The Journal of Laryngology & Otology (2006), 120, 200–204.

Main Article
# 2005 JLO (1984) Limited
doi:10.1017/S0022215105007383
Printed in the United Kingdom

Effect of nasal septal deviation on total ethmoid

cell volume
AHMET KEMAL FIRAT , MD, MURAT CEM MIMAN , MD*, YEZDAN FIRAT , MD*,
MUAMMER KARAKAS , MD, ORHAN OZTURAN , MD*, TAYFUN ALTINOK , MD

Abstract
Background: The aim of this study was to evaluate the effect of nasal septal deviation (NSD) on ethmoid
cell volume and to determine whether there was any correlation between NSD grade and ethmoid cell
volume.
Methods: Forty computerized tomography (CT) scans from patients with rhinosinusitis symptoms with
NSD were evaluated. Septal deviations were classified into three groups according to the degree of
deviation on CT. Ethmoid cell volumes were measured and the relationship between NSD and
ethmoid cell volume was investigated.
Results: There was a moderate but significant negative correlation between the septal deviation angle
and the percentage of the ethmoid cell volumes ( p ¼ 0.001, r ¼ 20.5152, r2 ¼ 0.2654). Total ethmoid cell
volume on the ipsilateral side compared with the contralateral side was found to decrease as the degree of
NSD increased.
Conclusions: Nasal septal deviation affects the total ethmoid cell volume of the nasal cavity. The results
of our study underline the role of ethmoid cell volume in the compensation mechanism equalizing the
nasal cavity airflow changes due to NSD.

Key words: Nasal Septum; Ethmoid Sinuses; Computed Tomography

Introduction convex side. This difference in airflow between the

Nasal septal deviation (NSD), either congenital or two nasal cavities initiates the differentiation of
acquired, is one of the most common diagnoses in the paraseptal structures.6 Nasal cavity volume
daily otolaryngology practice. Prenatal, natal or post- decreases on the ipsilateral (convex) side of the
natal traumas, external traumas, growth anomalies of septal deviation and increases on the contralateral
the maxilla and incisors, facial anomalies, congenital (concave) side. Paraseptal structural changes occur
deformities, breathing through the mouth, applying in an attempt to compensate for such volume
pressure to the palate with the tongue, and finger- changes. The lateral nasal wall and the middle turbi-
sucking are among the common causes of septal nate play major roles in this compensation mechan-
deviation.1 The incidence of NSD is lower in ism. The increased incidence of Haller and agger
neonates than in adults, showing an increasing nasi cells, pneumatization of uncinates and turbi-
trend with age.1 – 5 Embryologically, the ethmoid nates, and hypertrophy of turbinates have been
cells appear after nasal septal development; their reported previously as a consequence of NSD.1,7,8
development starts at 11 – 12 weeks with epithelial To the best of our knowledge, changes in ethmoid
invagination, and 94 per cent of the ethmoid cell volume in association with NSD have not pre-
cells are present at birth. Active pneumatization of viously been investigated. Elahi et al. 7 reported a
the ethmoid cells occurs between 0 –4 and 8 – 12 positive correlation between the angle of nasal devi-
years. Pneumatization continues until 20 years ation and ethmoid bulla prominence but did not
of age. undertake volumetric measurement. In this study,
The nasal structures attempt to equalize the we aimed to evaluate the effect of NSD on ethmoid
amount of air passing through the nasal cavities. cell volume and to determine if there was any corre-
An NSD jeopardizes nasal aerodynamics and lation between the grade of the NSD and the
diminishes the amount of nasal airflow at the ethmoid cell volume.

From the Departments of Radiology and
Otorhinolaryngology, Inonu University School of Medicine, Malatya, Turkey.
Accepted for publication: 5 August 2005.

