A radiological anatomic study of the cribriform plate compared with constant structures.

Abstract

Understanding of the anterior skull base anatomy is crucial to avoid intracranial violations during endoscopic surgery. The aims of this study were to define the normative data about cribriform plate depth and the relationship between this dimension and the measurements of the adjacent anatomical structures such as middle turbinate length, maximal vertical orbital height and distance between the ethmoid roof and the nasal floor.
Paranasal computerized tomographic scans of 136 healthy adults were included into the study. The cribriform plate depth compared to the ethmoid roof and the adjacent anatomical structures mentioned above were measured bilaterally.
The maximal vertical orbital height was detected as the most constant anatomic measurement. We found the mean level difference between the ethmoid roof and the cribriform plate as 6.1 +/- 2.3 (range 1-12 mm) on the left side and 6.1 +/- 2.2 (1-15 mm) on the right side. The middle turbinate was significantly longer in the Keros Type I group than in the other groups (p<0.05). Furthermore, the distance between the ethmoid roof and the nasal floor was lowest in the Keros Type I group (p<0.01). The distance between the ethmoid roof and the nasal floor was statistically higher in Keros group 3 among all groups (p<0.01). The deeper the cribriform plate, the higher the nasal cavity.
To the best of our knowledge, our study has a unique feature by including the data of the constant anatomical structures comparing with the cribriform plate depth. Since in the group with excessive cribriform plate depth, the middle turbinate was short, care should be taken especially during middle turbinate resections.

Full text

Page 1

Rhinology, 43, 225-229, 2004
*Received for publication: February 20, 2004; accepted: July 7, 2004
INTRODUCTION
The surgeons dealing with endoscopic sinus surgery should
understand the crucial complex anatomy of the anterior cranial
base to avoid the intracranial violation during endoscopic
surgery. The cribriform plate forms the vault of the nasal cavi-
ty and is located lower than the ethmoid roof. The medial
extension of the orbital plate of the frontal bone forms the eth-
moidal roof. The ethmoidal roof joins to the lateral lamella of
the cribriform plate medially and ascends like a dome laterally.
The greater difference between the depth of the cribriform
plate and the fovea ethmoidalis, the wider relationship
between the uppermost ethmoidal cells and endocranium will
be. The ethmoid roof is formed by the frontal bone and is
located 7- 16 mm superiorly to the cribriform plate (Keros,
1962; Lang, 1998). The relationship between the olfactory fossa
and the ethmoid roof was classified into three types by Keros
(1962). According to the Keros classification, the difference
between the height of the cribriform plate and ethmoid roof is
1 to 3 mm in Keros Type I, 4-7 mm in Keros Type II and more
than 8 mm in Keros Type III (Keros, 1962).
The computerized tomographic (CT) evaluation of the
paranasal sinuses providing detailed knowledge of the
paranasal sinus anatomy is a must and the gold standard
method prior to endoscopic surgery. The coronal CT of the
paranasal sinuses additionally evaluates the anterior skull base
and the relationships between the paranasal sinuses and the
central nervous system effectively.
The aims of this study were to define the normative data about
the cribriform plate depth in comparison to the ethmoid roof and
the relationship between this dimension and the measurements
of the adjacent anatomical structures such as middle turbinate
length, maximal height of the orbit at the vertical plane and dis-
tance between the ethmoid roof and the nasal floor.
PATIENTS AND METHODS
Paranasal sinus CT scans of 136 (75 females, 61 males) healthy
Background: Understanding of the anterior skull base anatomy is crucial to avoid intracra-
nial violations during endoscopic surgery. The aims of this study were to define the normative
data about cribriform plate depth and the relationship between this dimension and the mea-
surements of the adjacent anatomical structures such as middle turbinate length, maximal
vertical orbital height and distance between the ethmoid roof and the nasal floor.
Patients and Methods: Paranasal computerized tomographic scans of 136 healthy adults
were included into the study. The cribriform plate depth compared to the ethmoid roof, and
the adjacent anatomical structures mentioned above were measured bilaterally.
Results: The maximal vertical orbital height was detected as the most constant anatomic
measurement. We found the mean level difference between the ethmoid roof and the cribri-
form plate as 6.1 ± 2.3 (range 1-12 mm) on the left side and 6.1 ± 2.2 (1-15 mm) on the
right side. The middle turbinate was significantly longer in the Keros Type I group than in the
other groups (p<0,05). Furthermore, the distance between the ethmoid roof and the nasal
floor was lowest in the Keros Type I group (p<0,01). The distance between the ethmoid roof
and the nasal floor was statistically higher in Keros group 3 among all groups (p<0,01). The
deeper the cribriform plate, the higher the nasal cavity.
Conclusion: To the best of our knowledge, our study has a unique feature by including the
data of the constant anatomical structures comparing with the cribriform plate depth. Since
in the group with excessive cribriform plate depth, the middle turbinate was short, care
should be taken especially during middle turbinate resections.
Key words: cribriform plate, ethmoid roof, orbit, nasal cavity height, middle turbinate length,
constant anatomic structures.
SUMMARY
A radiological anatomic study of the cribriform
plate compared with constant structures*
Gulnur Erdem1, Tamer Erdem2, Murat Cem Miman2, Orhan Ozturan2
1
Radiology Department, Inonu University, School of Medicine, Malatya, Turkey
Otorhinolaryngology Department, Inonu University, School of Medicine, Malatya, Turkey
2

