597

COPYRIGHT © 2013 BY THE JOURNAL OF BONE AND JOINT SURGERY, INCORPORATED

The Mechanism of Action of Induced

Membranes in Bone Repair
Olli-Matti Aho, BM, Petri Lehenkari, MD, PhD, Jukka Ristiniemi, MD, PhD, Siri Lehtonen, PhD,
Juha Risteli, MD, PhD, and Hannu-Ville Leskelä, MD, PhD

Investigation performed at the University ofOulu, Oulu, Finland

Background: Inducement of foreign-body granulation tissue is a relatively novel therapeutic modality in bone repair. A
two-stage bone reconstruction method, known as the Masquelet technique, combines inducement of a granulation tissue
membrane and subsequent bone autografting as a biphasic technique allowing reconstruction of large bone defects. In
light of their already well-characterized osteogenesis-improving capabilities in animals, we performed this translational
study to investigate these membranes in patients.

iVIethods: Fourteen patients with complicated fractures and bone defects were randomly selected for this study. Biopsy
samples of foreign-body-induced membranes were collected at different time points during scheduled surgical proce-
dures. The membranes were co-cultured with mesenchymal stromal cells, and differentiation into the osteoblastic lineage
was assessed by measuring alkaline phosphatase activity, aminoterminal propeptide of type-l procollagen (PINP) pro-
duction, and Ca2+ concentration. Histological characteristics were evaluated with image analysis. Quantitative reverse
transcription polymerase chain reaction was used to measure vascular endothelial growth factor (VEGF), interleukin-6
(IL-6), and type-l collagen (Col-1) expression.

Resuits: The induced membranes were characterized histologically by maturating vascularized fibrous tissue. The
vascularization was greatest in one-month-old samples and decreased to <60% in three-month-old samples. One-month-
old membrane samples had the highest expression of VEGF, IL-6, and Col-1, whereas two-month-old membranes ex-
pressed <40% of the levels of the one-month-old membranes. Specific alkaline phosphatase activity, PINP production, and
Ca2+ concentration were increased in co-cultures when a membrane sample was present. In cultures of one-month-old
membranes, PINP production was more than two times and Ca^^ deposition was four times higher than that in cultures
of two-month-old membranes.

Conciusions: The induced membranes have osteogenesis-improving capabilities. These capabilities, however, appear
to decrease over time. We speculate that the optimal time for performing second-stage surgery may be within a month
after implantation of foreign material.

Ciinicai Reievance: The two-stage bone reconstruction method is an easily adaptable technique for repair of large bone
defects. Further understanding of membrane bioiogy and maturation in humans can help to optimize current procedures
and improve their overall success rate.

T he treatment of large bone defects is challenging, and

there is no consensus on the optimal treatment method.
Limited defects can usually be treated with traditional
bone autografts, whereas defects exceeding 4 to 5 cm are more
likely to have complications''. Vascularized bone transfer" and
a method based on the progressive migration of a bone seg-
ment' are two of the more widely used techniques in the
treatment of large bone defects. A more novel method, devel-
oped by Masquelet et al., utilizes the body's natural foreign-
body response and regenerative capabilities\

Disclosure: One or more of the authors received payments or services, either directly or indirectly (i.e., via his or her institution) trom a third party in
support of an aspect of this work. In addition, one or more of the authors, or his or her institution, has had a financial relationship, in the thirty-six months
prior to submission of this work, with an entity in the biomédical arena that could be perceived to influence or have the potential to influence what is written
in this work. No author has had any other relationships, or has engaged in any other activities, that could be perceived to influence or have the potential to
influence what is wntten in this work. The complete Disclosures of Potential Conflicts of Interest submitted by authors are always provided with the
online version of the article.

