topo and imaging - coavision.org
TRANSCRIPT
1
University Hospitals Eye Institute 1Please no cameras or recorders during the class presentation. You will be asked to leave the room if request is not followed.
201Corneal Imaging in the Modern
Optometric PracticeLoretta Szczotka-Flynn, OD, PhD
Room: Crystal C
#OptoWest
University Hospitals Eye Institute 2
#OptoWest
Make sure to check in and out for each course – you must do this even if you are staying in the same room for multiple courses.
Hand in your course ticket as you leave the class in order to receive credit. The ticket is the ONLY way to receive credit for this course.
If you must leave the class for any amount of time keep in mind if you are out of the room for more than 10 minutes, you will not receive any CE credit.
There will be a “coffee break” in the exhibit hall, sponsored by VSP Global, Friday and Saturday from 10:30am to 11:30am.
University Hospitals Eye Institute 3
Please remember to complete the evaluation forms on the back of your course tickets.
A special thank you goes to our industry sponsors: Vision West, VSP Global, Vision West, VSP Global, AllerganAllergan, , CooperVisionCooperVision, Alcon, Practice Management, , Alcon, Practice Management, EssilorEssilor and and GenzymeGenzyme for their support of this conference.
If you have a cell phone or pager, please turn it off. If you must take a call, do so outside the room.
#OptoWest
University Hospitals Eye Institute 4
201 – Corneal Imaging in the Modern Optometric PracticeLoretta Szczotka-Flynn, OD, PhD
This course material and information was developed independently of any assistance.
I have the following financial relationships to disclose:• Alcon: Honorarium/Writing & Speaking• Bausch & Lomb: Honorarium & Travel/Speaking• Vistakon: Research Grant/Research
University Hospitals Eye Institute 5Please no cameras or recorders during the class presentation. You will be asked to leave the room if request is not followed.
201Corneal Imaging in the Modern
Optometric PracticeLoretta Szczotka-Flynn, OD, PhD
Room: Crystal C
#OptoWest
LORETTA SZCZOTKA-FLYNN OD, PhD
CORNEAL IMAGING IN THE MODERN OPTOMETRIC PRACTICE
2
University Hospitals Eye Institute
LORETTA SZCZOTKA-FLYNN O.D.,Ph.D.,F.A.A.O.(Dipl)
Professor
Dept. of Ophthalmology
Case Western
Reserve University
Director: Contact Lens Service
UH Eye Institute
University Hospitals Eye Institute 8
TOPICS FOR TODAY
• Corneal Topography
• Anterior Segment OCT
• Confocal Microscopy
University Hospitals Eye Institute
Corneal Topography:
• Is the gold standard in contact lens practices• Acceptance credited to:
– lowered prices
– smaller instruments
– portable instruments
– software enhancements
University Hospitals Eye Institute
Contact Lens Applications
• Standard map displays – Select qualitative designs
• spherical, toric, aspheric• appraisal of corneal apex; bifocals
– Select quantitative parameters• BC, PC, OAD, OZ
• Contact lens software modules• Other software modules
– KC detection, shape indices, irregularity
• Monitor CL induced change • Pachymetry• Wavefront analysis•
University Hospitals Eye Institute
LIMITATIONS OF THE KERATOMETER
• Power and curvature at visual axis only• Qualitative assessment of regularity
– No interpretation or quantitative data
• Measures 4 points 3.6mm apart– 5% of the corneal surface (depends on K’s)– Lacks central & peripheral data
• Assumes sphero-cylindrical surface• Assumes orthogonal symmetry
University Hospitals Eye Institute 12
Subjective Keratometry Errors
3
University Hospitals Eye Institute 13
Mapping the Cornea
• Placido Disc Topography
• Rasterstereography– Par Vision Systems; elevation; based on stereo-photography
• Interferometry– Euclid System; Elevation topography
• Scanning Slit Topography– Orbscan II (includes placido)
• Scheimpflug– Pentacam
University Hospitals Eye Institute 14
Reflective and Slit-scan Technologies
• One image, one surface.• Angle-dependent specular
reflection.• Measures slope (as a
function of distance).
• Multiple images, multiple surfaces.
• Omni-directional diffuse backscatter.
• Triangulates elevation.
Two prevailing technologies have been used for corneal topography. Both have advantages and disadvantages. The overwhelming advantage of slit-scan systems is that they measure multiple ocular surfaces.
Placido reflective systems can only measure the anterior tear film. ORBSCAN measures the anterior cornea, posterior cornea, and the anterior lens and iris.
