ Biomaterials 21 (2000) 2529-2543 2537

up to 9 5 % . At present, only p o r e sizes of up to 100 g m used to tissue e n g i n e e r b o n e a n d cartilage [62-64]. Excel-

can be r e p r o d u c i b l y f a b r i c a t e d by t h e r m a l l y i n d u c e d lent results in tissue e n g i n e e r i n g cartilage h a v e been
phase-separation technology. achieved by using n o n - w o v e n m e s h e s c o m p o s e d of poly-
A n u m b e r of textile t e c h n o l o g i e s h a v e the p o t e n t i a l to m e r fibers of P G A , P G A / P D L A , a n d P G A / P L L A . This
design a n d fabricate highly p o r o u s scaffolds [61]. Yet, w o r k has been reviewed by F r e e d et al. [65] a n d will n o t
only so-called n o n - w o v e n mesh-like designs h a v e been be discussed here. In general, n o n - w o v e n c o n s t r u c t s can

Fig. 6. (a) 3D scaffold systems of various porosity and pore geometry fabricated by FDM. Magnification x 7.5, scale bar represents 1 mm. (i)-(iii)
lay-down pattern: 0/90~ nozzle tip: 0.016"; porosity: 50, 68, 75%; (iv)-(vi)0/90~ 0.010"; 50, 68, 75%; (vii)-(viii)0/60/120~ 0.016"; 68, 75%; (ix) 0/60/120~
0.010"; 80%; (x)-(xii) 0/60/120~ 0.010"; 50, 68, 75%. (b) Left: cross-sectional view of freeze-fractured PCL scaffold with lay-down pattern
0/72/144/36/108 ~ Right: plan view of same specimen. (c) Left: top view of PCL-HA scaffold with lay-down pattern 0/60/120 ~ The material
composition consists of 80% PCL and 20% HA by weight. Right: a close-up view of the same specimen at a higher magnification, shows that HA
particles are at the scaffold surface. (d) Freeze-fracture cross-sectional surface ( x 23) of a bioerodable scaffold designed for a bone/cartilage interface.
A 0/60/120 ~ and a 0/90 ~lay-down pattern of the roads shown in the upper and lower portion respectively of the SEM picture. Despite being fabricated
with different lay-down patterns, the porosity (67%) of both portions could be made identical. However, by applying FDM the porosity of each
portion of the 3D scaffold can be varied according to the road spacing for individual layers.
 184 The Biomaterials Silver Jubilee Compendium
2538 D.W. Hutmacher / Biomaterials 21 (2000) 2529-2543

Fig. 6. (Continued).

be only used for Strategy II since their physical proper- Rapid prototyping technologies as well as so-called 'wa-
ties do not allow load-bearing applications. fer stacking systems' [66] (Fig. 4) have the potential to
All the above-described technologies except the mem- design a 3-D construct in a multi-layer design within the
brane-lamination method, do not allow the fabrication of same gross architectural structure.
a 3-D scaffold with a varying multiple layer design. Such In engineering literature [67] Rapid prototyping Tech-
a matrix architecture is advantageous in instances where nologies (RP) also called Solid Free Form fabrication
tissue engineers want to grow a bi- or multiple tissue (SFF) methods are defined as a set of manufacturing
interface, e.g. an articular cartilage/bone transplant. processes that are capable of producing complex-free
 The Biomaterials Silver Jubilee Compendium 185
D.W. Hutmacher / Biomaterials 21 (2000) 2529-2543 2539