200
EFFECT OF NASAL SEPTAL DEVIATION ON TOTAL ETHMOID CELL VOLUME 201

Materials and methods rule of Ikeda et al.:9

In this prospective study, we evaluated a total of
40 patients with NSD and sinusitis symptoms, who Volume ¼ (A1 þ A2)h=2 þ (A2 þ A3)h=2 þ   
underwent paranasal computerized tomography
(CT) scans (24 women, 16 men; mean age of where A1, A2 . . . ¼ cross-sectional area of ethmoid
35 + 13.62, ranging between 20 and 70 years old). cell in cm2 and h ¼ slice thickness.
The following exclusion criteria were chosen: prior Normal distribution of the groups and homogeneity
paranasal surgery, history of sinonasal trauma, of variances were analysed by the Kolmogorof–
maxillofacial anomalies, nasal polyposis disease, Simirnov test and the Levene test, respectively. The
mass lesions of the sinonasal region and acute rhino- unpaired t-test was used in the analysis of difference
sinusitis attacks. between the right and left nasal cavity volumes.
Computed tomography examinations (Philips One-way analysis of variance (ANOVA) and the
Medical Systems, Secura, Netherlands) were per- further Scheffe test were used to compare the percen-
formed in the prone position with the head hyper- tage of ipsilateral and contralateral nasal volumes
extended. Coronal CT images with contiguous, (Vipsilateral, Vcontralateral) and the differences
2 mm-thick slices angled perpendicular to the hard between the ipsilateral and contralateral volumes
palate were obtained. The bone algorithm with (Vcontralateral 2 Vipsilateral) between the groups.
wide window (300/2350 HU) was used to obtain The Pearson correlation coefficient test was preferred
appropriate images. The image files were transferred for evaluating the correlation between the angles of
to an Easy Vision computerized workstation (Philips septal deviation and the percentage of bilateral
Medical Systems) for evaluation of NSD and volumes (Vipsilateral/Vcontralateral). The level of
ethmoid cell volumes. statistical significance was determined as p , 0.05.
The direction of the septal deviation was noted
and the angle of deviation calculated utilizing a pro-
tractor by measuring the angle between the apex of Results
the septal deviation and the plate crossing the Nasal septal deviations were towards the right in
anterior nasal spine and the crista Galli.7,8 Patients 25 patients and towards the left in the remaining
were grouped by their NSD angles according to 15 patients. Nasal septal deviation angles were
the Elahi et al. grading system: Group I, found to range between 4.1 and 20.28 (mean
NSD 4 98; group II, 9 , NSD 4 158, and group 10.46 + 4.218). Patients were grouped according to
III, NSD . 158. the severity of their NSD, as previously described,
The ethmoid cell volumes of the 80 nasal cavities to give the following: group I, 20 patients; group II,
belonging to the 40 patients were evaluated using 11 patients; group III, nine patients. The NSD angle
the volume measurement system of the Easy characteristics of the groups are described in Table I.
Vision computer workstation. Anterior and pos- The total ethmoid cell volumes, independent of
terior ethmoid cells, including infraorbital and group and side of NSD, were 5.83 + 1.57 cm3 and
supraorbital cells, agger nasi and Onodi cells, were 5.69 + 1.82 cm3, on the right and left side, respect-
taken into account in calculating total ethmoid cell ively. There was no statistically significant difference
volumes. Ethmoid cell variations, such as supraorbi- between the ethmoid cell volumes of each side
tal and infraorbital cells, were included in the assess- (p . 0.05). On the convex (ipsilateral) side the
ment of the effects of septal deviation on total total ethmoid cell volume was 5.42 + 1.64 cm3 and
ethmoid cell volume. Including these volumetric on the concave (contralateral) side it was
variations in the calculation of total ethmoid cell 6.10 + 1.69 cm3. No statistically significant differ-
volume is logical as an NSD could cause these cells ence was noted between the total ethmoid volumes
to appear. Volume was calculated by defining the on the ipsilateral and contralateral sides of the
borders of the ethmoid cell boundaries in each NSD (p ¼ 0.075). The ethmoid volume character-
slice, and margins were determined by a semi- istics on the ipsilateral and contralateral sides of
automated system. Border points were set and the NSD for each group are given in Table II.
these points connected with parabolic lines genera- Within each group, there was no significant differ-
ted by the software programme. After determining ence between the total ethmoid cell volumes on
the area of the ethmoid cells in each slice, the total the ipsilateral and contralateral sides of the NSD
volume was calculated according to the trapezoid (p . 0.05).

