Hydroxylapatite as an Alloplastic Graft in the Treatment

of Human Periodontal Osseous Defects

Roland M. Meffert,* Jeffery R. Thomas,f Kent M. Hamilton^ and
Carol N. Brownstein§

Accepted for publication 23 July 1984

Twelve patients, 32 60 years of age, received a polycrystalline ceramic form of pure

dense hydroxylapatite as an alloplastic bone implant material in intrabony defects following
reflection of full mucoperiosteal flaps, root planing and defect-curettment. The defects were
measured from an acrylic stent, using an endodontic silver point which was placed to the
base of the defect. Similarly, debrided and curetted defects in the same patients were not
implanted and served as controls. Recalls for documentation and plaque control were at 1,
2, and 4 weeks, and at 3, 6 and 9 months. Measurements relating to changes in defect-depth
were made upon reentry at 9 months.
The twelve defects, serving as controls, showed very little difference between the pretherapy
and 9-month measurements. The initial mean measurement from the base of the defect to
the highest alveolar crest was 4.27 mm and the 9-month mean measurement after curettage
only was 3.36 mm. In terms of resolution of the original defect this amounted to 19.49%
reduction, but a 0.46-mm mean loss in height of the alveolar crest provided an actual
percentage fill of the original defect of 9.91%.
Of sixteen experimental defects, the same initial mean measurement from the base of the
defect to the highest alveolar crest was 5.18 mrn and the 9-month mean measurement after
grafting was 2.43 mm. In terms of resolution of the original defect, this amounted to a
53.57% reduction, but in contradistinction to the curettage sites, a mean increase in height
of the highest alveolar crest of 0.61 mm gave a true percentage fill of the original defect of
66.89%.
At the 9-month reentry, the implanted mass seemed to be partially "calcified" and was
resistant to penetration with a probe or removal with a curette. The data and clinical
impression strongly suggest that hydroxylapatite has a potential as an alloplastic implant
with clinically apparent acceptance by the soft and hard tissues.

In the past few years, periodontists have used various
alloplastic grafting materials in an attempt to regenerate
bone lost from periodontal disease. There has been
much confusion regarding the efficacy of such materials
and techniques.
The calcium phosphate system, and in particular
hydroxylapatite (HA), has been the subject of intensive
investigation because it is from this system that verte-
brate tooth and bone mineral is derived.1 The most
widely investigated calcium phosphate biomaterials are
*
Professor and Director, Postdoctoral Division, Department of
Periodontics, Louisiana State University, New Orleans, LA 70119.
t Graduate Student, Department of Periodontics, University of
Texas Health Science Center at San Antonio, San Antonio, TX
78284.
t Private Practice.
§ Associate Professor, Department of Periodontics, The University
of Texas Health Science Center at San Antonio.
composed of either HA [CAi0(PO4)6(OH)2] or trical-
cium phosphate, TCP, [Ca3(P04)2]. Virtually all cal-
cium phosphate biomaterials can be classified as poly-
crystalline ceramics since their material structure is
derived from individual crystals of a highly oxidized
substance which have become fused together at the
crystal grain boundaries by a high temperature process
called sintering.
The factors which determine the rate and degree (if
any) of bioresorption of calcium phosphate bioceramics
can be summarized as follows:
1. High density implants, be they HA, TCP, or mix-
tures of the two will show little tendency to bioresorb
due to their small surface area.
2. Porous TCP implants will resorb much more rap-
idly than porous HA implants of very similar struc-
ture.
The hydroxylapatite is radiopaque and as the ma-