form parts directly from a computer-aided design

(CAD) model of an object without part specific toolin
g or knowledge. Unlike machining processes such as
milling, drilling, etc., which are subtractive in nature, RP
systems join together liquid, powder and sheet materials
to form parts. Layer by layer, RP machines fabricate
plastic, wood, ceramic and metal objects using thin hori-
zontal cross sections directly from a computer-generated
model.
Rapid prototyping technologies, such as 3-D printing
(3-DP) and fused deposition modeling (FDM) allow the
development of manufacturing processes to create por-
ous scaffolds that mimic the microstructure of living
tissue. Three-dimensional printing--developed at the
Massachusetts Institute of Technology [68]--is also
a rapid prototyping technology which has been used to
process bioresorbable scaffolds for tissue engineering ap-
plications [69,70]. The technology is based on the print-
ing of a binder through a print head nozzle onto
a powder bed, with no tooling required. The part is built
sequentially in layers. The binder is delivered to the
powder bed producing the first layer, the bed is then
lowered to a fixed distance, powder is deposited and
spread evenly across the bed, and a second layer is built.
This is repeated until the entire part, e.g. a porous scaf-
fold, is fabricated. Following treatment, the object is
retrieved from the powder bed and excess unbound pow-
der is removed. The speed, flow rate and even drop
position can be computer controlled to produce complex
3-D objects [63]. This printing technique permits CAD)
and custom-made fabrication of bioresorbable hybrid
scaffold systems. The entire process is performed under
room-temperature conditions. Hence, this technology
has great potential in tissue engineering applications.
Biological agents, such as cells, growth factors, etc., can
be incorporated into a porous scaffold without inactiva-
tion if non-toxic binders, e.g. water can be used [71].
Unfortunately, aliphatic polyesters can be only dissolved
in highly toxic solvents, such as chloroform, methylene
chloride, etc. To date, only bioresorbable scaffolds with-
out biological agents within the polymer matrix and in
combination with particle leaching have been processed
by 3-D printing. In addition, the mechanical properties
Fig. 7. Schematic illustration of the surgical placement of a tissue
and accuracy of the specimen manufactured by 3-D engineered a bone/cartilage interphase. The dental cylinder implant-
printing have to be significantly improved [72]. like design of the bony part of the scaffold allows to apply a press-fit
The FDM process forms 3-D objects from a CAD file between the host bone and the cylindrical portion of the scaffold.
as well as digital data produced by an imaging source
such as computer tomography (CT) or magnetic reson-
ance imaging (MRI) [73]. The process begins with the In effect, the process draws the designed model (scaffold)
design of a conceptual geometric model on a CAD work- one layer at a time.
station. The design is imported into a software, which Thermoplastic polymer material, 1.78 mm in diameter,
mathematically slices the conceptual model into horizon- feeds into the temperature-controlled FDM extrusion
tal layers. Toolpaths are generated before the data is head where it is heated to a semi-liquid state. The head
downloaded to the FDM hardware. The FDM extrusion extrudes and deposits the material in ultra-thin layers
head (Fig. 5) operates in the X and Y axes while the onto a fixture less base. The head directs the material
platform lowers in the Z-axis for each new layer to form. precisely into place. The material solidifies, laminating to
 186 The Biomaterials Silver Jubilee C o m p e n d i u m
2540 D.W. Hutmacher/ Biomaterials 21 (2000) 2529-2543

the preceding layer. Parts are fabricated in layers, where the research and application scope of F D M results in
a layer is built by extruding a small bead of material, or a need/demand to evaluate the critical factors for using
road, in a particular lay-down pattern, such that the layer the new polymeric and composite materials on F D M
is covered with the adjacent roads. After a layer is com- systems [74].
pleted, the height of the extrusion head is increased and At present, the author's multidisciplinary group has
the subsequent layers are built to construct the part. In been able to evaluate the parameters to process a number
the past, non-medical and medical F D M users could of potential scaffold materials, such as P C L and
only use a few non-resorbable polymeric materials, such P C L / H A by F D M . To our knowledge, we are the first
as polyamide, ABS, and other resins. The broadening of group to report the processing of bioresorbable scaffolds

Fig. 8. (a) Schematic illustration of a perfusion culture chamber for tissue engineering a bone/cartilage interphase; (b) schematic sketch of the
bioreactor concept for the tissue engineering of bone and cartilage simultaneously, in the one device like scaffold architecture.
 The Biomaterials Silver Jubilee C o m p e n d i u m 187
D.W. Hutmacher/ Biomaterials 21 (2000) 2529-2543 2541

