Download Central Corneal Thickness

JOBNAME: corn 26#3 2007 PAGE: 1 OUTPUT: Friday February 16 11:29:44 2007
tsp/corn/135872/ICO200493
CLINICAL SCIENCE
Central Corneal Thickness
Z-Ring Corneal Confocal Microscopy Versus Ultrasound Pachymetry
Erica Brugin, MD,* Alessandra Ghirlando, MD,* Catia Gambato, MD,* and Edoardo Midena, MD*†
Purpose: To compare the repeatability and validity of corneal
pachymetry by a corneal confocal microscope with a z-axis adapter
(Confoscan 4.0 with z-ring adapter: z-CS4) versus ultrasound (US)
pachymetry in the measurement of central corneal thickness (CCT).
Methods: CCT in 44 eyes of 44 subjects was determined with
z-CS4. Z-CS4 exams were used to estimate the repeatability of
thickness measurement by z-ring adapter for this confocal
microscope. Intraclass Correlation Coefficient (ICC) between two
different z-CS4 users was also determined. CCT in the same 44 eyes
was determined with US pachymetry and measurements were compared with z-CS4 CCT.
Results: Z-CS4 CCT showed high intrainstrument reproducibility
(ICC = 0.989; 95%CI 0.982-0.993; P , 0.0001). Mean difference
among three CCT consecutive measures, in the same eye, was 0.8 6
11.1 mm. High correlation was found between two users (ICC =
0.896; 95%IC 0.830-0.937; P , 0.0001). Z-CS4 CCT showed high
correlation with US pachymetry (ICC = 0.921; 95%CI 0.851-0.958;
P , 0.0001). Mean corneal thickness determined was statistically
different with the two methods (US: 512.6 6 65.8 mm; z-CS4:
487.8 6 60.1 mm; P , 0.0001).
Conclusion: Z-CS4 seems an accurate, noninvasive and reproducible technique for CCT evaluation and confirms that central cornea is
thinner when measured with confocal microscopy compared to
ultrasounds.
Key Words: corneal pachymetry, ultrasound pachymetry, corneal
confocal microscopy, z-ring adapter
(Cornea 2007;26:303–307)
C
orneal thickness is a major factor for planning and
monitoring photorefractive procedures and monitoring
several corneal diseases.1–10 Techniques for measuring central
corneal thickness (CCT) include optical pachymetry, ultrasound pachymetry, confocal microscopy, ultrasound
Received for publication April 4, 2006; revision received September 11, 2006;
accepted November 3, 2006.
From the *Department of Ophthalmology, University of Padova, Padova,
Italy; and the †Fondazione GB Bietti, IRCCS, Roma, Italy.
The authors state that they have no proprietary interest in the products named
in this article.
Reprints: Edoardo Midena, Department of Ophthalmology, University of
Padova, Via Giustiniani 2, 35128 Padova, Italy (e-mail: edoardo.midena@
unipd.it).
Copyright Ó 2007 by Lippincott Williams & Wilkins
biomicroscopy, optical ray path analysis or scanning-slit
corneal topography, and optical coherence tomography.11–20
Ultrasound pachymetry is the current standard for corneal
thickness measurement.21–24 In vivo corneal confocal microscopy (CCM) is a diagnostic technique of the cornea that allows
the visualization of a small area (;340 3 255 mm) of the
cornea resolved in thin confocal slices.25,26 CCM allows one
to analyze and record the different corneal layers with magnification up to 31000. Unfortunately, current corneal confocal
microscope configuration is unable to provide completely
accurate quantification along the eye z-axis because of eye
movements in an antero-posterior direction.
Recently, a corneal confocal microscope with a z-ring
adapter (z-ring Confoscan 4.0; Nidek Technologies, Padova,
Italy) claiming to accurately measure corneal layer distance
along the z-axis has been introduced in clinical practice. The
aim of this study was twofold: to test z-ring CCM for
repeatability and accuracy and to compare corneal pachymetry
performed by confocal microscope with z-ring adapter versus
standard ultrasound pachymetry.
