ARTICLE IN PRESS

The closure of oroantral communications with resorbable

PLGA-coated ␤-TCP root analogs, hemostatic gauze, or
buccal flaps: A prospective study
Bojan Gacic, DDS, PhD,a Ljubomir Todorovic, DDS, PhD,b Vladimir Kokovic, DDS, PhD,c
Vesna Danilovic, DDS, PhD,d Ljiljana Stojcev-Stajcic, DDS, PhD,e
Radojica Drazic, DDS, PhD,f and Aleksa Markovic, DDS, PhD,g Belgrade, Serbia
UNIVERSITY OF BELGRADE

Objective. The aim of this study was to compare the treatment of oroantral communications (OACs) with
bioresorbable root analogs made of poly(lactide-co-glycolide) (PLGA)– coated ␤-tricalcium phosphate (␤-TCP),
hemostatic gauze or a buccal flap technique.
Study design. In this prospective clinical study, 30 patients with oroantral communications were randomly assigned to
a treatment. Clinical success, vestibular depth at the defect site, pain, and swelling were monitored.
Results. The OAC closure was successful in all cases. The vestibular depth stayed constant in the groups treated with
the PLGA–␤-TCP composite or hemostatic gauze. In contrast, a vestibular depth reduction of 1.2 ⫾ 0.2 mm was
observed in the buccal flap group, indicating atrophy of the alveolar ridge in these patients. Furthermore, pain and
swelling were more pronounced in this group.
Conclusion. Closures of OACs with PLGA–␤-TCP composite or hemostatic gauze are reliable minimally invasive
methods that minimize atrophy of the alveolar ridge, swelling, and pain compared with a buccal flap technique. (Oral
Surg Oral Med Oral Pathol Oral Radiol Endod 2009;108:844-850)

Communications between the oral cavity and the max-
illary sinus occur with frequencies between 0.31% and
4.7% after the extraction of upper teeth.1-4 If untreated,
oroantral fistulas may form, which can result in chronic
infection of the maxillary sinus. Smaller oroantral com-
munications (OACs), with a diameter of ⬍2 mm, are
often closed by a coagulum and heal spontaneously.5,6
However, in the case of larger defect sizes or disturbed
blood coagulation, iatrogenic closure of the defect be-
comes indispensable. Mucoperiosteal flaps of different
designs are widely used to seal off the sinus cavity.7-10
To stabilize the coagulum in the extraction socket,
hemostatic materials such as hemostatic gauze of ly-
ophilized fibrin glue have been applied with or without
flaps.11-13 Flap-based methods have several disadvan-
tages, e.g., flap mobilization may leave a denuded area
and lead to scarring and changes in the mucosal topog-
a
Lecturer, Clinic of Oral Surgery, Faculty of Dentistry.
b
Professor, Clinic of Oral Surgery, Faculty of Dentistry.
c
Lecturer, Clinic of Oral Surgery, Faculty of Dentistry.
d
Professor, Department of Histology and Embryology.
e
Assistant Professor, Clinic of Oral Surgery, Faculty of Dentistry.
f
Assistant Professor, Clinic of Oral Surgery, Faculty of Dentistry.
g
Associate Professor, Clinic of Oral Surgery, Faculty of Dentistry.
Received for publication Apr 30, 2009; returned for revision Jul 10,
2009; accepted for publication Jul 17, 2009.
1079-2104/$ - see front matter
© 2009 Published by Mosby, Inc.
doi:10.1016/j.tripleo.2009.07.026
raphy. Furthermore, preparation, adaptation, and fixa-
tion of flaps take a significant amount of practice and
time.1
The closure of OACs with synthetic bone graft sub-
stitutes constitutes an alternative to flap-based tech-
niques.1,14-16 Besides closure of the opening, these ma-
terials may positively affect the surrounding hard tissue,
because the insertion of biomaterials into extraction sock-
ets has been reported to reduce alveolar ridge resorp-
tion.17 However, this potential benefit of perforation
closure with biomaterials has not been assessed yet.
The use of synthetic bone graft materials is limited
owing to the dislocation of grafting material into the
sinus cavity. Thoma et al.1 solved this problem by
inserting thermally molded poly(lactide-co-glycolic)
acid (PLGA)– coated porous ␤-tricalcium phosphate
(␤-TCP) granulate into oroantral perforations in 14
patients. The inherently stable fully degradable root
analog successfully closed off the communication.
Here, we present a prospective clinical study where
this method was compared with the closure of OACs
using hemostatic gauze and with a standard flap-based
approach. In addition to the clinical results, the impact
of the 3 methods on the vestibular depth was compared
and short-term postoperative effects, such as pain and
swelling, were assessed. A histologic analysis of a
patient who was treated with a ␤-TCP root analog is
also presented.

