ARTICLE

Ray-tracing analysis of intraocular

lens power in situ
Thomas Olsen, MD, PhD, Mikkel Funding, MD

PURPOSE: To describe a method for back-solving the power of an intraocular lens (IOL) in situ
based on laser biometry and ray-tracing analysis of the pseudophakic eye.
SETTING: University Eye Clinic, Aarhus Hospital, Aarhus, Denmark.
DESIGN: Evaluation of diagnostic test or technology.
METHODS: This study comprised pseudophakic eyes with an IOL power ranging from

2.00 to C36.00 diopters (D). Preoperatively, the corneal radius was measured with conventional
autokeratometry and the axial length (AL) with optical biometry. After surgery, the position of the
IOL was recorded using laser interferometry. Based on the postoperative refraction and the biometric measurements, a ray-tracing analysis was performed back-solving for the power of the IOL in
situ. The analysis was performed assuming pupil diameters from 0.0 to 8.0 mm with and
without correction for the Stiles-Crawford effect.
RESULTS: The study evaluated 767 pseudophakic eyes (583 patients). Assuming a 3.0 mm pupil,
the mean prediction error between the labeled and the calculated IOL power (G1 standard deviation
[SD]) was
0.26 D G 0.65 (SD) (range
2.4 to C1.8 D). The prediction error showed no bias with
IOL power or with AL. The calculated IOL power depended on the assumed pupil size and the StilesCrawford effect. However, the latter had a modulatory effect on the prediction error for large pupil
diameters (>5.0 mm) only.
CONCLUSION: The optics of the pseudophakic eye can be accurately described using exact ray
tracing and modern biometric techniques.
Financial Disclosure: Dr. Olsen is a shareholder of IOL Innovations Aps, manufacturer of the
PhacoOptics IOL calculation software. Dr. Funding has no financial or proprietary interest in any
material or method mentioned.
J Cataract Refract Surg 2012; 38:641647 Q 2012 ASCRS and ESCRS

Numerous methods and formulas have been described to determine the intraocular lens (IOL) power
to be implanted from preoperative data of the patient,
and these methods have improved significantly over
the years.14 However, not every patient ends up
exactly as expected and the question then arises as to
what caused this refractive surprise. Often, a measurement error is the cause. In this case, the error can
usually be identified by repeating all biometric measurements and comparing them with the preoperative
data. In some cases, the formula or the IOL constant is
to blame, especially when the formula or IOL constant
has not been updated recently. In other cases, the
patient had laser in situ keratomileusis (LASIK)
many years ago and does not tell the surgeon.
However, in some situations the reason for the
refractive surprise is not obvious and the following
question then arises: Do you actually know what
Q 2012 ASCRS and ESCRS
Published by Elsevier Inc.

power was implanted? Did your staff (and you) take

the wrong box? And consider this: What if the power
of the IOL actually implanted differed from the power
that was on the label? The aim of the present study was
to describe a method to measure the actual effective
IOL power in the pseudophakic eye based on modern
biometry and ray-tracing analysis.
PATIENTS AND METHODS
This retrospective consecutive series comprised patients
referred for cataract or refractive IOL surgery who had
primary IOL implantation from January to December 2010.
Cases with preoperative or postoperative astigmatism
greater than 4.0 diopters (D), with keratoconus, or with previous trauma or surgery (ie, post-LASIK or corneal transplantation) were excluded from the series.
Before surgery, the corneal radius was measured in 2 meridians with an automated keratometer (ARK700, Nidek,
Ltd.) and averaged. The axial length (AL) was measured
0886-3350/$ - see front matter
doi:10.1016/j.jcrs.2011.10.035

