Experimental posterolateral spinal fusion with
porous ceramics and mesenchymal stem cells
G. Cinotti, A. M. Patti, A. Vulcano, C. Della Rocca, G. Polveroni,
G. Giannicola, F. Postacchini
From The University La Sapienza, Rome, Italy
lternatives to autogenous bone graft for spinal
A fusion have been investigated for many years. It has
been shown that osteoconductive materials alone do not
give a rate of fusion which is comparable to that of
autogenous bone graft. We analysed the effectiveness of
porous ceramic loaded with cultured mesenchymal stem
cells as a new graft material for spinal fusion in an
animal model.
Posterolateral fusion was carried out at the L4/L5
level in 40 White New Zealand rabbits using one of the
following graft materials: porous ceramic granules plus
cultured mesenchymal stem cells (group I); ceramic
granules plus fresh autogenous bone marrow (group II);
ceramic granules alone (group III); and autogenous
bone graft (group IV). The animals were killed eight
weeks after surgery and the spines were evaluated
radiographically, by a manual palpation test and by
histological analysis.
The rate of fusion was significantly higher in group I
compared with group III and higher, but not
significantly, in group I compared with groups II and IV.
In group I histological analysis showed newly formed
bone in contact with the implanted granules and highly
cellular bone marrow between the newly formed
trabecular bone. In group II, thin trabeculae of newly
formed bone were present in the peripheral portion of
the fusion mass. In group III, there was a reduced
G. Cinotti, MD, Orthopaedic Surgeon
G. Giannicola, MD, Orthopaedic Surgeon
F. Postacchini, MD, Professor of Orthopaedics
Department of Orthopaedics and Traumatology
A. M Patti, MD, Biologist
A. Vulcano, Biologist
Department of Public Health Sciences
C. Della Rocca, MD, Associate Professor
Department of Experimental Medicine and Pathology
University La Sapienza, Piazzale Aldo Moro 5, Rome 00185, Italy.
G. Polveroni, Veterinary Surgeon
Veterinary Clinic, Via AM Ricci 109, 02100 Rieti, Italy.
Correspondence should be sent to Dr G. Cinotti at Piazza Acilia 4, 00199
Rome, Italy.
©2004 British Editorial Society of Bone and Joint Surgery
doi:10.1302/0301-620X.86B1.14308 $2.00
VOL. 86-B, No. 1, JANUARY 2004
amount of newly formed bone and abundant fibrous
tissue. In group IV, there were thin trabeculae of newly
formed bone close to the decorticated transverse
processes and dead trabecular bone in the central
portion of the fusion mass.
In vitro cultured mesenchymal stem cells may be
loaded into porous ceramic to make a graft material for
spinal fusion which appears to be more effective than
porous ceramic alone. Further studies are needed to
investigate the medium- to long-term results of this
procedure, its feasibility in the clinical setting and the
most appropriate carrier for mesenchymal stem cells.
J Bone Joint Surg [Br] 2004;86-B:135-42.
Received 7 March 2003; Accepted after revision 15 September 2003
Posterolateral fusion with autogenous bone graft is com-
monly used in patients with developmental or degenerative
spinal conditions which are unresponsive to conservative
therapy. It has been shown that the procedure is safe and
that, in properly selected patients, high rates of bony fusion
and successful clinical outcomes can be obtained.1-3 How-
ever, several limitations, including pseudarthrosis, pain at
the donor site, the need for blood transfusion and prolonged
operating times are known to be associated with the pro-
cedure.4-7
Alternatives to autogenous bone graft to achieve spinal
fusion have been investigated over many years.8-14 It has
been shown that bone morphogenetic proteins (BMPs) with
adequate carriers may achieve similar, or even higher, rates
of fusion compared with posterolateral fusion with auto-
genous bone graft.8,10,12,15 However, the cost of the dose of
BMPs which is required is high and there are possible
adverse effects of these growth factors.
Porous ceramic cylinders loaded with cultured mesen-
chymal stem cells have been tested as an alternative to
autogenous bone graft in critical-sized defects of long
bones.16 The results have suggested that healing of segmen-
tal bone defects was more likely to occur with porous
ceramic loaded with cultured mesenchymal stem cells than
with porous ceramics alone.16 Since the healing of a criti-
cal-sized defect of long bones is considered to be more chal-
lenging than that of other conditions, such as bone cavities,
we hypothesised that porous ceramic loaded with cultured
135
136 G. CINOTTI, A. M. PATTI, A. VULCANO, C. DELLA ROCCA, G. POLVERONI, G. GIANNICOLA, F. POSTACCHINI
Fig. 1a
Photomicrographs showing cell adhesion to ceramic granules not pre-soaked (a) or pre-soaked (b) in fibronectin solution.