844
 ARTICLE IN PRESS
OOOOE
Volume 108, Number 6 Gacic et al. 845

Table I. Basic characteristics of patients included in Table II. Diameters of oroantral communications
the study (OACs)
Gender Age (mean ⫾ SD) Teeth* Method OAC diameter
16 (53%) female 36 ⫾ 16 years 4 (13%) second premolar PLGA-coated ␤-TCP 1-3 mm: 1
14 (47%) male 21 (70%) first molar 3-5 mm: 7
4 (13%) second molar ⱖ5 mm: 2
1 (3%) third molar Hemostatic gauze 1-3 mm: 3
3-5 mm: 5
*Oroantral communication created following the removal of corre-
ⱖ5 mm: 1
sponding tooth.
Buccal flap 1-3 mm: 1
3-5 mm: 6
ⱖ5 mm: 3

MATERIAL AND METHODS

All procedures and materials described in this pro-
spective clinical study were approved by the Ethics
Committee at the Faculty of Dentistry, University of
Belgrade. Treatment strategy of our clinic is that OACs
should not be left to heal spontaneously, owing to
ethical reasons. Accordingly, 30 patients, with OACs
created after tooth extraction, were randomly selected
into 3 groups. Closure of perforations was achieved
either with PLGA-coated ␤-TCP granulate, with hemo-
static degradable gauze, or by application of a buccal
flap. Only patients that were in good general health were
included. Smokers and pregnant or lactating woman and
patients under any medication were excluded. An in-
formed consent was obtained from each of the patients.
The patients’ ages ranged from 17 to 70 years, with an
average of 36.9 years (Table I). For extraction of mul-
tirooted teeth, the dental crown was separated from the
roots and the roots were extracted one at a time. Second
premolars and single-rooted teeth were extracted in 1
step. After tooth extraction, the nose blowing test was
performed, and when positive, an OAC was diagnosed.
The minimal diameter of the communication was deter-
mined using specially devised blunt probes with diameters
of 1, 3, and 5 mm1 (Table II).
OAC closure using PLGA-coated ␤-TCP
Root analogs were made according to the manufac-
turer’s instructions with PLGA-coated synthetic, phase-
pure, porous ␤-TCP granules (Degradable Solutions,
Schlieren, Switzerland). In brief, the root above which
the OAC was located was cleaned of granulation tissue
with curette, incubated in 96% H2O2 for 1 minute, and
rinsed with sterile physiologic saline solution. The
cleaned root was pressed into impression material (De-
gradable Solutions) in the chamber of a heating device.
After the impression material had set, the mold was cut
and the tooth removed. The mold was replaced in the
heating device and was filled with PLGA-coated ␤-
TCP granules. After 1 minute of heating, the granules
condensed to form a solid but porous copy of the
extracted root. Polylactide powder (Degradable Solu-
tions) was applied onto the coronal end of the root
analog and sealed using a heated condenser (Degrad-
able Solutions), thus forming an integrated membrane.
The mold was removed from the heater and the root
analog was removed from the mould after a short
cooling period (Fig. 1, A). Fresh bleeding was induced
by curettage of the extraction site. The root analog was
placed in the extraction socket and carefully pushed up
until proper seating was achieved (Fig. 1, B). The
wound edges were approximated by a single endless
suture (Fig. 1, C).
OAC closure with hemostatic gauze
Resorbable hemostatic gauze composed of reconsti-
tuted oxidized cellulose (Surgicel; Johnson and John-
son, Somerville, NJ) was used for closure of the OAC.
The gauze was cut into small pieces (Fig. 1, D), which
were inserted one at a time into the extraction socket
(Fig. 1, E). Care was taken not to compress the gauze,
because the material quells by taking up blood. The
wound edges were approximated digitally and secured
with a single endless suture (Fig. 1, F).
OAC closure with buccal flaps
Buccal sliding flaps were applied as described by
Rehrmann9 (Fig. 1, G and H).
All patients were advised to restrain from blowing
their nose, blowing up balloons, playing brass instru-
ments, drinking through a straw, or taking antibiotics or
analgesics that contain acetylsalicylic acid. Any sutures
were removed after 7 days.
Postoperative follow-up
Control appointments were made 1, 2, and 7 days
and 1, 3, and 6 months after the intervention. All
patients received 4 mg dexamethasone intramuscularly,
nonsteroidal antiinflammatory drugs for pain relief, and
detailed instructions concerning oral hygiene. During
the first week, the vestibular depth, swelling, and pain
intensity were assessed. The vestibular depth was mea-
sured with a compass as the distance between the
 ARTICLE IN PRESS
OOOOE
846 Gacic et al. December 2009

