Does the tear film removehigher order wavefrontcorrections?
Charles Campbell
The role of tear film on therefractive state of the eyen The strongest refracting surface of the eye is
the air/tear film interface
n However, if the tear film has a constantthickness it may be considered to be a zeropower element
The role of tear film on therefractive state of the eye
n If the tear film is constant thickness, i.e.a zero power element, the cornealsurface may be considered as the firstrefracting element
n In refractive surgery it is the cornealsurface that is changed and this is theeffect that is usually considered - thetear film is neglected
The role of tear film on therefractive state of the eye
n But suppose the tear film is not constantthickness and so the air/tear surfacedoes not replicate the corneal surface
n Now the tear film is no longer a zeropower element and must be considered
n When can this happen? How will effectrefractive corrections?
Effect of a tear film of non-constant thickness
n If a tear film initially of constantthickness covering a local correction inform a local depression
n Flows and fills in that depression,much the corrective effect is removed
The flow of thin fluid films
n Will the tear film flow to partially eraselocal refractive corrections ablated intothe corneal surface in the time betweenblinks?
n To answer this question we need toconsider how and why thin fluid filmsflow
The flow of thin fluid films
n Parameters effecting flown Surface tension (s)
n Mean surface curvature gradient (—M)
n Viscosity (m)
n Film thickness (h)
The flow of thin fluid films -surface tension (s)
n The surface tension at a gas/fluid interface -such as the air/tear film surface - acts tominimize surface area
n Surface area is minimized when the localmean surface curvature, M, is constant overthe interface surface
n Surface tension in a curved interface causesa pressure difference, P,between the gas andthe fluid to be exist
†
P = 2sM (Young-Laplace formula)
The flow of thin fluid films - meansurface curvature gradient (—M)
n A gradient in the mean surfacecurvature,—M, creates a pressuregradient,—P,in the fluid film
n The pressure gradient causes fluid toflow from areas of higher curvature toareas of lower curvature until meancurvature, and hence the pressure, isequalized
The flow of thin fluid films -viscosity (m)
n In a flowing fluid film in contact with airon one side and solid on the othern the fluid in contact with the solid surface
moves at exactly the same speed as thatsurface
n the fluid at the air interface flows withhighest velocity with respect to the solidsurface
The flow of thin fluid films -viscosity (m)n We may think of the thin flowing film as being
made of thin layers sliding with respect to oneanother
n Viscosity may be thought of as a ‘friction’between the layer that slows flow by inhibitingmovement between layers
n The flow restricting force due to viscosity isproportional to the velocity gradient,F=m∂V/∂h air
corneaFlow rate
The flow of thin fluid films -film thickness (h)
n For a given surface velocity, the thinner thefilm, the higher the velocity gradient must beand the greater the viscosity will resist flow
n Film flow rate is found to be proportional to h3
- as the film gets thinner the flow rate quicklyslows
Tear film dynamics – initialconditions – change with time
n The initial film after a blink – can thetear film be modeled as initially having aconstant thickness?
n How does the thickness change withtime?
Can we assume the tear film startslife with a constant thickness?
n To answer this question, we turn to coatingflow theory, the theory of creating thin film viaa wiper or dipping action widely usedindustrially forn Flow of paint
n Creation of thin emulsion films
n Coating surfaces - such as spectacle lenses
n For the eye, the wiper is the upper eye lidmoving over the corneal surface
Creation of a thin film by amoving wiper- coating flow
upper lid
+
upper lipmeniscus
Rm
hf tear film thickness
U
h(x)x
hl lubrication film thickness
y
n For coating flow the shape of the underlying surfacedoes not effect film thickness
†
hf = 1.3375Rm mU /s( )2 / 3
Thin film flow dynamics – changein thickness with time
†
∂h x, y( )∂t
=s3m
∂∂x
∂M x, y( )∂x
h x,y( )3( )Ê
Ë Á
ˆ
¯ ˜ +
∂∂y
∂M x, y( )∂y
h x, y( )3( )Ê
Ë Á
ˆ
¯ ˜
Ï Ì Ó
¸ ˝ ˛
n Equation giving the time rate of change of thesurface air/fluid surface height
n This equation allows the change in surfaceheight over short periods of time to be directlycalculated
Simulation of tear film flown Assume an initial constant tear thickness (~ 6 m)n Assume the tear film surface assumes the initial shape
and local curvature of the corrected wavefront errorn Calculate the local curvature values from the surface
elevation valuesn NOTE – second order surfaces have constant mean
curvature i.e. —M = 0
n Using the differential thickness change equationcalculate the rate of thickness change for a short periodof time ~ 10 ms
n Find the new surface shape and curvature valuesn Iterate the process for a length of time representative of
the time between blinks
†
∂h x, y( )∂t
=s3m
∂∂x
∂M x, y( )∂x
h x,y( )3( )Ê
Ë Á
ˆ
¯ ˜ +
∂∂y
∂M x,y( )∂y
h x, y( )3( )Ê
Ë Á
ˆ
¯ ˜
Ï Ì Ó
¸ ˝ ˛
Simulation for an eye with an aboveaverage amount of higher orderaberrations.
n The change in tearfilm thickness fromthe initial conditionof constantthickness
Simulation for an eye with an aboveaverage amount of higher orderaberrations.
n The change in tearfilm thickness fromthe initial conditionof constantthickness
RMS error change = 0.009micron @ 1/61 l
Experiment observation of tear filmflow via live corneal topography video
n Simulation is fine but can these effectsbe physically measured?
n Flow effects can be observed byrecording live corneal topographyvideo and observing surface curvaturechanges as the corneal topographermires change shape.
Tear film relaxation following blink ina 4 day post op PRK corneal surface
Motion just after a blink -4 dayspost-op PRK
Untreated cornea having a 50 yearold scar at about 165° that induceshigher order aberrations
Conclusionsn Tear film flow following initial film formation by
a blink can be demonstrated by computersimulation and observed with live cornealtopography video
n For surface variations of the size expectedfrom wavefront corrections the change in thetear film surface is too small to causenoticeable changes in visual image quality
n Tear film dynamics will not erase theeffects of ablative corrections for higherorder aberrations.