J Bone Joint Surg Am. 2013;95:597-604 • http://dx.doi.org/lO.21O6/JBJS.LOO31O


T H E JOURNAL OF BONE & JOINT SURGERY - JBJS.ORG
VOLUME 95-A - NUMBER 7 - APRIL 3,
The formation of granulation tissue is an essential part
of soft-tissue healing and can also be observed during the
foreign-body response". The foreign-body response is ini-
tiated by an injury to vascularized soft tissue caused by for-
eign material. The injury always initiates a phase of acute
inflammation, which can lead to tissue regeneration and
maturation of the granulation tissue into dense fibrosis,
bone, or other mesenchymal tissue. In unfavorable condi-
tions, chronic inflammation can develop and can influence
the tissue maturation process. The extent, quality, and du-
ration of the inflammatory reaction can be modified with
optimization of the tissue environment by debridement of
all dead tissue and infection, by suitable selection of bio-
materials^ and by sufficient vascularization. Both the acute
and the transient inflammatory processes in the formation of
granulation tissue are characterized by a large presence of
macrophages and infiltration of macrophages and fibroblasts.
Capillary formation and neovascularization are also key fea-
tures of the maturating granulation tissue. Cranulation, orig-
inally formed from the hematoma and loose extracellular
matrix components, is organized into a form of a tissue pre-
cursor, which in time is transformed into a neovascularized
connective-tissue membrane. Eventually, the granulation tissue
encapsulates the foreign material and later a mature fibrous
capsule is formed at the end stage of the inflammatory
response'''^
This foreign-body-induced granulation tissue is a useful
therapeutic reagent for reconstructive surgery and has been
applied to the reconstruction of large bone defects'*". In ad-
dition, inducing the granulation tissue membrane may, in
procedures such as nerve repair, help to prevent the impractical
formation of the nonfunctional or fibrotic granulation tis-
sue'''''^ When the membrane is induced over a specific spacer,
the surgeon can create space for nerve growth and prevent the
more rapid process of fibrosis that usually fills the space and
prevents nerve tissue regeneration. The use of induced gran-
ulation tissue membranes in bone reconstruction is based
on a two-stage-surgery technique'. In the first stage, an anti-
biotic cement spacer is inserted into the bone defect. In the
second stage, usually carried out one to two months later,
the spacer is replaced with cancellous bone autograft (Fig. 1).
The cement spacer induces the formation ofa fibrous capsule
Fig-l
Radiographs showing treatment of a large bone defect with the two-staged technique of bone-grafting. Fig. 1-A An antibiotic cement spacer stabiiized with
internal fixation after bone resection. Fig. 1-B Four weeks later, the spacer was removed and the granulation tissue cavity was filled with cancellous
autograft bone. Fig. 1-C Newly formed bone two months after bone-grafting.
598
T H E MECHANISM OF ACTION OF INDUCED MEMBRANES
2013 JN BONE REPAIR
through the classic foreign-body reaction and releases anti-
biotics locally, which contributes to eradication of a possible
infection. The capsule is later used to ( 1 ) confine the area of
the defect, (2) support the implanted autograft, and (3)
provide the graft with neocirculation. The overall time needed
for the bone defect to heal varies according to the defect
location and complicating clinical factors, such as vascularity.
For example, the time to healing of tibial fractures has ranged
from twenty to seventy-nine weeks'".The extent of the bone
defect, however, does not seem to affect the time needed
for healing, although the availability of autograft bone
does restrict the usage of the two-stage method for larger
defects'''.
The induced granulation tissue membrane has been
shown to have anabolic capabflities in animal models. The
membranes have rich vascularization and high production of
growth factors, and they may possess the ability to differentiate
co-cultured bone-marrow mesenchymal stromal cells into a
specific lineage". High production of osteoinductive factors has
also been described, and the authors of one study reported that
the highest expression of bone morphogenetic protein-2
(BMP-2) was found in membranes that were four weeks of
age'". One animal study demonstrated that the membranes
can house stem cells, which can be obtained by culturing and
propagating cells from the induced membranes of normal rats'.
These so-called granulation-tissue-derived stem cells expressed
the characteristics of both embryonic pluripotent cells and
adult stem cells. Another animal study identified core-binding
factor alpha 1 (CBFAl)-positive cells in the induced mem-
brane, which could indicate that the membrane can work as a
repository of osteoprogenitor cells". These findings have not
been confirmed in humans.
As the induced-membrane technique is being used more
widely and more often, it is essential to obtain information on
this process in humans, in whom the healing process differs
from that in smaller vertebrates in many ways, such as pro-
longed healing time and a different inflammation process. An
earlier study of rabbits suggested that the induced membrane
reaches the highest biochemical activity in terms of bone re-
generation at four weeks after spacer implantation". This
would suggest, if confirmed in human tissues, that the second
stage of the two-stage-surgery technique should be carried out
 599
THE JOURNAL OF BONE & JOINT SURGERY - JBJS.ORG THE MECHANISM OF ACTION OF INDUCED MEMBRANES
VOLUME 95-A -NUMBER 7 -APRIL 3, 2013 IN BONE REPAIR

TABLE 1 Patient Characteristics

Length of Bone Duration Antibiotic

Patient Age (yr) Sex Location of Bone Defect Defect [mm) Spacer in Place (mo)
22 iViale Distal tibial 45 7.5
10 Male Distal femoral 40 1
77 - Female Proximal tibial 22 1.5
65 Male Distal tibial 0 22
59 Maie Distal tibial 30 1.5
26 Male Distal tibial 29 2
27 Male Medial tibial 23 1
58 Male Medial tibial 47 3
43 Male Distal femoral 31 1
65 Fennale Distal tibial 63 2
85 Female Distai femoral 32 1
72 Female Proximal femoral 0 33
38 Male Distai tibial 43 2
17 Male Distal femoral 39 5
and proximal tibial

