J Oral Maxillofac Surg

571667.671, 1999

The Incidence, Location, and Height of

Maxillary Sinus Septa in the Edentulous
and Dentate Maxilla
Gerald Krennmair, MD, DMD, * Christian W Ulm, MD, DMD,+
Herbert Lugmayr, MD,+ and Peter Solar, MD, DMDJ

Purpose: This study evaluated the incidence, location, and height of antral septa and demonstrates their
clinical implications.
Materials and methods: One hundred ninety-four maxillary posterior regions, subdivided into four
groups (group 1, 61 clinically examined atrophic ridges; group 2, 41 anatomically examined atrophic
ridges; group 3,42 radiographically [CT] examined atrophic ridges; and group 4,50 CT examined dentate
maxillary ridges), were examined for the incidence, location, and height of antral septa.
Results: The incidence of antral septa was significantly greater (P < .Ol) in atrophic edentulous regions
(groups 1,2, and 3) than in dentate regions (group 4). However, the septa were much lower (P < .Ol>. In
atrophic maxillae, about 70% of antral septa were located in the anterior (premolar) region.
Conclusions: Antral septa are more commonly found in edentulous atrophic maxillae than in dentate
maxillae. The septae in edentulous atrophic maxillae are shorter than those found in dentate maxillae.
When present, maxillary sinus septae are more common anteriorly than posteriorly. CT scanning is the
preferred radiographic method for detecting the presence (or absence) of sinus septae. Panoramic
radiography has less sensitivity and specility than CT scanning for the detection of sinus septa.

After tooth loss, the edentulous alveolar process of the zones of resorption, which allows transfer of mastica-
posterior maxilla is often affected by resorption. Such tory pressure.4,5
resorption results in a loss of vertical bone volume, Antral septa of varying shape and size were first
whereas progressive sinus pneumatization leads to a mentioned in a detailed description of maxillary sinus
decrease in the alveolar process from the cranial anatomy by Underwood* in 1910. For decades, these
aspect. The magnitude of these changes varies from septa were considered clinically insignificant ana-
one person to another, often leaving inadequate bone tomic variations.6 However, new diagnostic methods
height for placement of endosseous implants.‘-* Be- for verification of sinus disorders, such as endoscopy
cause teeth are lost gradually, atrophy-related resorp- and intraoperative observation, have led to a different
tion may occur differently in different areas of the attitude toward the maxillary sinus and its anatomic
alveolar process.*J For biomechanical reasons, a bony variations.7-9
septum remains in the region between two such Detailed knowledge of maxillary sinus anatomy is
becoming increasingly important during sinus augmen-
tation surgery. lo-l2 Such augmentation allows anchor-
age of dental implants even when the posterior
Received from the University of Vienna, School of Dentistry, maxillary region has undergone severe bone resorp-
Vienna, Austria. tion.ls,14 In this surgical technique, a hinge door is
*Associate Professor, Clinical Lecturer, School of Dentistry, Univer- made in the facial antral wall and inverted to create
sity of Vienna, Austria; and Private Practice, Wels. space for the grafting material.*O~ll Either an autolo-
tAssistant Professor, School of Dentistry, Department of Oral gous or a xenogenic bone graft is placed between the
Surgery, University of Vienna, Austria. antral floor and the elevated sinus membrane (includ-
*Associate Professor, Department of Radiology, University of ing the inverted bone plate).lO Although there have
Vienna, Austria. been some moditications of this surgical technique
$Assistant Clinical Fellow, School of Dentistry, University of during the past few years, either with a supplementary
Vienna, Austria. or a simultaneous Le Fort I osteotomy or horseshoe
Address correspondence and reprint requests to Dr Krennmair: osteotomy, or a nasal floor elevation,t5x16 the original
A-4600 W&-Austria, Trauncggsiedlung 8, AWlrid. technique described by Boyne and JameslO is still valid
o 1999 Americain Association of Oral and Maxillofaclal Surgeons today.
0278.2391,‘99/5706-0006$3.00,‘0 Surgical interventions in the posterior maxillary