Figure 1a – There is no evidence of cell adhesion to the surface or inner pores of ceramic granules soaked in the cell sus-
pension. Figure 1b – Ceramic granules pre-soaked in a fibronectin solution showing cell adhesion to the surface and inner
pores of the granules (arrows) (20 x).
mesenchymal stem cells could be used successfully even in
an unfavourable environment such as the graft bed in
posterolateral spinal fusion.
Materials and Methods
Posterolateral fusion was performed in 40 adult White New
Zealand rabbits weighing 4.5 kg. The animals were ran-
domly allocated to receive one of the following graft materi-
als: a) porous ceramic granules plus cultured mesenchymal
stem cells (group I); b) ceramic granules plus fresh auto-
genous bone marrow (group II); c) porous ceramic granules
alone (group III) and d) autogenous bone graft (group IV).
The surgical procedure was the same in all animals.
Briefly, under general anaesthesia, through a dorsal midline
skin incision two paramedian fascial incisions were made
and through the intermuscular plane between multifidus and
longissimus the transverse processes of the fourth and fifth
lumbar vertebrae were exposed on both sides. With a small
rongeur the transverse processes were decorticated and 2g
of graft material were positioned on the graft bed on each
side.
Graft materials
Group I. Porous ceramic granules and cultured mesenchy-
mal stem cells were used. The procedure included the fol-
lowing phases.
Collection of bone marrow. After preparation of the skin
and under general anaesthesia (10 mg/kg of intramuscular
ketamine and 20 mg/kg of intravenous sodium pentobarbit-
urate), 8 to 10 ml of bone marrow were aspirated from one
or both iliac crests with an 18-gauge needle. The material
was placed in heparinised tubes and sent for culture in vitro.
Culture and differentiation of mesenchymal stem cells. The
bone marrow was mixed with one volume of phosphate-
buffered saline and the nucleated cell fraction was enriched
Fig. 1b
for mesenchymal stem cells by density-gradient centrifuga-
tion over a Lympholyte-H (Cedarlane, Hornby, Ontario,
Canada) cushion. The cells at the medium-Lympholyte-H
interface were collected, washed three times with culture
medium (DMEM/HAMS F-12 (Dulbecco’s MEM/Nutrient
Mix F12), and seeded into culture flasks with a standard cul-
ture medium which consisted of DMEM/HAMS F-12 sup-
plemented with 15% fetal calf serum, 25 mM HEPES,
50 µg/ml of gentamicin, 2 µg/ml of amphotericin B and
50 µg/ml of L-ascorbic acid. The cells were harvested at
37˚C. On the sixth day of culture, the non-adherent cells
were removed along with the culture medium. The medium
was changed twice weekly. Confluent monolayers were
obtained about one week after the start of cell division.
Mesenchymal stem cells were subcultured at ten to 14 days.
To obtain subpassages, cell monolayers were recovered by
digestion with 0.25% trypsin-versene solution. Before treat-
ment with trypsin, careful washing of monolayers with
phosphate-buffered saline was performed to eliminate
trypsin inhibitors. Subcultivation was performed by replat-
ing at 6 x 103 cells per cm2 flasks. For each passage, one
control culture was performed.
To promote osteogenic differentiation, cells were stimu-
lated during subpassages by cultivation in standard medium
supplemented with 100 nM dexamethasone. At the comple-
tion of cell culture, the medium was removed, the cell layer
washed in phosphate-buffered saline and digested as previ-
ously described. Samples of cell suspension were diluted
with 0.5% Trypan Blue and the number of viable cells deter-
mined. An evaluation of the activity of alkaline phosphatase
by an enzymic immunoassay technique was carried out to
assess the osteogenic differentiation of cell cultures. Mor-
phological examination on the stained monolayered cells
and histochemical evaluation using the Von Kossa method
were also performed.
THE JOURNAL OF BONE AND JOINT SURGERY
 EXPERIMENTAL POSTEROLATERAL SPINAL FUSION WITH POROUS CERAMICS AND MESENCHYMAL STEM CELLS
Loading of ceramic granules with mesenchymal stem cells.