PLGA-coated
fl-TC P

A B C
haemostatic
gauze

D E F
buccal flap

G H I
Fig. 1. Methods for the closure of oroantral communications. A, A root analog was made by thermally molded poly(lactide-co-
glycolide) (PLGA)– coated ␤-tricalcium phosphate (␤-TCP). B, The composite material was inserted into the oroantral commu-
nications and (C) the wound edges were approximated. D, E, Alternatively, hemostatic gauze was inserted into the extraction
socket and (F) the wound edges were approximated. G, For closure with a buccal flap, a mucoperiosteal flap was mobilized and
(H, I) sutured over the opening.

marginal gingiva of the extraction site to the highest
point in the vestibule in the same frontal plane (Fig. 2,
A and B). Swelling was measured extraorally by assess-
ing the distance between the intertragic notch and the
corner of the mouth (Fig. 2, C). The pain intensity was
quantified using a visual analog scale18 that ranged
from 0 to 100 arbitrary units. Whereas vestibular depth
measurements were continued during the 6 month
observation period, swelling and pain were assessed
only during the first week. Furthermore, the patients
were advised to perform the nose blowing test at each
follow-up appointment.
Histologic analysis
A Straumann wide neck implant (Straumann, Basel,
Switzerland) was placed 6 months after closure of the
OAC with PLGA-coated ␤-TCP granulates. During this
process, tissue samples were taken with a trephine drill
from the site where the alloplastic material had been
inserted. The sample was fixed in 10% formalin for 48
hours, decalcified in 10% formic acid (pH 5) for 2-3
weeks, and embedded in Paraplast. Histologic sections
(8 ␮m; McCormic Scientific, LLC, Maryland Heights,
MO, USA) were cut with a microtome (Leica; SM
2000), stained with hematoxylin and eosin, and ob-
served by light microscopy (Leica DM1000).
RESULTS
Clinical outcome
In 70% of all cases, the OAC had occurred after
extraction of first maxillary molars, the other cases
occurred after extraction of second premolars and
second and third molars (Table I). The average min-
imum diameter of the OAC did not differ among the
3 groups that were assigned to the 3 methods
(Kruskal-Wallis test: P ⫽ .320; Table II). Healing of
 ARTICLE IN PRESS
OOOOE
Volume 108, Number 6
A B
C each comparison). After 7 days, the groups treated with
Fig. 2. Assessment of the vestibular depth and swelling. A,
B, Vestibular depth was assessed with a compass. C, Swelling
was assessed extraorally by measuring the distance between
the mandibular angle and the chin tip (MC) and the intertra-
gus insection and the corner of the mouth (IC).
the OAC was controlled at every appointment by
pressurizing the nasal and the oral cavity. Closure
was successful in all 30 patients and took place
without complications. An implant was set in 1 pa-
tient at the site where the communication had oc-
curred (Fig. 3). The implant showed a good primary
stability. The stability quotient (ISQ) measured with
an Osstell mentor device (Osstell, Gothenburg, Sweden)
was 64 immediately after implant insertion. Within 6
weeks, the ISQ increased to 80.
Vestibular depth is preserved in patients treated
with a ␤-TCP root analog or hemostatic gauze
Vestibular depth was measured before, immediately
after, after 7 days, and after 6 months. During the 6
months’ observation period, vestibular depth did not
change in the group treated with PLGA-coated ␤-TCP
(Table III) (analysis of variance [ANOVA]: P ⬍ 1.00).
In patients treated with surgical gauze, vestibular depth
was slightly decreased on day 7 and did not change any
more thereafter. However, these slight variations were
not significant (ANOVA: P ⬍ .98). In contrast, appli-
cation of a buccal flap led to a pronounced change in
vestibular depth (ANOVA: P ⬍ .02). A decrease of
vestibular depth immediately after the intervention was
observed. The patients treated with a buccal flap had a
significantly reduced vestibular depth after 6 months
(paired t test; P ⬍ .01). The average depth loss was 1.2 ⫾
0.2 mm (mean ⫾ SEM), varying between 0 and 2.5.
Gacic et al. 847
Subjective pain intensity was the lowest for the
group treated with a ␤-TCP root analog
The subjective pain intensity was the highest for the
first day after the intervention independent of the method
that was used for OAC closure. However, pain intensities
differed among the 3 groups. Patients with buccal flaps
reported the highest pain, and pain was lower in the
groups treated with hemostatic gauze or ␤-TCP, with
the latter showing the lowest value (Table IV). On the
second day after the intervention, pain intensities were
reduced. Again, the patients treated with a buccal flap
suffered the most and the patients treated with ␤-TCP
the least pain. The differences between the treatments
were significant (Mann-Whitney U tests: P ⬍ .01 for
a ␤-TCP root analog or hemostatic gauze were painless
whereas pain was still reported by patients with buccal
flaps.
Swelling was negligible in patients treated with a
␤-TCP root analog
Swelling was assessed by measuring the distances
between the mandibular angle and the chin tip (MC)
and from the intertragic notch to the angle of the mouth
(IC) before, immediately after surgery, and on days 1
and 7. Results are shown in Table V. Distances did not
change over time in patients who were treated with
PLGA-coated ␤-TCP (ANOVA: P ⬍ 1.00 [MC]; P ⬍
.42 [IC]), indicating that the treatment did not induce swell-
ing. A slight but significant increase in the IC distance was
observed in the hemostatic gauze group (ANOVA: P ⬍
.01), indicating that swelling was triggered by the treat-
ment. The patients with the buccal flap showed the most
pronounced swelling (ANOVA; P ⬍ .01).
Histologic analysis
Histologic analysis of the samples taken during the
implant placement procedure did not show signs of
necrosis or inflammation, which is in accordance with
the reported biocompatibility of PLGA-coated ␤-TCP
granulate.19 Mature lamellar bone was observed (Fig. 4,
A), and numerous osteocytes were present in lacunae (Fig.
4, B). Osteoblasts and newly formed bone matrix could be
observed, indicating an ongoing osteogenesis (Fig. 4, C).
DISCUSSION
Oroantral communications should be treated by es-
tablishing a physical barrier to prevent infection of the
maxillary sinus and fistula formation. Despite the pos-
sibility for small-sized OAC to heal spontaneously, von
Wowern20 reported that only 2 out of 9 OACs heal on
their own in a study on 116 patients with OAC. Based
on the results from that study, it was concluded that
every OAC should be treated irrespective of its size.20
 ARTICLE IN PRESS
OOOOE
848 Gacic et al. December 2009

