Method to Estimate the Complete and Two-Turn Cochlear Duct Length

Abstract

Hypothesis:
Using a linear measurement of the cochlea on a single radiographic image can reliably estimate the complete and two-turn cochlear duct length (CDL) in a normal human temporal bone.

Background:
CDL is measured from the middle of the round window to the helicotrema. Histologic studies have shown the length of the organ of Corti (OC) to range from 25 to 35 mm. CDL measurements, performed either radiographically or histologically, are quite tedious and time-consuming. We propose equations that can reliably estimate both two-turn and complete CDL using a single computed tomography (CT) image.

Methods:
Prior studies of CDL, measured either histologically or radiographically, were reviewed, which yielded distributions of CDL measured at the OC and the lateral wall of the cochlea. Using Escudé's third equation as a basis, we were able to extrapolate complete and two-turn CDL based on a CT scan measurement of the diameter of the basal turn (A).

Results:
Using measurement A, the relationship of two-turn CDL measured at the OC is 2TL(oc) = 3.65(A-1) and for 2TL(i) = 3.65(A-0.7). The equation for estimation of complete CDL is CDL(oc) = 4.16A - 4 and for CDL(i) = 4.16A - 2.7.

Conclusion:
Using a single linear measurement from a CT scan image can reliably estimate the two-turn and complete CDLs in human temporal bones. The two-turn length represents the best compromise of cochlear coverage while minimizing intracochlear trauma for electrode insertions.

Full text

Method to Estimate the Complete and Two-Turn
Cochlear Duct Length
*George Alexiades, †Anandhan Dhanasingh, and †Claude Jolly
*Icahn School of Medicine at Mount Sinai, New York, New York, U.S.A.; and ÞMED-EL Corporation,
Innsbruck, Austria
Hypothesis: Using a linear measurement of the cochlea on a
single radiographic image can reliably estimate the complete
and two-turn cochlear duct length (CDL) in a normal human
temporal bone.
Background: CDL is measured from the middle of the round
window to the helicotrema. Histologic studies have shown the
length of the organ of Corti (OC) to range from 25 to 35 mm.
CDL measurements, performed either radiographically or his-
tologically, are quite tedious and time-consuming. We propose
equations that can reliably estimate both two-turn and complete
CDL using a single computed tomography (CT) image.
Methods: Prior studies of CDL, measured either histologically
or radiographically, were reviewed, which yielded distributions
of CDL measured at the OC and the lateral wall of the cochlea.
Using Escude
´’s third equation as a basis, we were able to ex-
trapolate complete and two-turn CDL based on a CT scan
measurement of the diameter of the basal turn (A).
Results: Using measurement A, the relationship of two-turn CDL
measured at the OC is 2TL(oc) = 3.65(A-1) and for 2TL(i) =
3.65(A-0.7). The equation for estimation of complete CDL is
CDL(oc) = 4.16A j4 and for CDL(i) = 4.16A j2.7.
Conclusion: Using a single linear measurement from a CT scan
image can reliably estimate the two-turn and complete CDLs in
human temporal bones. The two-turn length represents the best
compromise of cochlear coverage while minimizing intracochlear
trauma for electrode insertions. Key Words: Cochlear duct
lengthVCochleaVOrgan of Corti.
Otol Neurotol 00:00Y00, 2014.
The human cochlea is the full adult size at birth. Cochlear
duct length (CDL) is defined as the length of the scala media
measured from the middle of the round window to the
helicotrema. Mary Hardy (1), in 1938, first described his-
tologic measurements of CDL in 68 specimens. Since that
time, there have been numerous publications (2Y14) on the
variability of the size of the human cochlea, with lengths
measuring from 25 to 45 mm. These measurements have
been performed both histologically as well as radiograph-
ically and have been measured at both the bony lateral wall
(LW) and at the level of the organ of Corti (OC).
Measurement of the CDL in a patient can be valuable
in the preoperative stage of cochlear implantation. With
variable length electrodes available for implantation and
reports of incomplete insertions of the longer electrodes,
variability in the length of the cochlear duct can be a
significant variable in the depth of insertions. In addition,
as we move to a soft-surgery technique in an attempt at
hearing and structure preservation, correctly identifying
the implantable length of the cochlea can assist in the
selection of the correct length of electrode for that par-
ticular patient.
Measurements of CDL are often tedious and time-
consuming. Histologic measurements are performed by
sectioning the cochlea and hand measuring each segment
under the microscope manually (1). Another method
described for measuring CDL on CT scans involved
measuring the diameter of the basal and middle turns as
well as the axial height of the cochlea. These measure-
mentswerethenusedintheArchimedeanspiralequation
to determine the CDL on the LW (8). In addition, there is
variability in measuring the CDL because of where along
the width of the scala media the measurements are taken.
Traditionally, CDL has been measured at the LW or at
theleveloftheOC.Weareproposingamethodtoes-
timate the complete CDL and two-turn length (2TL)
using a single measurement off of a multiplanar
reconstructed CT scan image, measured at the OC and at
a level named ‘‘i,’’ which represents the implantable
CDL of a cochlear implant electrode. The use of the LW
length may overestimate the implantable depth of an LW
electrode array because the electrode trajectory for an
LW electrode is somewhere between the bony LW and
the OC caused by the diameter of the array.
Address correspondence and reprint requests to George Alexiades,
M.D., 380 Second Ave., 9th Floor, New York, NY 10010, U.S.A.; E-
mail: galexiades@nyee.edu
The authors disclose no conflicts of interest.
Otology & Neurotology
00:00Y00 Ó2014, Otology & Neurotology, Inc.
1
Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.