63

64 Meffert, Thomas, Hamilton, Brownstein
terial is already in its highest oxidation state, will not
oxidize further or resorb. It was theorized that résorp-
tion might lead to deterioration by-products which
might promote foreign body or immunological reac-
tions.2
The histologie sequelae associated with implantation
of nonporous calcium phosphate materials in bone can
be generally characterized as being representative of
normal bone healing processes on and around the im-
plant.3"8 At the interface of both dense or porous ma-
terials, bone is usually deposited directly on the surfaces
without the presence of an intervening fibrous tissue
capsule. Normal calcification also takes place at the
implant sites. In one study of dense HA implants placed
in surgically created defects in dog femurs, the implant
sites were examined along with adjacent old bone sites
for calcium and phosphorous content and Ca/P ratios
with the electron microprobe using point counts, line
ratemeter scans and pulse images. Normal calcification
processes were observed immediately adjacent to the
implants, as evidenced by increasing Ca/P ratios (from
1.50 at 1 month to 1.62 at 6 months) and increasing
calcium and phosphorous concentrations; at 6 months
mineralization with the implant sites was comparable
to that of surrounding bone.5
Calcium phosphate implants are inert, biocompatible
and evoke no inflammatory or immune response when
implanted in osseous or nonosseous tissues.9"'1 In a dog
study determining the suitability of a porous TCP im-
plant for use in spinal fusions, most of the implants
became crushed and/or dislodged shortly after implan-
tation (3-6 weeks), indicating the mechanical unsuita-
bility of the material for this use. Histologically, how-
ever, examination of the implant sites revealed an ab-
sence of inflammatory or foreign body response to the
material.12
The osteogenicity of calcium phosphate implants has
also been the subject of some investigation. Although
it is not uncommon to read statements to the effect
that these materials may stimulate Osteogenesis,1314 the
origin of these claims may be due to the high biocom-
patibility of the calcium phosphates relative to other
materials. Examination of these materials in traditional
bone induction models15 produced negative results. In
one millipore chamber study, a dense HA implant
material, in the form of plugs or particles admixed with
hematopoietic marrow, was compared with control
chambers containing marrow alone. After 6 weeks of
implantation, more of the control chambers contained
newly formed bone (90%) than did the HA-marrow
containing chambers (35%).16
Although it is clear that calcium phosphate implants
do not induce bone formation in the traditional sense,
they do sometimes display bone growth "guiding" prop-
erties, causing bone to grow into areas which it would
otherwise not occupy. This property of the calcium
phosphates has been characterized by various individ-
uals with terms such as "osteoconductive" and "osteo-
J. Periodontol.
February. 1985
philic." In series of controlled experiments by Fer-
a
raro,17 surgically created defects were made in the or-
bital rims, mandibles and iliac crests of dogs. When left
unfilled, only minimal evidence of repair was observed
during periods of up to 1 year. When similar defects
were filled with porous bioresorbable TCP implants,
total growth of bone across the prosthesis was observed
1 year postimplantation with bone being present from
one bone-prosthesis interface to the other. Some of the
resorbable implant was still present at that time, how-
ever.
Other authors have reported that empty control de-
fects repair more rapidly than calcium phosphate im-
planted defects. This observation has been a fairly
consistent finding in investigations using porous forms
of calcium phosphate (TCP) as the implant material.
In one study evaluating a porous nonresorbable HA
implant material in surgically created periodontal de-
fects in dogs, the empty control sites were completely
filled with new bone after 16 weeks but the porous
blocks of HA in the experimental sites still contained
significant amounts of proliferating fibrovascular tis-
sue.18 Other investigators, using a porous paniculate
bioresorbable TCP implant in essentially the same
model, obtained similar results.19,20 In contrast to the
above, surgically created periodontal defect implant
sites containing dense HA implants in particulate form
appear to allow for healing rates more or less compa-
j, rabie to those observed in empty controls.21
The difference between the two materials may lie in
the need for the résorption rate of the implant material
to match closely the tissue infiltration rate. If it is not,
it would be expected that healing with the porous
ceramic matrix, as evidenced by ultimate appearance
of mature bone, could be slower than an empty defect.
A study was reported in which the calvaría of rabbits
implanted with porous blocks of bioresorbable TCP
material, demonstrated an uptake within the porous
implant approximately 1/10 that of surrounding bone
but 1 /2 that of surrounding bone at the periphery. The
results were obtained after 3 months postimplantation
and were determined by injecting the rabbit with radi-
olabeled calcium 24 hours prior to killing.22 In contrast
to porous implants, implant sites filled with dense
calcium phosphate particles impose fewer restrictions
to investing tissues which can grow over and around
the particles according to their own dictates.5 6
A wide variety of surgical models has been used to
investigate calcium phosphate implants with many of
them using autografts and/or empty defects as controls.
Autogenous material consistently has outperformed
calcium phosphate implants with regard to histologi-
cally determined healing rates in such areas as perio-
dontal lesions (dogs),19 spinal fusions (dogs)12 and seg-
mentai replacements (rabbits).23 In a study designed to
determine the time required to effect bony bridging of
rabbit tibia lesions, for example, a porous TCP material,
in granular form, was compared to autogenous cancel-
Volume 56
Number 2
lous bone as well as to mixtures of the two. The auto-
graft effected bridging at 4 to 6 weeks, the ceramic at
14 to 16 weeks, and a 50/50 mixture of the two at 6
weeks.23 While this study revealed the superiority of the
autogenous material, the results clearly point to the
utility of the material as an autografi extender.
In terms of studies exploring the efficacy of the
hydroxylapatite in the treatment of periodontal disease,
Rabalais et al.24 implanted the material into intrabony
lesions, using nonimplanted sites as controls. Measure-
ments obtained at surgical reentry after 6 months in-
dicated greater fill in the experimental sites. As a result,
they proposed that hydroxylapatite had a definite po-
tential in treating periodontal osseous defects. Boyne
and Fremming25 placed ceramic particles in surgically
created wounds in the maxillary lateral incisor region
of Rhesus monkeys. Again, the contralateral side was
the unimplanted control. Their results demonstrated
the lack of inflammatory response around the ceramic
particles and a surrounding of the particles in the defect
by new bone. Particles superior to the defect and in the
soft tissue aspect of the pocket were simply surrounded
by connective tissue with no evidence of new bone
formation.
Froum et al.26 in a case report to determine the
clinical and histologie response to hydroxylapatite im-
plants in intraosseous lesions, evaluated sections re-
moved between 8 weeks and 8 months postgraft surgery
and reported pocket depth decreased but no indication
of new periodontal attachment, Osteogenesis or cemen-
togenesis adjacent to the graft particles. They felt the
material was a biocompatible "fill." In another report,
Stahl et al.27 found that the allograft (hydroxylapatite)
showed a similar regeneration of lost periodontal at-
tachment to the sites treated with debridement and
autogenous grafts but it was slower. Again, they theo-
rized the synthetic graft acted as a "filler." Finally,
Moskow and Lubarr28 in a case report, in which a
combination of hydroxylapatite particles with autogen-
ous bone chips was used to treat an extensive periodon-
tal defect, reported the ceramic particles were compat-
ible with the periodontal tissues, showed no evidence
of extrusion or rejection and were completely encap-
sulated by connective tissue fibers. But this was only 9
weeks after therapy and could be related to the earlier
reference of delayed healing around the ceramic parti-
cles.
The purpose of this reentry study was to evaluate the
efficacy of dense hydroxylapatite particles as an implant
material in human intrabony periodontal defects as
compared to control defects treated by debridement
only.
MATERIALS AND METHODS
Patients. Twelve volunteer adult Periodontitis pa-
tients, being routinely treated in the postdoctoral clinic
were selected: eight females and four males, from 32 to
Hydroxylapatite as an Alloplastic Graft 65
60 years of age, all in good systemic health, and having
at least two or more advanced vertical osseous defects,
as verified by radiographie and clinical analysis. The
defects were either 1-wall, 2-wall, wide 3-wall or com-
bination defects; narrow 3-wall areas were excluded
from the study due to their relative predictability in
healing.
Presurgical Evaluation. Full mouth radiographs were
exposed and study casts prepared; the location of each
defect was noted as well as corresponding tooth mobil-
ity (using the clinical assessment of 0-3 of S.C. Miller).
The patient's oral hygiene was evaluated and graded at
time of surgery, at réévaluation and at each réévaluation
appointment, according to the index of Loe, 1967.
An acrylic stent was prepared to fit over the teeth in
the treated site and a hole cut in the stent so that an
endodontic silver point could be inserted through the
hole and placed to the greatest depth of the defect. The
point was held by locking pliers with the beaks parallel
to the occlusal surface and a Boley gauge used to
measure distances to the nearest tenth of a millimeter.
Intraoral radiographs and photographs were exposed
of each defect site with the standardization of radi-
ographs established with the Fixott-Everett grid.
Surgical Phase and Clinical Measurement. Each de-
fect site was exposed by reflection of full mucoperiosteal
flaps, following sulcular incisions and retention of the
marginal gingiva. The defects were cleared of granula-
tion tissue (Fig. 1 ) and the exposed root surfaces thor-
oughly planed to a smooth hard surface. Upon debride-
ment of the site, the stent was placed, the silver point
inserted through the stent to the depth of the cleaned-
out defect (Fig. 2) and the following measurements (by
notching the silver point with a sharp instrument)
taken:
•
from bottom of stent to bottom of defect
•
from CEJ to bottom of defect
•from highest alveolar crest to bottom of defect
Since each patient had to have at least two defect
sites (for control and HA treatment), a coin toss was
used to determine which therapeutic modality would
be used for a given area. Upon such determination, the
Figure 1. Osseous defect degranulated.