667
668
region require detailed knowledge of maxillary sinus
anatomy and the possible anatomic variations.‘@ For
example, the presence of septa can limit the creation
of a window in the lateral antral wall and elevation of a
hinged door. Antral septa also make it difficult to
prevent perforation of the Schneiderian membrane
when elevating it from an alveolar recess containing
several septa.‘OJ* Precise knowledge of the patient’s
maxillary sinus morphology allows exact planning of
surgery and helps to avoid such complications. The
aim of this study was to examine the incidence,
location, and height of antral septa by using clinical,
anatomic, and imaging (panoramic radiography, com-
puted tomography [CT]) findings.
Materials and Methods
This study was based on data from 194 sinuses in
edentulous and dentate maxillae that were subdivided
into four groups according to the presence and
absence of teeth and the examination method used.
Group 1 consisted of 61 totally or partially edentulous
maxillae (23F/38M; 58.3 t 5.7 years) with clinically
and radiographically (panoramic radiography) verified
atrophy-related resorption of the alveolar process.
Sinus augmentation combined with placement of
endosseous implants was performed on all of these
maxillary sinuses. The assumed incidence of antral
septa in the sinuses was evaluated using panoramic
radiography. This was then compared with the intraop-
eratively verified number of septa. The height of the
septa was measured intraoperatively using a calibrated
periodontal probe. Each sinus floor that exhibited a
septum was divided into three portions (anterior
[premolar region]; middle [first molar region], and
posterior [second molar region]) to allow precise
evaluation of the topographic location of the septa.
Group 2 consisted of 41 cadaveric edentulous
maxillae (26F/15M, 76.4 ir 9.3 years) obtained from
the collection of the Department of Anatomy of the
University of Vienna. After the maxillary sinus had
been exposed from the cranial aspect by a horizontal
section below the floor of the orbits, the alveolar
recess of the antrum was examined for the presence
of septa. A calibrated periodontal probe was used as a
measuring device, and the location of the septa was
assessed in a similar fashion as in group 1 (anterior,
middle, posterior portion).
Group 3 consisted of 42 radiographically (CT)
examined patients with edentulous atrophic maxillary
regions (19F/23M; 61.5 -+ 16.3 years), and group 4
consisted of 50 partially/completely dentate patients
with nonatrophic posterior maxillary regions (22F/
28M; 36.2 +- 14.2 years). The CT examinations were
performed using a Somatom Plus (Siemens, Erlangen,
feed and slice thickness: 1 mm; window width: 2,600;
MAXILLARY SINUS SEPTA
middle position: 500 HU; 2s 120 kV, 500 mAs, image
matrix: 512 X 512). The incidence of maxillae show-
ing antral septa was evaluated using an axial plane of
section. The height of the septa (mm) was assessed
using subsequent sagittal/coronal reconstructions. The
location (anterior, middle, posterior portion) of the
antral septa in groups 3 and 4 was determined in the
same way as in groups 1 and 2.
The incidence of maxillae with septa and the
intramaxillary location and height of the antral septa
were compared between groups. Nonparametric data
were compared using the x2 (chi square) test. Mean
values were compared using the Student’s t-test. P
values less than .05 were considered statistically
significant.
Results
Antral septa were found in 17 (27.8%) of 61 patients
in group 1 who underwent sinus augmenation. Table
1 shows a comparison between the incidence of septa
correctly or falsely diagnosed using panoramic radiog-
raphy and the intraoperatively verified number of
septa. Panoramic radiography led to a false diagnosis
regarding the presence or absence of antral septa in
13 (21.3%) of the 61 cases. On the panoramic radio-
graphs, the antral septa were located in the following
regions (Table 2): 12 septa (77.5%) were found in the
anterior portion of the alveolar recess that corre-
sponded to the premolar region; five septa were
located in the middle portion (corresponding to the
region of the first molar); and no septa were clinically
found in the posterior portion. Figure 1 shows a
panoramic radiograph with an antral septum. Figure 2
shows the corresponding operative field after infrac-
ture of the facial antral wall, confirming the presence
of a septum on the antral floor.
In the anatomic study (group 2), 13 (31.7%) of 41
maxilla had 15 antral septa (36.6%) on the sinus floor.
Eleven maxillae (26.8%) had one septum (Fig 3)
whereas two maxillae (4.8%) had two septa. Eleven
septa were in the anterior portion, two in the middle
portion, and one in the posterior portion (Table 2). In
maxillae exhibiting two septa, they were located in
the anterior and middle portions.
RadiogmphicaIIy - Intraoperatively
Presence of septa: Correct positive: 9 (14.7%)
Positive 14 (22.9%) False positive: 5 (8.2%)
Negative 47 ~77.1%) Correct negative: 39 (63.9%)
False negative: 8 (13.1%)
JOHN D. STOVER 671

