Abstract
Background: We are not aware of any in vitro study comparing the
biomechanical, biochemical, and cellular properties of commercial
extracellular matrix materials marketed for rotator cuff tendon repair. In
this study, the properties of GraftJacket, TissueMend, Restore, and CuffPatch
were quantified and compared with each other. The elastic moduli were also
compared with that of normal canine infraspinatus tendon.
Methods: Samples were tested from different manufacturing lots of
four materials: GraftJacket (ten lots), TissueMend (six), Restore (ten), and
CuffPatch (six). The Kruskal-Wallis test was used to compare thickness,
stiffness, and modulus as well as hydroxyproline, chondroitin/dermatan sulfate
glycosaminoglycan, hyaluronan, and DNA contents among these matrices. The
moduli of the extracellular matrices were also compared with those of normal
canine infraspinatus tendon.
Results: All four extracellular matrices required 10% to 30% stretch
before they began to carry substantial load. Their maximum moduli were
realized in their linear region at 30% to 80% strain. The elastic moduli of
all four commercial matrices were an order of magnitude lower than that of
canine infraspinatus tendon. TissueMend had significantly higher DNA content
than the other three matrices (p < 0.0001), although both Restore
and GraftJacket also had measurable amounts of DNA.
Conclusions: Our data demonstrate chemical and mechanical
differences among the four commercial extracellular matrices that we
evaluated. Probably, the source (dermis or small intestine submucosa), species
(human, porcine, or bovine), age of the donor (fetal or adult), and processing
of these matrices all contribute to the unique biophysical properties of the
delivered product. The biochemical composition of commercial extracellular
matrices is similar to that of tendon. However, the elastic moduli of these
materials are an order of magnitude lower than that of tendon, suggesting a
limited mechanical role in augmentation of tendon repair.
Clinical Relevance: These data will help inform and guide the
clinical community with regard to the appropriate use of commercially
available extracellular matrix products for augmentation of rotator cuff
tendon repair. Knowledge of the biophysical properties of these materials is
fundamental to making an educated decision about whether a given matrix might
provide mechanical augmentation and/or enhance the biology of tendon-to-bone
healing.
Currently, the most common method for enhancing the healing of rotator cuff
repair biologically is the use of natural extracellular matrices. Several
matrices are commercially marketed as patches to reinforce soft-tissue repair
during rotator cuff surgery (Table
I). These products include collagen-rich extracellular matrices
such as dermis (GraftJacket, TissueMend, and Zimmer Collagen Repair Patch) and
small intestine submucosa (Restore and CuffPatch). All of the animal-derived
products are regulated by the United States Food and Drug Administration (FDA)
through the 510(k) device mechanism rather than through biologic or drug
mechanisms. In general, these products have FDA approval "for
reinforcement of the soft tissues, which are repaired by suture or suture
anchors, during rotator cuff repair surgery"
().
Consistent with many other allograft products, GraftJacket Regenerative Tissue
Matrix is classified as human tissue for transplantation. Under these
regulations, no premarket regulatory review is required.
On the basis of sales reports, it is estimated that thousands of
extracellular matrix scaffolds are used annually in the United States alone.
Although the use of biologic scaffolds is becoming more popular, we are not
aware of any human clinical trial demonstrating their efficacy in improving
the healing of rotator cuff tendons. Furthermore, there is little
retrospective data describing the complications or adverse events associated
with the use of these products.
GraftJacket Regenerative Tissue Matrix is manufactured by LifeCell
(Branchburg, New Jersey) and is distributed by Wright Medical Technology
(Arlington, Tennessee) for the orthopaedic and podiatric markets. GraftJacket
is derived from human allograft skin that is processed with use of a patented
technique to remove the epidermis, cells, and cell remnants. The remaining
acellular, dermal layer is preserved by utilizing a proprietary freeze-drying
method, which retains the native extracellular architecture and vascular
channels. The matrix contains biochemical components including collagen,
elastin, and proteoglycans and is not artificially cross-linked. It is
packaged dry. The material is a single layer and is provided in a variety of
thicknesses (0.5 to 2 mm) and sizes for targeted surgical indications.