MATERIALS AND METHODS
This cross-sectional study included 44 eyes of 44
consecutive subjects referred to the Laboratory of Advanced
Ocular Diagnostic of the Department of Ophthalmology of the
University of Padova. All subjects had a complete ophthalmologic examination including anterior segment biomicroscopy,
funduscopy, corneal pachymetry, and corneal topography. The
last procedure was performed primarily to exclude keratoconus.
CCT was measured by standard ultrasound pachymetry and
CCM. One eye of each subject was randomly selected for
measurement, and the study eye was assigned to ultrasound
and confocal microscopy pachymetry in randomized order,
according to a 2 3 2 Latin Square to ensure complete balance
between instruments’ evaluation. Informed consent was
obtained from all subjects after the nature and purpose of
the study were explained. All measurements were taken in late
morning/early afternoon, when corneal thickness is considered
stable.27 The interval between measurements was selected to
maintain standard condition for both measurements (the 2 tests
were performed within 6 hours).28
In vivo CCM was performed by using z-ring Confoscan
4.0, a scanning slit confocal microscope equipped with
an Achroplan (Zeiss, Oberkochen, Germany) nonapplanating 340 immersion objective lens designed for full-thickness
examination of the cornea, with a working distance of 1.92 mm
and a motorized focusing device. Confoscan 4.0 is now
Cornea Volume 26, Number 3, April 2007
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
303
JOBNAME: corn 26#3 2007 PAGE: 2 OUTPUT: Friday February 16 11:29:44 2007
tsp/corn/135872/ICO200493
Brugin et al
equipped with a z-ring adapter system (z-CS4; Fig. 1). The
z-ring adapter is an optomagnetic sensor that makes the
examined eye integral with the tip of the confocal microscope
during the examination. Therefore, with the z-ring sensor, the
confocal microscope tip position turns out to be fixed in regard
to the examined cornea, compensating for eye movements
along the z-axis. The z-ring adapter is made of a part settled on
the optical tip and of a removable part including a metallic
stand and a contact ring in polymethylmethacrylate (PMMA;
external diameter, 11 mm; internal diameter, 9 mm). The
z-ring adapter is related to the cornea through a thin layer of
polyacrylic gel, and during the scan, it is able to keep uniform
pressure (maximum pressure of 30 mm Hg) on the cornea by
proper sensor.
Reliability of measures by z-CS4 was previously checked
with a set of PMMA contact lenses with known thickness
between 350 and 650 mm, and differences were within 65 mm
(E. Midena, unpublished data).
The center of the cornea was studied in all examinations.
Before the examination, a drop of topical anesthetic 0.4%
oxybuprocaine chlorohydrate (Novesina; Novartis Farma,
Varese, Italy) was instilled in the lower conjunctival fornix
of the eye to reduce blinking. The patient was seated in front of
the microscope, with the aid of a chin rest and a forehead
support, while fixating, with the examined eye, on a bright
blue target inside the instrument to minimize eye movement
during the examination. One drop of 0.2% polyacrylic gel
(Viscotirs gel; Ciba Vision Ophthalmics, Venezia, Italy) was
applied onto the objective tip to serve as an immersion fluid.
Baseline macrometric alignment to the corneal apex was made
manually by the operator, who moves the z-ring adapter to
the apex of cornea. This procedure is automatically completed
by a micrometric motor-driven system. The focal plane is
automatically moved to reach the anterior chamber and begins
scanning and recording of corneal images. The images are
FIGURE 1. Schematic drawing showing the z-ring adapter
system: The constant pressure system continuously records the
pressure of the objective lens onto the eye (which may change
depending on the eye movement), and the software
automatically moves the optical head forward or backward.
This feedback allows to maintain a constant contact pressure.
In the bottom right-hand, a real picture of the z-ring adapter.
(M: motor)
304
Cornea Volume 26, Number 3, April 2007
always taken from the endothelium to the corneal epithelium.23
Image intensity depth profiles are generated from confocal
microscope video recordings by averaging the pixel intensity
in the center of each consecutive video frame image and
plotting data as a function of the z-depth in the z-scan curve.