earlier than previously thought. The aim of the current ob-
servational study was to assess biochemical and histological
data of human induced membranes collected during two-stage
surgery. We also conducted a series of in vitro cell-culture ex-
periments using mesenchymal stromal cells and samples of
membranes to assess the effect of the induced membrane on
mesenchymal stromal cell differentiation into bone cells.
Materiais and iVIethods
mesenchymal stromal cells were maintained in culture as previously de-
scribed . Briefly, mesenchymal stromal cells from the bone marrow samples
were allowed to attach to the culture flasks. After two days, unattached cells
were washed away and the cells on theflaskswere cultured for one to two weeks
until near-confluence. Mesenchymal stromal cells were cultured in a-MEM
(minimum essential medium; GIBCO, Paisley, United Kingdom), supple-
mented with 10% heat-inactivated FBS (fetal bovine serum) (Autogen Bioclear,
WUtshire, United Kingdom), 20 mM HEPES (GIBCO), 100 U/mL of penicillin,
0.1 mg/mL of streptomycin (GIBCO), and 2 mM L-g!utamine (GIBCO) at
37°C in 5% CO2 and 95% air. Human mesenchymal stromal cells were de-

T he Ethical Committee of The Northern Ostrobothnia Hospital District

approved the study protocol for collection and use of human cells and
tissues.
Patients and Induced Membranes
In our hospital, patients with large segmental bone defects are treated with two-
stage surgery '". Fourteen patients were randomly selected for the study (Table
I). In addition, control membrane samples were collected from the surface of
osteosynthesis material from three patients. At the second stage of the surgery,
antibiotic cement beads were removed and autogenous cancellous bone graft
was applied to the bone defect. After removal of the spacer, the ends of the bone
were debrided, but the other parts of the foreign-body membrane around the
beads were preserved. For the study, the membrane was harvested from both
the bone ends and the body of the induced membranes. Samples of subcuta-
neous tissue were taken as controls for cell culture experiments. Immediately
after harvesting, the membranes were divided for the different treatments;
formalin was used for the histological analysis, liquid nitrogen was used for the
polymerase chain reaction analysis, and fresh membranes were washed in
phosphate-buffered saline solution (PBS) for the culture. Membrane pieces (4 X
4 mm) and subcutaneous tissue samples were cultured with mesenchymal
stromal cells as described below.
Human Mesenchymal Stromal Cell Cultures
Human bone marrow was harvested from patients who were operated on for
hip osteoarthrosis and who were randomly selected for this study. Bone marrow
was obtained from the proximal femoral diaphysis during the operation, and
tached with use of trypsin-EDTA solution (GIBCO) digestion. Cells were then
counted and seeded at a density of 5 X 10' with a membrane sample (4x4 mm)
in twenty-four-well plates. Cells used in this study were from the first or the
second passage, and the serum batch in all cell cultures was the same. The
culture media were changed twice a week. After the membranes were removed
from the culture, osteoblastic differentiation was evaluated on the basis of the
activity of alkaline phosphatase in cell lysates after three weeks of culture. Bone
formation was evaluated after three weeks by assessing the production of
aminoterminal propeptide of type-I procoUagen (PINP) and quantifying cal-
cium deposition after flve weeks. The separately used osteosynthesis-inducing
medium contained the above described medium with 0.1 |xM dexamethasone,
10 mM sodium ß-glycerophosphate, and 0.05 mM ascorbic acid-2-phosphate
(Sigma-Aldrich, St. Louis, Missouri).
Specific Alkaline Phosphatase Activity
Specific alkaline phosphatase activity was determined from mesenchymal
stromal cells after a three-week culture period with or without membranes.
Samples were extracted into an assay buffer containing 50 mM Tris-HCl, 0.1 %
Triton X-100, and 0.9% NaCl (pH 7.6), and lysate was frozen. Lysate samples
were then thawed, and enzyme activity was determined in duplicates with use of
0.1 M 4-p-nitrophenyI phosphate (Sigma-Aldrich) as a substrate. Absorbance
was read at 405 nm in a plate reader (Victor^ PerkinElmer Life Sciences/Wallac
Oy, Turku, Finland). The total protein content of the cultured mesenchymal
stromal cells was determined by BIO-RAD Protein Assay (Bio-Rad Laborato-
ries, Hercules, California). The enzyme activities were normalized with protein
concentration and were expressed as absorbance per muligram of protein.
 6oo
THE JOURNAL OF BONE & JOINT SURGERY • JBIS.ORG THE MECHANISM OF ACTION OF INDUCED MEMBRANES
VOLUME 95-A -NUMBER 7 -APRIL 3, 2013 IN BONE REPAIR