Reabsorbable granules of calcium phosphate (80%) and
hydroxyapatite (20%), with a dimension of 1 to 4 mm and a
pore size ranging between 200 and 400 µm (Pro-Osteon
500R, Irvine, California) were used. Briefly, once approxi-
mately 30 million cells had been obtained from the in vitro
culture, the porous ceramic granules were soaked in a solu-
tion containing 100 mg/ml of fibronectin (Fibronectin cellu-
lar from human foreskin fibroblast; Sigma-Aldrich, Milan,
Italy) for 16 hours at 4˚C. This procedure was found to be
necessary since, without it, little cell adhesion occurred on
the surface and within the pores of the granules (Fig. 1). The
solution was then discharged and the granules were placed
in 5 ml of cell suspension in culture medium at 37˚C and
gently agitated for three hours to facilitate the flow of cell
suspension into the pores. A number of residual cells,
approximately 10% of the initial cell population, were
found in the suspension medium after the removal of
ceramic granules.
Implantation of the graft material at surgery. Under general
anaesthesia, animals from which bone marrow had been
collected underwent spinal fusion as previously described.
Once the graft bed had been prepared, 2 g of porous ceramic
granules loaded with mesenchymal stem cells were
implanted as graft material on each side.
Group II. Porous ceramic granules and fresh bone marrow
were used. Before implantation, the ceramic granules were
soaked in a solution of fibronectin as in group I. Under gen-
eral anaesthesia the graft bed was prepared and 2 g of
ceramic granules were placed on the decorticated transverse
processes. Then 10 ml of bone marrow were aspirated from
one or both iliac crests with an 18-gauge needle and imme-
diately placed on the graft material (5 ml on each side).
Group III. Porous ceramic granules alone were used.
Before implantation, they were soaked in a fibronectin solu-
tion as in groups I and II.
Group IV. Autogenous bone graft was harvested from both
iliac crests (2 g of corticocancellous bone graft on each
side) and used as graft material.
Post-operative treatment. All the animals were housed
and treated according to university-approved guidelines.
Antibiotics and analgesic medications were given from the
first to the third post-operative day. The animals were able
to move on the first post-operative day and gained weight
between the first and second week after surgery.
Assessment of fusion. The animals were killed eight weeks
after surgery by an intraperitoneal injection of phenobarbi-
tal. The lumbar spine was harvested en-bloc and the fusion
mass was evaluated radiographically, by a manual palpation
test and by histological analysis.
Radiographic evaluation. Anteroposterior plain films of the
lumbar spine were evaluated blindly by two of the authors
(GC, GG) according to the subjective grading scale used by
Curylo et al.17 In summary, a radiographic score of four
points was given when a continuous bony mass was present
on both sides without lucency, three points when a continu-
VOL. 86-B, No. 1, JANUARY 2004
137
ous bony mass was present bilaterally with lucency on one
side, two points when a bony mass was present on both
sides with lucency bilaterally, one point when a bony mass
was present on one side only and zero points when a bony
mass was not seen on either side. A solid fusion was diag-
nosed when the radiographic score was three or four and
pseudarthrosis when the score was two or less. On axial CT
(2 mm slice thickness) the presence of newly formed bone
between ceramic granules and, in the control group,
between corticocancellous bone chips, was evaluated.
Manual palpation test. After the division of the supra-
spinous and interspinous ligaments, the posterior joints and
the intervertebral disc were excised, thus leaving the graft
material as the only tissue connecting the adjacent verte-
brae. This was performed at the operated level and, for com-
parison, at the adjacent non-operated levels above and
below. The result of the manual palpation test was classified
as solid fusion, when no vertebral movement was detected
at the operated level and, as uncertain, when vertebral
movement was present at the operated level but markedly
reduced compared with the adjacent levels. When vertebral
movement was equal to, or slightly reduced, compared with
the adjacent levels, the result was classified as a pseudar-
throsis.
Histological analysis. Specimens including the newly
formed bony mass and the transverse processes were fixed,
at least overnight, in 4% formaldehyde freshly prepared
from paraformaldehyde in phosphate buffer at pH 7.2 and
then cut longitudinally into macroscopic consecutive sec-
tions 5 mm thick. One of the most representative sections
was decalcified and routinely processed for paraffin embed-
ding. Another was embedded in glycolmethachrylate with-
out decalcification. Serial sections 5 µm and 2 µm thick
were cut from paraffin blocks and from plastic blocks,
respectively. The sections were stained with haematoxylin
and eosin (HE), and azure II-Methylene Blue and examined
under light microscopy.
Statistical analysis. Non-parametric tests were used for sta-
tistical analysis. In particular, the Kruskal-Wallis test was
used to evaluate the overall differences between the four
groups and the Mann-Whitney test to analyse differences
between each group of treatment. A p value of less than 0.05
was considered to be statistically significant.