641

642

RAY-TRACING ANALYSIS OF THE IOL POWER IN SITU

with partial coherence interferometry (PCI) optical biometry

(IOLMaster, software version 3.0 or higher, Carl Zeiss Meditec AG). The instrument was calibrated at weekly intervals.
The surgical technique was small-incision sutureless clear
corneal or scleral tunnel phacoemulsification. The IOL
was implanted in the bag after a continuous curvilinear capsulorhexis was created. The IOL types included for the present study were from the Acrysof nonaspheric series
(SA60AT, MA60AC, MA60MA, Alcon Laboratories Inc.).
The IOL powers ranged from
2.00 to C36.00 D.
Postoperatively, the visual acuity and refraction (performed at 6 meters) were recorded 1 to 3 weeks after surgery
as part of the routine follow-up. Only cases with good postoperative visual acuity (O20/40) were included in the study.
To correct for distance refraction, 0.17 D was subtracted from
the spherical equivalent measured at 6 m. At the time the
refraction was taken, the actual position of the IOL (the postoperative anterior chamber depth [ACD]) was measured
using the Lenstar LS900 biometer (Haag-Streit AG). This
optical biometer uses laser interferometry not only for the
measurement of the AL but also for the corneal thickness,
the ACD, and the lens thickness (Figure 1).

Figure 1. Laser biometric scan of a pseudophakic eye showing the

position of the IOL, the corneal thickness, the IOL thickness, and
the AL. The scan includes 3 individual scans indicated by separate
colors (red, yellow, and blue), showing almost identical readings as
indicated by the small SD between them (bottom of figure) (AD Z
anterior chamber depth; AL Z axial length; CCT Z central corneal
thickness; LT Z intraocular lens thickness; OS Z left eye; RT Z
retinal thickness).

Eye Model
Ray-tracing analysis of any optical system requires the
physical dimensions and refractive indices of all optical
media to be clearly defined. For the pseudophakic eye, the
cornea, the IOL, the distances between the interfaces, and
the AL have to be defined.
The refractive power of the cornea is often obtained by
keratometry, by which the anterior corneal radius of curvature is measured. However, the dioptric reading given by
the keratometer assumes a thin-lens calculation for corneal
power, as follows:
nc
1
r
where nc Z 1.3375 and r is the corneal radius in meters. This
keratometer value is not the true physiologic power, however.5 The correct optical model of the cornea requires it to
be assigned an anterior surface as well as a posterior surface,
each with a certain curvature. In most corneal models, the
posterior curvature is assumed to be a fixed ratio of the anterior curvature assuming a standard corneal shape. For many
years, the standard shape, and hence the radius of the posterior surface, was assumed to be as proposed by Gullstrand.6
It was not until recently that modern studies using Scheimpflug imaging and other techniques provided detailed information not only on the curvatures of both surfaces of the
cornea but also on their asphericity.79 Therefore, in the present study, the posterior surface of the cornea was assumed to
be a fixed ratio of the anterior surface according to the model
described by Dubbelman et al.8 so that
DZ

R2 Z 0:84  R1
where R2 is the radius of the posterior surface of the cornea
and R1 is the radius of the anterior surface of the cornea.
Also from the work of Dubbelman et al., the asphericity of
the corneal surfaces was assumed to depend on the age of
the patient according to the following equations:
Ka Z 0:76 0:003  age
Kp Z 0:76 0:325  Ka
0:0072  age
where Ka is the asphericity of the anterior surface of the
cornea, Kp is the asphericity of the posterior surface of the
cornea, and age is the age of the patient in years.
The refractive index of the cornea was assumed to be
a constant value of 1.376, and the thickness of the cornea
was assumed to be a constant value of 0.5 mm. (Initial experiments showed negligible effect when using the individually
measured corneal thickness.)
For more than a decade, the use of PCI10 has significantly
improved the precision by which the AL can be measured
compared with ultrasound. The first commercially available
PCI technique was the IOLMaster instrument. However,
precision is not the same as accuracy. For an accurate interpretation of the AL, it should be realized that the output
reading of the IOLMaster is not the true optical path length
of the eye. The readings given by the commercial version
of the IOLMaster device have been calibrated against immersion ultrasound according to the following formula11,12:
AxZeiss Z OPL=1:3549
1:3033=0:9571
where AxZeiss is the output reading of the PCI instrument
and OPL is the optical path length measured by PCI. Accordingly, the optical path length can be found as

Submitted: August 5, 2011.