Page 2

226
Erdem et al.
adult volunteers were included in the study. The mean age was
34.1 years old (between 18-71 years). The mean age of the 61
males was 35.2 (range: 18-71) and 75 females was 33.2 (range:
18-65). Patients with massive nasal polyposis, benign or malign
tumors, previous nasal surgery or trauma were excluded.
CT examinations were performed on General Electric CT
ProSpeed and Philips CT Secura. Scans were obtained by 3
mm increments from the glabella to the posterior boundary of
the sphenoid sinus in the coronal plane. The paranasal coronal
CT scans were evaluated for measurements. The cribriform
plate depth compared to the ethmoid roof, maximal height of
the orbit in vertical plane, the middle turbinate length, and the
distance between the ethmoid roof and the nasal floor at the
anterior ethmoidal region were measured bilaterally. As
already determined by Meyers and Valvassori (1998), the hori-
zontal foveal plane, which passes medially between the point
of the junction roof and the medial orbital walls laterally
through the orbit, was used for the determination of the height
of the fovea ethmoidalis (Figure 1). Then, the data of the crib-
riform plate depth were classified according to the Keros classi-
fication system (Keros, 1962).
Statistical methods
The possible relationships between the cribriform plate depth
and the other anatomical measurements were analyzed statisti-
cally by the unpaired t-test. The significance of differences of
the data from male subjects and female subjects; from the left
and right sides and from the Keros groups was compared sta-
tistically by the unpaired t-test with equal variances. The inci-
dences of the asymmetry in the Keros groups were also com-
pared using the chi-square test. In all analyses, the SPSS 5.0
statistical analysis program was utilized and p<0,05 was
defined as the level of significance.
RESULTS
Since, no statistical significant gender difference between the
measurements obtained from females and males of the cribri-
form plate depth, middle turbinate length, distance between
the ethmoid roof and the nasal floor and orbital height
(p>0,05) were found, the results were evaluated for the overall
data of the 136 cases. The mean values and standard deviations
of the measurements are given in Table I. The mean values
and standard deviations of the measurements are given in
Table 2 according to the Keros classification. No statistical sig-
nificant differences were found between the right and left
sides’ cribriform plate depth, orbital height, distance between
the ethmoid roof and the floor of nasal cavity and middle
turbinate lengths after the Keros classification (p>0,05).
The asymmetry of the opposite side cribriform plates were also
evaluated. Table 3 shows the incidence of asymmetry of our
subjects grouped according to the Keros classification. For
Table 1. The measurements of some anatomic structures using paranasal sinus computed tomography in coronal position. Values are in millimeters.
Cribriform plate
depth
The length of the
middle turbinate
The maximal height
of the orbit
Distance between
the ethmoid roof
and the nasal floor
Left side
6.1±2.3
Right side
6.1±2.2
Left side
26.2±3.9
Right side
26.3±4.4
Left side
39.2±3.8
Right side
39.1±3.6
Left side
52.3±5.1
Right side
51.8±5.2
Table 2. The measurements of some anatomic structures using paranasal sinus computed tomography in the coronal position according to the Keros
classification. (n: number of cases). Values are in millimeters
The length of the
middle turbinate
The maximal height
of the orbit
Distance between
the ethmoid roof
and the nasal floor
Left side Right sideLeft side Right sideLeft side Right side
KEROS I [n:11 (%8.1)]
KEROS II [n:81(%59.6)]
KEROS III [n:44(%32.3)]
27.7±3.8
26.1±3.3
26.0±4.8
28.5±4.2
25.9±4.4
26.5±4.4
39.9±8.9
39.0±3.3
39.2±2.6
39.7±8.3
39±3.2
39.3±2.4
47.8±3.7
51.5±4.6
54.9±5.1
47.1±3.0
51.2±4.8
54.2±5.1
Table 3. The distribution of asymmetry seen in our cases according to
the Keros classification (**p<0.01).
1 mm2 mm more than 2 mmTotal
KEROS I
KEROS II
(Asymmetry
range 1-4 mm)
KEROS III
(Asymmetry
range 1-6 mm)
4-- 4 (%36.7)
20156 41 (%50.6)
312**1025 (%56.8)