Results
The surgical procedure was well tolerated in all groups. In
group I, two rabbits died as a result of anaesthetic complica-
tions and one developed a deep infection. In group II, one
animal died from anaesthetic complications and one devel-
oped a subcutaneous infection which did not involve the
graft material. Two animals died in group IV, one from
anaesthetic complications and one from a gastrointestinal
infection four weeks after surgery.
Gross inspection. In groups I, II and III, the implanted gran-
ules were visible in the external aspect of the fusion mass and
138 G. CINOTTI, A. M. PATTI, A. VULCANO, C. DELLA ROCCA, G. POLVERONI, G. GIANNICOLA, F. POSTACCHINI

Table I. The radiological findings. A score of 3 + 4 corresponds to solid Table II. The results of the manual palpation test. The total number of
fusion. The total number of specimens is given in parentheses specimens is given in parentheses
Radiological score Uncertain
Group Solid fusion fusion Pseudarthrosis
Group 0 1 2 3 4 3+4
I 4 3 0 (7)
I 0 0 1 4 2 6 (7)
II 3 4 1 (8)
II 0 1 3 4 0 4 (8)
III 1 4 5 (10)
III 2 2 3 3 0 3 (10)
IV 2 2 4 (8)
IV 1 0 5 2 0 2 (8)

variously incorporated within it. In specimens of group I and
II ceramic granules were tenaciously adherent to the sur-
rounding tissue, while in group III they were less adherent to,
and more easily detachable from, the adjacent tissue com-
pared with groups I and II. In all specimens of groups I, II and
III, ceramic granules showed evidence of partial reabsorption,
i.e., they were usually smaller and had round edges compared
with their initial size and shape, and pores were less evident
than initially. No specimen showed signs of inflammation
around the implanted material on gross inspection.
Radiological findings. These are given in Table I and
Figure 2. There was a significant difference between the
percentage of solid fusion observed in the four groups
(p = 0.039). In group I the rate of fusion was significantly
higher than that in group III (p = 0.015) and higher, but not
significantly, than that in group IV (p < 0.06). No significant
difference was found between groups I and II (p = 0.06) and
between group II and groups III and IV. On both radio-
graphs and CT scans the fusion mass appeared to be larger
and more homogenous in group I than in groups II and III.
Moreover, in both groups II and III, granules of graft mate-
rial tended to be less adherent to the transverse processes
and to extend beyond the fused level compared with group I.
Manual palpation test. The results of the manual palpation
test are given in Table II. In group I, the rate of pseudarthro-
sis was significantly lower than that in group III (p = 0.015)
and lower, but not significantly, than that in group IV
(p ≤ 0.06). No significant differences were found between
the other groups.
Histological analysis
Group I. Newly formed bone was observed in contact with
the implanted granules (Fig. 3a). In many specimens the
entire surface of the granules and pores was in contact with
the newly formed bone, while in a few the surface of the
granules was in contact with tight fibrous tissue. Most of the
new bone appeared to be without cartilaginous formation,
although islands of cartilaginous tissue and endochondral
ossification were also present (Fig. 3b). Highly cellular
bone marrow was found between newly formed trabecular
bone. The remodelling processes included appositional
bone formation with active osteoblasts aligned on the sur-
face of bone trabeculae.
Group II. New trabecular bone was observed adjacent to the
implanted granules. There were several differences from
group I. First, the apposition of new bone was more abun-
dant in the peripheral portion of the fusion mass than in the

Fig. 2a Fig. 2b Fig. 2c Fig. 2d

Radiological results in groups I (a), II (b), III (c) and IV(d). Figures 2a and 2b – A continuous bony mass is present bilaterally with no evidence of lucent
lines. Figure 2c – A lucent line is present between the transverse process and the fusion mass (arrow). A discontinuous bony mass is present on the contral-
ateral side. Figure 2d – A lucent line is present on one side (arrow). A thin, apparently continuous, bony mass is present on the contralateral side.

THE JOURNAL OF BONE AND JOINT SURGERY

 EXPERIMENTAL POSTEROLATERAL SPINAL FUSION WITH POROUS CERAMICS AND MESENCHYMAL STEM CELLS 139

Fig. 3a Fig. 3b

Fig. 3c Fig. 3d

Photomicrographs of specimens of groups I (a, b), III (c), and IV (d). Figures 3a and 3b – Formation of new bone is seen adjacent to the
ceramic. Highly cellular bone marrow is present along with islands of cartilaginous cells. Figure 3c – Fibrous tissue is present between
ceramic. Figure 3d – Dead trabecular bone along with abundant fibrous tissue is present (haematoxylin and eosin x 25).