University Hospitals Eye Institute 15
What are the choices?Placido Disc Systems
• Optikon/Eyequip– Keratron; Scout
• Humphrey Instr.– Atlas
• Medmont– Medmont E300
• Technomed Tech– Technomed, colored rings
• Tomey Tech.– TMS-1, 2, 3
• Topcon– Infrared Placido unit
• Nidek– combo Placido, spatial refractometer
• Orbscan II
• Others
* discontinued* discontinued
University Hospitals Eye Institute 16
Placido curvature may be ambiguous.d
ista
nt p
atte
rnd
ista
nt p
atte
rn
convex surface
concave surface
R > 0
image is smalland erect
image is smalland inverted
R < 0
Placido images
To see why, compare convex and concave test surfaces. The only difference is the direction (erect or inverted) of the specular image.Curvature ambiguity arises precisely because erect and inverted Placido images look identical.
Reflective devices can suffer other ambiguities.
Because of its inherent symmetry, Placido reflective corneal topography can NOT disambiguate a central hill from a central depression.
Both appear as a local increase in curvature, that is, as “central islands”.
University Hospitals Eye Institute 17
Avoiding Artifacts and Misinterpretation
• 1) Eyelid position
• 2)Ring digitization errors
• 3) Color scales
• 4) Corneal apex alignment
• 5) Selecting the correct map
University Hospitals Eye Institute 18
1) Eyelid Position
• Keratron only– eyelash breaks ring pattern
– acquires image when not in focus
4
University Hospitals Eye Institute 19
2) Ring Digitization
• Long vs Short Working Distance Systems
Long Working Distance SystemsLong Working Distance Systems
University Hospitals Eye Institute 20
• Short Working Distance System
University Hospitals Eye Institute 21
View the placido image
University Hospitals Eye Institute
3) Decoding Color Scales
• Absolute Scale– Standard
• Normalized Scale– Color Map
– Autosize
• Adjustable Scale– Customized
University Hospitals Eye Institute
Absolute Scale
• Fixed Color Coded Scheme– Always assigns same color to a given dioptric interval
– Forces data to fit within a predetermined range
University Hospitals Eye Institute 24
Saturated scales
• Can saturate extreme ends of scales• Tomey 9-100 D
• Keratron 9-101.5 D
• EyeSys/Premier 35-52 D
• Humphrey 39-50 D
5
University Hospitals Eye Institute
Absolute Scale
• Advantages– Visually compare eyes or occasions
– Quickly visualize significant corneal changes
– Quick insight for best CL fitting approach
• Disadvantages– Large dioptric range
– Large intervals• Local irregularities may be masked
– Less detail if scale saturates
University Hospitals Eye Institute
Normalized Scale
• Automatic subdivision of each cornea into equal, adjusted, dioptric intervals – Set number of colors to fill dioptric range
– Color intervals vary between eyes/occasions
– Usually smaller intervals than axial maps• Min 0.40 D steps Tomey
• Min 0.50 D steps EyeSys/Premier
• Min 0.25D steps Humphrey
University Hospitals Eye Institute 27 University Hospitals Eye Institute
Normalized Scale
• Advantages– Significantly more detail
– Highlights local irregularities
• Disadvantages– Can be misleading
– Cannot visually compare maps without referring to associated color scales
University Hospitals Eye Institute 29 University Hospitals Eye Institute 30
6
University Hospitals Eye Institute 31 University Hospitals Eye Institute
Scale selection in Contact Lens Fitting
• Absolute Scale– Quickly determine fitting approach based on
• Corneal Astigmatism
• Average corneal curvature
• Normalized – May sway the fitting approach if scale not referred to
University Hospitals Eye Institute
Lower left:Spherical RGP orSoft toric
Lower right:Bitoric RGP withmachinable polymer
Upper left:Bitoric/LidAttachment
Upper right:Bitoric/InterpalpebralUniversity
Host
University Hospitals Eye Institute 34
Relatively normal surfaces can appear complex due to minor variations.
University Hospitals Eye Institute 35
Here are two anterior elevation maps from normal corneas that could be confusing to read. The color changes are presented using a standard scale in 5 micron steps. The problem with large step sizes is that it tends to mask large changes by lumping them into a limited number of color steps categories. Also by changing step sizes and scaling for different cases, one can easily get confused and under or over diagnose pathology.
Anterior Elevation Maps:Standard Color Scale
University Hospitals Eye Institute 36
Nor
mal B
and
Nor
mal B
and
Scale
Scale
Stan
dard
Color
St
anda
rd C
olor
Sc
ale
Scale
A different approach is to set up a “normal band” or series of uniform color steps within which changes would not be considered pathological. In this way scale changes within a certain physiological range would all be represented by a single color and changes outside of the
range would show up as different colors.