Page 3

Radiology of the cribriform plate
227
more than 2 mm asymmetry, the Keros 3 group showed a sta-
tistical difference when compared to the other groups (p<0,01).
Since, no difference was found between the data of the oppo-
site sides obtained from the groups, they have been unified at
each Keros group (Table 4). The maximal orbital height in the
vertical plane was detected as the most constant anatomic
measurement. The middle turbinate was significantly longer in
Keros Type I group compared to the other groups (p<0,05).
Furthermore, the distance between the ethmoid roof and the
nasal floor was the lowest in the Keros Type I group (p<0,01).
The distance between the ethmoid roof and the nasal floor was
statistically higher in Keros group 3 among all groups (p<0,01).
DISCUSSION
The frontal bone is thick and dense in the ethmoid roof area.
But medially, the transition from the thick bony part of the
frontal bone to the thinner lamellae of the ethmoid is observed
(Stammberger, 1991). The thinner medial wall of the ethmoid
roof is formed by the lamina lateralis of the lamina cribrosa.
Because of the lateral lamella of the lamina cribrosa is the
thinnest part of the ethmoid roof, the intracranial violation risk
is higher in that area during endoscopic surgery. The lateral
lamella is seen as a long segment in cases with a deeper cribri-
form plate. Only paranasal sinus CT-scans taken in the coronal
plane can provide adequate information about the patient’s
individual conditions and variations (Stammberger, 1991).
Keros introduced a classification system for the depth of the crib-
riform plate in comparison to the ethmoid roof in 1962 (Keros,
1962). The relationship between the olfactory fossa and the eth-
moid roof was classified into three types by Keros (1962). Type I
relationship shows that the olfactory fossa is flat and the lateral
lamella is higher (Figure 2). In Type II, the lateral lamella is high-
er and the olfactory fossa is deeper (Figure 3). The Type III,
characteristics are a considerable high ethmoid roof and a deeper
olfactory fossa (Figure 4). The Type III relationship is the most
dangerous type for endoscopic sinus surgery because of the high
penetration possibility through the lateral lamella of the lamina
cribrosa. The more asymmetry in ethmoid roof height, the high-
er incidence of intracranial penetration is on the side where the
position of the ethmoidal roof is lower (Dessi et al., 1994).
The variability in the depth of the cribriform plate may be
Table 4. The measurements of some anatomic structures using paranasal sinus computed tomography in the coronal position according to the Keros
classification. (n: number of sides, * p<0.05, **p<0.01). Values are in millimeters.
The length of the
middle turbinate
The maximal height
of the orbit
Distance between
the ethmoid roof
and the nasal floor
KEROS I (n:22)
KEROS II (n:162)
KEROS III (n:88)
*28.1±3.9
26.0±3.9
26.2±4.6
39.8±8.4
39.0±3.2
39.2±2.5
**47.5±3.3
51.4±4.7
**54.6±5.1
Figure 1. Anatomic measurements performed in our report are shown
on a drawing paranasal sinus tomography obtained in the coronal posi-
tion.
A-B: Maximal height of the orbit in the vertical plane
C-D: Olfactory fossa depth
E-F: Middle turbinate height
G-H: Distance between the ethmoid roof and the nasal floor
Figure 2. Type I cribriform plate according to Keros classification.