TABLE I
NASAL SEPTAL DEVIATION: GROUP CHARACTERISTICS

Group I Group II Group III Total

n 20 11 9 40
% 50 27 23 100
NSD angle mean + SD (8) 7.00 + 1.32 11.53 + 1.50 16.83 + 1.57 10.46 + 4.21
NSD angle range (8) 4.1 –8.8 9.1– 14.1 15.7– 20.2 4.1 –20.2

Group I, NSD 4 98; group II, 9 , NSD 4 158, and group III, NSD . 158.
202 A K FIRAT, M C MIMAN, Y FIRAT et al.

TABLE II
TOTAL ETHMOID CELL VOLUME: GROUP CHARACTERISTICS

Group I Group II Group III

3

TECV ipsilateral (cm ) 5.40 + 1.88 5.36 + 1.18 5.55 + 1.74
TECV contralateral (cm3)
5.80 + 1.88 6.05 + 1.28 6.82 + 1.66
Vipsilateral/Vcontralateral† 0.93 + 0.12 0.89 + 0.07 0.80 + 0.09
Vcontralateral 2 Vipsilateral‡ (cm3) 0.40 + 0.70 0.69 + 0.48 1.26 + 0.33


p . 0.05 for differences between ipsilateral and contralateral total ethmoid cell volume (TECV) within all groups. †p ¼ 0.013 for
difference between group I and other groups. ‡p ¼ 0.028 for difference between group III and other groups.

These data could be explained by the extreme

variation in patients’ total ethmoid cell volumes,
which ranged between 1.58 and 9.40 cm3. In order
to eliminate the influence of this variation, the
differences between the sides (Vcontralateral 2
Vipsilateral) and the ratio of ethmoid volumes
between the sides (Vipsilateral/Vcontralateral)
were calculated (Table II). As a percentage of the
total ethmoid cell volume, ipsilateral volumes were
93 per cent of contralateral volumes in group I, 89
per cent of contralateral volumes in group II and
80 per cent of contralateral volumes in group III
(Table II). When comparing the groups according
to the percentage of the volumes, a statistically sig-
nificant difference ( p ¼ 0.013) was found, with the
volume ratio of group I being higher than that of
the other groups. When the differences between
the total ethmoid cell volumes were compared
(Vcontralateral 2 Vipsilateral), statistically signifi-
cant differences were also found between the
groups (p ¼ 0.028), with higher values in group III.
The correlation between the septal deviation angle
and the ratio of the ethmoid volumes (Vipsilateral/
Vcontralateral) is shown on Figure 1; there was
moderate but significant negative correlation
between these variables (p ¼ 0.001, r ¼ 20.5152,
r2 ¼ 0.2654). The total ethmoid cell volume on the
ipsilateral side compared with the contralateral side
was found to decrease in correlation with the
increase in degree of NSD.
Coronal CT images from patient group III, reveal-
ing the NSD angle and the difference in total
ethmoid cell volumes between the sides of the
NSD, are shown in Figure 2.

FIG . 2
Coronal paranasal computed tomography images of a patient
FIG . 1 from group III. (a) nasal septal deviation angle measurement
The correlation between nasal septal deviation angle and the method; (b) evaluation of ethmoidal cell volumes. The
Vipsilateral/Vcontralateral (Vipsi/Vcontra) ratio of the total ethmoidal cell area difference between the nasal cavities was
ethmoid cell volumes (p ¼ 0.001, r ¼ 20.5152, r2 ¼ 0.2654). noted.
EFFECT OF NASAL SEPTAL DEVIATION ON TOTAL ETHMOID CELL VOLUME 203