Final revision submitted: October 19, 2011.
Accepted: October 20, 2011.
From University Eye Clinic, Aarhus Hospital NBG, Aarhus, Denmark.
Corresponding author: Thomas Olsen, MD, PhD, University
Eye Clinic, Aarhus Hospital NBG, Aarhus C, Denmark. E-mail: t@
olsen.dk.

OPL Z AxZeiss  0:9571 1:3033  1:3549

According to Haigis,12 the group refractive index of the
phakic eye is 1.3574. Using this value, however, a small inconsistency between the preoperative and postoperative
readings with the IOLMaster device was previously found

J CATARACT REFRACT SURG - VOL 38, APRIL 2012

643

RAY-TRACING ANALYSIS OF THE IOL POWER IN SITU

by our group; for this reason, the index of 1.3616 was suggested.13 Using this latter value, an estimate of the geometric
distance was obtained according to the formula

The measured power of the IOL was finally calculated

from the curvatures according to the paraxial thick-lens
formula

Axtrue Z AxZeiss  0:9571 1:3033  1:3549=1:3616

T
 D1  D2
n
where D12 is the total dioptric power of the thick lens, D1 is
the dioptric power of front surface, D2 is the dioptric power
of back surface, T is the thickness of the lens in meters, and n
is the refractive index of the material. The dioptric power of
each IOL surface was calculated according to the thin-lens
formula

This conversion between the IOLMaster reading and the true

geometric AL was used in the present ray-tracing experiments. To study the possible bias, however, the ray-tracing
calculations were also performed using the unmodified
IOLMaster AL readings.

Ray-Tracing Procedure
Ray tracing is a well-known procedure to analyze the
optical properties of any optical system using the Snell law
as follows:
sini n2
Z
sinu n1
where i is the incident angle of ray on the surface, u is the
refracted angle, n1 is the refractive index of first medium,
and n2 is the refractive index of the second medium. To
study the optical properties of the pseudophakic eye,
a computer program was written in Pascal (Borland Delphi, Inprise Corp.) to trace the ray intersections of an
incoming beam of rays through the optical media and
study the distribution of the intersections with the retina.
To validate the calculations, model eye data were exported
to the Zemax software (Radiant Zemax LLC) for optical
design.
When an optical engineer wants to build a complex optical
system, he or she may do so by solving for 1 or more variables, giving the best outcome while keeping the other conditions equal. The same principle can be applied to the
pseudophakic eye: To solve for the optical properties of the
IOL, we can use the actual refraction, the vertex distance,
the corneal curvature, the position of the IOL, and the optical
path length (rather, the geometric AL) to solve for the curvatures of the IOL optic that most effectively describes the total
optical system of the pseudophakic eye.
To solve for the curvatures of the IOL, something about
the shape of the optic must be known. Most IOL manufacturers will provide a cutting chart stating the diopter specific
values of refractive index, thickness, and curvatures of the
front surface and back surface of the IOL optic. Usually
such a chart will show that the IOL has a certain configuration; that is, a 1:1, 1:2, or 1:3 biconvex design. The manner
by which the IOL changes its physical curvatures over the
power range (ie, the shape factor) varies between manufacturers and between IOL models; however, in some cases,
the posterior curvature of the IOL optic is kept constant
over a certain power range while the anterior curvature
changes to produce the labeled power. In other models, the
IOL optic may keep its overall configuration constant over
the entire diopter range. The thickness of the IOL often follows a linear relationship with the power (minus-powered
IOLs excluded).
In the present study, the algorithm solving for the front
and back curvatures of the IOL was constructed from the
manufacturers stated information on the shape factor of
the IOL over the diopter range. The IOL curvatures (and
power-dependent thickness) were found by computer iterations while observing the width of the point-spread function
(PSF) at the retina until a minimum of the root-mean-square
deviations from the center was found.