PINP Masson Trichrome and Hematoxylin and Eosin Staining

The amount of type-I procoJlagen syntJiesis was quantified from tJie culture me-
dium. TJie œncentration of PINP was measured in duplicates witJi use of a com-
merciaUy available radioimmunoassay (Orion Diagnostica, Espoo, Finland). TJie
assay was done in sequential saturation, and we used duplicate 100-(JLL aliquots of
1:20 or 1:50 diluted culture medium. The intra-assay and interassay coefficients of
variation of PINP in samples were 4.5% to 10.3% and 3.1% to 10.8%, respectively.
Calcium Quantification
For calcium quantification, the cultures were stopped at five weeks and the co-
ctiltured membranes were removed. The cells and matrix on the bottom of the wells
were washed three times with Ca^+ and Mg^^-fi-ee PBS and incubated overnight at
room temperature in 0.6 M HCl to dissolve the deposited calcium. Calcium deter-
mination was based on the reaction of calcium with o-cresolphthalein complexone
according to the manufacturer's instructions (Roche Diagnostics, Mannheim, Ger-
many). The colorimetric reaction was read at 570 nm in a plate reader (Victor ).
and Image Analysis of Vascularity
Membrane samples were rinsed in PBS and were formalin-fixed, paraffin-
embedded, and sectioned. Sections of 8 |xm were stained with the Masson tri-
chrome method. Some membrane samples were also stained with hematoxylin
and eosin for imaging purposes. Subsequent quantification of the proportional
vascularity area was performed with a digital image analysis system (MCID-
M5-1-; Imaging Research, Canada). Proportional vascularity was quantified by
selectingfiverepresentative similarly sized areas of each membrane and counting
the average proportional area of vascular tissue in contrast to membrane stroma.
Extraction of RNA and Quantitative Reverse Transcription
Polymerase Chain Reaction (RT-qPCR) Analysis
Pieces of membranes were immediately frozen after collection and were stored
in liquid nitrogen until they were used for RNA analysis. Two hundred to 600 mg
of frozen membrane was powdered with a pestle and homogenized into

Fig. 2
Representative histological sections of induced membranes stained with Masson trichrome and hematoxyiin and eosin (original magnification xlOO). Fig.
2-A Granulation tissue from a one-month-old membrane. Some foreign-body giant cells (arrow) can be seen. Fig. 2-B Newly formed vessel tissue (arrows)
from a 1.5-month-old membrane. Fig. 2-C Mature fibrous tissue of a twenty-two-month-old membrane. Only a few blood vessels (arrows) are seen. Fig. 2-D
Endochondral ossification takes place in induced membranes (arrows indicate cartilage). Fig. 2-E Newly formed bone tissue can be seen (arrow indicates
lamellar bone). Fig. 2-F Foreign-body giant cell (arrow) in a one-month-old membrane stained with hematoxylin and eosin.
 6oi
T H E JOURNAL OF BONE & JOINT SURGERY - JBKS.ORG T H E MECHANISM OF ACTION OF INDUCED MEMBRANES
VOLUME 9 5 - A - N U M B E R 7 - A P R I L 3 , 2 0 1 3 IN BONE REPAIR

TRIzol Reagent (Invitrogen). RNA was isolated according to the manufacturer's
instructions, and the isolated RNA was further purified by the RNeasy MinElute
Cleanup Kit (Qiagen, Valencia, California), after which the amount of RNA was
measured spectrophotometrically. One microgram of RNA was used for cDNA
synthesis with the RevertAid First Strand Synthesis Kit (Fermentas, subsidiary
of Thermo Fisher Scientific, Waltham, Massachusetts) by using oligo-dT as
a primer. One microliter of the cDNA product was used for RT-qPCR with
the SYBER Green technique (iQ SYBR Green; Bio-Rad Laboratories) by
quantifying the product at each cycle in a real-time manner. The amount of
the product was monitored for forty cycles, after which the specificity of the
product was confirmed by melting curve and in some cases by visualizing the
products with agarose gel electrophoresis.
The expression of gIyceraldehyde-3-phosphate dehydrogenase (GAPDH)
was studied by using primer sequences 5'-GAGTCAACGGATTTGGTCGT-3' and
5'-GACAAGCTTCCCGTTCTCAG-3' with 58°C as an annealing temperature.
Primer sequences were 5'-GGTTT CGACT TCAGCTTCCT G-3' and 5'-TC-
ACCAGTCTGCATGTTGCA-G-3' for type-I collagen (Col-1), 5'-GGGCAGA-
ATCATCACGAAGTG-3' and 5'-ATTGGATGGCAGTAGCTGCG-3' for vascular
endothelial growth factor (VEGF), and 5'-GGTACATCCTCGACGGCATCT-3'
and 5'-GTGCCTCTTTGCTGCTTTCAC-3' for interIeukin-6 (IL-6), and they
were analyzed with use of 66°C as an annealing temperature. The expression of
each compound was related to GAPDFl expression by using the AACt method" by
considering GAPDH as a housekeeping gene and comparing the expression with
one selected control sample.
Statistical Analysis
Statistical analysis was performed with the Statistical Package for the Social Sci-
ences (SPSS) version 13.0 (Chicago, Illinois). The Student t test was used to
compare two means. The multiple-comparison data were studied with one-way
variance analysis followed by post hoc comparisons (two-tailed t test). Graphs
were created with Origin 6.0 software (OriginLab, Northampton, Massachusetts).
Source of Funding
of the vascularity of the one-month-old membranes (Figs. 3-A,
3-C, and 3-D). Also, RT-qPCR showed significantly decreased
production of VEGF, IL-6, and Col-1 after one month: two-
month-old membranes expressed <40% of the levels of one-
month-old membranes.
Influence of Induced Membrane on Cultured Human
Mesenchymal Stromal Cells
In this study, we used a previously described'" model of
mesenchymal stromal cell differentiation into osteoblasts
and measurement of their activity as bone-forming cells.
Human mesenchymal stromal cells can be induced in vitro
to differentiate into mature osteoblasts by glucocorticoids,
ß-glycerophosphate, and ascorbate. For the study of the
osteoinducement of the induced membranes, human mem-
branes and mesenchymal stromal cells were cultured together
without these osteoinducing reagents. At the age of four
weeks, harvested membranes significantly induced both
specific alkaline phosphatase activity and calcium concen-
tration in mesenchymal stromal cell cuhures. In osteoin-
duced cultures, both specific alkaline phosphatase activity
and calcium concentration of mesenchymal stromal cells
were induced significantly (Figs. 4-A and 4-B).
A _ B
120i 'or
100
i
1
5140
3 ••
A
IL6
VEGF
Con