The step size of the scale is user selected as is the center point and width of the normal band. For anterior and posterior corneal elevation maps I have chosen a step size of 5 microns. The center point of the normal band is zero deviation from the best fit sphere, and the width of the normal band is 50 microns, or 25 microns more elevated and 25 microns more depressed than the best fit sphere.
Elevation Color Scales
Everything in the normal band will be shown as green.
7
University Hospitals Eye Institute 37
Normal Band Scale
On the left the standard scale for the normal anterior corneal surface is used and on the right the normal band scale is used showing an all green normal surface.
Standard Color Scale
Anterior Elevation
University Hospitals Eye Institute 38
The normal band scale is a screening tool for gross pathology. Using the standard scale, the with-the-rule astigmatic pattern in this patient is obvious but is eliminated by use of the normal band scale.
Normal Band Scale
Standard Color Scale
Anterior Elevation
University Hospitals Eye Institute 39
The normal band scale can also be used with 3-D Orbscan maps. The image on the lower right is the normal band representation of the cornea on the left.
Anterior Corneal Elevation
University Hospitals Eye Institute 40
Here, the normal band scale on the lower right clearly demonstrates a posterior cone centrally in the 3-D posterior corneal surface elevation map.
Posterior Corneal Elevation
University Hospitals Eye Institute 41
For keratometric curvature maps we’ve chosen a normal band of from 40 to 48 keratometric diopters. Within this range all values will appear as green on the Normal Band scale.
Keratometric Curvature Color Scales
Nor
mal B
and
Nor
mal B
and
Scale
Scale
Stan
dard
Color
St
anda
rd C
olor
Sc
ale
Scale
Editor Note: A keratometric curvature is any anterior surface curvature expressed in keratometric diopters [K]. Because of its dioptric evaluation, it has often been referred to as a power, although it is really a scaled surface curvature. That is, it is a geometric property, not an optical University Hospitals Eye Institute 42
Normal Band ScaleStandard Color Scale
It should be emphasized that this scale could be changed for individual needs and as shown here some rather significant changes can be masked so its usefulness with such a wide scale is for screening.
Tangential Curvature [K]
8
University Hospitals Eye Institute 43
For corneal thickness we are using a normal band from 500 to 600 microns. Obviously this scale should be changed for post-Lasik or PRK eyes.
Corneal Thickness Color Scales
Nor
mal B
and
Nor
mal B
and
Scale
Scale
Stan
dard
Color
St
anda
rd C
olor
Sc
ale
Scale
University Hospitals Eye Institute 44
Normal Band Scale
Corneal Thickness Map
Standard Scale
For corneal thickness we are using a normal band from 500 to 600microns. Obviously this scale should be changed for post-Lasik or PRK eyes.
University Hospitals Eye Institute 45
Standard Scale Quad Map
This is a quad map of an untreated eye where the anterior corneal elevation map is shown on the upper left quadrant, the posterior corneal elevation map is on the upper right quadrant, corneal power map is on the lower left quadrant and corneal thickness map is on the lower right quadrant. Determining normal from abnormal features is certainly not obvious.
University Hospitals Eye Institute 46
Posterior Keratoconus:Normal Band Scale
Here is the same eye using the normal band scale. The anterior elevation map on the upper left is now all green indicating a normal pattern. The corneal power map in the lower
left quadrant is virtually all yellow centrally indicating corneal powers between 40-48D.
The first appearance of abnormality is on the posterior corneal elevation map on the upper right where an area of central elevation is present with 2 areas of peripheral depression. On the corneal thickness map on the lower right a central area of corneal thinning below 500 microns is seen. The most likely diagnosis is a posterior cone with mild central corneal thinning.
University Hospitals Eye Institute 47
The other eye of this patient underwent Lasik and here is that quad map using the standard scale.