66 Meffert, Thomas, Hamilton, Brownstein
Figure 2. Measurements taken with stem, silver point.
Figure 3. Defect implanted with hydroxylapatile particles.
site either filled with HA (hydroxylapatite) parti-
was
cles, 40-60 mesh,* or left empty after curettage as a
control. The HA particles were transferred from the
sterile vial to a dappen dish, moistened with sterile
saline, and placed into the defect with a Woodson
plastic instrument. The particles were placed into the
intraosseous wound to the approximate level of the
crest or the remaining osseous wall(s) (Fig. 3). The
operative site was closed with 4-0 black silk sutures and
protected with a noneugenol dressing. Prior to dressing-
placement, postoperative radiographs were taken of the
experimental and/or control sites as baseline data (Fig.
4). Antibiotics were prescribed immediately prior to
surgery and for a minimum of 6 days postoperative.
Medication for analgesia was administered on an "as
needed" basis.
At approximately 1 week following surgery, the dress-
ing and sutures were removed, the area debrided and
the dressing replaced. Each defect site was examined
visually and radiographically for healing. Any sign of
sloughing or particle-migration was noted. The dressing
was usually removed at 14 days postsurgery and oral
hygiene instructions delivered and demonstrated to the
patient.
*
Calcitite 4060. Calcitek, Inc.. San Diego, CA.
J. Periodontol.
February, 1985
Figure 4. Post implantation of hydroxylapatite particles.
The patients were seen at 1 -week, 2-week, 4-week, 3-
month, 6-month and 9 month intervals, with visual,
radiographie and photographic examinations accom-
plished at these times, along with oral hygiene indices
and mobility evaluation.
At 9 months posttherapy, a mucoperiosteal flap was
reflected at each site, the area cleared of granulation
tissue down to hard structure, photographs exposed and
measurements made, again using the same stent de-
scribed previously. After recording of the data and
completing any additional therapy deemed necessary,
îe flaps were reapproximated, sutured with 4-0 black
lk interrupted sutures, and the patient placed on a
.ecall schedule of 12, 18 and 24 months for documen-
tation, observation and maintenance.
The / statistic for means of unpaired samples was
used in the statistical analysis of HA grafted defects
versus control (curettement only) defects. The defects
were also split into four subsets: those less than or equal
to 4 mm, control and experimental; and those greater
than 4 mm up to 6.2 mm, control and experimental.
These were analyzed by a one-way analysis of variance
and by the Newman-Keul's test.
Since the study deals with continuous data, subtrac-
tive procedures, division and use of t- tests with per-
centages are valid.
RESULTS
At the time of reentry at 9 months postimplantation
or postcurettage, the following clinical findings were
evidenced. In the areas implanted with the HA particles,
tissue health was excellent with no clinical evidence of
inflammation. Upon reflection of the mucoperiosteal
flaps in these areas, however, it was noted that the tissue
was very tightly bound down and careful reflection with
selective undermining was essential to separate the soft
tissue from the underlying bone. Upon reflection, there
was some evidence of particles embedded in the con-
nective tissue on the inner surface of the flap and a very
distinct demarcation of the implanted site from the
surrounding osseous tissues (Figs. 5-10). Exposure of
Volume 56
Number 2 Hydroxylapatite as an Alloplastic Graft 67