Data obtained from these scans are averaged and used for
subsequent calculations. Every point in the z-scan curve is
quantified in international units (IUs). Because of the z-ring
adapter, all points on the z-scan curve are directly correlated
to high-resolution images, and the exact z-axis position of
specific tissue landmarks (ie, the surface epithelium, subepithelial nerve plexus, endothelium) are considered to
calculate the distance between individual corneal layers. The
confocal microscopy z-scan profile shows 2 peaks (identified
in the z-scan curve) corresponding to major reflection
originating from the endothelium (first peak) and superficial
epithelial cells (second peak; Fig. 2). Whereas the first peak is
sharp and well defined, the second one appears as a wide curve
corresponding to several epithelial images. Corneal thickness
was defined as the distance between the first peak (endothelium) and the last clear and centered frame of all epithelial
images within the second peak. Each corneal confocal
microscope examination was preplanned with the following
parameters: a z-axis range of movement of 900 mm (3 passes
for each complete corneal scan), thus giving a theoretical
z-axis distance between images in the scans of 7 mm. Z-CS4
acquires 350 images per examination at a rate of 25 frames per
second. Therefore, capture time is 14 seconds. The speed of
the image plane at the cornea is approximately 72 mm/s. To
define the accuracy of corneal confocal pachymetry by the
z-ring adapter system, the 3 CCTs for each examination were
FIGURE 2. Corneal confocal microscopy frame with z-scan
curve. Z-scan curve (bottom): the first peak corresponds to the
endothelium (high reflectivity) peak; the second peak corresponds to the last high quality epithelium image.
q 2007 Lippincott Williams & Wilkins
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
JOBNAME: corn 26#3 2007 PAGE: 3 OUTPUT: Friday February 16 11:29:48 2007
tsp/corn/135872/ICO200493
Cornea Volume 26, Number 3, April 2007
analyzed. To determine interobserver repeatability of z-ring
corneal confocal pachymetry, all subjects were tested by 2
independent and masked operators, both experienced in
the use of the corneal confocal microscope.
Each cornea was also measured using an ultrasonic
pachymeter (UP1000; Nidek, Gamagori, Japan), calibrated by
the manufacturer. This instrument uses an acoustic index of
1640 m/s to determine corneal thickness. One drop of 0.4%
oxybuprocaine chlorohydrate (Novesina; Novartis Farma) was
instilled in the conjunctival fornix. The probe was lightly
touched to the central corneal 5 times, a thickness measurement was recorded for each touch, and these measurements
were averaged.
Z-CS4 uses an autoalignment system, and recording
starts when the tip of the objective lens is aligned with the
center of the cornea. The operator is able to document this
phase by using the real-time monitor image that seems
perfectly centered without lateral dark bands. The system is
non–operator dependent. For ultrasound pachymetry examination, patients were asked to fixate on a small target with the
fellow eye, and the tip of the ultrasound probe was manually
centered on the pupil visually, as done in previous studies.
The intraclass correlation coefficient (ICC) and its
95% confidence interval (95% CI) were used to determine the
reliability of the confocal z-CS4 readings. For the interobserver reliability assessment and the comparison of z-CS4
with ultrasound pachymetry, a Bland–Altman graph was also
evaluated. Paired student t test was used to evaluate the statistical difference between corneal thickness as measured
by the 2 operators and by the 2 instruments.
To determine the interobserver repeatability of z-CS4,
the average of the 3 peak values obtained on each examination
by each operator was used (44 by 2 operators). Intrainstrument
repeatability of the CCT measurements performed by z-CS4
was tested comparing the 3 peak values obtained from the
scans of each examined eye by both operators (3 values by
88 exams). To evaluate the agreement between z-CS4 and
ultrasound pachymetry (‘‘gold standard’’) in measuring
corneal thickness, the average of the 6 values obtained from
the z-CS4 examinations of observer 1 and 2 and ultrasound
values were used (44 values by 2 instruments).
All statistical analyses were performed on a computer
using a SPSS software package (SPSS for Windows, version
12.0; SPSS, Chicago, IL). Values of #0.05 (2-sided) were
considered to indicate statistical significance.
CCT: Z-Ring CCM Versus Ultrasound Pachymetry
CI, 0.982–0.993; P , 0.0001). The mean difference among
3 CCT measurements in the same eye was 0.8 6 11.1 mm.
The mean difference in CCT between scans taken by
2 different operators with the z-CS4 was not statistically
significant (first operator: 500.1 6 72.0 mm; second operator:
496.2 6 67.3 mm; P = 0.356). Figure 3 shows interobserver
repeatability data (ICC = 0.896; 95% CI, 0.830–0.937;
P , 0.0001).