D12 Z D1

n2
n1
r
where D is the dioptric power of the single surface, r is the
radius of curvature in meters, n1 is the refractive index of
the first medium, and n2 is the refractive index of the
second medium.
DZ

Pupil and the Stiles-Crawford Effect

Although the eye on first sight might look like a physical
system, there could be modifications due to biology. One
such modification is the directional sensitivity of the retina,
which was discovered by Stiles and Crawford in 193314 as
a discrepancy between the objective and the functional
area of the pupil in terms of luminous effectiveness. The
effect was found to depend on the point of entry of rays
through the pupil so that peripheral rays had less stimulus
effect on the retina than central rays. For a complete
description of the sensory representation of the physical
image, the Stiles-Crawford effect therefore has to be taken
into account because of the modulatory effect on the spherical aberration, as previously described using ray tracing.15
In the present study, the ray-tracing analysis was performed with and without correction for the StilesCrawford effect assuming a pupil diameter of 0.1 mm,
1.0 mm, 2.0 mm, 3.0 mm, 4.0 mm, 5.0 mm, 6.0 mm,
7.0 mm, and 8.0 mm.

Outcome Measures
The error of the prediction was defined as the difference
between the calculated power and the implanted (labeled)
power of the IOL (labeled value minus predicted value). Statistical analysis was performed using distributional methods
and the statistical service of an Excel spreadsheet (Microsoft
Corp.). Distribution methods were used where appropriate.
Linear regression analysis was performed using the method
of least squares. A probability level of 0.05 (2 tailed) was
considered statistically significant.

RESULTS
The study included 767 eyes of 583 patients. The mean
age of the 234 men and 349 women was 69.7 years
(range 20 to 95 years). The mean IOL power was
C20.69 D.
The IOL power calculated by ray tracing was
linearly correlated with the implanted (labeled) power
(correlation coefficient r O 0.99, P!.00001) (Figure 2).
Assuming a 3.0 mm pupil, the mean difference between the calculated and the observed IOL power

J CATARACT REFRACT SURG - VOL 38, APRIL 2012

644

RAY-TRACING ANALYSIS OF THE IOL POWER IN SITU

Figure 2. Correlation between implanted IOL power (labeled value)

and the IOL power calculated by ray tracing assuming a 3.0 mm
pupil in 767 cases (correlation coefficient r Z 0.99; P!.00001; slope
of y on x linear regression line Z1.0) (IOL Z intraocular lens).

was
0.26 D G 0.65 (SD). The range of error was from

2.4 to C1.8 D. Table 1 shows the results for other pupil diameters.
No significant bias between the estimated IOL
power and the labeled IOL power was found, as
shown by a difference plot (Figure 3). Also no bias between the IOL power and the AL was found (Figure 4).
However, when the unmodified AL readings provided by the IOLMaster were used directly in the
ray-tracing calculations, a significant bias was observed with the IOL power and the AL (Figure 5).
The estimated IOL power was dependent on the assumed pupil size. A small overestimation of the labeled power was seen for pupil sizes close to zero,
while an underestimation was found for larger pupils
(Figure 6). The effect of the assumed pupil diameter
was found to be dependent on the Stiles-Crawford effect, which tended to correct the underestimation with
larger pupils (O5.0 mm). For a 6.0 mm pupil, the difference plus/minus Stiles-Crawford correction was
approximately 0.25 D of the estimated IOL power.
DISCUSSION
In the present study, we applied exact ray tracing to
the pseudophakic eye to show the accuracy of back-

calculating the IOL power in situ. Using the present

algorithms, the error between the calculated and the
actual (labeled) IOL power showed no bias with the
AL or IOL power over the entire range of powers
and the mean offset errors were very small. Assuming
a 3.0 mm pupil, the overall prediction error was