the control sites demonstrated very little change in most

cases from the pretherapy defect.
The implanted mass seemed to be partially "calci-
fied" and was impossible to probe without marked
resistance. One could not probe between the radicular
surface and the implanted area. Instrumentation with
a curette failed to remove or dislodge any of the graft
mass.

Figure 8. Intrabony osseous defect after debridement.

Figure 5. Osseous defect after debridement.

Figure 9. Osseous defect immediately after implantation with hydrox-

ylapatite.

Figure 6. Osseous defect after HA implantation.

Figure 10. Re-entry of osseous defect 8V2 months postimplantation.

Blending of the implanted material into the sur-

Figure 7. Re-entry of osseous defect 9 months after implantation.
rounding bone occurred rather rapidly from a radi-
ographie standpoint (Figs. 11-14), but immediate
postimplantation and 9-month radiographs revealed an
estimated loss of less than one-fourth of the particles,
especially where the flap reapproximation was not op-
timum on the proximal surfaces. This always seemed
 J. Periodontol.
68 Mejfert, Thomas, Hamilton, Brownstein February. 1985

Diagram I illustrates how the reentry measurements

were obtained at the time of surgery and it identifies
the symbols used in determining various measure-
ments, means and percentages in Tables 1, 2 and 3.
Table 1, analyzing the results of the twelve control
sites in which curettage only was the therapy employed,

Figure 11. Radiograph of osseous defects before therapy.

Figure 14. Radiograph 8 weeks postlherapy, showing gradual blend-

ing of HA particles into normal osseous structures.

Figure 12. Radiograph of defects immediately after therapy (premo-

lar-molar interdental test site, molar-molar interdental control site).

Figure 13. Radiograph 2 weeks positherapy.

to occur within the first few weeks postoperatively (Figs.

15-18).
The mobility in all teeth seemed to decrease slightly
and was not directly related to the modality of therapy.
The decreased patterns seemed to be a reflection of
decreasing the inflammatory state rather than any sup-
port from the material. Figure 16. Mesial defect 1st molar immediately after augmentation.
Volume 56
Number 2 Hydroxylapatite as an Alloplastic Graft 69

Figure 17. Mesial defect ist molar two months post-augmentation. Figure 18. Mesial defect 1st molar 9 months after augmentation.