(Fig. 6a-d) for tissue engineering applications using References

FDM.
[1] Patrick Jr CW, Mikos AG, McIntire. Prospectus of tissue engin-
eering. In: Patrick Jr CW, Mikos AG, McIntire LV, editors.
Frontiers in tissue engineering. New York, USA: Elsevier Science,
5. Tissue engineering of a articular cartilage-bone 1998. p. 3-14.
interface [-2] Naughton GK, et al. Emerging developments in tissue engineer-
ing and cell technology. Tissue Eng 1995;1(2):211-9.
Articular cartilage repair, regeneration and generation [3] Langer R, Vacanti JP. Tissue engineering. Science 1993;260: 920-6.
have been reviewed by Buckwalter and M a n k i n [75] as [4] Agrawal C, Niederauer G, Micallef D, Athanasiou K. The use of
well as N e w m a n [76]. Both the reports have concluded PLA-PGA polymers in orthopaedics. In: Encyclopedic hand-
book of biomaterials and bioengineering. New York: Marcel
that in the last two decades, clinical and basic scientific Dekker, 1995. p. 2081-115.
investigations have shown that the i m p l a n t a t i o n of [5] Bajpai PK. Biodegradable scaffolds in orthopedic, oral and maxi-
artificial matrices, growth factors, perichondrium, and llofacial surgery. In: Rubin LR, editor. Biomaterials in reconstruc-
periosteum, can stimulate the f o r m a t i o n of cartilaginous tive surgery. Mosby: St. Louis, 1983. p. 156.
tissue in o s t e o c h o n d r a l and chondral defects in synovial [6] Chaignaud BE, Langer R, Vacanti JP. The history of tissue
engineering using synthetic biodegradable polymer scaffolds and
fluids. However, the available evidence indicates that the
cells. In: Atala A, Mooney DJ, editors. Synthetic biodegradable
results vary considerably a m o n g the individuals, and that polymer scaffolds. Boston: Birkhauser, 1997. p. 1-14.
the tissues formed using these t r e a t m e n t regimes do not [-7] Hollinger JO, Chaudhari A. Bone regeneration materials for the
duplicate the composition, structure, and mechanical mandibular and craniofacial complex. Cells Mater 1992;2:143-51.
properties of n o r m a l articular cartilage [77-81]. In addi- [-8] Hutmacher D, Hurzeler M, Schliephake H. A review of material
tion, none of those matrix designs could be securely fixed properties of biodegradable and bioresorbable polymers and de-
vices for GTR and GBR applications. Int J Oral Maxillofac
under load-bearing conditions to the o s t e o c h o n d r a l bone Implants 1996;11:667-78.
which is a conditio sine qua non from a biomechanical [9] Hutmacher D, Kirsch A, Ackermann KL, Huerzeler MB. Matrix
and clinical point of view. and carrier materials for bone growth factors--state of the art and
The author's multidisciplinary g r o u p has concep- future perspectives. In: Stark GB, Horch R, Tancos E, editors.
tualized a rationale for tissue engineering a load-bearing Biological matrices and tissue reconstruction. Heidelberg, Ger-
many: Springer, 1998. p. 197-206.
o s t e o c h o n d r a l t r a n s p l a n t (Fig. 7). The scaffold design,
[10] Pachence JM, Kohn J. Biodegradable polymers for tissue
material and fabrication technology, as well as the engineering. In: Lanza RP, Langer R, Chick WL, editors. Prin-