Page 4

228
Erdem et al.
related to developmental factors. Krmpotic-Nemanic et al.
(1997) studied 150 skull specimens from fetal to perinatal life,
150 skull specimens from newborn to 30 years of age and 100
skull base specimens from 30 to 90 years of age. The ethmoid
labyrinth development was relatively slow in the first decade of
life. The cribriform lamina first lowers in its frontal part
because of the development of the frontal bone. The lowering
of the cribriform plate continues in an anteroposterior direc-
tion depending on the pneumatisation degree of the frontal
sinus (Krmpotic-Nemanic et al., 1997). The orbital parts of the
frontal bone were formed by the cerebral and orbital lamina.
The cerebral lamina had the tendency to grow more quickly
during the first decade of the life. The intensity of pneumatisa-
tion of the ethmoid labyrinth and the frontal sinus determines
the relations of the subunits of the anterior cranial fossa
(Krmpotic-Nemanic et al., 1997).
The literature contains large series of papers about the cribri-
form plate depth and our results are in accordance with them.
The depth of the cribriform plate compared to the ethmoid
roof was found between 4 mm to 7 mm in 70.2 % of the cases
where the right side was a little bit lower than the left side in
the Keros’ study (Krmpotic-Nemanic et al., 1997). Lang (1998)
observed that the average of the depth was 5.03 mm (range 0-
15.6 mm). Meloni et al. (1992) found that this measurement
was 5.94 mm on average with a range from 1.3 mm to 17 mm
in 106 patients. Arslan et al. (1999) detected a mean difference
between the ethmoid roof and the vault of the nasal cavity of 8
mm (range 2-14 mm) on the right side and of 9.5 mm (range 3-
16 mm) on the left side. We found the mean level difference
between the ethmoid roof and the cribriform plate as 6.1 ± 2.3
(range 1-12 mm) on the left side and 6.1 ± 2.2 (1-15 mm) on
the right side. These data showed no statistical significant dif-
ference between the opposite sides.
In our study, as defined by Meyers and Valvassori (1998), we
used the horizontal foveal plane which passes medially
between the point of the junction roof and the medial orbital
walls laterally through the orbit for determination of the height
of the fovea ethmoidalis.
Besides the amount of the depth of the cribriform plate, the
asymmetry between the opposite sides was found investigated
by several authors. Many reports indicated that the mean cribri-
form plate depth on the right side was lower than the left side
(Keros, 1962; Teatini et al., 1987; Meloni et al., 1992; Dessi et
al., 1994; Lebowitz et al., 2001). Lebowitz et al. (2001) evaluated
the fovea ethmoidalis for height and contour asymmetry on the
left and right sides and found that in the 86 of the 200 (43%)
cases, the foveae of both sides were symmetric in height and
contour. Nineteen cases (9.5%) showed a height asymmetry.
Twelve (63.2%) of them had low fovea on the right side and
seven (36.8%) of them had low fovea on the left side. The mean
difference in height was 2.53 mm (range 1-4mm). Ninety-six
(48%) of the CT scans demonstrated contour asymmetry of the
fovea ethmoidalis. Dessi et al. (1994) found asymmetry in 15 of
150 CT scans (10%) in a prospective study. In 13 cases, the right
fovea ethmoidalis was 2-7 (mean 3.9 mm) millimeters lower
than the left one. They found that the cribriform plate was 2-7
millimeters (mean 5 mm) lower than the ethmoidal roof. But
Jones et al. (2002) measuring cribriform plate height relative to
a supraorbital horizontal line did not find any statistically signif-
icant difference between the right and the left sides. Their
results showed that the mean depth of the cribriform plate to
supraorbital horizontal line was 14.3 mm. The clinical value of
this method for the measurement of the depth of the cribriform
plate to supraorbital horizontal line is limited. The mean level
difference between the ethmoid roof and the cribriform plate is
more meaningful for endoscopic sinus surgery to avoid the
intracranial violation. We found a left and right side asymmetry
between the cribriform plate and the ethmoid roof in 70 of the
136 cases (51,5%) with a statistically insignificant distribution
for the sides (36 at the left and 34 at the right side).
To the best of our knowledge, our study has the unique fea-
ture to include the data of the constant anatomical structures
compared with the cribriform plate depth. The maximal orbital
height in the vertical plane was detected as the most constant
anatomic measurement. The variability of the size of the well-
known and every-day confronted structures was reported for
Figure 3. Type II cribriform plate according to Keros classification. Figure 4. Type III cribriform plate according to Keros classification.