MATERIALS AND METHODS
Mary Hardy published the landmark article on the OC length
of the human cochlea in 1938. Performing a PubMed search
using the keywords ‘‘cochlear duct length’’ and ‘‘organ of Corti
length’’ revealed 13 other studies measuring the CDL in normal
human cochleae from 1975 to 2010 (2Y14). The data of the
maximum, minimum, and average CDL as well as the number
of specimens were recorded. The averages of CDL measure-
ments performed at the LW were compared with those mea-
sured at the OC.
Using the straight line equation y=mx +b, where mis the
slope and bis the intercept, regression analysis was performed
using Microsoft Excel to generate CDL and 2TL equations. R
2
values represent the goodness of fit of the yaxis parameter in
reference to the xaxis. A value of 1 represents a perfect fit. For
generation of the CDL and 2TL bell curves, the normal distri-
bution function in Microsoft Excel was used.
The value of Ais defined as the linear measurement from the
round window to the farthest point on the opposite wall of the
cochlea on a reformatted CT scan slice. This Avalue is used for
CDL and 2TL calculations. Martinez-Monedero et al. (15) ex-
amined 124 normal human cochleae and described three-
dimensional morphology of the cochlea. The principal author
measured Avalues in 105 of the specimens and supplied the
values to the authors of this study. These values were then used
for CDL and 2TL calculations as a test for equation validity by
comparing them to Hardy and Lee’s data (for CDL) and Hardy’s
data (for 2TL). The Independent Sample Ttest was used to
assess statistical significance between these results.
RESULTS
CDL Distribution
In reviewing the literature, eight studies measured the
CDL at the level of the organ of Corti (1,2,4Y8,14), five
studies measured CDL at the bony LW (9Y13), and one
study measured CDL at both levels (3). The CDL mea-
sured at the level of the OC (CDL
(oc)
) (n = 296; range,
25.2Y40.1 mm; mean, 32.89 mm) was, as a group, shorter
than those measured at the LW (CDL
(lw)
) (n = 198; range,
32.6Y45.6 mm; mean, 38.9 mm). Having access to the
individual specimen measurements in Hardy’s 68 speci-
mens, the CDL
(oc)
range was 25.26 to 35.45 mm, and the
mean was 31.52 mm, and Lee’s 27 specimens had a
CDL
(oc)
range of 25.5 to 35.1 mm and a mean of 30.8 mm,
showing that these two studies are representative samples
of the entire series. Taking Hardy’s and Lee’s specimens,
the distribution curve of CDL is shown in Figure 1 and
2TL in Figure 2. In addition, Hardy measured basal turn,
middle turn, and apical turn lengths in her specimens,
which revealed the basal turn comprising 58% of CDL,
middle turn 29%, and apical turn making up 13% (1).
Relationship of CDL to Basal Turn Length
Linear regression analysis was performed on the data
from Hardy’s 68 specimens. Hardy’s measurements of
CDL were performed at the level of the OC. The results
are shown in Figures 3 and 4, where Figure 3 shows the
relationship of the basal turn length (BTL) to CDL
(oc)
and Figure 4 shows the relationship of BTL to 2TL. For
CDL, the regression analysis yielded an equation CDL =
1.71(BTL) + 0.18, with an r
2
value of 0.79 ( p=0.086)
(Fig. 3). For 2TL, the regression analysis yielded an
FIG. 1. CDL distribution curve for Hardy and Lee’s combined
specimens (n = 95).
FIG. 2. 2TL distribution curve for Hardy’s specimens (n = 68).
FIG. 3. Regression analysis comparing BTL with CDL (n = 68)
using a straight line equation y=mx +b.R
2
= 0.79, p= 0.086.
2 G. ALEXIADES ET AL.
Otology & Neurotology, Vol. 00, No. 00, 2014
Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.