bioreactor design (Fig. 8a and b), enables the tissue ciples of tissue engineering. Austin, Texas, USA: R.G. Landes Co.,
engineering of bone and cartilage simultaneously, in the 1997. p. 273-93.
one scaffold architecture. The device-like design of the [11] West JL, Hubbell JA. Bioactive polymers. In: Atala A, Mooney
bony scaffold part allows a secure b o n e - t o - b o n e fixation DJ, editors. Synthetic biodegradable polymer scaffolds. Boston:
Birkhauser, 1997. p. 83-95.
of the articular b o n e - c a r t i l a g e interface. The concept has [12] Wong WH, Mooney DJ. Synthesis and properties of biodegrad-
been discussed in detail elsewhere E82]. able polymers used as synthetic matrices for tissue engineering. In:
Atala A, Mooney DJ, editors. Synthetic biodegradable polymer
scaffolds. Boston: Birkhauser, 1997. p. 51-82.
6. Conclusions [13] Vert M, Li MS, Spenlehauer G, Guerin P. Bioresorbability and
biocompatibility of aliphatic polyesters. J Mater Sci 1992;3:
432-46.
The application of regulatory a p p r o v e d biomaterials [-14] Kronenthal RL. Biodegradable polymers in medicine and sur-
to design and fabricate 3-D scaffolds has strongly gery. Polym Sci Technol 1975;8:119-37.
s u p p o r t e d the drive for the establishment of tissue [15] Agrawal CM, Athanasiou KA. Techniques to control pH in
engineering research. Reviewing the experimental and vicinity of biodegrading PLA-PGA implants. J Biomed Mater
Res Appl Biomater 1997;38(2):105-14.
clinical studies, it can be concluded that the ideal
[-16] Li S, Garreau H, Vert M. Structure-property relationships in the
scaffold and matrix material for tissue engineering case of the degradation of massive aliphatic poly(~-hydroxy-acids)
bone and cartilage has not yet been developed. In in aqueous media. Part 1: poly(D,L-lacticacid). J Mater Sci: Mater
general, the tissue engineers do not i m p l e m e n t inno- Med 1990;1:123.
vative scaffold fixation features into their design concept. [17] Li S, Garreau H, Vert M. Structure-property relationships in the
However, from a clinical point of view, the secure case of the degradation of massive aliphatic poly(~-hydroxy-acids)
in aqueous media. Part 2: degradation of lactide-glycolide
and user friendly transplant fixation is a conditio sine PLA37.5 : GA25 and PLA75 :GA poly(D,L-lactic acid). J Mater
qua non. The shape of a m e s e n c h y m a l tissue such as Sci: Mater Med 1990;1:131.
a bone and articular cartilage is often critical to its [18] Li S, Garreau H, Vert M. Structure-property relationships in the
function. Ideally, a polymeric scaffold material should case of the degradation of massive aliphatic poly(~-hydroxy-acids)
permit application of a solid-free form fabrication tech- in aqueous media. Part 3: influence of the morphology of poly(L-
lactic acid). J Mater Sci: Mater Med 1990;1:198.
nology, so that a porous scaffold with any desired 3-D
[19] Bergsma EJ, Brujn W, Rozema FR, Bos RM, Boering G. Late
g e o m e t r y can be designed and fabricated by using CT tissue response to poly(L-lactide) bone plates and screws. Bio-
and M R I data. materials 1995;16(1):25-31.
 188 The Biomaterials Silver Jubilee Compendium
2542 D.W. Hutmacher / Biomaterials 21 (2000) 2529-2543