RE-ENTRY MEASUREMENTS INITIAL MEASUREMENTS

Taken
during
surgery
A':
B':
CEJ to base of defect
Alveolar crest to base of defect
Taken
during
surgery
{£ CEJ to base of defect
Alveolar crest to base of defect

Arithmetic Determinations
CEJ to alveolar crest C: CEJ to alveolar crest
(A' minus B') (A minus B)
D: Base of original defect
to Base of re-entry defect
(A minus A')
E: Initial alveolar crest to
re-entry alveolar crest
(C minus C)
Diagram I

shows a minor difference in the pretherapy and 9-
month measurements. The initial mean measurement
(± SEM) from the base of the defect to the highest
alveolar crest was 4.27 ± 0.39 mm and the 9-month
mean measurement after curettage only was 3.36 ±
0.31 mm. Table 3 shows that this was a 19.49 ± 4.52
mean per cent resolution of the original defect.
Table 2 outlines the results of the 16 experimental
sites in which HA particles were implanted into the
defects. There were four patients who had two sites
implanted, hence the difference in numbers of subjects
between control and experimental. The initial mean
measurement from the base of the defect to the highest
alveolar crest was 5.18 ± 0.44 mm and the 9-month
mean measurement after grafting was 2.43 ± 0.27 mm.
Table 3 shows this to be 53.57 ± 4.79% resolution of
the original defect. Means and standard error were not
included for A (CEJ to base of defect—initial) and A'
 J. Periodontol.
70 Meffert, Thomas, Hamilton, Brownstein February, 1985

TABLE I
DEFECT DEPTH MEASURED BEFORE and AFTER
CURETTAGE ONLY (CONTROL SITES)*
Initial Reentry Initial Reentry
Alveolar Crest Alveolar Crest
CEJ to Base CEJ to Base to Base of to Base of
Case No. of Defect of Defect Defect Defect
(A) ( ') (B) ( ')

5.4 4.5 4.3 1.9

6.3 6.2 2.3 2.1

8.6 8.2 4.0 3.6

9.7 7.3 6.2 4 0
7.3 7.0 5.2 5.1
10.0 9.8 5.7 4.2
7.2 7.0 4.0 3.4
7.2 6.5 5.5 3.7
8.4 7.8 4.0 3.3
10 6.4 6.2 2.8 2.3
11 4.5 4.3 2.1 2.0
12 6.1 5.5 5.1 4.7

Mean ·
S.E.M.: N.C. N.C. 4 27 ± 0.39 3.36 i 0.31
'Measurements in millimeters; reentry measurements 9 months later
**N.C: Mean and S.E.M. not computed as data in A and A' were used for individual
subtraction purposes only

TABLE II
DEFECT DEPTH MEASURED BEFORE and AFTER
HA ALLOPLASTIC GRAFTS (EXPERIMENTAL SITES)*

Initial Reentry Initial Reentry

Alveolar Crest Alveolar Crest
CEJ to Base CEJ to Base to Base of to Base of
Case No. of Defect of Defect Defect Defect
(A) ( ') (B) ( ')
6.4 3.3 5.0 2.1
9.7 4.6 6.1 3.5
7.5 5.2 3.9 2 8

9.9 6.0 7.2 3 0

6.6 4.8 4.1 3.0
10.5 7.7 4.7 3.1
10.2 6.2 7.3 3.6
6.5 3.2 2.8 0.5
10.1 6.4 7.7 3.5
10) 7.9 3.9 62 2.7
11 ( »» 8.8 4.0 3 4 1.2
12 9.2 5.7 5.5 2.6
13 7.4 4.9 4.5 2.0
14 6.8 6.1 3.1 2.1
15 5.0 2.9 3.3 0.0
16 12.1 6.0 8.0 3.2

Mean ± S.E.M. N.C. N.C. 5.18 ± 0.44 2.43 0.27

—

•Measurements in millimeters; reentry measurements 9 months later

"Two experimental sites in same patient
***N.C: Mean and S.E.M. not computed as data in A and A' were used for individual
subtraction purposes only
Volume 56
Number 2 Hydwxylapatite as an Alloplastic Graft 71

TABLE III
COMPARATIVE RESULTS OF CLINICAL MEASUREMENTS OF HA GRAFT SITES AND
DEBRIDEMENT ONLY (CONTROL) SITES AT INITIAL AND NINE MONTHS POST SURGERY

HA Grafts Site Controls

Measurements (mm.) (n 16)=
(n 12)
=
P-value
or Percentages Symbols* Mean ± S.E.M. Mean ± S.E.M. (unpaired t test)
Initial Defect 0.44
5.18 ± 4.27 ± 0.39 0.144
Depth (mm.)
Reentry Defect B' 2.43 ± 0.27 3.36 ± 0.31
Depth (mm.)
Percentage B-B'
Original Defect (X 100) 53.57 ± 4.79 19.49 ± 4.52 <0.001
Resolved
Amount of Defect
3.36 ± 0.34 0.45 ± 0.21
Fill (mm.)