0.26 G 0.65 D in the IOL plane. If one assumes
a mean error of zero (which can be obtained retrospectively by correcting for the mean offset error), the standard deviation of 0.65 D is the equivalent of
determining the IOL power within G1.00 D and
G2.00 D in 88% and 99% of cases, respectively.
Considering the accuracy of the current method, we
assume that a large part of the spread is the result of
measurement errors. However, other sources of error,
such as deviations between the labeled and the actual
IOL power, should also be considered. The current International Organization for Standardization (ISO)16
requires the tolerance limits of the labeled IOL power
to be within G0.30 D, G0.40 D, G0.50 D, and G1.00 D
of the true power for the power ranges 0.00 to 15.00 D,
15.00 to 25.00 D, 25.00 to 30.00 D, and more than
30.00 D, respectively. If the tolerances of the IOL types
used in the present study were no better than that, it is
possible the mislabeling constitutes a considerable
source of error and that the true accuracy of the current
ray-tracing method is better than stated in this paper.
At present, however, we do not know the exact tolerance limits of the IOL type used in the present study.
Pupil size is another source of error when calculating IOL power using ray tracing. We made no attempt
in the present study to correct for individual pupil
diameter. Due to higher-order aberrations (HOAs)
(see below), the actual pupil size is likely to influence
the final refraction, especially with spherical IOLs. It
is therefore possible that the overall prediction accuracy might have been improved if the pupil diameter
had been measured and used for ray-tracing analysis
in the individual case.
Ray-tracing analysis is very sensitive to the assumptions of the physical model of the pseudophakic eye. In
the present study, the cornea was modeled using the
result of recent studies of the radius and asphericity
of the anterior and the posterior surfaces of the normal
cornea. This model predicts the curvature of the

Table 1. The mean error (calculated minus implanted value) predicting the implanted IOL power by ray tracing assuming different pupil
diameters in 767 cases.
Pupil (mm)
0.01

1.00

2.00

3.00

4.00

5.00

6.00

7.00

8.00

Mean error +0.21 G 0.65 +0.16 G 0.65 +0.00 G 0.65
0.26 G 0.65
0.64 G 0.66
1.16 G 0.67
1.79 G 0.69
2.54 G 0.71
3.45 G 0.75
(D) G SD
Range

2.01, +2.34
2.07, +2.28
2.21, +2.08
2.44, +1.78
2.86, +1.38
3.41, +0.91
4.04, +0.47
4.69, +0.11
5.40,
0,27

J CATARACT REFRACT SURG - VOL 38, APRIL 2012

RAY-TRACING ANALYSIS OF THE IOL POWER IN SITU

645

Figure 3. Difference plot (Bland-Altman) showing the difference

between the labeled and the ray tracingcalculated IOL power versus the mean of the IOL power (767 cases) (IOL Z intraocular lens).

Figure 4. Prediction error (difference between the labeled and calculated IOL power) versus the AL (767 cases).

posterior surface to be 84% of the front curvature,

which is lower than the old, assumed Gullstrand ratio
of 88%. The difference between 84% and 88% radius
ratio is the equivalent of approximately 0.30 D in corneal power, which is the equivalent of 0.50 D in IOL
power. Hence, if the calculations were performed in
the paraxial domain (equivalent to ray tracing with
an infinitely small pupil size), we would expect
a mean difference of approximately 0.50 D in estimated IOL power between the 2 ratio models. However, for a complete description of the corneal optics,
HOAs should also be taken into account. One HOA
is spherical aberration caused by corneal asphericity;
this is a significant factor in normal corneas and especially in diseased corneas or corneas with previous laser ablation treatment (ie, laser in situ keratomileusis)
for refractive errors. In such eyes, paraxial ray tracing
does not give a complete description of the corneal