[20] Bergsma EJ, Rozema FR, Bos RM, Brujn W. Foreign body [41] Reece GP, Patrick Jr CW. Tissue engineered construct design
reactions to resorbable poly(L-lactide) bone plates and screws principles. In: Patrick Jr CW, Mikos AG, McIntire LV, editors.
used for the fixation of unstable zygomatic fractures. J Maxillofac Frontiers in tissue engineering. New York, USA: Elsevier Science,
Surg 1993;51:666-70. 1998. p. 166-96.
[21] BiSstmann O, Hirvensalo E, M~ikinen J, Rokkanen P. Foreign [42] Vander AJ, Shermann JH, Luciano DS. Human physiology. New
body reactions to fracture fixation implants of biodegradable York: McGraw-Hill, 1985. p. 341-66.
synthetic polymers. J Bone Jt Surg 1990;B72:592. [43] Lu L, Mikos AG. The importance of new processing techniques in
[22] Rehm KE, Claes L, Helling HJ, Hutmacher D. Application of tissue engineering. MRS Bull 1996;11:28-32.
a polylactide pin. An open clinical prospective study. In: Leung [44] Thomson RC, Yaszemski MJ, Mikos AG. Polymer scaffold pro-
KS, Hung LK, Leung PC, editors. Biodegradable implants in cessing. In: Lanza RP, Langer R, Chick WL, editors. Principles of
fracture fixation. Hong Kong: World Scientific, 1994. p. 54. tissue engineering. Austin, TX, USA: R.G. Landes Co., 1997.
[23] Hutmacher DW, Kirsch A, Ackermann KL, Huerzeler MB. A tis- p. 263-72.
sue engineered cell occlusive device for hard tissue regenera- [45] Widmer MS, Mikos AG. Fabrication of biodegradable polymer
tion--a preliminary report. Int J Periodontics Restorative Dent, scaffolds for tissue engineering. In: Patrick Jr CW, Mikos AG,
accepted for publication. McIntire LV, editors. Frontiers in tissue engineering. New York,
[24] Shikinami Y, Okuno M. Bioresorbable devices made of forged USA: Elsevier Science, 1998. p. 107-20.
composites of hydroxyapatite (HA) particles and poly-L-lactide [46] Mikos AG, Thorsen AJ, Czerwonka LA, Bao Y, Langer R, Win-
(PLLA): part I. Basic characteristics. Biomaterials 1998;20: 859-77. slow DN, Vacanti JP. Preparation and characterization of
[25] Freed LE, Vunjak-Novakovic G. Tissue culture bioreactors: poly(L-lactic acid) foams. Polymer 1994;35(5):1068-77.
chondrogenesis as a model system. In: Lanza RP, Langer R, [47] Mikos AG, Sarakinos G, Leite SM, Vacanti JP, Langer R.
Chick WL, editors. Principles of tissue engineering. Austin. Texas, Laminated three-dimensional biodegradable foams for use in
USA: R.G. Landes Co, 1997. p. 151-65. tissue engineering. Biomaterials 1993;14:323-30.
[26] Young JH, Teumer J, Kemp PD, Parenteau NL. Approaches to [48] Widmer MS, Gupta PK, Lu L, Meszlenyi RK, Evans GRD,
transplanting engineered cells and tissues. In: Lanza RP, Langer Brandt K, Savel T, Gurlek A, Patrick CW, Mikos AG. Manufac-
R, Chick WL, editors. Principles of tissue engineering. Austin. ture of porous biodegradable polymer conduits by an extrusion
Texas, USA: R.G. Landes Co., 1997. p. 297-307. process for guided tissue regeneration. Biomaterials 1998;19:
[27] Hillsley MV, Frangos JA. Review: bone tissue engineering: the 1945-55.
role of interstitial fluid flow. Biotechnol Bioeng 1994;43:573-81. [49] Vieth WR. Diffusion in and through polymers: principles and
[28] Thompson DE, Agrawal CM, Athanasiou K. The effects of dy- applications. M finchen: Carl Hanser Verlag, 1991.
namic compressive loading on biodegradable implants of 50-50% [50] Tom JW, Debenedetti PG. Particle formation with supercritical
polylactic acid-polyglycolic acid. Tissue Eng 1996;2(1):61-74. fluids--a review. J Aerosol Sci 1991;22(5):555-84.
[29] Dunkelman NS, Zimber MP, LeBaron RG, Pavelec R, Kwan M, [51] Michaeli W, Seibt S. Molding of resorbable polymers at low