Percentage Fil
of Original
Defect f (X 100) 66.89 ± 7.05 9.91 ± 3.50 <0.001

Change in Alveolar
0.61 ± 0.26 -0.46 ± 0.25** <0.01
Crest (mm.)
Percentage Change in I (X 100) 13.32 ± 6.39 -9.58 ± 3.46** 0.012
Alveolar Crest
Height***
Symbols are identified in Diagram 1 ; means of determining various measurements/percentages are noted above
*

**Negative numbers signify bone résorption

***Percentage is based on the depth of the initial defect in millimeters

(CEJ to base of defect—reentry) since it was data used 100 I Defects Less Than
I or Equal to 4 millimeters
for subtractive purposes only to obtain values for the Defects Greater Than
amount of defect fill (D on Diagram I). 80-1 4millimeters but Less
Than 6.2 millimeters
The same Table 3, however, shows a significant n=5
change in the height of the alveolar crest in relation to Percent 60
the original depth of the defect from the original alveo- Defect n=7
Resolution 40
lar crest. A 0.46 ± 0.25 mm mean loss in height of the
n=6
alveolar crest in the control areas, compared to a mean
20
increase in height of 0.61 ± 0.26 mm in the HA grafted =6

sites was observed. This would give a true percentage

fill of the original defect of 9.91 ± 3.50% for the control Control Hydroxylapatite
Defects Defects
sites and 66.89 ± 7.05% for the experimental grafted
Diagram III
sites. The per cent defect resolution, defect fill and
change in alveolar crest of the control and grafted areas [5^3 Defects Less Than
is depicted in Diagram II. Diagrams III and IV outline 100 V///A orEqual to 4 millimeters

100 80
n =
5 Defects Greater Than
4 millimeters but Less
Than 6.2 millimeters
Hydroxylapatite
n =
7
80 Percent 60
J Control Defects Defect
60- Fill 40-
Percent
40- 20-
n = 6 n=6

20 0
Control Hydroxylapatite
Defects Defects
0 Diagram IV

-20 Percent Defect Percent Fill Percent Change the per cent defect resolution and per cent fill, respec-
Resolution of Defect of Alveolar Crest tively, of 12 defects which broke the statistical base
Diagram II down into those defects less than or equal to 4 mm and