Posterior Keratoconus:Standard Color Scale
University Hospitals Eye Institute
4) Corneal Apex Position
• Classification of Normal Corneal Topography ; Bogan et al; Arch Ophthal 1990
• 399 normal corneas; Axial Maps• 22.6% round• 20.8% oval• 17.5% symmetric bow tie• 32.1 % asymmetric bow tie• 7.1% irregular
9
University Hospitals Eye Institute 49
Primary GazePrimary Gaze UpgazeUpgaze
University Hospitals Eye Institute
5) Selecting the Correct Display
• Placido Image
• Eye Image Overlay
• Axial
• Tangential
• Elevation
• Refractive
• Distortion/Irregularity
• Difference
• Serial Images
• Isometric
• Keratometric
• Meridional
• Numeric
University Hospitals Eye Institute 51
Eye Image Overlay
University Hospitals Eye Institute
Understanding the Maps
• Curvature Maps– Axial
• Sagittal
• Color
• Default
– Tangential• Instantaneous
• Local
• True
• Refractive Maps• Snell’s Law
• Power
• Elevation Maps• Height
• 3-D
University Hospitals Eye Institute
Axial “radius of curvature”
• Distance along the normal from the corneal surface to the optic axis
• A reference distance, not a true curvature
• Most recognizable display
• Running avg/excludes extreme values
• Spherically biased
• Used in keratometry
University Hospitals Eye Institute 54
Axial Curvature Map
Axial Curvature ( a.k.a. Color, Power, or Sagittal Map) displays a kind of radially-averaged curvature.This eye has with-the-rule astigmatism. The steep meridian is aligned with the red-shifted bow-tie.Axial curvature is measured exclusively in meridional planes (radial directions).The bow-tie is actually an axial artifact generated by the central intersection of the meridional planes.
Axial Curvature
10
University Hospitals Eye Institute
Tangential radius of curvature
• Based on std. mathematical formula for local radius at a point along a curve
• Includes extreme curvature values
• Proportional to local curvature
• Derivative of axial data
• Axis independent
• More detail
University Hospitals Eye Institute 56
University Hospitals Eye Institute
Tangential maps for CL fittingTangential maps for CL fitting
• More accurate corneal shape– Correlated to slit lamp observations– RGP fluorescein patterns– Optical zone selection
• More detail– Detect irregular astigmatism– CL induced warpage– Monitor disease progression
University Hospitals Eye Institute 58
Clinical Appearance of KC
University Hospitals Eye Institute
Axial maps for CL fittingAxial maps for CL fitting
• Larger, diffuse patterns
• Less noise, but less detail
• Global representation of shape– More appropriate for spherical RGP curve selection
• Normals
• KC
• Post-Surgical/Oblate shapes
University Hospitals Eye Institute
Axial maps in keratoconus
• Not sensitive enough for localized curvature changes off VK axis; which effects:
– Apex curvature and position• How does this effect RGP fitting?
– KC detection
– KC progression
11
University Hospitals Eye Institute
Tangential maps in keratoconus
• POSSIBLE APPLICATIONS:– KC Progression
– Early KC detection (for PRK screening)
– KC Subset classification based on cone shape, size, and position
– KC Statistical Indices
– Contact Lens Fitting??
University Hospitals Eye Institute 62
University Hospitals Eye Institute 63 University Hospitals Eye Institute
Elevation MapsElevation Maps
• Depict relative height differences from a reference sphere
• Reds: higher than reference sphere
• Blues: lower than reference sphere
University Hospitals Eye Institute 65
high
sea-level
low
color contour map
terrain profile
Elevation, whether of the earth or the cornea, is measured relative to some reference surface. The terrestrial reference surface is the “mean sea level”.
Relative Elevation
University Hospitals Eye Institute 66
With-the-Rule Astigmatism MapsAxial CurvatureElevation (sphere)
Elevation and axial curvature maps look different, almost reversed.On the axial map, the steep axis follows the sharpest curvature (red line segments) and aligns to the red-shifted bow-tie.On the relative elevation map, the steep axis dips below the reference surface and runs from “sea” to “sea”.
12
University Hospitals Eye Institute 67
Topographical Elevation is Relative
5.6 mi 5.6 miles
12,000 miles
On a local scale these features are significant
University Hospitals Eye Institute 68
Local clinical features (microns high) are a thousand times smaller than the cornea (millimeters in radius).This disparity in scale makes it IMPOSSIBLE to simultaneously map both the global curvature and clinically significant local elevation features of the cornea.To see clinically relevant features, corneal global curvature must first be removed. This is done by measuring elevation relative to some close-fitting reference surface, a process that inevitably introduces distortion.
Removing Global Curvature
University Hospitals Eye Institute 69
Elevation Topology: Central Hill
The normal cornea is prolate, meaning that meridional curvature decreases from center to periphery.Prolateness of the normal cornea causes it to rise centrally above the reference sphere. The result is a central hill.
Sharp center
Flat periphery
Immediately surrounding the central hill is an annular seawhere the cornea dips below the reference surface.In the far periphery, the prolate cornea again rises above the reference surface, producing peripheral highlands.
University Hospitals Eye Institute 70
Spherical reference surface
Elevation Distortion
Post Lasik profile
As an example of distortion, consider the corneal surface following myopic lasik correction. It is centrally flattened bythe surgery.To see surface features, elevation must be measured with respect to some reference surface.This relative elevation peak is NOT the highest point on the cornea.This apparent central "concavity" does NOT exist.