central portion. The new trabecular bone appeared to be
thinner compared with that of specimens of group I; and
bone marrow between trabecular bone showed a reduced
number of nucleated cells.
Group III. A reduced amount of new bone was present com-
pared with the other groups. The newly formed bone was
mainly located close to the transverse processes. Fibrous
tissue was often present adjacent to the ceramic (Fig. 3c).
Scant haematopoietic marrow tissue was present.
Group IV. Thin trabeculae of newly formed bone were
present close to the decorticated transverse processes and,
more rarely, in the central portion of the graft bed. Dead
trabecular bone showing no evidence of revascularisation
was frequently seen at the central portion of the fusion mass
(Fig. 3d).
The quantitative differences in the amount of bone for-
mation between group I and the other groups were particu-
larly evident in the central areas of the graft. In such areas a
VOL. 86-B, No. 1, JANUARY 2004
prevalence of fibrous tissue was seen in the experimental
groups to which mesenchymal cells had not been added,
while newly formed bone was present in group I. In the
peripheral portion of the graft, the quantitative difference in
bone formation was evident between groups I and II com-
pared with groups III and IV. In the latter, formation of new
bone was only present close to the transverse process. Dif-
ferences between group I and group II were less pronounced
in the peripheral portion of the graft. They were mainly due
to the thinner conformation of the bony trabeculae in group
II.
Discussion
Autogenous bone graft is known to supply the necessary
materials to promote the formation of new bone including
bone-forming cells, growth factors and a scaffold to support
bone cells and intercellular matrix. As a result, autogenous
140 G. CINOTTI, A. M. PATTI, A. VULCANO, C. DELLA ROCCA, G. POLVERONI, G. GIANNICOLA, F. POSTACCHINI
bone graft has been successfully used even in challenging
procedures, such as arthrodesis of the spine, in which a
brace or plaster cast was the only device used to stabilise the
graft material.3,18 However, in specific surgical procedures,
such as spinal fusion at multiple levels or revision surgery in
which bone graft has already been harvested at the time of
the initial operation, the amount of autogenous bone graft
may be insufficient to promote bony fusion and a pseud-
arthrosis may occur. Further limitations of autogenous bone
graft are the prolonged operating time and the increased
perioperative and post-operative morbidity associated with
the surgical procedure.
Substitutes of autogenous bone graft for spinal fusion
have been investigated for many years.8,14 However, some
of these studies provided little information since different
animal models were used.9,13,14 More recently, an animal
model for posterolateral fusion yielding rates of fusion com-
parable with the clinical setting has been developed.19
Using this model, materials with osteoconductive properties
have been tested, including carriers with bony or porous
structures (allograft, porous ceramics, sintered bovine
bone)8,12 or without bony structures (collagen sheet or poly-
lactic acid polymer).10,20 The results of these studies have
shown that osteoconductive materials do not give a high rate
of fusion when used alone in the posterolateral fusion
model.10,15,20,21 The addition of osteoinductive substances,
particularly BMPs, led to a significantly higher rate of
fusion than was seen with osteoconductive materials
alone.8,10,12,15 However, the high manufacturing cost and
possible adverse effects of BMPs, including the release of
factors which may stimulate the production of osteoclasts,22
may limit the clinical use of these substances. In addition,
when BMPs were used in clinical trials, the results were not
better than those obtained with autogenous bone graft23 or
demineralised bone matrix was used.24
In our present study we investigated the effectiveness of
a new graft material for spinal fusion consisting of an osteo-
conductive carrier loaded with cultured mesenchymal stem
cells. The latter are capable of giving rise to bone,25 carti-
lage,26 muscle27 and fat28 once exposed to different growth
conditions. Since osteogenic differentiation of mesenchy-
mal stem cells has been found to occur in vitro in response
to various bioactive factors,29,31 we hypothesised that cul-
tured mesenchymal stem cells could provide adequate
osteoinductive material for the carrier, either by inducing
formation of new bone per se or by releasing osteogenic
growth factors which could stimulate the recruitment and
differentiation of further mesenchymal stem cells.