TissueMend Soft Tissue Repair Matrix is manufactured by TEI Biosciences
(Boston, Massachusetts) and marketed by Stryker Orthopaedics (Mahwah, New
Jersey). TissueMend is marketed as an acellular, nondenatured collagen
membrane derived from fetal bovine dermis. It is not artificially
cross-linked. The device is one layer and nominally 1 mm thick. TissueMend is
composed primarily of type-I and type-III collagen fibers. It is processed to
remove carbohydrate, lipid, and fat cells. It is lyophilized and packaged
dry.
The Zimmer Collagen Repair (ZCR) Patch is manufactured by Tissue Science
Laboratories (Aldershot, Hampshire, United Kingdom). It is distributed by
Zimmer (Warsaw, Indiana) for rotator-cuff-related indications. The ZCR Patch
is marketed as an acellular sheet of cross-linked porcine dermis. Organic and
enzymatic extractions are undertaken to remove fat, cellular material, and
soluble proteins. The material is cross-linked with diisocyanate and is thus
resistant to enzymatic degradation. It is one layer and nominally 1.5 mm
thick. It is packaged hydrated.
The Restore Orthobiologic Implant is manufactured and marketed by DePuy
Orthopaedics (Warsaw, Indiana). Restore is a disk composed of ten layers of
porcine small intestine submucosa. The small intestine submucosa is
disinfected with peracetic acid and ethanol and does not contain any viable
cells. The small intestine submucosa extracellular matrix contains
predominately type-I collagen, fibronectin, chondroitin sulfate, heparin,
heparin sulfate, hyaluronan, and growth factors (such as fibroblast growth
factor-2 [FGF-2], transforming growth factor-ß [TGF-ß], and vascular
endothelial growth factor
[VEGF])1-4.
To produce the Restore Orthobiologic Implant, ten individual small intestine
submucosa layers are oriented at approximately 20° relative to each other
and laminated together under a vacuum press. Restore is not artificially
cross-linked. The implant is packaged dry and is nominally 0.8 to 1 mm
thick.
CuffPatch Bioengineered Tissue Reinforcement is manufactured by
Organogenesis (Canton, Massachusetts) and is marketed by Arthrotek (Warsaw,
Indiana). CuffPatch is an eight-layer, acellularized, porcine small intestine
submucosa sheet. A nondetergent, nonenzymatic chemical cleaning protocol
removes cells and cellular debris from small intestine submucosa without
damaging the native collagen structure. Following lamination of the individual
small intestine submucosa layers, the product is cross-linked with
water-soluble carbodiimide. CuffPatch is packaged hydrated and is nominally
0.6 mm thick.
We are not aware of any comparative in vitro study describing the
biomechanical, biochemical, and cellular properties of these materials. The
objective of this study was to quantify and compare the properties of
GraftJacket, TissueMend, Restore, and CuffPatch with each other and, when
possible, with those of normal tendon. We were unable to obtain the ZCR Patch
for inclusion in this study.
Ten lots of GraftJacket, six of TissueMend, ten of Restore, and six of
CuffPatch were obtained from the respective manufacturers
(Table I). The geometry of the
sample lots of GraftJacket, TissueMend, and CuffPatch was approximately 5
× 5 cm, and Restore was 10 cm round. Two test strips, 12 × 45 mm
and 4 × 45 mm, were cut from each lot of each material for mechanical
testing. To determine stiffness, 12-mm-wide strips were tested because wide
strips or patches are used clinically. To determine the elastic modulus,
4-mm-wide samples were used, as this achieved a 7.5-to-one aspect ratio (after
gripping) for determination of material
properties5. The
remaining material from each lot was used for the biochemical and cellular
assays. All samples were maintained at room temperature in their commercial
packaging until analysis.
Mechanical Testing
Mechanical test samples were soaked in 0.9% normal saline solution at
4°C overnight prior to testing. Samples were clamped in custom grips under
a uniform torque of 0.14 N-m, leaving a nominal gauge length of 30 mm.