Fig. 3. Implant placement at the site of the oroantral communication. Radiographs before and 3, 6, and 7.5 months after tooth
extraction and treatment with poly(lactide-co-glycolide)– coated ␤-tricalcium phosphate. The last image shows the tooth after
completion of the prosthetic treatment (6 weeks after implant placement).

Table III. The effects of methods of closure of oroantral communication on the values of the vestibular depth
Follow-up period
Baseline Postop 7 days 6 months
Method of
closure X ⫾ SD Median Min. Max. X ⫾ SD Median Min. Max. X ⫾ SD Median Min. Max. X ⫾ SD Median Min. Max.
Root analog 12.65 ⫾ 1.49 12.75 10 15 12.55 ⫾ 1.42 12.75 10 15 12.65 ⫾ 1.49 12.75 10 15 12.65 ⫾ 1.49 12.75 10 15
Hemostatic 13.55 ⫾ 1.23 13.75 11 15 13.55 ⫾ 1.23 13.75 11 15 13.55 ⫾ 1.23 13.75 11 15 13.55 ⫾ 1.23 13.75 11 15
gauze
Buccal flap 13.4 ⫾ 1.61 13.75 10 15.5 11.7 ⫾ 1.62 11.25 9 14 10.85 ⫾ 1.73 11 7 13 12.2 ⫾ 1.74 12.5 8 14.5

NOTE. Analysis of variance test: P ⬍ .02 for buccal flap.