The comparison of CCT between ultrasound pachymetry and z-CS4 in 44 eyes showed a significantly high ICC of
0.921 (95% CI, 0.851–0.958; P , 0.0001). The difference
between the mean values of corneal thickness calculated using
the CCM and ultrasonic pachymeter was 24.8 mm. In fact,
mean corneal thickness determined with the ultrasound
pachymeter was 512.6 6 65.8 versus 487.8 6 60.1 mm with
the z-CS4 (paired t test; P , 0.0001; Fig. 4).
DISCUSSION
The measurement of CCT has become increasingly
important in ophthalmic practice. Refractive surgery is
routinely planned according to preoperative measurement of
CCT,8 and accurate determination of intraocular pressure may
need to be modified according to CCT.5,7,9,10
Currently, the most used clinical method to measure CCT
is ultrasound pachymetry. Ultrasound pachymeters are portable,
objective, and simple-to-use instruments. Recent studies have
documented that this method has a high degree of intraoperator,
interoperator, and intrainstrument reproducibility.17–19,22,23
However, ultrasound pachymeter repeatability, even if good,
varies from one instrument to another. Moreover, ultrasound
pachymetry has the disadvantage that the placement of the
probe exactly on the center of the cornea is crude and operator
dependent. This method also requires direct cornea–probe
contact.21 Mild patient discomfort and risk for infection are
RESULTS
Forty-four eyes of 44 subjects (21 men and 23 women;
mean age, 34 6 7 years; range, 22–49 years) were examined
by ultrasound pachymeter and CCM with z-ring adapter
(z-CS4). Mean corneal thickness quantified by these instruments was calculated for all subjects. Twenty of 44 eyes had
previously undergone (.5 years) refractive surgery. CCM
images, examined by an expert operator, never showed
pathologic corneal reflectivity (normal corneal reflectivity:
,30 IU, as previously reported by Gambato et al29).
The intrainstrument reproducibility of CCT measurements performed by the z-CS4 was ICC = 0.989 (95%
FIGURE 3. Central corneal thickness measured by confocal
microscopy (Z-CS4) by two different operators. (Bland–Altman
graph) Data are reported in microns (Intraclass correlation
coefficient = 0.896; 95% CI 0.830-0.937; P , 0.0001).
q 2007 Lippincott Williams & Wilkins
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
305
JOBNAME: corn 26#3 2007 PAGE: 4 OUTPUT: Friday February 16 11:29:52 2007
tsp/corn/135872/ICO200493
Brugin et al
FIGURE 4. Central corneal thickness measured by confocal
microscopy (Z-CS4) vs ultrasound pachymetry. (Bland–Altman
graph) Data are reported in microns. (Intraclass correlation
coefficient = 0.921; 95% CI 0.851-0.958; P , 0.0001. Mean
central corneal thickness 6 SD: z-ring CS4 = 487.8 6 60.1 mm
versus ultrasound = 512.6 6 65.8 mm; P , 0.0001).
additional concerns with a direct contact technique.21 A new
instrument, the corneal confocal microscope, enables scanning
of the different layers of the central cornea with magnification
up to 31000 and analyzing individual corneal layers. This
instrument allows central corneal analysis without direct
cornea–instrument contact because of a thin layer of polyacrylic
gel between the cornea and probe during the scan.
Until now, the greatest limitation of CCM pachymetry
compared with ultrasound pachymetry was the lack of precision because of the time required to scan through the cornea
and the inevitable corneal movements during the scan
(ie, the anterior–posterior motion caused by ocular pulse).1
Antero-posterior eye movements during the scan lead to an
unpredictable error in the estimate of corneal thickness.1 The
new z-ring adapter used with corneal confocal microscope
Confoscan 4.0 has an eye–tip system that avoids the error
produced by eye movements along the z-axis. Therefore, the
corneal confocal microscope equipped with a z-ring allows
calculation of the distance between corneal layer images and
measurement of corneal thickness.