Thickness was determined at five points along the length of the samples with
use of a constant-pressure linear variable displacement transducer (LVDT)
probe (~0.001 MPa). Cross-sectional area was estimated as the product of
the nominal sample width and the average tissue thickness. Gripped samples
were then mounted in a testing tank containing physiologic saline solution at
37°C. Uniaxial tension tests were performed with use of an Instron 5543
series testing system (Instron, Canton, Massachusetts) fitted with either a
50-N or a 500-N load cell (Honeywell Sensotec, Columbus, Ohio), depending on
the material. Samples were preconditioned by cycling for five cycles to loads
representing ~10% grip-to-grip strain. Samples were then tested at 10
mm/min to failure. A 0.02-N load was used to zero the data.
Stiffness was determined from the load-displacement data of 12-mm-wide test
strips. Linear stiffness was estimated as the slope of the load-displacement
curves in the linear region for each material. The grip-to-grip displacement
used to define the linear region ranged from 16 to 28 mm for Graft-Jacket, 11
to 16 mm for TissueMend, 6.6 to 7.5 mm for Restore, and 6 to 6.6 mm for
CuffPatch.
Modulus was determined from the stress-strain data of 4-mm-wide strips.
Uniaxial grip-to-grip strain was used to estimate modulus. Modulus was
determined from the slope of the stress-strain curve in two regions: (1) 7.5%
to 8.5% grip-to-grip strain, which is in the linear region for canine
infraspinatus tendon, hereafter referred to as the "8% modulus"
and (2) the linear region for each material, hereafter referred to as the
"linear modulus." The region of grip-to-grip strain used to define
the linear region for the various materials fell in the range of 53% to 93%
for GraftJacket, 37% to 53% for TissueMend, 22% to 25% for Restore, and 20% to
22% for CuffPatch (Fig. 1).
For comparison with tendon, 4-mm-wide infraspinatus tendon strips were cut
from nine fresh-frozen canine shoulders. The tendon samples were clamped as
described for the extracellular matrix samples, leaving a nominal gauge length
of 30 mm. The only difference between the methods for the mechanical testing
of the tendon and the extracellular matrix was the level of preconditioning.
Because the tendons were stiffer than the extracellular matrix products, they
were preconditioned by cycling for five cycles to loads representing about 1%
grip-to-grip strain. The "8% modulus" and "linear
modulus" are the same outcome parameter for tendon and were determined
from the slope of the stress-strain curve between 7.5% to 8.5% grip-to-grip
strain (Fig. 1).
Hydroxyproline Content
One sample (5 to 8 mg dry weight, 0.5 cm2) from each lot from
each group was randomly cut and was analyzed for hydroxyproline content
according to previously described
methods6,7.
CuffPatch samples were lyophilized prior to analysis. Briefly, tissue samples
and standards were hydrolyzed in 6 N constant boiling hydrochloric acid
(Pierce Biotechnology, Rockford, Illinois) at 110°C. The hydrolyzed
samples were then dried, resuspended in water, and incubated with 0.062 M
chloramine-T reagent followed by 1.2 M Ehrlich reagent. Samples were run in
triplicate on a ninety-six-well plate and read at absorbance of 557 nm.
Chondroitin/Dermatan Sulfate and Hyaluronan Content
One sample (5 to 8 mg dry weight, 0.5 cm2) from each lot from
each group was randomly cut and was analyzed for chondroitin/dermatan sulfate
and hyaluronan content according to previously described
methods8. CuffPatch
samples were lyophilized prior to analysis. All samples were digested with 2.5
mg/mL of proteinase K (Fisher Biotech, Fair Lawn, New Jersey). Aliquots for
chondroitin/dermatan sulfate glycosaminoglycan (GAG) analysis were assessed
with fluorophore-assisted carbohydrate electrophoresis
(FACE)9,10.
Briefly, samples were precipitated in 77% ethanol and digested with 50 mU/mL
of hyaluronidase SD (Streptococcus dysgalactiae) followed by 100
mU/mL of chondroitinase ABC (Seikagaku America, Falmouth, Massachusetts).