The cultured cells were loaded into ceramic granules,
with a pore size ranging from 200 to 400 µm. This material
appeared to be a suitable carrier for mesenchymal stem cells
since approximately 90% of the initial cell population was
found to adhere to the ceramic. However, it was interesting
to note that cell adhesion to both the external surface and
inner pores of the ceramic was observed only when the
granules were presoaked in a solution with fibronectin. This
extracellular matrix glycoprotein may play a relevant role in
the process of the formation of new bone since it has been
shown to regulate the adhesion and migration of mesenchy-
mal stem cells32-34 and affect the differentiation and sur-
vival of osteoblast-like cells.34
A rate of fusion ranging between 50% and 70% and
between 0% and 100% has been reported, respectively,
when autogenous bone and alternative graft materials were
used in experimental posterolateral fusion.8,10,12,15,17,19 In
our study, autogenous bone graft gave a rate of fusion com-
parable with that of previous investigations. Cultured mes-
enchymal stem cells loaded into porous ceramic achieved a
significantly higher rate of solid fusion compared with
porous ceramic alone, and a trend towards a higher rate than
with autogenous bone. Although the radiographic results
may be affected by the radiopacity of non-reabsorbed
ceramic, which may render it more difficult to diagnose the
presence of pseudarthrosis, both plain radiography and CT
showed that even when a continuous bony mass was
present, it appeared to be more homogenous and ceramic
granules were more adherent to the transverse process in
rabbits treated with ceramic and mesenchymal stem cells
than in those which had ceramic alone. This result was also
confirmed on gross inspection, which showed that in the
animals treated with ceramic alone, porous granules were
more easily detachable from the adjacent tissue of the
fusion mass than in those treated with ceramic plus mesen-
chymal stem cells or plus fresh bone marrow.
The osteoinductive properties of fresh bone marrow
have been documented in studies using a model of a long
bone defect.35,36 However, when fresh bone marrow was
used as osteoinductive material in experimental postero-
lateral fusion the results were controversial.8,17 A compari-
son of the osteogenic potential of fresh bone marrow and
mesenchymal stem cells has shown that the latter induced
faster and more extensive formation of new bone than fresh
bone marrow.37 In our study, rabbits treated with ceramic
plus mesenchymal stem cells showed a higher rate of
fusion than those treated with ceramic and fresh bone
marrow but the difference was not significant, probably
because of the small number of animals included in each
treatment group. However, evident differences between the
two groups were found on histological analysis. In particu-
lar, the newly formed bone showed thicker trabeculae and
more abundant bone-marrow cells in animals treated with
ceramic and mesenchymal stem cells than in those treated
with ceramic and fresh bone marrow. Moreover, in the
latter group, the new bone was essentially present in the
peripheral portion of the fusion mass, while in the animals
treated with ceramic and mesenchymal stem cells it was
seen in both the peripheral and inner portion of the fusion
mass. It can be speculated that osteogenic growth factors,
which are essential to promote differentiation of bone-
marrow cells into osteogenic cells, are likely to be present
in the peripheral portion of the graft bed while their pres-
ence may be inadequate in the inner portion, probably
THE JOURNAL OF BONE AND JOINT SURGERY
 EXPERIMENTAL POSTEROLATERAL SPINAL FUSION WITH POROUS CERAMICS AND MESENCHYMAL STEM CELLS
because of insufficient blood supply in this area. Con-
versely, mesenchymal stem cells, which have been stimu-
lated to differentiate in vitro into osteoprogenitor cells,
could be less dependent on local growth factors in inducing
the formation of new bone.
The results of our study seem to encourage further inves-
tigations on the use of mesenchymal stem cells loaded on
porous ceramic as graft material for spinal fusion. However,
several factors which may affect the success of the pro-
cedure should be considered. First, in the clinical setting,
mesenchymal stem cells may show different capabilities of
growth and differentiation depending on constitutional fac-
tors and donor age, and therefore the procedure may be
unsuitable for some patients. Secondly, the degree of differ-
entiation of mesenchymal stem cells which is more likely to
promote the formation of new bone in the host tissue, is not
well known. In particular, it has been hypothesised that less
differentiated mesenchymal stem cells, which have greater
potential for division and differentiation than more highly
differentiated mesenchymal cells, could be more effective in
promoting the formation of new bone.38 Thirdly, the sur-
vival of the implanted mesenchymal stem cells may be
affected by several factors, including the type of carriers,
local conditions of blood supply or other factors which at
present are not known.
Finally, according to the manufacturer’s indications, the
ceramic carrier used in our study should undergo reabsorp-
tion after six months. However, since in this study, the ani-
mals were killed eight weeks after the implantation of the
ceramic, our results do not provide information regarding
the long-term behaviour of the graft material, in particular,
whether its complete reabsorption occurs and, if this is the
case, whether the reabsorbed ceramic are replaced by bone
or fibrous tissue.