Table IV. Pain intensity after closure of oroantral communication in the postoperative period
Follow-up period
1st day 2nd day 7th day
Method of closure X ⫾ SD Median Min. Max. X ⫾ SD Median Min. Max. X ⫾ SD Median Min. Max.
Root analog 6.4 ⫾ 4.3 6 0 13 2.5 ⫾ 2.37 2 0 7 0 0 0 0
Hemostatic gauze 8.6 ⫾ 3.66 7 4 16 3.8 ⫾ 2.74 4 0 9 0 0 0 0
Buccal flap 19.3 ⫾ 3.4 19.5 15 25 11.4 ⫾ 5.21 10 5 19 2.8 ⫾ 2.57 2 0 9
NOTE. Mann-Whitney U test: P ⬍ .01.

In most cases, OACs are covered with mucoperiostal
flaps of different designs. However, these methods have
inherent disadvantages like postoperative pain and swell-
ing. Long-term effects may include scar formation and
alveolar ridge resorption,21,22 which may impede implant
placement and prosthetic treatments.
In the present study, 3 different methods for the
closure of OAC were compared by assessment of
posttreatment pain, swelling, and loss of alveolar ridge
depth. Oroantral perforations were closed with either a
porous custom-fit root analog made of phase-pure PLGA-
coated ␤-TCP granules, hemostatic gauze, or a buccal
flap.
The closure of oroantral perforations was successful in
all of the patients, and fistula formation was prevented.
For buccal flaps following Rehrmann, success rates be-
tween 84% and 94% are reported,8,23,24 which is in agree-
ment with our findings. Solid custom-made hydroxyapa-
tite blocks have been reported for the closure of oroantral
fistulas, which was successful in all 6 patients enrolled in
the present study. However, hydroxyapatite blocks be-
came loose and were eventually lost as the fistulas closed.
In contrast, all ␤-TCP root analogs were maintained. This
positive outcome may be due to the osteoconductive prop-
erties and the biodegradability of porous ␤-TCP, which is
resorbed slowly by solution in body fluids and by cellular
 ARTICLE IN PRESS
OOOOE
Volume 108, Number 6 Gacic et al. 849

Table V. Amount of swelling after closure of oroantral communication according to applied methods, using metric
measurement from the inter-tragic notch to the angle of the mouth
Follow-up period
Baseline Postop 1st day 7th day
Method of
closure X ⫾ SD Median Min. Max. X ⫾ SD Median Min. Max. X ⫾ SD Median Min. Max. X ⫾ SD Median Min. Max.
Root analog 15.05 ⫾ 0.8 15 13.5 16 15.05 ⫾ 0.8 15 13.5 16 15.1 ⫾ 0.81 15.25 13.5 16 15.05 ⫾ 0.8 15 13.5 16
Hemostatic 14.55 ⫾ 1.23 15 12 16 14.55 ⫾ 1.23 15 12 16 14.85 ⫾ 1.25 15 12 16 14.55 ⫾ 1.23 15 12 16
gauze
Buccal flap 15.25 ⫾ 1.01 15 14 17 15.25 ⫾ 1.01 15 14 17 16.45 ⫾ 1.04 16.75 15 18 15.45 ⫾ 0.96 15 14.5 17

NOTE. Analysis of variance test: P ⬍ .01 for buccal flap and hemostatic gauze at first postoperative day.

oc

wb

oc
o

A B C
Fig. 4. Histologic analysis. Hematoxylin-eosin–stained thin sections of tissue samples taken at the site 6 months after poly(lactide-
co-glycolide)– coated ␤-tricalcium phosphate had been applied. A, Major lamellar bone is observed. B, Numerous osteocytes (oc)
are present. C, Woven bone (wb) with associated osteoblast-like cells. Magnifications: ⫻40 for A, ⫻400 for B, and ⫻200 for C.