The Finnish Medical Foundation and Northern Ostrobothnia Hospital District

funding was used for laboratory costs and salary. In addition, funding for the
80
60
11

•
12
o
ËIOO
80-
\
PCR

salary of the corresponding author (O.-M.A.) was received from the National
Doctoral Programme of Musculoskeletal Disorders and Biomaterials (TBDP)
40
20
•
1^ • J1
graduate school. A personal grant for the corresponding author was granted by 1 20
0 o 0 t
The Finnish Medical Foundation. None of the study sponsors had any role in
1 1.5 3 5 22 - 1 15 2 7.5 22
data collection or analysis, in manuscript writing, or manuscript publishing. Membrane age (months) Membrane age (months)

Results
Induced Membrane Consists of Maturating Fibrous Tissue
and Differentiates into Bone Tissue

M asson trichrome and hematoxylin and eosin-stained in-

duced membranes showed some foreign-body reaction
with macrophages and foreign-body giant cells at cement
spacer interfaces. Induced membrane includes newly formed
primitive vascular tissue and has the capability to differentiate 1 Month 5 Months
toward calcified tissue by endochondral ossification. We could Fig. 3

also see newly formed bone tissue, even mature lamellar bone,
in induced membranes. Whereas granulation tissue was char-
acteristic for younger membranes, more uniform fibrous tissue
dominated the histological pattern of older samples (Fig. 2).
Vascularity of Induced Membrane Decreases During Aging
Histological analysis of the membranes showed a clear differ-
ence between early and older membranes. At four weeks,
membranes appeared richly vascularized by numerous small
capillaries and larger veins. Over time, the vascularity rate
decreased. Already at three months, the membranes had <60%
The effect of induced-membrane age on membrane vascularity and se-
creted growth factors. Fig. 3-A Proportional area of vascularity in harvested
membranes of different ages. The results were normalized against the one-
month values for clarity. Fig. 3-B mRNA was investigated with the quan-
titative reverse transcription polymerase chain reaction (PCR) method.
Vascular endothelial growth factor (VEGF), interleukin-6 (1L6), and type-I
collagen (Coll) results were determined for membranes of different ages,
and the values were normalized against the average of the one-month
samples. Fig. 3-C A one-month membrane. Fig. 3-D A five-month mem-
brane. The specimens were stained with Masson trichrome. Arrows indi-
cate blood vessels.