Discussion patients with NSD and sinusitis. We hope that this

The incidence and type of anatomic variations of will stimulate new research on this topic. Further
paraseptal structures related to NSD have been com- studies investigating the incidence of complications
prehensively described in the literature.1,7,8 The pro- of endoscopic sinus surgery and of infection in the
minence of the ethmoid bulla on the contralateral nasal cavities with the smaller ethmoid cell volume
side of the NSD was found to be the most frequent and the importance of ethmoid volume in the
ethmoid cell variation associated with NSD.7 We education by virtual endoscopic sinus surgery. Pre-
reviewed the literature and found no study analysing operatively evaluated, exact anatomic data including
the ethmoid cell volumes in patients with NSD; such volumetric measurements could be expected to
evaluation can be valuable in pre-operative assess- improve the success rates of functional endoscopic
ment and in determining an appropriate surgical sinus surgery by giving the surgeon better knowledge
approach. of the patient’s individual anatomy. As a result, this
will decrease the incidence of complications.
The relationship between the structural variations
In this study, ethmoid cell volume was estimated
associated with septal deviation, and chronic sinusitis
by a semi-automatic, computerized measurement
had been already investigated with qualitative and
system, which could also measure the volume of
quantitative evaluations. Stammberger et al. 10 pro-
paranasal sinuses, nasal cavity and paraseptal struc-
posed agger nasi cells, middle concha and uncinate
tures. This method allows no gap between slices, so
process variations and Haller cell and septal devi-
volumetric estimation is closer to the real volume.
ations as a etiologic factors in the development of
It is, however, time-consuming and has some restric-
sinus disease. However, these authors reached no
tions. For example, there were problems in delinea-
consensus about the role of septal deviation in the
tion of the borders of ethmoid cells, especially at
pathogenesis of chronic sinusitis. Collet et al. 11
the boundaries adjacent to the superior meatus,
reviewed literature of the past 20 years investigating
fronto-ethmoidal recess and sinus lateralis. In order
the relationship between NSD and chronic sinusitis
to overcome these problems, a radiologist and an
and found few studies revealing a statistically signifi-
otolaryngologist worked together to establish con-
cant association (they also stated that the results and
sensus on the delineated ethmoid cell borders.
methodologies of those studies were contradictory).
In cases of disagreement, the previous coronal
This conclusion is supported by recent studies
slices and saggital reconstructions were examined.
showing no evidence for a contribution from NSD
Despite these disadvantages, this system is more
to rhinosinusitis disease (incidences of NSD were
accurate for paranasal sinus measurements than
the same in rhinosinusitis patients and the control
applying an elipsoid volume calculation formula,
group).11,12 However, anatomic variations and NSD
which requires only the three dimensions of the
located near the osteomeatal area may lead to
ellipsoid:
obstruction and play a role in the pathogenesis of
chronic sinusitis.12
It can be speculated that the location of anatomic Sinus volume index ¼ 12A  B  C
variations is more important than their presence.
The volumetric increase on the concave side of an where A ¼ height, B ¼ length and C ¼ width.14
NSD could result in osteomeatal complex (OMC) The similarity of the total ethmoid cell volume cal-
obstruction. In our study, ethmoid cell variations culations reported in our study and those reported in
were not studied separately but were included previous studies supports the accuracy and reliability
in the calculation of the total ethmoid cell volume. of our measurement method.
The volumes of ethmoid cell variations obstruc-
ting the OMC could be separately evaluated;
however, this was outside the scope of our current
Conclusion
research.
The total ethmoid cell volume differences on Nasal septal deviation affects the total ethmoid cell
either side of an NSD are a routinely observed clini- volumes of the nasal cavity. In this study, ethmoid
cal issue in daily rhinology practice. However, we cell volume was estimated by a semi-automatic, com-
could find in the literature no volumetric investi- puterized measurement system. The angle of septal
gation clarifying the role of NSD-related ethmoid deviation was calculated utilizing a protractor by
cell volume differences in the pathogenesis and treat- measuring the angle between the apex of the septal
ment of rhinosinusitis. There were few studies inves- deviation and the plate crossing the anterior nasal
tigating the relationship between chronic sinusitis spine and crista Galli. The results of our study eluci-
and variations and mucosal thickening of ethmoid date the role of ethmoid cell volume in the compen-
cells. Bolger et al. 13 studied paranasal CT scans of sation mechanism equalizing the nasal cavity airflow
patients with chronic sinusitis, investigating mucosal changes due to NSD, especially when severe (this
thickening of the ethmoid infundibulum and the effect has been previously reported for other para-
anterior ethmoidal cells. Aktas et al. 1 reported a septal structural changes).
greater incidence of ethmoidal sinusitis on the Studies with larger patient and control groups are
same side of the septal deviation. However, needed to eliminate the effect of variations of the
ethmoid cell volume was not evaluated by these total ethmoid cell volumes. Further studies could
studies. Our study is unique in presenting primary investigate the consequences of such ethmoid cell
data on changes in total ethmoid cell volume in volume variation for endoscopic sinus surgery.
204 A K FIRAT, M C MIMAN, Y FIRAT et al.