equation of 2TL = 1.5(BTL), with an r
2
value of 0.92
(p=0.018)(Fig.4).
Linear AMeasurement in Relation to CDL/2TL
Escude
´described the angle of implant insertion in re-
lation to BTL and a linear measurement Ain 2006 (3). His
third equation, L= 2.62 Alog
e
(1.0 + 5/235), de-
scribes the length of the LW of the cochlea for a certain
insertion angle 5and the letter Adenoting the measure-
ment from the round window to the farthest point on the
opposite wall of the cochlea on a reformatted CT scan
slice (Fig. 5). By simple substitution, using an insertion
angle of 360 degrees, the length of the basal turn at the
LW (BTL
(lw)
) is defined as:
BTL lwðÞ
¼L360o
ðÞ¼2:62 Aloge1þ360=235 ¼2:43 AðÞ
To reconcile LW measurements with OC measure-
ments, the Avalue needs to be adjusted. Prior histologic
studies have shown that the organ of Corti lies approxi-
mately 0.5 mm off of the lateral cochlear wall (3).
Therefore, to adjust the Avalue for OC measurements,
0.5 mm needs to be subtracted from both ends of the A
value measurement, which results in:
AocðÞ
¼AlwðÞ
j2*0:5mmðÞ¼AlwðÞ
j1
Therefore, the proper equation for CDL measurements
at the organ of Corti (with rounding to the nearest 10th
decimal place) is:
CDL ¼1:71 BTLðÞþ0:18
CDL ocðÞ¼1:71 2:43 Aj1ðÞðÞþ0:18
¼4:16Aj4ðEquation 1Þ
2TL CDL at the organ of Corti is:
2TL ¼1:5BTLðÞ
2TL ocðÞ
¼1:52:43Aj1ðÞðÞ¼3:65 Aj1ðÞðEquation 2Þ
To better approximate effective CDL for a particular
cochlear implant electrode, the average radius of the
electrode should be substituted for the distance of the OC
displacement from the bony LW. For example, using the
MED-EL FLEXsoft electrode, the average radius is 0.35 mm
(range, 0.2Y0.65). Using this average radius, we end with the
following equations for CDL and 2TL and are heretofore
referred to as CDL
(i)
and 2TL
(i)
:
FIG. 4. Regression analysis comparing BTL with 2TL (n = 68)
using a straight line equation y=mx +b.R
2
= 0.92, p= 0.018.
FIG. 5. Multiplanar reconstructed image of the left cochlea
showing the full basal turn and round window. Arrowed line in-
dicates Ameasurement from the middle of the round window to
the opposite wall.
FIG. 6. Distribution curves for CDL and 2TL of 104 specimens
using the equation plotted against curves measured by Hardy and
Lee via histologic measurements. The curves to the left are for
2TL, and the curves to the right are for CDL.
3METHOD TO ESTIMATE THE COCHLEAR DUCT LENGTH
Otology & Neurotology, Vol. 00, No. 00, 2014
Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.