Samples were lyophilized until dry and fluorotagged with 12.5 mM
2-aminoacridone HCl (Molecular Probes, Eugene, Oregon) and 1.25 M sodium
cyanoborohydride (Sigma-Aldrich, St. Louis, Missouri). After fluorotagging,
samples were mixed with glycerol, then mixed 1:1 with an AMAC-derivatized
trisaccharide standard (maltotriose), and diluted 1:1 with 0.5 M
4-morpholinoethanesulfonic acid (pH = 7.0; Acros Organics, Geel, Belgium).
The FACE gels were run with use of a commercially available electrophoresis
apparatus (Glyko, San Leandro, California) and a custom gel formulation.
Briefly, 25 µL of 10% ammonium persulfate (Bio-Rad Laboratories, Hercules,
California) and 7.5 µL of TEMED (N,N,N',N'-tetramethylethylenediamine)
(Bio-Rad) were added to 5 mL of gel solution comprising 20% acrylamide-Bis
(37.5:1; Bio-Rad), 0.045 M Tris-acetate (pH 7.0) and 2.25% glycerol
(Sigma-Aldrich). A resolving gel was poured between 10 × 10-cm glass
plates, with use of a 0.5-mm spacer. A stacking gel was then poured with use
of the same solution composition. The gels were run with use of standard 1
× Tris-borate-EDTA buffer for eighty minutes at 500 V. Gels were imaged
under ultraviolet light (365 nm), and peak areas for the various saccharide
derivatives present were quantitated (?DiHA, ?Di0S, ?Di4S,
?Di6S) with Gel Pro software (Media Cybernetics, Silver Spring,
Maryland). The total chondroitin/dermatan sulfate GAG content was defined as
?Di0S+?Di4S+?Di6S per milligram dry weight. The hyaluronan
content was defined as ?DiHA per milligram dry weight.
DNA Content
Approximately 60 mg (dry weight, ~6 cm2) from each lot from
each group was analyzed for DNA content. CuffPatch samples were lyophilized
prior to analysis. Samples were rehydrated in 1 × TE (Tris-EDTA) buffer
and digested with 5 mg/mL of proteinase K (Fisher Biotech). DNA content was
determined with use of the PicoGreen dsDNA Assay (Molecular Probes) according
to the manufacturer's instructions. Briefly, samples were incubated with and
without 50 µg/mL of DNase I (Sigma-Aldrich), then mixed 1:1 with the
PicoGreen dsDNA quantitation reagent in 1 × TE buffer, and incubated at
room temperature for five minutes, shielded from light. Samples were run in
triplicate on black ninety-six-well assay plates (Applied Biosystems, Foster
City, California) at an excitation wavelength of 485 nm and an emission
wavelength of 525 nm. Absorbance readings of DNase-treated samples were
subtracted from the paired, nontreated samples to correct for background
autofluorescence.
Statistical Analysis
The Kruskal-Wallis test was used to compare thickness, stiffness, and
moduli as well as hydroxyproline, chondroitin/dermatan sulfate GAG,
hyaluronan, and DNA contents among the extracellular matrices. The moduli of
the extracellular matrices were also compared with those of normal canine
infraspinatus tendon. A p value of =0.05 was considered significant.
All four extracellular matrices required 10% to 30% stretch before they
began to carry substantial load (Fig.
1). The maximum moduli of the extracellular matrices were realized
in their linear region at 30% to 80% strain, whereas the linear region for the
canine infraspinatus tendon was approximately 7% to 10% grip-to-grip strain.
All specimens failed at a grip during mechanical testing. Hence, failure
stress and strain could not be calculated.
There were significant differences (p < 0.05) in thickness
between all extracellular matrices except between Restore and CuffPatch
(Fig. 2, A, and
Appendix). GraftJacket was the thickest extracellular matrix (mean and
standard deviation, 1.58 ± 0.15 mm) and CuffPatch was the thinnest
(0.40 ± 0.04 mm). The linear stiffness of 12-mm-wide-strips of
GraftJacket was significantly greater (p < 0.05) than that of all
of the other extracellular matrices (Fig.
2, B, and Appendix).