T H E JOURNAL OF BONE & JOINT SURGERY - IBJS.ORG
VOLUME 95-A - N U M B E R 7 - A P R I L 3,
Muscle BM CB Muscle BM
1 month 2 months 1 month 2 months
Membrane age Membrane age
Fig. 4
Figs. 4-A and 4-B Influences of induced membrane on bone formation of
cultured human mesenchymal stromal cells (MSCs). Cells were cultured
for three weeks to determine specific alkaline phosphatase (ALP) activity
(Fig. 4-A) and for five weeks tor calcium quantification assays (Fig. 4-B).
Figs. 4-C and 4-D Influence of the harvesting sites (muscle, bone marrow
[BM], and cancellous bone [CB]) of the induced membranes on cultured
human mesenchymal stromal cells. Mesenchymal stromal cells and
membranes were cultured without osteosynthesis-inducing medium (OS),
as described in Materials and Methods. Mesenchymal stromal cells were
cultured for three weeks for measurement of aminoterminal propeptide of
type-l procollagen (PINP) production (Fig. 4-C) and for five weeks for cal-
cium quantification (Fig. 4-D). Figs. 4-E and 4-FThe effect of induced-
membrane age on mesenchymal stromal cells. Mesenchymal stromal cells
and membranes were cultured without osteosynthesis-inducing medium,
as described in Materials and Methods. PINP production (Fig. 4-E) was
measured and calcium quantification (Fig. 4-F) was performed. The exis-
tence of mesenchymal stromal cells, subcutaneous control tissue (Sub-
cutis), membrane, and/or osteosynthesis-induced medium in the cell
culture assay are indicated with (-f ) or (-). Each bar represents the mean
and standard deviation (I bar). *p<0.05, * * p < 0.01, and * * * p < 0.001.
6O2
T H E MECHANISM OF ACTION OF INDUCED MEMBRANES
2013 IN BONE REPAIR
Influence of Harvesting Site of Induced Membranes on
Cultured Human Mesenchymal Stromal Cells
At the second stage of treatment, the harvested membranes
were collected at the different areas of the operative site. The
memhranes from the sites of hone marrow and cancellous hone
were significantly more osteoinductive than the memhrane
at the site of muscle as measured hy PINP production. Bone
formation measured hy calcium content did not change sig-
nificantly hetween memhranes at different sites (Figs. 4-C and
4-D).
Effect of Time of Second Stage of Surgery on Induced
Membrane and Co-Cultured Human Mesenchymal
Stromal Cells
We studied whether the time point of the second stage of the
operation influenced the activity and osteoinducement of
memhrane in human mesenchymal stromal cell cultures. We
harvested memhranes after one month and two months and
co-cultured them with mesenchymal stromal cells. Interest-
ingly, when memhranes were harvested at one month after a
cement spacer was introduced into the hone defect, the mem-
CB
hranes spontaneously induced mesenchymal stromal cell dif-
ferentiation and activation as demonstrated hy measurement of
PINP synthesis and calcium quantification. In contrast, mem-
hranes harvested after two months did not increase mesenchy-
mal stromal cell differentiation and activation. PINP levels were
more than two times and Ca^* deposition was four times higher
in one-month-old memhranes (Figs. 4-E and 4-F).
Discussion
T he treatment of large hone defects in general remains an
issue of dehate. The two-stage hone reconstruction tech-
nique is a novel method that has heen proven to he successftil
for large-defect reconstruction"'''"'^'^". In this study, we inves-
tigated the histological and hiochemical characteristics of the
induced memhranes utilized in this technique.
Earlier studies have shown that foreign-hody-induced
memhranes contain well-vascularized fihrous tissue that is
mostly preserved during follow-up"*'"''^'. In contrast, our re-
sults demonstrated that, in human memhranes, the propor-
tional area of vascular tissue clearly decreases after the spacer
has heen in place for two months. By two months, the mem-
hrane is composed of more mature fihrous tissue with ahun-
dant collagen and fewer hlood vessels. Three to six months after
spacer implantation, the memhrane matures to its final com-
position in terms of extracellular matrix density.
We ohserved some intramemhranous ossification and
even cartilage formation in some specimens. As far as we
know, this has not heen previously descrihed, hut foreign-
hody reaction in humans appears to produce immature
multipotent mucofihrous tissue that even has some similari-
ties to emhryogenic mesodermal tissue. In previous studies,
this memhrane has heen descrihed as heing capahle of pro-
ducing and containing undifferentiated mesenchymal stromal
cells'. We did not expect this finding and did not attempt to
collect and isolate multipotent mesenchymal stromal cells
 6 03
THE JOURNAL OF BONE & JOINT SURGERY -JBJS.ORG
VOLUME 95-A - NUMBER 7 - APRIL 3, 2013
from the membranes, but our observation encourages doing so
in the future.