uptake25,26 as bone ingrowth closes the defect. If left
untreated, small OACs may close spontaneously. Healing
depends on the maintenance of a coagulum in the perfo-
ration, which was the rational for closure of the OACs
with hemostatic gauze. To our knowledge, the success
rate of this method never has been assessed. The present
results suggest that the clinical outcome of the closure with
hemostatic gauze is similar to the other treatments, although
estimation of the success rate requires a larger study.
The vestibular depth was taken as a measure for the
vertical dimension of the alveolar process. It is obvious
that the vestibular depth is reduced immediately after
application of a buccal flap. In the course of 6 months, the
vestibular depth gradually increased in these patients but
did not reach pretreatment dimensions, which strongly
suggests that the application of a buccal flap resulted in a
decreased vestibular depth. In contrast, the vestibular
depth of the patients treated with PLGA-coated ␤-TCP or
hemostatic gauze was unchanged after 6 months; there-
fore, these methods should be preferred if preservation of
the alveolar process is necessary. Indeed, the filling of
extraction sockets with various bone graft substitutes has
been reported to prevent postextraction alveolar atrophy
and may thus render augmentative measures that would be
required to treat the atrophied alveolar ridges unneces-
sary.17 Using ␤-TCP for the treatment of an OAC thus
may fulfil 2 clinical needs: to close the OACs efficiently
and to preserve the alveolar ridge.
Pain was most intense for patients that were treated
with a buccal flap. Patients that had received a degradable
root analog reported the least pain, whereas pain intensi-
ties were intermediate for the group treated with hemo-
static gauze. Also, the patients treated with coated ␤-TCP
or hemostatic gauze were free of pain after 7 days,
whereas patients with a buccal flap were still suffering
from pain. Similarly, swelling, which was measured ex-
traorally, was intense for the patient group treated with
buccal flaps and moderate for the hemostatic gauze group.
Remarkably, swelling was absent in patients treated with
PLGA-coated ␤-TCP. These findings agree with Thoma
et al.,1 who reported that postoperative pain and swelling
were markedly reduced if OACs were closed with PLGA-
coated ␤-TCP compared with buccal sliding flaps. The
closure of an oroantral opening with PLGA-coated ␤-TCP
or hemostatic gauze is less invasive than procedures re-
quiring flap mobilization, which explains the low pain
 ARTICLE IN PRESS
850 Gacic et al.
intensities and, in the case of ␤-TCP, the absence of
extraoral swelling. Consequently, these procedures may
reduce the need for antibiotic or analgesic therapies.
Porous ␤-TCP has osteoregenerative properties.27
Because a solid bony support is a prerequisite for stable
placement of an implant, treatment of an oroantral
perforation with ␤-TCP may be advantageous if a pros-
thetic treatment is planned. Histologic analysis after 6
months showed that bone had formed at the site where
the ␤-TCP had been inserted and that remaining ␤-TCP
granules were in close contact with bone tissue, under-
scoring the biocompatibility of this biocomposite. The
implant during whose placement the tissue sample for
the histology was collected was stably integrated. In
this study, no comparisons of bone quality and quantity
at the former defect sites were made. However, the
observation that the vestibular depth was reduced in the
patients who were treated with buccal flaps suggests
that bone loss may have occurred in these cases.
In conclusion, we found that closure of OACs with
hemostatic gauze or PLGA-coated ␤-TCP was as ef-
fective as the well established buccal flap approach.
However, in contrast to buccal flaps, pain intensities
were lower for these 2 minimally invasive methods,
and swelling was virtually absent in patients treated
with PLGA-coated ␤-TCP granules. Together with the
osteoconductive properties of porous ␤-TCP, the latter,
therefore, may form a valuable alternative for the clo-
sure of oroantral openings with flaps, especially if
prosthetic treatments are anticipated.
The authors thank Professor Doctor Zoran Stajcic for his
open heart and good will in adding his great knowledge to
this article. Also, the authors thank Mr. Michael Koehli for
his unselfish help in this study. The PLGA-coated ␤-
TCP and the instruments used to make the root analog