In conclusion, our study has shown that mesenchymal
stem cells, which have been stimulated to differentiate into
preosteogenic cells in vitro, may be loaded into porous
ceramic to give rise to a graft material which may promote
the formation of new bone in experimental spinal fusion.
This graft material appeared to be more effective than
porous ceramic alone and at least as effective as autogenous
bone graft. Further studies are needed to investigate the
medium- to long-term results of this procedure, its feasibil-
ity in the clinical setting and the most appropriate carrier for
mesenchymal stem cells.
No benefits in any form have been received or will be received from a com-
mercial party related directly or indirectly to the subject of this article.
References
1. Kiviluoto O, Santavirta S, Salenious P, Morri P, Pylkkannen P. Pos-
terolateral spine fusion: a 1-4 year follow-up of 80 consecutive patients.
Acta Orthop Scand 1985;56:152-4.
2. Savini R, Ghirardini GL, Morandi ML, Pellacci F. L’artrodesi poste-
rolaterale lombosacrale. Ital J Orthop Traumatol 1984;10:457-68.
3. Stauffer RN, Coventry MB. Posterolateral lumbar spine fusion: ana-
lysis of Mayo Clinic series. J Bone Joint Surg [Am] 1972;54-A:
1195-204.
VOL. 86-B, No. 1, JANUARY 2004
141
4. Cha CW, Deible C, Huzzonlgro T, et al. Allogenic transfusion require-
ments after autologous donations in posterior lumbar surgeries. Spine
2002;27:99-104.
5. Fernyhough JC, Schimandle JH, Weigel MC, Edwards CC, Lev-
ine AM. Chronic donor site pain complicating bone graft harvesting
from the posterior iliac crest for spinal fusion. Spine 1992;17:
1474-80.
6. Kurz L, Grafin SR, Booth RE Jr. Harvesting autogenous iliac bone
graft: a review of complications and techniques. Spine 1989;14:
1324-31.
7. Steinman JC, Herkowitz HN. Pseudarthrosis of the spine. Clin Orthop
1992;284:80-90.
8. Boden SD, Martin GJ, Morone M, et al. The use of coralline
hydroxyapatite with bone marrow, autogenous bone graft, or osteoin-
ductive bone protein extract for posterolateral lumbar spine fusion.
Spine 1999;24:320-7.
9. Guigui P, Plais PY, Flautre B, et al. Experimental model of postero-
lateral spinal arthrodesis in sheep: part 2: application of the model: eval-
uation of vertebral fusion obtained with coral (Porites) or with a biphasic
ceramic (Triosite). Spine 1994:2798-803.
10. Itoh H, Ebara S, Kamimura M, et al. Experimental spinal fusion with
use of recombinant human bone morphogenetic protein 2. Spine 1999;
24:1402-5.
11. Martin GJ Jr, Boden SD, Titus L, Scarborough NL. New formulation
of demineralized bone matrix as a more effective graft alternative in
experimental posterolateral lumbar spine arthrodesis. Spine 1999;7:
637-45.
12. Minamide A, Tamaki T, Kawakami M, et al. Experimental spinal
fusion using sintered bovine bone coated with type I collagen and
recombinant human bone morphogenetic protein-2. Spine 1999;18:
1863-70.
13. Muschler GF, Huber B, Ullman T, et al. Evaluation of bone-grafting
materials in a new canine segmental spinal fusion model. J Orthop Res
1993;11:514-24.
14. Sandhu HS, Kanim LEA, Kabo JM, et al. Evaluation of rhBMP-2
with an OPLA carrier in a canine posterolateral (transverse process) spi-
nal fusion model. Spine 1995;20:2669-82.
15. Boden SD, Schimandle JH, Hutton WC. The use of an osteoinductive
growth factors for lumbar spinal fusion: part II. Study of dose, carrier
and species. Spine 1995;20:2633-44.
16. Bruder SP, Kraus KH, Goldberg VM, Kadiyala S. The effects of
implant loaded with mesenchymal stem cells on the healing of
canine segmental bone defects. J Bone Joint Surg [Am] 1998;80-A:
985-96.
17. Curylo LJ, Johnstone B, Petersilge CA, Janicki JA, Yoo JU. Aug-
mentation of spinal arthrodesis with autologous bone marrow in a rabbit
posterolateral spine fusion model. Spine 1999;5:434-8.