CDL iðÞ¼1:71 2:43 Aj0:7ðÞðÞþ0:18
¼4:16A<2:7ðEquation 3Þ
2TL iðÞ¼3:65 Aj0:7ðÞ ðEquation 4Þ
As shown in these equations, the electrode position in
the cochlea lies between the LW and the OC in this
particular case.
Validity Testing
Martinez-Monedero et al. (15) measured Avalues in
105 human temporal bone specimens (average, 8.55 mm;
range, 6.8Y10.3 mm) using a cone-beam CT scan of the
124 specimens cited in their 2011 article. Using these A
values, CDL
(oc)
and 2TL
(oc)
were calculated using equa-
tions 1 and 2 for all the specimens. The results are shown
in Figure 6 and plotted against distribution curves from
Hardy’s 68 specimens (for CDL
(oc)
and 2TL
(oc)
) and
Lee’s 27 specimens (for CDL
(oc)
) histologic measure-
ments in their specimens. Using the Independent Sample
Ttest, there was no statistical difference between Hardy
and Lee’s data and our equation for CDL
(oc)
measure-
ments ( p= 0.437). Comparing Hardy’s 2TL
(oc)
with our
equation also showed no statistical difference between the
two groups ( p= 0.923).
DISCUSSION
The size of the human cochlea has a wide variability,
and its distribution falls along a bell curve. The rela-
tionship of the middle turn to the basal turn (equation 2)
was highly correlated stat istically ( p= 0.018), whereas
the relationship of CDL to BTL (equation 1) showed a
trend but fell shy of statistical significan ce ( p= 0.086).
This shows a clear correlation of the middle turn con-
tributing 50% of the length of the basal turn to the CDL
and 2TL. The apical contribution is more variable as its
contribution to CDL makes estimation less accurate than
estimating 2TL alone, although it contributes only a small
percentage of the total CDL. There is a statistically sig-
nificant correlation between the diameter of the basal turn
(A) and 2TL and a correlation of Ato CDL, although this
does not show statistical significance with the current
sample size. This allows us to reliably estimate 2TL and
approximate complete CDL and simply by measuring A
length on a multiplanar reconstructed CT scan image.
The length of CDL measured at the OC is significantly
shorter as compared with that of the LW. It is important to
understand how the length of the cochlear duct shortens
as one moves away from the LW of the cochlea because
this significantly impacts effective implantable depth. As
evidenced in the example for CDL
(i)
and 2TL
(i)
,asyou
move to a thinner electrode, then effective implantable
length will be longer as compared with a thicker electrode
in the same cochlea. No validation studies were performed
on CDL
(i)
and 2TL
(i)
measurements and these will need to
be conducted in the future to corroborate these estimations.
Microdissection studies by Wright and Roland (16)
showed significant variability in the apical region of the
cochlea. In addition, the size of the scala tympani along
the LW suggests that current electrodes do not fit in this
region and either cause trauma to the inner ear or are
displaced medially where the scala tympani is larger. To
decrease the likelihood of trauma to the apical region, the
2TL is felt by the authors to be the best compromise of
cochlear coverage while minimizing intracochlear trau-
ma. Values of Aof 7.3, 8.4, and 9.2 mm equate to 2TL
(i)
of 24-, 28-, and 31-mm lengths, respectively, which can
be used for preoperative electrode selection.
REFERENCES
1. Hardy M. The length of the Organ of Corti in man. Am J Anat
1938;62:291Y311.
2. Walby AP. Scala tympani measurement. Ann Otol Rhinol Laryngol
1985;94(4 Pt 1):393Y7.
3. Ulehlova
´L, Voldrich L, Janisch R. Correlative study of sensory cell
density and cochlear length in humans. Hear Res. 1987;28:149Y51.
4. Wright A, Davis A, Friedberg G, Ulehlova
´L, Spencer H. Hair cell
distributions in the normal human cochlea. Acta Otolaryngol Suppl
1987;444:1Y48.
5. Bredberg G, Teti A, Zambonin Zallone A, Lundevall E, Lurato S.
Ultrastructural evaluation of the microslicing method for the study of
temporal bone pathology. Acta Otolaryngol Suppl 1987;436:7Y14.
6. Ariyasu L, Galey FR, Hilsinger R Jr, Byl FM. Computer-generated
three-dimensional reconstruction of the cochlea. Otolaryngol Head
Neck Surg 1989;100:87Y91.
7. Kawano A, Seldon HL, Clark GM. Computer-aided three-
dimensional reconstruction in human cochlear maps: measure-
ment of the lengths of organ of Corti, outer wall, inner wall, and
Rosenthal’s canal. Ann Otol Rhinol Laryngol 1996;105:701Y9.
8. Ketten DR, Skinner MW, Wang G, Vannier MW, Gates GA, Neely
JG. In vivo measures of cochlear length and insertion depth of
nucleus cochlear implant arrays. Ann Otol Rhinol Laryngol Suppl
1998;175:1Y16.
9. Adunka O, Unkelbach MH, Mack MG, Radeloff A, Gstoettner W.
Predicting basal cochlear length for electric-acoustic stimulation.
Arch Otolaryngol Head Neck Surg 2005;131:488Y92.
10. Escude
´B, James C, Deguine O, Cochard N, Eter E, Fraysse B. The
size of the cochlea and predictions of insertion depth angles for
cochlear implant electrodes. Audiol Neurotol 2006;11:27Y33.
11. Skinner MW, Holden TA, Whiting BR, et al. In vivo estimates of
the position of advanced bionics electrode arrays in the human
cochlea. Ann Otol Rhinol Laryngol Suppl 2007;197:2Y24.
12. Miller JD. Sex differences in the length of the Organ of Corti in
humans. J Acoust Am 2007;121:151Y5.
13. Erixon E, Hogstorp H, Wadin K, Rask-Andersen H. Variational
anatomy of the human cochlea: implicaitons for cochlear implan-
tation. Otol Neurotol 2009;30:14Y22.
14. Lee J, Nadol JB Jr, Eddington DK. Depth of electrode insertion and
postoperative performance in humans with cochlear implants: a
histopathologic study. Audiol Neurootol 2010;15:323Y31.
15. Martinez-Monedero R, Niparko JK, Aygun N. Cochlear coiling
pattern and orientation differences in cochlear implant candidates.
Otol Neurotol 2011;32:1086Y93.
16. Wright CG, Roland PS. Temporal bone microdissection for ana-
tomic study of cochlear implant electrodes. Cochlear Implants Int.
2005;6:159Y68.
4 G. ALEXIADES ET AL.
Otology & Neurotology, Vol. 00, No. 00, 2014
Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.