The 8% strain/linear modulus for canine infraspinatus tendon averaged 405.3
± 86.4 MPa. This modulus was significantly higher than that of all
extracellular matrix materials (p < 0.0001)
(Fig. 2, C and
D, and Appendix). The tendon data were then excluded from
the analysis-of-variance test to allow comparisons among the extracellular
matrices. Both the 8% modulus (Fig. 2,
C) and the linear modulus
(Fig. 2, D) for the
dermis-derived extracellular matrices (GraftJacket and TissueMend) were
significantly lower (p < 0.01) than those for the extracellular
matrices derived from small intestine submucosa (Restore and CuffPatch).
Neither the 8% nor the linear modulus differed significantly between the two
dermis extracellular matrices or between the two small-intestine-submucosa
extracellular matrices.
The hydroxyproline content of GraftJacket was significantly less than that
of the other three extracellular matrices (p < 0.0003)
(Fig. 3, A, and
Appendix), and Restore had significantly less hydroxyproline than TissueMend
(p < 0.02). Restore had significantly higher chondroitin/dermatan
sulfate GAG (Fig. 3,
B, and Appendix) and hyaluronan
(Fig. 3, C, and
Appendix) content than the other three extracellular matrices (p <
0.0001). GraftJacket had significantly higher chondroitin/dermatan sulfate GAG
content than CuffPatch (p < 0.04) and significantly higher
hyaluronan content than TissueMend and CuffPatch (p < 0.0002).
TissueMend had significantly higher DNA content than the other three
extracellular matrices (p < 0.0001)
(Fig. 3, D, and
Appendix). Restore had significantly higher DNA content than GraftJacket and
CuffPatch (p < 0.0001). CuffPatch had significantly less DNA than
the other three extracellular matrices (p < 0.03).
The extracellular matrices derived from small intestine submucosa (Restore
and CuffPatch) had higher moduli than the dermis-derived extracellular
matrices (GraftJacket and TissueMend), and they reached their maximum (linear
region) moduli at lower levels of stretch (~20% compared with ~60%).
The 12-mm-wide GraftJacket test strips were significantly stiffer than
similarly sized pieces of the other materials as a result of being
significantly thicker. These findings demonstrate the importance of
characterizing both the material and the functional properties of these
implant biomaterials. Our biomechanical data are comparable with those in
previous reports. The material properties of 5 × 1-cm AlloDerm (not
substantively different from GraftJacket) (LifeCell) test strips were reported
to be 0.01 MPa (for strains up to 40%) and 18.4 MPa (for strains in the linear
region, 40% to
100%)11. No
uniaxial tension data are available specifically for TissueMend, Restore, or
CuffPatch, to our knowledge. However, the elastic modulus of 6 × 1-cm
test strips of small intestine submucosa patches manufactured by Cook Biotech
(West Lafayette, Indiana) averaged 26.3 ± 14.1
MPa12. Furthermore,
our modulus data are comparable with the properties of natural and processed
small intestine submucosa tested in biaxial tension and burst
tests13-15.
Our data demonstrate that the elastic moduli of commercial extracellular
matrices are an order of magnitude lower than that of canine infraspinatus
tendon. In addition, the extracellular matrix moduli are an order of magnitude
lower than the moduli (grip-to-grip strain) reported for different regions of
human infraspinatus tendon (from individuals between the ages of fifty-eight
and ninety-eight years), which ranged from approximately 84 to 187 MPa
depending on tendon region and arm
position16. The
disparity between the moduli of these extracellular matrices and those of
canine and human tendons suggests that, as a load-sharing augmentation device
for tendon repair, these extracellular matrices would likely carry only small
loads. If used as a primary graft to connect tendon to bone, these
extracellular matrices would stretch appreciably under the associated muscle
and joint loads. One should keep in mind that a tendon repair is less stiff
than the tendon proper with which the above comparisons were made; however,
the same conclusions can be inferred. While prestretching at implantation may
improve the functional contribution of these extracellular matrices, they may
offer more of a biologic advantage than a mechanical one for tendon
repair.