Purchio AF. Cartilage production by rabbit articular chon- temperatures. ANTEC 1995. p. 3397-9.
drocytes on polyglycolic acid scaffolds in a closed bioreactor [52] Mooney DJ, Baldwin DF, Suh NP, Vacanti JP, Langer R. Novel
system. Biotechnol Bioeng 1995;46:299-305. approach to fabricate porous sponges of poly(D,L-lactic-co-
[30] Freed L, Vunjak-Novakovic G, Biron RJ, Eagles DB, Lesnoy DC, glycolic acid) without the use of organic solvents. Biomaterials
Barlow SK, Langer R. Biodegradable polymer scaffolds for tissue 1996;17:1417-22.
engineering. Biotechnology 1994;12:689-93. [53] Harris LD, Kim BS, Mooney DJ. Open pore biodegradable
[31] Grande DA, Halberstadt C, Naughton G, Schwartz R, Manji R. matrices formed with gas foaming. J Biomed Mater Res 1998;
Evaluation of matrix scaffolds for tissue engineering of cartilage 42:396-402.
grafts. J Biomed Mater Res 1997;34:211-20. [54] Whang K, Thomas CH, Healy KE, Nuber G. A novel
[32] Martin I, Padera RF, Langer R, Vunjak-Novakovic G, Freed LE. method to fabricate bioabsorbable scaffolds. Polymers
In vitro differentiation of chick embryo bone marrow stromal 1995;36:837-42.
cells into cartilaginous and bone-like tissues. J Orthop Res [55] Whang K, Tsai DC, Nam EK, Aitken M, Sprague SM, Patel PK,
1998;16:181-9. Healy KE. Ectopic bone formation via rhBMP-2 delivery from
[33] Puelacher WC, Vacanti JP, Ferraro NF, Schloo, Vacanti CA. porous bioresorbable polymer scaffolds. J Biomed Mater Res
Femoral shaft reconstruction using tissue engineered growth of 1998;42:491-9.
bone. Int J Oral Maxillofac Surg 1996;25:223-8. [56] Whang K, Healy KE, Elenz DR, Nam EK, Tsai DC, Thomas CH,
[34] Ma PX, Langer R. Morphology and mechanical function of Nuber MD, Glorieux FH, Travers R, Sprague SM. Engineering
long-term in vitro engineered cartilage. J Biomed Mater Res bone regeneration with bioabsorbable scaffolds with novel micro-
1999;44:217-21. architecture. Tissue Eng 1999;5(1):35-51.
[35] Freed LE, Hollander AP, Martin I, Barry JR, Langer R, Vunjak- [57] Nam SY, Park TG. Porous biodegradable polymeric scaffolds
Novakovic G. Chondrogenesis in a cell-polymer-bioreactor sys- prepared by thermally induced phase separation. J Biomed Mater
tem. Exp Cell Res 1998;240(1):58-65. Res 1999;47(1):8-16.
[36] Reddi AH, Wientrob S, Muthukumaran N. Biological principles [58] Zhang R, Ma PX. Poly(~-hydroxyl acids)/hydroxyapatite porous
of bone induction. Orthopedic Clinics North America composites for bone-tissue engineering. I. Preparation and mor-
1987;18:207-12. phology. J Biomed Mater Res 1999;44(4):446-55.
[37] Brekke JH. A rationale delivery of osteoconductive proteins [59] Schugens C, Maguet V, Grandfils C, Jerome R, Teyssie P. Poly-
(a review). Tissue Eng 1996;2(2):97-114. lactide macroporous biodegradable implants for cell transplanta-
[38] Chen CS, Milan Mrksich, Sui Huang, Whitesides GM, tion. 2. Preparation of polylactide foams by liquid-liquid phase
Ingber D. Geometric control of cell life and death. Science separation. J Biomed Mater Res 1996;30:449-61.
1997;276:1425-8. [60] Lo H, Ponticiello MS, Leong KW. Fabrication of controlled
[39] Ingber DE. Tensegrity: the architectural basis of cellular release biodegradable foams by phase separation. Tissue Eng
mechanotransduction. Ann Rev Physiol 1997;59:575-99. 1995;1:15-28.
[40] Freed LE, Hollander AP, Martin I, Barry JR, Langer R, Vunjak- [61] Hutmacher DW, Teoh SH, Feng SS, Ramakrishna S, Zein I.
Novakovic G. Chondrogenesis in a cell-polymer-bioreactor sys- Tissue engineering research--the engineer's role. Medical Dev J,
tem. Exp Cell Res 1998;240(1):58-65. accepted for publication.
 The Biomaterials Silver Jubilee Compendium 189
D.W. Hutmacher / Biomaterials 21 (2000) 2529-2543 2543