and the floor is augmented with either autologous or 5. Ulm CW, Solar P, Krennmair G, et al: Incidence and suggested
surgical management of septa in sinus lift procedures. Int J Oral
xenogeneic bone grafts after infracturing of the facial
Maxillofac Implants 10:462, 1995
antral wall. 12:14,17If septa are present on the sinus floor 6. Belcher HE, Zammit-Maepel I, Rose JD: A sinus within a sinus.
or on the anterior antral wall, they can complicate Am J Roentgen01 162:237, 1994
both the inversion of the bone plate and elevation of 7. Kennedy DW, Zinreich SJ, Rosenbaum AE, et al: Functional
endoscopic sinus surgery: Therory and diagnostic evaluation.
the sinus membrane. If septa are encountered on the Arch Otolaryngol Head Neck Surg 111:576, 1985
antral floor, Boyne and JameslO recommend cutting 8. Wigand ME, Steiner W, Jaumann MP: Endonasal sinus surgery
them with a narrow chisel and removing them with a with endoscopical control: From radical operation to rehabilita-
tion of the mucosa. Endoscopy 10:255, 1978
hemostat so that the bone graft can be placed over the 9. NIafee MF, Chow JM, Meyers R: Functional endoscopic sinus
entire antral floor without interruption. Tidwell et aI’* surgery: Anatomy, CT-screening, indications and complica-
subdivided the facial bony wall into an anterior and a tions. Am J Radio1 160:735, 1993
10. Boyne P, James RA: Grafting of the maxillary sinus floor with
posterior part and inverted both trap doors. Septa left autogenous marrow and bone. J Oral Maxillofac Surg 38: 113,
in situ can be removed after such preparation. Septa 1980
left in situ may be a risk factor, because they might 11. Misch CE: Maxillary sinus augmentation for endosteal implants:
Organized alternative treatment plans. Int J Oral Implantol4:49,
tear the Schneiderian membrane during elevation.2,18s19 1987
Among possible complications of perforation of the 12. Tidwell JK, Blijdorp PA, Stoelinga PJ, et al: Composite grafting
sinus membrane is development of maxillary sinus- of the maxillary sinus for placement of endosteal implants. Int J
Oral Maxillofac Surg 21:204, 1992
itis.l*,‘9 Therefore, a modification of the conventional 13. Betts NJ, Miloro M: Modification of the sinus lift procedure for
surgical technique is required when septa are pre- septa in the maxillary antrum. J Oral Maxillofac Surg 52:332,
sent.13 Because atrophic septa are well discernible on 1994
14. Jensen OT, Greer R: Immediate placement of osseointegrating
CT, a routine presurgical CT examination is useful to implants into the maxillary sinus augmented with mineralized
obtain precise knowledge of their localization to avoid cancellous allograft and Gore-Tex: Second-stage surgical and
unexpected complications during surgery. histologic findings, in Laney WR, Tolman DE (eds): Tissue
Integration in Oral Orthopedic and Maxillofacial Reconstruc-
tion. Chicago, IL, Quintessence, 1992, pp 321-333
15 Farmand M: Horse-shoe sandwich osteotomy of the edentulous
References maxilla as a preprosthetic procedure. J Maxillofac Surg 14:238,
1. Chanavaz M: Maxillary sinus: Anatomy, physiology, surgery, 1986
and bone grafting related to implantology-Eleven years of 16. Sailer HF: A new method of inserting endosseous implants in
surgical experience (1979.1990). J Oral Implant01 16:199, 1990 totally atrophic maxillae. J Craniomaxillofac Surg 17:299, 1989
2. Cawood JI, Howell RA: A classification of the edentulous jaws. 17
_ I .
Wood RM, Moore DL: Grafting of the maxillary sinus with
Int J Oral Maxillofac Surg 17:232, 1988 intraorally harvested autogenous bone prior to implant placema-
3. Vinter I, Krmpotic-Nemanic J, Jalsovec D: Does the alveolar net. Int J Oral Maxillofac Implants 3:209, 1988
process of the maxilla always disappear after loss of teeth? 18. Quiney RE, Brimble E, Hodge M: Maxillary sinusitis from dental
Laryngorhinootologie 72:606, 1993 osseointegrated implants. J Laryngol Otol104:333, 1990
4. Underwood AS: An inquiry into the anatomy and pathology of 19. Ueda M, Kaneda T: Maxillary sinusitis caused by dental im-
the maxillary sinus. J Anat Physiol44:354, 1910 plants: Report of two cases. J OralMaxillofac Surg 50285, 1992

J Oral Max~llofac Surg

.57:671472, 1999

Discussion
The Incidence, location, and Height typically located. Particularly, the surgeon should know
where septa of sticient height to complicate sinus lift
of Maxillary Sinus Septa in the
procedures are typically located. Radiographic information (pan-
Edentulous and Dentate Maxilla ommic and computed tomography [m) is used to predict sinus
volume and degree of septation. Also of diagnostic and predic-
John D. Stover, DDS, PbD, MD
tive importance is information related to patient age, degree of
Assistant Professor, Department of Oral and Maxillofacial Surgery,
edentulism, and history and timing of past tooth loss.
Louisiana State University Medical Center, New Orleans,
This article provides important information with respect
Louisiana; e-mail:JSTOVE@lSUSD.lSUMC.EDU
to our ability to make intelligent predictions about the
For surgeons placing maxillary endosseous dental im- location and size of septa encountered in sinus lift proce-
plants and performing maxillary sinus lift procedures, knowl- dures. The authors have very nicely expanded an earlier
edge of general anatomic variations in the sinus is important. article published in 1995 examining the incidence, location,
The surgeon should know where, if present, sinus septa are and height of Underwood’s maxillary sinus septa using