Restore and CuffPatch are produced by laminating individual (anisotropic)
layers of small intestine submucosa relative to each other in order to obtain
"isotropic" material behavior. GraftJacket and TissueMend are
single layers of dermis and would be expected to exhibit anisotropic material
properties17. In
this study, no effort was made to orient the rectangular test samples with
respect to the material microstructure. Furthermore, the samples were tested
only in uniaxial tension. While these test conditions are not ideal for fully
characterizing the anisotropic material properties of these extracellular
matrices (biaxial tests would be preferable), not controlling for material
orientation and a uniaxial loading mode are considered clinically relevant
test conditions for tendon augmentation applications. It was not possible to
measure local uniaxial strain in GraftJacket or TissueMend samples with use of
a dye-marking (optical) technique because anisotropy and local structural
inhomogeneities (e.g., hair follicles) in the material resulted in nonuniform
strain patterns. Thus, we used grip-to-grip displacements to determine strain.
Certainly, using grip-to-grip displacements is a gross simplification of the
complex material strains in these extracellular matrices during loading, but
using them does not invalidate our comparisons or change our overall
conclusions.
In their processed form, commercial extracellular matrices have amounts of
hydroxyproline similar to the amounts in fresh canine flexor tendon, which
averages 0.104 ± 0.004 mg/mg dry
weight8. GraftJacket
had significantly less hydroxyproline than the other materials (0.078 ±
0.013 mg/mg dry weight), an amount that is similar to the hydroxyproline
content of fresh adult dermal biopsy specimens reported previously (0.072
± 0.008 mg/mg dry
weight)18. Assuming
that the hydroxyproline in these materials is solely derived from collagen and
represents 13% of collagen by
weight19, the
collagen content of these processed extracellular matrices can be estimated to
be 60% to 95% of their dry weight with use of our assay.
In their processed form, commercial extracellular matrices have amounts of
chondroitin/dermatan sulfate GAG similar to the amount in fresh canine flexor
tendon, which averages 0.84 ± 0.24 µg/mg dry
weight8. Restore had
significantly more chondroitin/dermatan sulfate GAG and hyaluronan than the
other materials, although all materials averaged <1 µg/mg dry weight of
either chondroitin/dermatan sulfate GAG or hyaluronan. These quantities are
significantly less than the total GAG content in unprocessed small intestine
submucosa (21 µg/mg dry weight, which includes heparin sulfate, keratin
sulfate, and
hyaluronan1). They
are also less than the average chondroitin/dermatan sulfate GAG content (1.33
µg/mg dry weight) or hyaluronan content (1.51 µg/mg dry weight) of fresh
adult dermal biopsy
specimens18 or the
average chondroitin/dermatan sulfate GAG content (4.06 µg/mg dry weight) or
hyaluronan content (4.19 µg/mg dry weight) in fetal
skin20. Presumably,
the processing steps involved in acellularizing these biomaterials degrade
and/or leach substantial amounts of the GAGs. In the case of CuffPatch,
carbodiimide cross-linking forms amide bonds bridging carboxyl and amine
groups, which could substantially modify the carboxyl groups within GAGs and
make them undetectable with our assay. The relative importance of GAGs in
extracellular matrix implant incorporation and tissue regeneration is unknown,
but native differences in GAG content between these materials appear to be
greatly diminished by processing.
Commercial extracellular matrix products are marketed as
"acellularized" biomaterials because they undergo proprietary
treatments to remove cellular elements. In this study, the DNA content of
TissueMend and Restore averaged 794.6 ± 97.8 and 526.8 ± 125.6
ng/mg dry weight, respectively. The DNA content of GraftJacket averaged 134.6
± 44.0 ng/mg dry weight. Only CuffPatch contained negligible amounts of
DNA. Whether the acellularization protocol used for CuffPatch is highly
effective or the cross-link processing affects the detection of DNA is not
known. Macroscopically, CuffPatch was fully digestible by proteinase K,
suggesting that remnant DNA should have been available for detection with our
assay. Furthermore, we were able to detect DNA contents as low as 0.2 ng/mg
dry weight with our assay and conditions. The literature provides little
quantitative DNA-content data with which to compare our results. One recent
paper confirmed the presence of porcine DNA in Restore with histological
analysis and nested polymerase chain
reaction21. In
addition, AlloDerm (same as GraftJacket) has been previously reported to have
a DNA content of 273 ± 169 ng/mg dry
weight22.