72 Meffert, Thomas, Hamilton, Brownstein
greater than 4 but less than 6.2 mm. A one-way analysis
of variance and a Newman-Keul's test failed to dem-
onstrate a significant difference between the groups of
experimental or the groups of control sites.
Table 3 demonstrates a significant difference in the
percentage of the original defect resolved—53.57% to
19.49%, test and control, and in the percentage fill of
the original defect—66.89% to 9.91%, test and control.
It also demonstrates a significant difference (P 0.012)
=
between the change in alveolar crest height in controls
and HA sites, a mean loss of 9.58% in controls and a
gain of 13.32% in grafted defects.
After measurements were recorded, any residual de-
fects were treated with sound therapeutic techniques.
DISCUSSION
The HA material was very well tolerated in the hard
and soft tissues and does not seem to evoke any inflam-
matory response; this has been confirmed by Jarcho et
al,9 Moskow and Lubarr" and many other investiga-
tors. The clinical appearance of the tissue covering the
grafted sites was one of exceptional gingival health while
bleeding upon routine probing was absent during the
course of the study.
At re-entry, the surface of the grafted areas appeared
pebbly and the HA particles were enmeshed in what
seemed to be a partially "calcified" matrix; the grafted
areas were hard to the touch and resisted penetration
by the probe. This was in accordance with the study of
Rabalais et al.24 when they reopened the sites in 6
months and reported the particles enmeshed in a soft
tissue matrix but hard and resistant to probing. Since
this study involved re-entry at 9 months postimplan-
tation, it is possible that a year or more is required for
optimum incorporation of the particles in the surround-
ing osseous structures. Delayed healing has been re-
ported around the HA particles in studies23 and im-
planted sites have healed slower than empty controls in
animal studies.21 The dense HA particles might evi-
dently have to be surrounded by connective tissue
initially and then possibly be incorporated into an
"osseous matrix" at some point in time.
All results demonstrated a very definite advantage to
the utilization and implantation of the hydroxylapatite
particles when compared to the unfilled controls. The
mean percentage in resolution of the original defect was
53.57% in the grafted sites and 19.49% in the control
areas; this compares very favorably with the figures of
48.5%, experimental, and 11%, control, in the study of
Rabalais et al.24 Of greater significance is the percentage
fill of 66.89% experimental and 9.91% control of the
original defect because there was a concurrent loss in
height of the alveolar crest osseous wall in the control
sites and an increase in height in the grafted areas. The
finding of changes in alveolar crest height was also
reported by Rabalais et al. and has been documented
J. Periodontol.
February, 1985
by many other investigators who have re-entered pre-
viously treated defects. Most of these prior reports have
consistently noted a loss of crestal bone height. The
findings of this study have shown a loss of crestal height
in the majority of the control defects, as expected, but
a surprising and significant gain in crestal bone height
in the HA grafted sites. In spite of the rather broad
range of values in the experimental group, the data
shows only four of the 16 sites as having a net loss of
crestal bone height. As Diagram I illustrates (symbols
C and C), these values are determined by measure-
ments from the CEJ. The statistical results suggest a
positive effect by the HA particles on the peripheral
crestal bone of the osseous defects. It can only be
speculated that the synthetic particles have a "bone
matrix maintenance" influence on the crestal bone
which allows regeneration of this bone, if it does indeed
resorb, and possibly further allows bone to grow around
any adjacent particles. This is additionally supported
by the radiographie findings that show the material to
be most stable in areas adjacent to the osseous bound-
aries of the defects.
When the defects are placed into subsets of less than
or equal to 4 mm, and greater than 4 mm but equal to
or less than 6.2 mm, control and experimental, the
results are not consistent with previous reports.
Rabalais et al.,24 using slightly different values in
groupings, found a decrease in per cent defect fill and
% slight increase in per cent defect resolved as defect depth
increased in the grafted sites. The results of this study
show similar changes in per cent defect fill but a slight
decrease in defect resolution as defect depth increased.
It should be noted that the differences in either param-
eter of these two groups in this investigation were not
found to be statistically significant in either the grafted
or control sites. The debrided-only sites revealed an
exact opposite relationship as compared to grafted de-
fects. An increased per cent defect resolution was noted
with increasing depth of defect as well as an increase in
per cent fill. The differences in this study and the
Rabalais et al.24 study could be due to the differences
in groups, size of sample and measurement techniques
(e.g., multiple point measurement of defect vs. single
point, and use of the nearest half millimeter measure
with a probe vs. silver points with a stent).
The apparent radiographie disappearance of the im-
planted particles may have been due to loss through
the sulcus because of the difficulty in reapproximating
the flap to the radicular surface interproximally. Rela-
tive radiodensity changes, as have been described by
Yukna,29 may also be, in part, responsible for this
apparent loss.
In conclusion, the clinical hard and soft tissue re-
sponse to the alloplastic ceramic particles at 9 months
is excellent and suggests that implantation of the ma-
terial into periodontal osseous defects leads to signifi-
cantly better defect-elimination and fill than debride-
Volume 56
Number 2
ment alone. Additional histologie evidence must now
be produced in order to evaluate properly the nature of
the interface between the implanted material and the
radicular surface; some investigators feel the success-
fully implanted mass may act as a mechanical block to
the apical migration of the junctional epithelium30 and
that there may be a partial ankylosis at the implant-
tooth interface with the absence of any functional per-
iodontium.