Page 5

Radiology of the cribriform plate
229
the first time. The deeper the cribriform plate, the higher the
nasal cavity. The middle turbinate was longer in height in shal-
low cribriform plates. The discussion of these data from the
practical viewpoint is limited to speculate that nobody’s nose is
the same. But, since in a patient with an excessive cribriform
plate depth, the middle turbinate was short, care should be
taken especially during middle turbinate resections.
REFERENCES
1. Arslan H, Ayd´ ynl´ yoglu A, Bozkurt M, Egeli E (1999) Anatomic
variations of the paranasal sinuses: CT examination for endoscop-
ic sinus surgery. Auris Nasus Larynx 26: 39-48.
2. Dessi P, Moulin G, Triglia JM, Zanaret M, Cannoni M (1994)
Difference in the height of the right and the left ethmoidal roofs: a
possible risk factor for ethmoidal surgery. Prospective study of 150
CT scans. J Laryngol Otol 108: 261-262.
3.Jones TM, Almahdi JMD, Bhalla RK, Lewis-Jones H, Swift AC
(2002) The radiological anatomy of the anterior skull base. Clin
Otolaryngol 27: 101-105.
4.Keros P (1962) Über die praktische Bedeutung der
Niveauunterschiede der lamina cribrosa des ethmoids. Z Laryngol
Rhinol Otol 41: 808-813.
5.Krmpotic-Nemanic J, Vinter I, Judas M (1997) Transformation of
the shape of the ethmoid bone during the course of life. Eur Arch
Otorhinolaryngol 254: 347-349.
6.Lang J (1998) Klinische Anatomie der Nase, Nasenhöhle und
Nebenhöhlen. Grundlagen für Diagnostik und Operation. Thieme
Stuttgart New York.
7.Lebowitz RA, Terk A, Jacobs JB, Holliday RA (2001) Asymmetry
of the ethmoid roof: analysis using coronal computed tomography.
Laryngoscope 111: 2122-2124.
Meloni F, Mini R, Rovasio S, Stomeo F, Teatini GP (1992)
Anatomic variations of surgical importance in ethmoid labyrinth
and sphenoid sinus. A study of radiologic anatomy. Surg Radiol
Anat 14: 65-70.
Meyers RM, Valvassori G (1998) Interpretation of anatomic varia-
tions of computed tomography scans of the sinuses: a surgeon‘s
perspective. Laryngoscope 108: 422-425.
10. Stammberger H (1991) Special endoscopic anatomy of the lateral
nasal wall and ethmoidal sinuses. In Functional Sinus Surgery.
Stammberger H (eds). Philadelphia: BC Decker 49-87.
11. Teatini G, Simonetti G, Salvolini U, Masala W, Meloni F, Rovasio
S, Dedola GL (1987) Computed tomography of the ethmoid
labyrinth and adjacent structures. Ann Otol Rhinol Laryngol 96:
239-250.
8.
9.
Gulnur Erdem, MD
Radiologist, Inonu University, School of Medicine
Turgut Ozal Medical Center, Radiology Department
Malatya
Turkey
Tel: +90-532-452-0065
Fax: +90-422-341-0728
E-mail: erdemgulnur@hotmail.com