The clinical implications of incomplete acellularization of these
biomaterials are not known. Acellularization treatment is intended to disrupt
cells and remove water-soluble cellular proteins in order to reduce
antigenicity. Acellularization may also enhance host cell infiltration with
phenotypically appropriate
cells23 and
possibly prevent transmission of infectious genomic
vectors22. Rigorous
immunologic and biochemical studies are needed to determine if and how much
acellularization treatment increases the safety and efficacy of these
implants.
The use of extracellular matrices is a strategy that is fundamentally
different from previous orthopaedic approaches, and the modification,
characterization, use, and clinical investigation of extracellular matrices
are rapidly evolving. In contrast to traditional orthopaedic devices composed
of polymers and metals that are intended to restore mechanical function and
remain unchanged for the life of the patient, natural extracellular matrices
are a temporary scaffold intended to enhance and accelerate the biology of
tissue repair. Extracellular matrices undergo host cell infiltration and
constructive tissue remodeling at variable rates. For instance, it has been
quantitatively shown that small intestine submucosa is rapidly replaced within
weeks following
implantation24.
Therefore, the properties described in this report are representative of these
materials only at the time of implantation.
We cannot engineer optimal repair strategies using extracellular matrix
materials until we better understand their mechanism of action. For example,
fabricating extracellular matrices in parallel with other materials may
increase their mechanical role. Natural extracellular matrices could be
modified prior to implantation to add growth factors or autologous stem cells
or tenocytes. Cell-seeded extracellular matrix grafts could be preconditioned
in bioreactors to improve their mechanical and biological properties.
In conclusion, our data demonstrated chemical and mechanical differences
among the four commercial extracellular matrices evaluated. It is likely that
the source (dermis or small intestine submucosa), species (human, porcine, or
bovine), age of the donor (fetal or adult), and processing of extracellular
matrices all contribute to the unique biophysical properties of the delivered
product. The data presented in this study will help inform and guide the
clinical community with regard to the appropriate use of these extracellular
matrix products for augmentation of rotator cuff tendon repair.
A table showing all study data is available with the electronic versions of
this article, on our web site at
(go to
the article citation and click on "Supplementary Material") and on
our quarterly CD-ROM (call our subscription department, at 781-449-9780, to
order the CD-ROM). ?
Note: The authors thank Robert S. Butler for statistical
assistance. DePuy Orthopaedics, Arthrotek, Stryker Orthopaedics, and Wright
Medical Technology donated the materials used in this study.
Hodde JP, Badylak SF, Brightman AO,
Voytik-Harbin SL. Glycosaminoglycan content of small intestinal submucosa: a
bioscaffold for tissue replacement. Tissue Eng.
1996;2:
209-17.2209
1996
[PubMed][CrossRef]
Hodde JP, Record RD, Liang HA, Badylak
SF. Vascular endothelial growth factor in porcine-derived extracellular
matrix. Endothelium.
2001;81:
11-24.8111
2001
McPherson TB, Badylak SF.
Characterization of fibronectin derived from porcine small intestinal
submucosa. Tissue Eng.
1998;4:
75-83.475
1998
[CrossRef]
Voytik-Harbin SL, Brightman AO, Kraine
MR, Waisner B, Badylak SF. Identification of extractable growth factors from
small intestinal submucosa. J Cell Biochem.
1997;674:
478-91.674478
1997
Jimenez ML, Brown TD, Brand RA. The
effects of grip proximity on perceived local in vitro tendon strain. J
Biomech. 1989;22:
949-55.22949
1989
[CrossRef]
Stegemann H, Stalder K. Determination of
hydroxyproline. Clin Chim Acta.
1967;18:
267-73.18267
1967
[PubMed][CrossRef]
Reddy GK, Enwemeka CS. A simplified
method for the analysis of hydroxyproline in biological tissues. Clin
Biochem. 1996;293:
225-9.293225
1996
[CrossRef]
Baker AR, Abreu EL, Mascha E, Derwin KA.
Homotypic variation of canine flexor tendons: implications for the design of
experimental studies in animal models. J Biomech.
2004;37: 959-68.
Erratum in: J Biomech. 2004;37:1955.37959
2004
[PubMed][CrossRef]
Calabro A, Hascall VC, Midura RJ.