The histological changes that were seen in the mem-
branes matched well the changes in growth-factor expression.
Expression of VECF decreases drastically in membranes after
one month. This finding is in accordance with the observed
decrease in the vascularity of the aging membranes. A more
steadily decreasing pattern of expression of VEGF was found in
rabbits by Pelissier et al.'". To obtain a deeper insight into the
early stages of foreign-body-induced membrane development
and the first stages of maturation, we measured type-I collagen
and IL-6 in membranes of different ages. Expression patterns
were similarly highest in the earlier samples and very small in
the membranes that were more than two months old. This
transition is probably caused by the general dampening of local
inflammation and the maturation process in the induced
membrane.
Foreign-body-induced membranes of animals directly af-
fect stem cells and cause them to be susceptible to difterentiation
into the osteoblastic lineage in vitro^'"'. The co-cultures in our
study indicate that this effect is present in human tissue also. We
used quantitative measurements of Ca^"^ and PINP to assess
in vitro stem-cell diflsrentiation into the osteoblastic lineage and
obtained conclusive results from both of these compounds: os-
teoblastic differentiation was greatly increased when a membrane
sample was present in the culture. The effect was considerably
greater when an osteogenesis-increasing medium was used.
In co-cultures with one and two-month-old membranes, the
membrane's age greatly affected the rate of stem-cell differenti-
ation. The two-month-old membranes had only a fraction of the
activity seen in the one-month-old samples.
When membrane samples were collected, we paid special
attention to the location of the membrane in relation to the
surrounding tissue to determine the effects on the membrane's
development and characteristics. We did not find conclusive
differences with histological analysis. However, our in vitro co-
cultures with mesenchymal stromal cells suggested that mem-
branes that develop adjacent to bone marrow possess the greatest
ability to stimulate osteogenesis. Clinical correlation for this
remains nonexistent and is difficult to study conclusively. It
seems that the surrounding tissue induces membrane that has
the capability to direct the repair process to favor formation of
the surrounding mesenchymal tissue—bone in this case.
Fractures leading to bone defects, which require more
complex means of treatment, are rare. In a prospective study
done in Scotland, these cases accounted for only 0.4% of all
patients admitted to the hospital because of a fracture^'.
References
1. Han CS, Wood MB, Bishop AT, Cooney WP 3rd. Vascularized bone transfer.
J Bone Joint Surg Am. 1992;74(10):1441-9.
2. May JW Jr, Jupiter JB, Weiiand AJ, Byrd HS. Clinicai classification of post-traumatic
tibial osteomyeiitis. J Bone Joint Surg Am. 1989:71(9):1422-8.
3. MasqueietAC, ntoussi F, BegueT, MuiierGP. [Reconstruction of the iong bones by the
induced membrane ahd spongy autograft]. Ann Chir Piast Esthet. 2000;45{3):346-53.
4. Weiland AJ. Current concepts review: vascularized free bone transplants. J Bone
Joint Surg Am. 1981;63(l):166-9.
THE MECHANISM OF ACTION OF INDUCED MEMBRANES
IN BONE REPAIR
Therefore, our sample collection was limited to a very select
group of patients. The tissue sampling size was also limited to
avoid compromising the treatment. This impaired our ability
to achieve a truly representative tissue-sample series at all de-
sired treatment time points. The limitations in sample material
and the tissue structure itself caused another problem in our
study, especially when the RT-qPCR was used, and hence we
were not able to measure all of the variables that we desired. For
a more comprehensive study, much more tissue sample ma-
terial is needed. Despite these shortcomings, we believe that
there was sufficient material for us to reach the conclusions
presented here. This idea is supported by the fact that most of
our observations are backed up by earlier experimental
k"'"'^'
Our results indicate that the biological activity of mem-
branes is a powerful reagent that should not be disregarded in
bone surgery. Ourfindingssuggest that the membrane has a local
osteoinductive effect and provides the bone defect area with rich
vascularization and neocirculation. The membrane's abilities do,
however, diminish rather quickly as it reaches a more mature
form. We can conclude that foreign-body-induced membranes
may reach their highest biological activity before one month has
passed after foreign-body implantation. •
NOTE: The authors thahk Minna Savilampi and Katja Koukkula for their expert technicai assistance.
OUi-Matti Aho, BM
Petri Lehenkari, MD, PhD
Siri Lehtonen, PhD
Juha Risteli, MD, PhD
Hannu-Ville Leskelä, MD, PhD