[62] Rotter N, Aigner J, Naumann A, Planck H, Hammer C, Burmes- fabricated by three-dimensional printing. J Biomater Sci Polym
ter G, Sittinger M. Cartilage reconstruction in head and neck Ed 1996;8(1):63-75.
surgery: comparison of resorbable polymer scaffolds for tissue [73] Agarwala MK, Jamalabad VR, Langrana NA, Safari A, Whalen
engineering of human septal cartilage. J Biomed Mater Res PJ, Danforth SC. Structural quality of parts processed by fused
1998;42:347-56. deposition. Rapid Prototyping J 1996;2(4):4-19.
[63] Sittinger M, Reitzel D, Dauner M, Hierlemann H, Hammer C, [74] Gray IV RW, Baird DG, Bohn JH. Effects of processing condi-
Kastenbauer E, Planck H. Resorbable polyesters in cartilage tions on short TCLP fiber reinforced FDM parts. Rapid Proto-
engineering: affinity and biocompatibility of polymer fiber struc- typing J 1998;1(4):14-25.
tures to chondrocytes. J Biomed Mater Res Appl Biomater [75] Buckwalter JA, Mankin HJ, Articular cartilage. Part II: degener-
1996;33(2):57-63. ation and osteoarthrosis, repair, regeneration, and transplanta-
[64] Puelacher WC, Vacanti JP, Ferraro NF, Schloo, Vacanti CA. tion. J Bone Jt Surg 1997;A79(4):612-32.
Femoral shaft reconstruction using tissue engineered growth of [76] Newman AP. Articular cartilage repair. Amer J Sports Med
bone. Int J Oral Maxillofac Surg 1996;25:223-8. 1998;26(2):309-24.
[65] Freed LE, Vunjak-Novakovic G. Culture of organized cell com- [77] Glowacki J, Yates K, Little G, Mizuno S. Induced chondroblastic
munities. Adv Drug Deliver Rev 1998;33(1-2):15-30. differentiation of human fibroblasts by three-dimensional
[66] Hutmacher DW, Burdet E, Schantz TJ, Robotic micro-assembly culture with demineralized bone matrix. Mater Sci Eng 1998;
fabrication of three-dimensional bioresorbable scaffolds for tissue C6:199-203.
engineering, in press. [78] de Chalain T, Phillips JH, Hinek A. Bioengineering of elastic
[67] Beaman JJ. Backround and definitions. In: Beamann JJ, Barlow cartilage with aggregated porcine and human auricular chon-
JW, Bourell DL, Crawford RH, Marcus HL, McAlea KP, editors. drocytes and hydrogels containing alginate, collagen, and •-elas-
Solid free-form fabrication: a new direction in manufacturing. tin. J Biomed Mater Res 1999;44:280-8.
Boston, USA: Kluwer Academic Publishers, 1997. p. 1-20. [79] Lee EH, Chen F, Chan JWK, Bose K. Treatment of growth arrest
[68] Sachs E, Cima M, Williams P, Brancazio D, Cornie J. Three- by transfer of cultured chondrocytes into physeal defects. J Paedi-
dimensional printing: rapid tooling and prototypes directly from atric Orthoped 1998;18(2):155-60.
a CAD model. J Eng Industry 1992;114:481-8. [80] ten Koppel PG, van Osch GJ, Verwoerd CD, Verwoerd-Verhoef
[69] Cima LG, Vacanti JP, Vacanti C, Ingber D, Mooney D, Langer HL. Efficacy of perichondrium and a trabecular demineralized
R. Tissue engineering by cell transplantation using degradable bone matrix for generating cartilage. Plast Reconstruct Surg
polymer substrates. J Biom Eng 1991;113:143-51. 1998; 102(6):2012-20.
[70] Park A, Wu B, Griffith LG. Integration of surface modification [81] Bruns J, Kahrs J, Kampen J, Behrens P, Plitz W. Autologous
and 3-D fabrication techniques to prepare patterned poly(L-lact- perichondral tissue for meniscal replacement. J Bone Jt Surg--
ide) substrates allowing regionally selective cell adhesion. J Bio- British Volume 1998;80(5):918-23.
mater Sci Polym Ed 1998;9(2):89-110. [82] Hutmacher DW, Zein I, Teoh SH, Ng KW, Schantz JT, Leahy JC.
[71] Wu BM, Borland SW, Giordano RA, Cima LG, Sachs EM, Cima Design and fabrication of a 3D scaffold for tissue engineering
MJ. Solid free-form fabrication of drug delivery devices. J Contr bone. In: Agrawal CM, Parr JE, Lin ST, editors. Synthetic bioab-
Rel 1996;40:77-87. sorbable polymers for implants, STP 1396. American Society for
[72] Giordano RA, Wu BM, Borland SW, Cima LG, Sachs EM, Cima Testing and Materials, West Conshohocken, PA, 2000.
MJ. Mechanical properties of dense polylactic acid structures p. 152-67.