were kindly provided by Degradable Solutions.
REFERENCES
1. Thoma K, Pajarola GF, Gratz KW, Schmidlin PR. Bioabsorbable
root analogue for closure of oroantral communications after tooth
extraction: a prospective case-cohort study. Oral Surg Oral Med
Oral Pathol Oral Radiol Endod 2006;101:558-64.
2. Beckedorf H, Sonnabend E. [The incidence of maxillary sinus
perforations in teeth extractions]. Zahnarztl Rundsch 1954;63:
566-9. German.
3. Punwutikorn J, Waikakul A, Pairuchvej V. Clinically significant
oroantral communications—a study of incidence and site. Int
J Oral Maxillofac Surg 1994;23:19-21.
4. Hirata Y, Kino K, Nagaoka S, Miyamoto R, Yoshimasu H,
Amagasa T. [A clinical investigation of oromaxillary sinus per-
foration due to tooth extraction]. Kokubyo Gakkai Zasshi 2001;
68:249-53. Japanese.
5. Guven O. A clinical study on oroantral fistulae. J Craniomaxil-
lofac Surg 1998;26:267-71.
6. Keresztesi K. [Conservative therapy of oroantral fistula]. Osterr
Z Stomatol 1954;51:317-23. German.
OOOOE
December 2009
7. Haanaes HR, Pedersen KN. Treatment of oroantral communica-
tion. Int J Oral Surg 1974;3:124-32.
8. Killey HC, Kay LW. An analysis of 250 cases of oro-antral
fistula treated by the buccal flap operation. Oral Surg Oral Med
Oral Pathol Oral Radiol Endod 1967;24:726-39.
9. Rehrmann A. Eine methode zur schliessung von kieferhöhlen-
perforationen. Dtsch Zahnarztl Z 1936:1136-9.
10. Ziemba RB. Combined buccal and reverse palatal flap for closure
of oral-antral fistula. J Oral Surg 1972;30:727-9.
11. Gotzfried HF, Kaduk B. [Closure of oroantral openings with an alco-
holic prolamine solution. Experimental animal study and initial clinical
results]. Dtsch Z Mund Kiefer Gesichtschir 1985;9:390-3. German.
12. Stajcic Z, Todorovic LJ, Petrovic V. Tissucol in closure of oroantral
communication. A pilot study. Int J Oral Surg 1985;14:444-6.
13. Bjoklund A, Kochk HJ, K.I. P. A new method of strenghtening palatal
closure defects of hard palate. Acta Otolaryngol 1976;82:147-50.
14. Becker J, Kuntz A, Reichart P. [Closure of oroantral perforations
with hydroxylapatite ceramic]. Dtsch Z Mund Kiefer Gesich-
tschir 1987;11:92-5. German.
15. Ogunsalu C. A new surgical management for oro-antral communi-
cation: the resorbable guided tissue regeneration membrane— bone
substitute sandwich technique. West Indian Med J 2005;54:261-3.
16. Zide MF, Karas ND. Hydroxylapatite block closure of oroantral
fistulas: report of cases. J Oral Maxillofac Surg 1992;50:71-5.
17. Schmidlin PR, Jung RE, Schug J. [Prevention of alveolar ridge
resorption after tooth extraction—a review]. Schweiz Monatss-
chr Zahnmed 2004;114:328-36. German.
18. Wewers ME, Lowe NK. A critical review of visual analogue
scales in the measurement of clinical phenomena. Res Nurs
Health 1990;13:227-36.
19. Nair PN, Luder HU, Maspero FA, Fischer JH, Schug J. Biocom-
patibility of beta-tricalcium phosphate root replicas in porcine tooth
extraction sockets—a correlative histological, ultrastructural, and
X-ray microanalytical pilot study. J Biomater Appl 2006;20:307-24.
20. von Wowern N. Frequency of oro-antral fistulae after perforation
to the maxillary sinus. Scand J Dent Res 1970;78:394-6.
21. Ericson S, Finne K, Persson G. A clinical-radiographic review of
treated oro-antral communications. Int J Oral Surg 1973:185-95.
22. von Wowern N. Closure of oroantral fistula with buccal flap:
Rehrmann versus Moczar. Int J Oral Surg 1982;11:156-65.
23. Amaratunga NA. Oro-antral fistulae—a study of clinical, radiological
and treatment aspects. Br J Oral Maxillofac Surg 1986;24:433-7.
24. von Wowern N. Correlation between the development of an
oroantral fistula and the size of the corresponding bony defect.
J Oral Surg 1973;31:98-102.
25. Knabe C, Koch C, Rack A, Stiller M. Effect of beta-tricalcium
phosphate particles with varying porosity on osteogenesis after sinus
floor augmentation in humans. Biomaterials 2008;29:2249-58.
26. Zerbo IR, Zijderveld SA, de Boer A, Bronckers AL, de Lange G,
ten Bruggenkate CM, et al. Histomorphometry of human sinus
floor augmentation using a porous beta-tricalcium phosphate: a
prospective study. Clin Oral Implants Res 2004;15:724-32.
27. Kamitakahara M, Ohtsuki C, Miyazaki T. Review paper: behav-
ior of ceramic biomaterials derived from tricalcium phosphate in
physiological condition. J Biomater Appl 2008;23:197-212.
Reprint requests:
Aleksa Markovic
Clinic of Oral Surgery
Faculty of Dentistry
University of Belgrade
Dr Subotica No 4
11000 Belgrade
Serbia
maleksa@sbb.rs