Relative elevation profile
University Hospitals Eye Institute 71
Elevation (sphere) Elevation (sphere)
These are pre-op (left) and post-op (right) elevation maps of a myopic with-the-rule astigmatic eye corrected with LASIK.The post-operative central “sea” is not a concavity but a central flattening.The ring of relatively highest terrain is not absolutely higher (more anterior) than the “sea” bottom near the map center.
University Hospitals Eye Institute 72
Close-Fitting Reference Surfaces
Topographic maps of terrestrial landscapes are displayed in the form of constant-elevation contours, measured from the “mean sea-level” of the earth.Corneal topography differs from terrestrial topography in that the reference surface is not some fixed “mean sea-level”, but is movable.
For the cornea, a reference surface (typically, a sphere) is constructed by fitting the reference surface as close as possible to the data surface.A best-fit minimizes the square difference (always a positive number) between the two surfaces, but only within a specified region known as the fit-zone.
Fit-zoneReference surface (sphere)
Data surface(cornea)
13
University Hospitals Eye Institute 73
Elevation (sphere)10 mm fit-zone
Elevation (sphere)5 mm fit-zone
These elevation maps are the same exam of a normal eye having with-the-rule astigmatism. They differ only by the size of the fit-zone used to construct the reference sphere.Both maps contain exactly the same information. They look different, because altering the fit-zone changes the size and alignment of the reference sphere.
7.16 mm radius 6.88 mm radius University Hospitals Eye Institute 74
Default settings:• shape = sphere
Elevation Relativity
Changing the shape, size, or alignment of the reference surface changes the relative elevation, just like changing the mean sea level would change the heights of mountains.
The default reference surface is the best-fit sphere. Other reference surface shapes are used for special applications.The sphere is especially useful, because it is only reference surface with no unique symmetry axis (i.e., every diameter is a symmetry axis).
Two parameters define the best-fit sphere. Radius specifies size. Center location specifies alignment.The default sphere alignment is floating, which means its center location is unconstrained. Other alignments may force the reference sphere to lie on the map axis.
• alignment = floating
University Hospitals Eye Institute 75
Float Center (C) Pin (P)same data surface
unconstrainedsphere center
constrainedsphere center
vie
w
a
xis
vie
w a
xis
vie
w
a
xis
Floating alignment imposes no additional constraints on the reference sphere. The sphere moves and changes size until it best-fits the data surface within the specified fit-zone.
Axial alignment imposes the center (C) constraint. The reference sphere center is forced onto the view axis, although it may move up or down the axis.
Pinned alignment imposes the pin (P) constraint, which forces the reference sphere surface to intersect the data surface on the view axis. The sphere center may be off axis.
“Apex” alignment imposes both center and pin (C+P)constraints. The sphere center lies is on the view axis, and the sphere surface intersects the data surface on the view axis.
Reference Sphere Alignments
University Hospitals Eye Institute 76
Normal Eye: 4 Alignments
T NN T
Axial (C)Float ( )
Pin (P) Apex (C+P)
This is a normal eye with some with-the-rule astigmatism. The four elevation maps differ only by reference sphere alignment. The fit-zone is the largest 10 mm zone.Floating alignment (no constraints) reveals the natural symmetry of the eye. The reference sphere is slightly nasal and superior. This alignment displaces the central hill (arrow) inferiorly and temporally.
Applying the pin constraint raises the reference sphere. The sphere center still floats superior nasally, and the central hill (arrow) remains displaced inferior temporally.
Axial alignment centers the reference sphere, which artificially centers the central hill (arrow). Apex alignment centers and raises the reference sphere, which centers and lowers the central hill (arrow).
Elevation (sphere)
University Hospitals Eye Institute 77
Artificial Alignment MapsElevation (sphere)Floating alignment
Elevation (sphere)Axial (C) alignment
This is a post-op exam of an originally myopic eye corrected with Lasik. The two elevation maps differ only by the alignment of the reference sphere.Floating alignment (left) shows the ablation (yellow circle) to be slightly offset in the pupil centroid (white dot) direction.Axial alignment (right), which forces the reference sphere center onto the map axis, moves the apparent ablation pattern 0.3 mm towards the map axis.Neither map is wrong, but floating alignment (no center constraint) allows the reference sphere to follow the natural symmetry of the surface.Neither map depression tracks the actual ablation edge, but only the smoothly varying corneal contour as seen from the reference sphere.