Adaptation of FACE methodology for microanalysis of total hyaluronan and
chondroitin sulfate composition from cartilage. Glycobiology.
2000;103:
283-93.103283
2000
[CrossRef]
Calabro A, Benavides M, Tammi M, Hascall
VC, Midura RJ. Microanalysis of enzyme digests of hyaluronan and
chondroitin/dermatan sulfate by fluorophore-assisted carbohydrate
electophoresis FACE. Glycobiology. 2000;
103: 273-81.103273
2000
[CrossRef]
Silver FH, Freeman JW, DeVore D.
Viscoelastic properties of human skin and processed dermis. Skin Res
Technol. 2001;71:
18-23.7118
2001
[CrossRef]
Raghavan D, Kropp BP, Lin HK, Zhang Y,
Cowan R, Madihally SV. Physical characteristics of small intestinal submucosa
scaffolds are location-dependent. J Biomed Mater Res A.
2005;731:
90-6.73190
2005
Freytes DO, Badylak SF, Webster TJ,
Geddes LA, Rundell AE. Biaxial strength of multilaminated extracellular matrix
scaffolds. Biomaterials.
2004;25:
2353-61.252353
2004
[PubMed][CrossRef]
Roeder R, Wolfe J, Lianakis N, Hinson T,
Geddes LA, Obermiller J. Compliance, elastic modulus, and burst pressure of
small-intestine submucosa (SIS), small-diameter vascular grafts. J
Biomed Mater Res. 1999;47:
65-70.4765
1999
[CrossRef]
Gloeckner DC, Sacks MS, Billiar KL,
Bachrach N. Mechanical evaluation and design of a multilayered collagenous
repair biomaterial. J Biomed Mater Res.
2000;52:
365-73.52365
2000
[PubMed][CrossRef]
Halder A, Zobitz ME, Schultz F, An KN.
Mechanical properties of the posterior rotator cuff. Clin Biomech
(Bristol, Avon). 2000;15:
456-62.15456
2000
[CrossRef]
Daly CH. Biomechanical properties of
dermis. J Invest Dermatol.
1982;79Suppl 1:
17s-20s.7917s
1982
[PubMed][CrossRef]
Poulsen JH, Cramers MK. Determination of
hyaluronic acid, dermatan sulphate, heparan sulphate and chondroitin 4/6
sulphate in human dermis, and a material of reference. Scand J Clin Lab
Invest. 1982;42:
545-9.42545
1982
Cetta G, Tenni R, Zanaboni G, De Luca G,
Ippolito E, De Martino C, Castellani AA. Biochemical and morphological
modifications in rabbit Achilles tendon during maturation and ageing.
Biochem J. 1982;204:
61-7.20461
1982
[PubMed]
Breen M, Weinstein HG, Andersen M, Veis
A. Microanalysis and characterization of acidic glycosaminoglycans in human
tissues. Anal Biochem. 1970;
35: 146-59.35146
1970
[PubMed][CrossRef]
Zheng MH, Chen J, Kirilak Y, Willers C,
Xu J, Wood D. Porcine small intestine submucosa (SIS) is not an acellular
collagenous matrix and contains porcine DNA: possible implications in human
implantation. J Biomed Mater Res B Appl Biomater.
2005;73:
61-7.7361
2005
[PubMed]
Choe JM, Bell T. Genetic material is
present in cadaveric dermis and cadaveric fascia lata. J Urol.
2001;166:
122-4.166122
2001
[PubMed][CrossRef]
Ketchedjian A, Jones AL, Krueger P,
Robinson E, Crouch K, Wolfinbarger L Jr, Hopkins R. Recellularization of
decellularized allograft scaffolds in ovine great vessel reconstructions.
Ann Thorac Surg. 2005;79:
888-96.79888
2005
[PubMed][CrossRef]
Record RD, Hillegonds D, Simmons C,
Tullius R, Rickey FA, Elmore D, Badylak SF. In vivo degradation of 14C-labeled
small intestinal submucosa (SIS) when used for urinary bladder repair.
J Biomaterials. 2001;22:
2653-9.222653
2001
[CrossRef]