Histologie observations were not accomplished since
the object of the study was not to show the presence or
absence of connective tissue attachment; in all proba-
bility, new connective tissue attachment did not occur
in this study. It has been shown, however, that there is
no appreciable difference in resistance to disease be-
tween a long junctional epithelial adhesion and a true
connective tissue attachment.31,32 It is also possible that
the clinical and histologie picture at 9 months may not
be indicative of the potential appearance of the im-
planted area at 12 or 18 months. This must also be
studied and reported. And finally, the increase in eresiai
bone-height findings may enhance the possibilities of
supra-crestal bone grafting, if this synthetic material
can be used in conjunction with another grafting ma-
terial that has definite osteogenic capabilities.
REFERENCES
1. McConnell. D.: Apalile. New York, Springer-Verlag, 1973.
2. Hulbert, S. F.: Biomaterials—the case for ceramics. Paper
presented at National Conference on "Use of Ceramics in Surgical
Implants". Clemson University. Clemson, SC. 1969.
3. Denissen, H. W.. and deGroot. K.: Immediate dental root
implants from synthetic dense calcium hydroxylapatite. J Prosthe
Dent 42: 551. 1979.
4. Denissen, H. W., deGroot. .. Kakkes, P., et al.: Animal and
human studies of sintered hydroxylapatite as a material for tooth root
implants (Abstr.). Ist World Biomater Cong (Baden, Austria), 3.8.1.
1980.
5. Jarcho. M.. Kay. J. F.. Gumaer, K. I., et al.: Tissue, cellular
and subcellular events at a bone-ceramic hydroxylapatite interface, J
Bioengl: 79, 1977.
6. Kato. .. Aoki. H.. Tabata. T.. and Ogiso, M.: Biocompatibility
of apatite ceramics in mandibles. Biomater Med Devices Art ifOrgans
7: 291, 1979.
7. Niwa. S.. Sawai, K., Takahashi. S., et al.: Experimental studies
of the implantation of hydroxylapatite in the medullary canal of
rabbits (Abs). Ist World Biomater Cong (Baden, Austria), 4.10.4
1980.
8. Ogiso. M.. Kaneda. H.. Arasaki. J., et al.: Epithelial attachment
and bone tissue formation on the surface of hydroxylapatite ceramics
(Abstr.). /.si World Biomater Cong (Baden. Austria), 4.1.5. 1980.
9. Jarcho, M.. Bolen, C. H.. Thomas. M. B., et al.: Hydroxylapatite
synthesis and characterization in dense polycrystalline form. J Mater
Sci 11:2027, 1976.
10. Cohen, D. W„ Yukna, R. ., Scopp, I. W., et al.: A symposium
on a new bone-grafting implant material for the treatment of perio-
dontal disease. Compendium of continuing education in dentistry
(Spec. Suppl), Jan-Feb, 1982.
11. Moskow, B. S., Lubarr, .: Histological assessment of human
periodontal defect after durapatite ceramic implant. J Periodont Dent
54:455. 1983.
12. Shima, T.. Keller. J. T.. Alvira. M. M., et al.: Anterior cervical
dissectomy and interbody fusion, an experimental study using a
Hydroxylapatite as an Alloplastic Graft 73
synthetic tricalcium phosphate, J Neurosurg 51: 533, 1979.
13. Getter, L„ Bhaskar, S. N., Cutright, D. E„ et al.: Three
biodegradable calcium phosphate slurry implants in bone, J Oral
SurgiO: 263, 1972.
14. Grower, M. F„ Horan, M„ Miller, R„ and Getter. L.: Bone
inductive potential of biodegradable ceramic in millipore fdter cham-
bers, J Dent Res 52: 160, 1973.
15. Urist, M. R„ and Strates, B. S.: Bone morphogenetic protein,
J Dent Res 50: 1392, 1971.
16. Boyne, P. J., Fremming, B. D., Walsh, R., Jarcho, M.: Eval-
uation of a ceramic hydroxylapatite in femoral defects, J Dent Res
57A: 108, 1978.
17. Ferraro, J. W.: Experimental evaluation of ceramic calcium
phosphate as a substitute for bone grafts, Plast Reconst Sitrg 63: 634,
1979.
18. Nery, E. B., Lynch, K. L„ Hirthe, W. M., and Mueller, K. H.:
Bioceramic implants in surgically produced infrabony defects, J
Periodontol46: 328, 1975.
19. Levin, M. P., Getter, L., and Cutright. D. E.: A comparison
of iliac marrow and biodegradable ceramic in periodontal defects, J
Biomed Mater Res 9: 183, 1975.
20. Levin, M. P.. Getter, L., Adrian, J„ and Cutright, D. E.:
Healing of periodontal defects with ceramic implants, J Clin Perio-
dontol 1: 197, 1974.
21. Boyne, P. J., and Shapton, . .: The response of surgical
periodontal defects to implantation with a hydroxylapatite ceramic,
p. 115. Trans 4th Ann Meet Soc Biomater and 10th Internat Biomater
Symp 1978.
22. Hassler, C. R., McCoy, L. G„ and Rotaru, J. H.: Long term
implants of solid tricalcium phosphate, Proc 27th Ann Conf Eng Med
Bio 16: 488, 1974.
23. Lemons, J. E., Ballard, J. B„ Culpepper, . I., and Niemann,
K. M. W.: Porous tricalcium phosphate ceramic for segmental lesions,
(Abstr.)/.« World Biomater Cong (Baden, Austria), 4.10:3. 1980.
·* 24. Rabalais, M. L.. Jr., Yukna, R. ., and
Mayer, E. T.: Evalu-
ation of durapatite ceramic as an alloplastic implant in periodontal
osseous defects. J Periodontol 52: 680, 1981.
25. Boyne, P. J.. Fremming, B. D.: Periodontal pocket elimination
using durapatite ceramic implant. Alpha Omegan 75: 17, 1982.
26. Froum, S. J., Kushner.L., Scopp, I. W., Stahl, S. S.: Human
clinical and histologie responses to durapatite implants in intraosseous
lesion—case reports. J Periodontol 53: 719, 1982.
27. Stahl, S. S., Froum, S. J„ and Kushner, L.: Healing responses
of human intraosseous lesions following the use of debridement,
grafting, and citric acid root treatment. II. Clinical and histologie
observations: One year postsurgery. J Periodontol 54: 325, 1983.
28. Moskow, B. S., and Lubarr, .: Histological assessment of
human periodontal defect after durapatite ceramic implant. J Perio-
dontol 54: 455, 1983.
29. Yukna, R. .: A symposium on a new bone grafting implant
material for the treatment of periodontal disease. Compendium of
Continuing Education, pp 15-18, January-February, 1982.
30. Cranin, N„ Shpuntoff, R.. and Satler, N.: Paniculate hydrox-
ylapatite (HA) as an implantable device to salvage failing endosteal
implants in dogs. J Dent Res 63: 200, 1984.
31. Magnusson, I., Runstad, L. Nyman, S., and Lindhe, J.: A long
junctional epithelium—a locus minoris resistentiae in plaque infec-
tion. J Clin Periodontol 10: 333, 1983.
32. Beaumont, R. H., O'Leary, T. J., and Kafrawy, A. H.: Relative
resistance of longjunctional epithelial adhesions and connective tissue
attachments to plaque-induced inflammation. J Periodontol 55: 213,
1984.
Send reprint requests to: Dr. Roland M. Meffert. Department of
Periodontics, Louisiana State University, 1100 Florida Ave, New
Orleans, LA 70119.