University Hospitals Eye Institute
Elevation MapsElevation Maps
• Colors identical to curvature displays, but represent height only
• Steep(red) curvature does not always mean high!• Flat (blue) curvature does not always mean low!
14
University Hospitals Eye Institute
Elevation maps for RGP fittingElevation maps for RGP fitting
• Most useful for RGP fluorescein patterns (vs. curvature maps):
– Reds: ALWAYS displace FL
– Blues: ALWAYS pool FL
University Hospitals Eye Institute 80
University Hospitals Eye Institute
Irregularity/Distortion MapsIrregularity/Distortion Maps
• Describes the visual limitations of the ocular surface– Humphrey Irregularity Map
• Local elevation differences between the actual corneal surface and the best fit regular toric surface
– EyeSys Holladay Display/Distortion Map
University Hospitals Eye Institute 82
University Hospitals Eye Institute 83
Difference Maps
Specialized software for Keratoconus Detection
15
University Hospitals Eye Institute 85
Keratoconus Detection Software
Yaron Rabinowitz MD
Normals KC
Central 44.2+-1.6 49.9+-4.8
I-S Value 0.12+-0.57 3.7+-3.81
R vs L 0.42+-0.25 3.66+-3.25
University Hospitals Eye Institute
Videokeratography PatternsVideokeratography Patterns
• Round
• Oval
• Superior Steep (SS)
• Inferior Steep (IS)
• Irregular
• Symmetric Bowtie (SB)
• SB/SRAX
• Asymmetric Bowtie AB/IS
• Asymmetric Bowtie AS/SS
• AB/SRAX
University Hospitals Eye Institute 87 University Hospitals Eye Institute 88
Videokeratographic Indices to Aid in Screening for KC
• Rabinowitz, Y. 1995– Central K > 47.20 D
– I-S Value >1.40 D
– SRAX >21 degrees
– Specificity=98%
University Hospitals Eye Institute 89
KISA% indexRabinowitz Y, Rasheed K. JCRS 1999
• KISA%= K x I-S x AST x SRAX x 100300
– Clinically detectable KC • KISA%>100
– KC suspect• KISA% 60-100
– Advanced KC• KISA%>5000
University Hospitals Eye Institute
MethodsMethods
• Quantitative Descriptors:– SRI; Surface Regularity Index
– SAI; Surface Assymetry Index
– SimK
– Central K
– I-S value
– C-P value
16
University Hospitals Eye Institute 91 University Hospitals Eye Institute 92
Defining Controls: Orbscan II
• In non-diseased eyes, the average amount of maximum posterior elevation using OrbscanII default settings compared to the best fit sphere is about 21-28 μm– Rao 2002
– Wei 2006
– Lim 2007
– Sonmez 2007
University Hospitals Eye Institute 93
Defining Controls: Orbscan II
• One study of 140 normal eyes documented the maximum posterior elevation was never greater than 46 μm– Wei 2006
• Another study of 50 normal eyes the posterior elevation was never greater than 40 μm– Rao 2002
University Hospitals Eye Institute 94
Defining Controls: Orbscan II
• Kollbaum (2006) delineated normals from clinically apparent KC– Posterior elevation was found to be the best performing single Orbscan
II index to detect KC• When highest elevation on the posterior surface was greater than 50 μm above the best-fit sphere.
– Sensitivity 0.93, Specificity 0.99, Accuracy 0.96.
Posterior Elevation on Orbscan II >50 microns indicative of Keratoconus
University Hospitals Eye Institute 95
kjkjkjkjkjkjkjkjkjkjkjkj
lklklklklklklklklklklklklklklllklklklklklklklklkl
Figure 3
University Hospitals Eye Institute 96
Defining Controls: Pentacam
• Posterior elevation ranges from -6 µm to + 18 µm in normals
• 15.5-32 μm of elevation on the posterior corneal surface can delineate forme frusteKC from normals
– de Su, Loiacono C, Richiardi L, et al Ophthalmology 2008 – Mihaltz K, Kovacs I, et al Cornea 2009 August 29.
• Feng et al recently published international values of corneal elevation in normals– Upper limits of normal for collective international data were 7.5 μm
and 13.5 μm at the posterior apex and posterior thinnest point, respectively.
17
University Hospitals Eye Institute 97
Figure 1 Orbscan posterior elevation of same patient in Figure 2
University Hospitals Eye Institute 98
Figure 2 Pentacamposterior elevation of same patient in Figure 1
University Hospitals Eye Institute 99
Scheimpflug Cut-points
• Pentacam– Mean posterior elevation greater than 29-32 μm (in the central 5
mm cornea) is predictive
• Galilei Scheimpflug-based analyzer – Elevation in the central 7 mm zone > 50.5 μm was predictive
Posterior Elevation >30 microns on Pentacamsuggests keratoconus
Anterior Segment OCT
Slides courtesy of Mark Harrod
UH Eye Institute Ophthalmic Photographer
University Hospitals Eye Institute 101
AS-OCT: What is it?