optics, and this is why exact ray tracing should be

considered the method of choice.
For a ray-tracing analysis of the pseudophakic eye to
be realistic, we need a realistic physical model of the
IOL optic. Before the modern evolution of aspheric
IOLs, the shape of most IOLs was spherical
biconvex. Due to the spherical aberration, we cannot
rely on paraxial ray tracing for a realistic representation of the effective power of these spherical IOLs.
The aim of modern aspheric IOLs is to counteract the
asphericity of the cornea and thereby enhance the
modulation transfer function of the eye; that is, narrowing the PSF at the retina. Nonetheless, by ISO definition, it is the paraxial power that is stated on the
label of that IOL. However, because the amount of
asphericity differs between different aspheric IOL
designs, the effective power in situ also differs. This
is the reason for differences in the A-constant that

Figure 5. Prediction error (difference between the labeled and calculated IOL power) versus the AL when the uncorrected IOLMaster
readings were used in the ray-tracing analysis (767 cases).

Figure 6. Prediction error (difference between the labeled and calculated IOL power) versus the assumed pupil diameter with and without correction for the Stiles-Crawford effect. The 2 curves differ
significantly from pupil diameters of 5.0 or more due to the modulatory effect of the Stiles-Crawford effect on spherical aberration. Bars
indicate standard deviations (767 cases).

J CATARACT REFRACT SURG - VOL 38, APRIL 2012

646

RAY-TRACING ANALYSIS OF THE IOL POWER IN SITU

can be found between IOLs with or without aspheric

correction, other things being equal. In the present
study, we only included IOLs of known spherical
design. The reason was that the shape factor of the
IOL design could be obtained from the manufacturer,
and this is essential for the ray-tracing analysis. It is
our experience that information on the physical characteristics of the aspheric IOL is more difficult to
obtain because it is generally considered proprietary
information. In the future, however, we expect this
information to be more readily obtainable as it will
increase our ability to accurately predict the optical
outcome of an IOL implant of any type.
The exact optical path length of the eye is another
important variable that is subject to interpretation.
As shown in the present study, the uncorrected AL
readings of the IOLMaster device could not be used directly in the ray-tracing analysis to give a satisfactory
unbiased estimation of the IOL power in situ. A significant bias with IOL power as well as with AL was
observed, indicating that the standard reading of the
IOLMaster device is not an ideal representation of
the optical path length. The reason for this effect
appears to be 2-fold. First, the output readings of the
IOLMaster device have been calibrated against
(immersion) ultrasound readings. Therefore, to obtain
the optical distances, one has to retransform the (ultrasound-like) readings to their original optical path
lengths. Second, to transform the measured optical
path length into the geometric length, it is necessary
to know the group refractive index valid for the given
eye. We previously showed that there is a small inconsistency between the preoperative and the postoperative geometric AL when these distances are based on
the group refractive indices originally recommended
by Haigis.12 To correct for this small discrepancy, we
suggested the use of a somewhat higher index for
the crystalline lens, giving a higher group refractive index for the phakic eye.13 The results in the present
study showing a high accuracy in predicting the in
situ IOL power with minimal offset error and no bias
are supportive of the current eye model. However,
we are aware that other models for the interpretation
of the AL readings, such as segmental calculation of
the overall optical path based on the individual refractive index of the cornea, the aqueous, the lens, and the
vitreous, could also be considered. Future studies are
needed to show whether the current eye model can
be improved.
Few studies have dealt with the measurement of
IOL power within the eye.17 Several years ago, one
of the authors (T.O.) described a method to measure
the power of an IOL in situ by using the PurkinjeSanson images from the IOL surfaces.18 This method
requires a special setup to detect the size of the images

produced by the convexconcave mirror constituted

by the IOL surface and may not be easy to perform
in daily clinical practice. The present method uses
modern optical biometry to capture the locations of
the IOL position in the eye using laser interferometry,
which is a very precise method of performing this
measurement. This, together with the known accuracy
of laser biometry for the AL measurement, makes the
basis for accurate analysis of the optical system constituted by the pseudophakic eye.
In conclusion, in the present study we have shown
how the optics of the pseudophakic eye can be accurately described by exact ray-tracing analysis of the
physical information provided by modern biometric
techniques. With the present method, it was possible
to achieve a very accurate estimation of the IOL power
over a large power range with no bias with IOL power
or AL. These results are encouraging for exact physical
methods to be applied to the field of IOL power
calculation.