Clinical Research Center (O.-M.A., RL., S.L., and H.-V. L),
Department of Clinical Chemistry, Institute of Diagnostics (J.R.),
Division of Anatomy and Cell Biology,
Department of Biomedicine (P.L, S.L., and H.-V. L),
RO. Box 8000, 90014,
University of Oulu, Oulu, Finland.
E-mail address for O.-M. Aho: olli-matti.aho@oulu.fi.
E-mail address for P. Lehenkari: Petri.Lehenkari@oulu.fi.
E-mail address for S. Lehtonen: Siri.Lehtonen@oulu.fi.
E-mail address for J. Risteli: Itiha.Risteli@oulu.fi.
E-mail address for H.-V. Leskelä: Hannu-Ville.Leskela@ppshp.fi
Jukka Ristiniemi, MD, PhD
Division of Orthopedic and Trauma Surgery,
Department of Surgery, P.O. Box 21, 90029,
Oulu University Hospital, Oulu, Finland.
E-mail address: Jukka.Ristiniemi@ppshp.ñ
5. Cattaneo R, Catagni M, Johnson EE. The treatment of infected nonunions and
segmentai defects of the tibia by the methods of Ilizarov. Clin Orthop Relat Res.
1992 Jul;(280)(280):143-52.
6. Anderson JM, Rodriguez A, Chahg DT. Foreign body reaction to biomateriais.
Semin immunoi. 2008;20(2):86-100.
7. Patei J, Gudehithiu KP, Dunea G, Arruda JAL, Singh AK. Foreign body-induced
granulation tissue is a source of adult stem cells. Transi Res. 2010;155(4):
191-9.
 THE JOURNAL OF BONE & JOINT SURGERY -JBJS.ORG
VOLUME 9 5 - A - N U M B E R 7 - A P R I L 3, 2013
8. Chong K, Woon CY, Wong M. Induced membranes—a staged technique of bone-
grafting for segmental bone loss: surgical technique. J Bone Joint Surg Am. 2011;
93(Suppl 1):85-91.
9. Gallucci G, Donndorff A, Boretto J, Constantini J, De Carii P. Infected nonunion of
the humérus treated with an antibiotic cement rod. Case report. Chir Main. 2007;
26(4-5)1242-6.
10. Masquelet AC. Muscle reconstruction in reconstructive surgery; soft tissue re-
pair and long bone reconstruction. Langenbecks Arch Surg. 2003;388(5);344-6.
11. Ristiniemi J, Fiinkkilä T, Pakarinen H, Jalovaara P. Articular bone loss treated
with a staged method using bone cement or beads as a spacer and subsequent
autografting; a reportof two cases. Arch Orthop Trauma Surg. 2009;129(l);79-82.
12. Ristiniemi J, Lakovaara M, Flinkkilâ T, Jalovaara P. Staged method using anti-
biotic beads and subsequent autografting for large traumatic tibial bone loss; 22
of 23 fractures healed after 5-20 months. Acta Orthop. 2007;78(4);520-7.
13. Woon CY, Chong KW, Wong MK. Induced membranes—a staged technique of
bone-grafting for segmental bone loss; a report of two cases and a literature review.
J Bone Joint Surg Am. 2010;92(l);196-201.
14. OzalpT, Masquelet AC. [The role of creating a biological membrane in expediting
nerve regeneration for peripheral nerve repairs]. Acta Orthop Traumatol Turc. 2008;
42(2);130-4. Turkish.
15. Mackinnon SE, Dellon AL, Hudson AR, Hunter DA. Nerve regeneration through
a pseudosynovial sheath in a primate model. Piast Reconstr Surg. 1985;75(6);
833-41.
THE MECHANISM OF ACTION OF INDUCED MEMBRANES
IN BONE REPAIR
16. Pelissier P, Masqueiet AC, Bareille R, Pelissier SM, Amedee J. Induced mem-
branes secrete growth factors including vascular and osteoinductive factors and
could stimuiate bone regeneration. J Orthop Res. 2004;22(l);73-9.
17. Viateau V, Guillemin G, Calando Y, Logeart D, Oudina K, Sedel L, Hannouche D,
Bousson V, Petite H. Induction of a barrier membrane to facilitate reconstruction of
massive segmental diaphyseal bone defects; an ovine model. Vet Surg. 2006;35(5);
445-52.
18. Leskelä HV, Risteii J, Niskanen S, Koivunen J, Ivaska KK, Lehenkari P. Osteo-
biast recruitment from stem cells does not decrease by age at late adulthood.
Biochem Biophys Res Commun. 2003;311(4);1008-13.
19. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time
quantitative PCR and the 2(-Deita Deita C(T)) Method. Methods. 2001;25(4);402-8.
20. Apard T, Bigorre N, Cronier P, Duteille F, Bizot P, Massin P. Two-stage recon-
struction of posttraumatie segmental tibia bone loss with nailing. Orthop Traumatol
Surg Res 2010 Sep;96(5);549-53.
21. Viateau V, Guiilemin G, Bousson V, Oudina K, Hannouche D, Sedei L, Logeart-
Avramoglou D, Petite H. Long-bone critical-size defects treated with tissue-engineered
grafts; a study on sheep. J Orthop Res. 2007;25(6);741-9.
22. Keating J, Simpson A, Robinson C. The management of fractures with bone
loss. J Bone Joint Surg Br. Voi 2005 Feb;87B(2);142-50.
23. Jung RE, Kokovic V, Jurisic M, Yaman D, Subramani K, Weber FE. Guided bone
regeneration with a synthetic biodegradabie membrane; a comparative study in
dogs. Clin Oral Implants Res. 2011;22(8);802-7.
Copyright of Journal of Bone & Joint Surgery, American Volume is the property of Journal of Bone & Joint
Surgery, Inc. and its content may not be copied or emailed to multiple sites or posted to a listserv without the
copyright holder's express written permission. However, users may print, download, or email articles for
individual use.