• An OCT with a lens adaptation
• Shows cross-sectional view of anatomy.– Cornea
– Angle
– Iris
– Lens
– Pachymetry
– Topography
• More detail than AS ultrasound
University Hospitals Eye Institute 102
AS-OCT: What is it?
• SD-OCT– Scan speeds: up to 26,000 a-scan/sec
– Axial resolution of 5 microns
– Scan depth of 2 mm• Optovue RTVue with Cornea Adapter Module
• Heidelberg Spectralis with Anterior Segment Module
• Bioptgen Envisu
• TD-OCT– Scan speeds of 2000 a-scan/sec
– Axial resolution of 18 microns
– Scan depth up to 6 mm• Zeiss Visante Omni
18
University Hospitals Eye Institute 103
AS-OCT: What is it?
• Comparison of TD-OCT vs SD-OCT
Figure from: Moyer S Cohen K. Anterior Segment OCT: A Comparison of Time Domain and Spectrial Domain Technologies J Ophthalmic Photography Summer 2009; Vol 31:90
University Hospitals Eye Institute 104
AS OCT: What Are We Looking At?
• Corneal Thickness– Normal
University Hospitals Eye Institute 105
AS OCT: What Are We Looking At?
• Corneal Thickness– Edema
University Hospitals Eye Institute 106
AS OCT: What Are We Looking At?
• Corneal Thickness– Edema
University Hospitals Eye Institute 107
AS OCT: What Are We Looking At?
• Corneal Thickness– LASIK Flap
Image Source: Optovue.
University Hospitals Eye Institute 108
AS OCT: What Are We Looking At?
• Corneal Thickness– Keratoconus: Mild
19
University Hospitals Eye Institute 109
AS OCT: What Are We Looking At?
• Corneal Thickness– Keratoconus: Moderate
University Hospitals Eye Institute 110
AS OCT: What Are We Looking At?
• Corneal Thickness– Keratoconus: Severe
University Hospitals Eye Institute 111
AS OCT: What Are We Looking At?
• Corneal Thickness– Keratoconus: Severe
University Hospitals Eye Institute 112
AS OCT: What Are We Looking At?
• Corneal Pathology– DESEK
Image Source: Optovue.
University Hospitals Eye Institute 113
AS OCT: What Are We Looking At?
• Corneal Pathology– Abrasion
University Hospitals Eye Institute 114
AS OCT: What Are We Looking At?
• Cornea Pathology– S/P PK
20
University Hospitals Eye Institute 115
AS OCT: What Are We Looking At?
• Scleral Contact Lens Fitting
University Hospitals Eye Institute 116
University Hospitals Eye Institute 117 University Hospitals Eye Institute 118
University Hospitals Eye Institute 119 University Hospitals Eye Institute 120
21
University Hospitals Eye Institute 121 University Hospitals Eye Institute 122
University Hospitals Eye Institute 123 University Hospitals Eye Institute 124
University Hospitals Eye Institute 125
AS OCT: What Are We Looking At?
• Lens and Capsule– Normal
Image Source: Optovue.
University Hospitals Eye Institute 126
AS OCT: What Are We Looking At?
• Lens and Capsule– Captured IOL
22
University Hospitals Eye Institute 127
AS OCT: What Are We Looking At?
• Iris Scans– Iris Cyst
University Hospitals Eye Institute 128
AS OCT: What Are We Looking At?
• Iris Scans– Iris cyst
University Hospitals Eye Institute 129
AS OCT: What Are We Looking At?
• Iris Scans– Iris Mass
University Hospitals Eye Institute 130
AS OCT: What Are We Looking At?
• Iris Scans– Open Angle
University Hospitals Eye Institute 131
AS OCT: What Are We Looking At?
• Iris Scans– Narrow Angle
University Hospitals Eye Institute 132
AS OCT: What Are We Looking At?
• Iris Scans– Closed Angle
23
University Hospitals Eye Institute 133
AS OCT: What Are We Looking At?
• Iris Scans– Closed Angle
University Hospitals Eye Institute 134
AS OCT: What Are We Looking At?
• Iris Scans– Shunt
University Hospitals Eye Institute 135
Confocal MicroscopyWhat Can it Show Me?
University Hospitals Eye Institute 136
Confocal Microscopy: What is It?
University Hospitals Eye Institute 137
Confocal Microscopy: What is It?