18. Cleveland M, Bosworth DM, Thompson FR. Pseudarthrosis in the
lumbosacral spine. J Bone Joint Surg [Am] 1948;30-A:302-12.
19. Boden SD, Schimandle JH, Hutton WC. An experimental lumbar
intertransverse process spinal fusion model: radiographic, histologic,
and biomechanical healing characteristics. Spine 1995;4:412-20.
20. Schimandle JH, Boden SD, Hutton WC. Experimental spinal fusion
with recombinant human bone morphogenetic protein-2. Spine 1995;20:
1326-37.
21. Boden SD, Schimandle JH, Hutton WC, et al. In vivo evaluation of a
resorbable osteoinductive composite as a graft substitute for lumbar spi-
nal fusion. J Spinal Disord 1997;10:1-11.
22. Lane JM. BMPs: why are not in everyday use? J Bone Joint Surg [Am]
2001;83:S161-2.
23. Friedlander GE, Clayton RP, Cole DJ, et al. Osteogenic Protein-1
(bone morphogenetic protein-7) in the treatment of tibial nonunions.
J Bone Joint Surg [Am] 2001;83-A:S151-8.
24. Geesink RG, Hoefnagels NH, Bulstra SK. Osteogenic activity of OP-
1 bone morphogenetic protein (BMP-7) in a human fibular defect.
J Bone Joint Surg [Br] 1999;81-B:710-8.
25. Haynesworth SE, Goshima J, Golberg VM, Caplan AI. Characteriza-
tion of cells with osteogenic potential from human bone marrow. Bone
1992;13:81-8.
26. Wakitani S, Goto T, Pineda SJ, et al. Mesenchymal cell-based repair
of large full-thickness defects of articular cartilage. J Bone Joint Surg
[Am] 1994;76-A:579-92.
27. Caplan AI. Mesenchymal stem cells and gene therapy. Clin Orthop
2000;379(Suppl):67-70.
142 G. CINOTTI, A. M. PATTI, A. VULCANO, C. DELLA ROCCA, G. POLVERONI, G. GIANNICOLA, F. POSTACCHINI
28. Beresford JN, Bennett JH, Devlin C, LeBoy PS, Owen M. Evidence
for an inverse relationship between the differentiation of adipocytic and
osteogenic cells in rat marrow stromal cells cultures. J Cell Sci 1992;
102:341-51.
29. Fried A, Benayahu D, Wientroub S. Marrow stroma-derived osteo-
genic clonal cell lines: putative stages in osteoblast differentiation.
J Cell Physiol 1993;155:472-82.
30. Jaiswal N, Haynesworth SE, Caplan AI, Bruder SP. Osteogenic dif-
ferentiation of purified, culture-expanded human stem cells in vitro.
J Cell Biochem 1997;64:295-312.
31. Maniatopoulos C, Sodek J, Melcher AH. Bone formation in vitro by
stromal cells obtained from bone marrow of young adult rats.
Cell Tissue Res 1998;254:317-30.
32. Dennis JE, Haynesworth SE, Young RG, Caplan AI. Osteogenesis
in marrow-derived mesenchymal cell porous ceramic composites
transplanted subcutaneously: effect of fibronectin and laminin on cell
retention and rate of osteogenic expression. Cell Transplant 1992;1:
23-32.
33. El-Ghannam A, Ducheyne P, Shapiro IM. Effect of serum proteins on
osteoblast adhesion to surface-modified bioactive glass and hydroxy-
apatite. J Orthop Res 1999;17:340-5.
34. Globus RJK, Doty SB, Lull JC, et al. Fibronectin is a survival factor
for differentiated osteoblasts. J Cell Sci 1998;111:1385-93.
35. Johnson KD, Frierson KE, Keller TS, et al. Porous ceramics as bone
graft substitutes in long bone defects: a biomechanical, histological, and
radiological analysis. J Orthop Res 1996;14:351-69.
36. Werntz JR, Lane JM, Burstein AH, et al. Qualitative and quantitative
analysis of orthotopic bone regeneration by marrow. J Orthop Res 1996;
14:85-93.
37. Kadiyala S, Jaiswa N, Bruder SP. Culture-expanded bone marrow-
derived mensenchymal stem cells can regenerate a critical-sized seg-
mental bone defect. Tissue Eng 1997;3:173-85.
38. Banfi A, Muraglia A, Dozin B, et al. Proliferation kinetics and differ-
entiation potential of ex vivo expanded human bone marrow stromal
cells: implications for their use in cell therapy. Exp Hematol 2000;28:
707-15.
THE JOURNAL OF BONE AND JOINT SURGERY