The corneal thickness of a consecutive series of referred
patients was measured with a z-CS4 and compared with a
standard ultrasound pachymeter. Our data show the high
intrainstrument repeatability and the high interobserver
repeatability of the z-CS4. Our data also show high validity
of the z-CS4 in reading pachymetric values compared with
the ultrasound pachymeter.
Therefore, the corneal confocal microscope with z-ring
adapter seems to be a diagnostic technique that allows an
accurate, reproducible measurement, without direct contact of
the tip of the lens to the center of the cornea. McLaren et al1
compared CCT determined by ultrasonic and corneal confocal
scanning slit (without z-ring adapter) pachymetry. Mean
306
Cornea Volume 26, Number 3, April 2007
thickness measured by calibrated CCM was statistically lower
than with ultrasonic pachymetry measurements.1 Javaloy et al3
found similar corneal thickness values with conventional
ultrasound, scanning-slit corneal topography, and CCM.
Different results may depend on the use of 2 different corneal
confocal microscopes, the different ultrasound pachymeters,
or the different method used to define corneal confocal
pachymetry. Moreover, the confocal microscope used by
Javaloy et al3 was not calibrated.
McLaren at al1 concluded that an ultrasound pachymeter
is suitable for measuring corneal thickness if relative changes in
thickness are relevant and if the same instrument is used
throughout the study. However, if true thickness is needed to
make clinical decisions or to compare with other studies, the
accuracy of a particular ultrasound pachymeter should be
evaluated by an independent measurement. The difference
between ultrasound and confocal microscope pachymetry
suggests that verification of clinical ultrasonic pachymeters
should be revised. McLaren et al1 also reported that their
method of confocal microscopy pachymetry was not a valid
alternative, because its reliability was limited by corneal motion
in the antero-posterior direction. Our method, which uses a
z-ring adapter, avoids this error. Our data confirm the difference
between ultrasound and confocal microscopy pachymetry, but
this difference is lower in our study than that in the report by
McLaren et al1 (24.8 vs. ;38 mm). Lower CCT differences may
depend on the use of the z-ring in this study, because other
variables, such as corneal refractive index and corneal radius of
curvature, have already been shown to be irrelevant. Many
studies have reported consistent differences when the same
cornea is measured with different approaches.11–15,18–20 We
believe that z-CS4 measurement approaches real corneal
thickness, because this technique overwhelms the major
limitation of CCM. When it is necessary to accurately determine
CCT for clinical intervention, it seems relevant to improve
current techniques. CCM with the z-ring adapter may represent
a new and valid approach.
REFERENCES
1. McLaren JW, Nau CB, Erie JC, et al. Corneal thickness measurement
by confocal microscopy, ultrasound, and scanning slit methods. Am J
Ophthalmol. 2004;137:1011–1020.
2. Holden BA, Mertz GW, McNally JJ. Corneal swelling response to contact
lenses worn under extended wear conditions. Invest Ophthalmol Vis Sci.
1983;24:218–226.
3. Javaloy J, Vidal MT, Villada JR, et al. Comparison of four corneal pachymetry
techniques in corneal refractive surgery. J Refract Surg. 2004;20:29–34.
4. Pflugfelder SC, Liu Z, Feuer W, et al. Corneal thickness indices
discriminate between keratoconus and contact-lens induced corneal
thinning. Ophthalmology. 2002;109:2336–2341.
5. Herndon LW, Weizer JS, Stinnett SS. Central corneal thickness as a risk
factor for advanced glaucoma damage. Arch Ophthalmol. 2004;122:
17–21.
6. Medeiros FA, Sample PA, Weinreb RN. Corneal thickness measurements
and frequency doubling technology perimetry abnormalities in ocular
hypertensive eyes. Ophthalmology. 2003;110:1903–1908.
7. Medeiros FA, Sample PA, Weinreb RN. Corneal thickness measurements
and visual function abnormalities in ocular hypertensive patients. Am J
Ophthalmol. 2003;135:131–137.
8. Price FW Jr, Koller DL, Price MO. Central corneal pachymetry in patients
undergoing laser in situ keratomileusis. Ophthalmology. 1999;106:2216–
2220.
q 2007 Lippincott Williams & Wilkins
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
JOBNAME: corn 26#3 2007 PAGE: 5 OUTPUT: Friday February 16 11:29:55 2007
tsp/corn/135872/ICO200493
Cornea Volume 26, Number 3, April 2007
9. Doughty MJ, Zaman ML. Human corneal thickness and its impact on
intraocular pressure measures: a review and meta-analysis approach.