• Light Microscopy– Nidek CS4
• SLO– Heidelbeg HRT with Rostock Camera Module
University Hospitals Eye Institute 138
Confocal Microscopy: What Are We Looking At?
• Normal– Z-Scan Showing Reflectance Peaks, Pachymetry
24
University Hospitals Eye Institute 139
Confocal Microscopy: What Are We Looking At?
• Normal
Endothelium Posterior Stroma Anterior Stroma
Nerve Plexus Basal Epithelium Epithelium
University Hospitals Eye Institute 140
Corneal Endothelial Image Analysis Parameters
• Cell density (cells/mm2)
• Cell area (µm2)
• Coefficient of variation – polymorphism or variation in cell
size
• % Hexagonal cells– pleomorphism or variation in cell
shape
Variable frame analysis
Corners Method/Morphometric Analysis
University Hospitals Eye Institute 141
Analytic methods: A Fixed-frame analysis. B Variable-frame analysis. C Corners analysis. D Center analysis. E Center-Flex analysis. F Automated University Hospitals Eye Institute 142
Confocal Microscopy: What Are We Looking At?
• Normal • Low Cell Density
University Hospitals Eye Institute 143
Confocal Microscopy: What Are We Looking At?
• Airbag Injury
University Hospitals Eye Institute 144
Confocal Microscopy: What Are We Looking At?
• Airbag Injury
25
University Hospitals Eye Institute 145
Confocal Microscopy: What Are We Looking At?
• Fuchs’ Dystrophy
University Hospitals Eye Institute 146
Confocal Microscopy: What Are We Looking At?
• Epithelial Staining
University Hospitals Eye Institute 147
Confocal Microscopy: What Are We Looking At?
• Acanthameoba Keratitis
University Hospitals Eye Institute 148
Confocal Microscopy: What Are We Looking At?
• Acanthameoba Keratitis
University Hospitals Eye Institute 149
Confocal Microscopy: What Are We Looking At?
• Fungal Keratitis
University Hospitals Eye Institute 150
Confocal Microscopy: What Are We Looking At?
• Fusarium Keratitis
26
University Hospitals Eye Institute 151
Confocal Images of Corneal Superficial EpitheliumPeriod 1 Day 1 Period 2 Day 35
10011024 L •photophobia •corneal thickness increase > 45m•corneal staining •Headache
University Hospitals Eye Institute 152
Confocal of Corneal Basal Epithelial CellsPeriod 1 Day 1 Period 2 Day 35
10011024 L •photophobia •corneal thickness increase > 45m•corneal staining •Headache
University Hospitals Eye Institute 153
Confocal of Cornea Mid-Stromal KeratocytesPeriod 1 Day 1 Period 2 Day 35
10011024 L •photophobia •corneal thickness increase > 45m•corneal staining •Headache
Stromal Reflectivity
University Hospitals Eye Institute 155
The screen shots of Navis software
University Hospitals Eye Institute 156
Lower reflective stroma
27
University Hospitals Eye Institute 157
higher reflective stroma
University Hospitals Eye Institute 158
University Hospitals Eye Institute 159
PHMB solution user grade 2.5 stain
OD
OS
University Hospitals Eye Institute 160
PHMB User NO STAIN
University Hospitals Eye Institute 161
PHMB User: Grade 3 Stain
OD Baseline no stain
OD 5 month 3+ stain
Immune cells
28
University Hospitals Eye Institute 163
100 microns
University Hospitals Eye Institute 164
Resources:•Huang D, Duker J, Fujimoto J, Lumbroso B, Schuman J, Weinreb R eds, 2010: Imaging the Eye from Front to Back with RTVue Fourier-Domain Optical Coherence Tomography: Thorofare, NJ: Slack•Mastropasqua L, Nubile M, 2002: Confocal Microscopy of the Cornea: Thorofare, NJ: Slack•Holz FG, Schmitz-Valkenberg S, Spaide RF, Bird AC eds, 2007: Atlas of Autofluorescence Imaging: Heidelberg: Springer•Schmitz-Valkenberg S, Holz FG, Bird AC Spaide RF. Fundus Autofluorescence Imaging Review and Perspectives. Retina 2008; 28:385-409•Holtz FG, Dithmar S. 2008: Fluorescence Angiography in Ophthalmology: Heidelberg: Springer•Barry C, Morris P. eds, Journal of Ophthalmic Photography: Autofluoescence Supplement 2007; 29•Barry C, Morris P. eds, Journal of Ophthalmic Photography: SD OCT Supplement 2009; 31
University Hospitals Eye Institute 165
Thank You.