8 Uygur K, Tuz M, Dogru H. The correlation between septal

. This study correlates ethmoid cell volume deviation and concha bullosa. Otolaryngol Head Neck Surg
2003;129:33– 7
with degree of nasal septal deviation. The 9 Ikeda A, Ikeda M, Komatsuzaki A. A CT study of the
authors assessed 40 sinus CT scans from course of growth of the maxillary sinus: normal subjects
patients with rhinosinusitis and subjects with chronic sinusitis. ORL J Otorhinolaryn-
gol Relat Spec 1998;60:147– 52
. Septal deviation was found to be significantly 10 Stammberger H, Wolf G. Headaches and sinus disease: the
related to ethmoid volume, with the side of endoscopic approach. Ann Otol Rhinol Laryngol Suppl
the septal deviation being associated with a 1988;134:3 –23
smaller ethmoid volume 11 Collet S, Bertrand B, Cornu S, Eloy P, Rombaux P.
Is septal deviation a risk factor for chronic
. These findings are relevant in the pre- sinusitis? Review of literature. Acta Otolaryngol Belg
operative planning of surgery in patients with 2001;55:299– 304
rhinosinusitis and nasal septal deviation 12 Jones NS, Strobl A, Holland I. A study of the CT findings
in 100 patients with rhinosinusitis and 100 controls. Clin
Otolaryngol 1997;22:47– 51
13 Bolger WE, Butzin CA, Parsons DS. Paranasal sinus
bony anatomic variations and mucosal abnormalities:
CT analysis for endoscopic sinus surgery. Laryngoscope
References 1991;101:56– 64
1 Aktas D, Kalcioglu MT, Kutlu R, Ozturan O, Oncel S. The 14 Barghouth G, Prior JO, Lepori D, Duvoisin B, Schynder P,
relationship between the concha bullosa, nasal septal devi- Guidinchet F. Paranasal sinuses in children: size evaluation
ation and sinusitis. Rhinology 2003;41:103 –6 of maxillary, sphenoid, and frontal sinuses by magnetic res-
2 Podoshin L, Gertner R, Fradis M, Berger A. Incidence and onance imaging and proposal of volume index percentile
treatment of deviation of nasal septum in newborns. Ear curves. Eur Radiol 2002;12:1451– 8
Nose Throat J 1991;70:485 –7
3 Min YG, Jung HW, Kim CS. Prevalence study of nasal Address for correspondence:
septal deformities in Korea: results of a nation-wide Yezdan Firat, MD,
survey. Rhinology 1995;33:61 –5 Inonu Universitesi,
4 Subaric M, Mladina R. Nasal septum deformities in chil- Turgut Ozal Tip Merkezi,
dren and adolescents: a cross sectional study of children KBB Hastaliklari Anabilim Dali,
from Zagreb, Croatia. Int J Pediatr Otolaryngol 44280 Malatya, Turkey.
2002;63:41 – 8
5 Yildirim I, Okur E. The prevalence of nasal septal devi-
ation in children from Kahramanmaraş, Turkey. Int J
Pediatr Otolaryngol 2003;67:1203– 6 Fax: 90 (422) 341 0128
6 Stammberger H. Functional Endoscopic Sinus Surgery, E-mail: [email protected]
the Messerklinger Technique. Philadelphia: BC Decker,
1991 Dr A K Firat takes responsibility for the integrity of the
7 Elahi MM, Frenkiel S, Fageeh N. Paraseptal structural content of the paper.
changes and chronic sinus disease in relation to the Competing interests: None declared
deviated septum. J Otolaryngol 1997;26:236– 40