Surv Ophthalmol. 2000;44:367–408.
10. Stodtmeister R. Applanation tonometry and correction according to
corneal thickness. Acta Ophthalmol Scand. 1998;76:319–324.
11. Yaylali V, Kaufman SC, Thompson HW. Corneal thickness measurement
with Orbscan Topography System and ultrasonic pachymetry. J Cataract
Refract Surg. 1997;23:1345–1350.
12. Izatt JA, Hee MR, Swanson EA, et al. Micrometer-scale resolution
imaging of the anterior eye in vivo with optical coherence tomography.
Arch Ophthalmol. 1994;112:1584–1589.
13. Muscat S, McKay N, Parks S, et al. Repeatability and reproducibility of
corneal thickness measurements by optical coherence tomography. Invest
Ophthalmol Vis Sci. 2002;43:1791–1795.
14. Wong AC-M, Wong C-C, Yuen NS-Y, et al. Correlational study of central
corneal thickness measurements on Hong Kong Chinese using optical
coherence tomography, Orbscan and ultrasound pachymetry. Eye. 2002;
16:715–721.
15. Bechmann M, Thiel MU, Neubauer AS, et al. Central corneal thickness
measurement with a retinal optical coherence tomography device versus
standard ultrasonic pachymetry. Cornea. 2001;20:50–54.
16. Fishman GR, Pons ME, Seedor JA, et al. Assessment of central corneal
thickness using optical coherence tomography. J Cataract Refract Surg.
2005;31:707–711.
17. Marsich MW, Bullimore M. The repeatability of corneal thickness
measures. Cornea. 2000;19:792–795.
18. Giasson C, Forthomme D. Comparison of central corneal thickness
measurements between optical and ultrasound pachometers. Optom Vis
Sci. 1992;69:236–241.
CCT: Z-Ring CCM Versus Ultrasound Pachymetry
19. Wheeler NC, Morantes CM, Kristensen RM, et al. Reliability coefficients
of three corneal pachymeters. Am J Ophthalmol. 1992;113:645–651.
20. Sanchis-Gimeno JA, Herrera M, Lleo-Perez A, et al. Quantitative
anatomical differences in central corneal thickness values determined with
scanning-slit corneal topography and noncontact specular microscopy.
Cornea. 2006;25:203–205.
21. Avitabile T, Marano F, Uva MG, et al. Evaluation of central and peripheral
corneal thickness with ultrasound biomicroscopy in normal and
keratoconic eyes. Cornea. 1997;16:639–644.
22. Miglior S, Albe E, Guareschi M, et al. Intraobserver and interobserver
reproducibility in the evaluation of ultrasonic pachymetry measurements
of central corneal thickness. Br J Ophthalmol. 2004;88:174–177.
23. Rainer G, Petternel V, Findl O, et al. Comparison of ultrasound
pachymetry and partial coherence interferometry in the measurement of
central corneal thickness. J Cataract Refract Surg. 2002;28:2142–2145.
24. Reader AL III, Salz JJ. Differences among ultrasonic pacymeters in
measuring corneal thickness. J Refract Surg. 1987;3:7–11.
25. Jalbert I, Stapleton F, Papas E, et al. In vivo confocal microscopy of the
human cornea. Br J Ophthalmol. 2003;87:225–236.
26. Ivarsen A, Stultiens BAT, Moller-Pedersen T. Validation of confocal
microscopy through focusing for corneal sublayer pachymetry. Cornea.
2002;21:700–704.
27. Mandell RB, Fatt I. Thinning of the human cornea on awakening. Nature.
1965;16:292–293
28. Harper CL, Boulton ME, Bennett D, et al. Diurnal variations in human
corneal thickness. Br J Ophthalmol. 1996;80:1068–1072.
29. Gambato C, Ghirlando A, Moretto E. Mitomicyn C modulation of corneal
wound healing after photorefractive cheratectomy in highly myopic eyes.
Ophthalmology. 2005;112:208–218.
q 2007 Lippincott Williams & Wilkins
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
307