REFERENCES
1. Holladay JT, Prager TC, Chandler TY, Musgrove KH, Lewis JW,
Ruiz RS. A three-part system for refining intraocular lens power
calculations. J Cataract Refract Surg 1988; 14:1724
2. Sanders DR, Retzlaff JA, Kraff MC, Gimbel HV,
Raanan MG. Comparison of the SRK/T formula and other
theoretical and regression formulas. J Cataract Refract
Surg 1990; 16:341346
3. Haigis W. The Haigis formula. In: Shammas HJ, ed, Intraocular
Lens Power Calculations. Thorofare, NJ, Slack, 2004; 4157
4. Olsen T. Calculation of intraocular lens power: a review. Acta Ophthalmol Scand 2007; 85:472485. Available at: http://
onlinelibrary.wiley.com/doi/10.1111/j.1600-0420.2007.00879.x/
pdf. Accessed November 15, 2011
5. Olsen T. On the calculation of power from curvature of the cornea. Br J Ophthalmol 1986; 70:152154. Available at: http://
www.ncbi.nlm.nih.gov/pmc/articles/PMC1040942/pdf/brjopthal
00624-0073.pdf. Accessed November 15, 2011
6. Gullstrand A. The optical system of the eye. In: Southall JPC, ed,
Helmholtzs Treatise on Physiological Optics, translated from
the third German edition. Rochester, NY, The Optical Society
of America, 1924; Vol. 1; Anatomy, Physiology, and Dioptrics
of the Eye, Appendix II.3; 351
352
7. Dunne MCM, Royston JM, Barnes DA. Normal variations of the
posterior corneal surface. Acta Ophthalmol (Copenh) 1992;
70:255261
lker8. Dubbelman M, Weeber HA, van der Heijde RGL, Vo
Dieben HJ. Radius and asphericity of the posterior corneal
surface determined by corrected Scheimpflug photography.
Acta Ophthalmol Scand 2002; 80:379383. Available at: http://
onlinelibrary.wiley.com/doi/10.1034/j.1600-0420.2002.800406.
x/pdf. Accessed November 15, 2011
9. Dubbelman M, Sicam VADP, van der Heijde GL. The shape of
the anterior and posterior surface of the aging human cornea.
Vision Res 2006; 46:9931001
10. Drexler W, Findl O, Menapace R, Rainer G, Vass C,
Hitzenberger CK, Fercher AF. Partial coherence interferometry:
a novel approach to biometry in cataract surgery. Am J Ophthalmol 1998; 126:524534

J CATARACT REFRACT SURG - VOL 38, APRIL 2012

647

RAY-TRACING ANALYSIS OF THE IOL POWER IN SITU

11. Haigis W, Lege B, Miller N, Schneider B. Comparison of immersion ultrasound biometry and partial coherence interferometry
for intraocular lens calculation according to Haigis. Graefes
Arch Clin Exp Ophthalmol 2000; 238:765773
12. Haigis W. Pseudophakic correction factors for optical biometry.
Graefes Arch Clin Exp Ophthalmol 2001; 239:589598
13. Olsen T, Thorwest M. Calibration of axial length measurements
with the Zeiss IOLMaster. J Cataract Refract Surg 2005;
31:13451350
14. Stiles WS, Crawford BH. The luminous efficiency of rays entering the eye pupil at different points. Proc R Soc Lond [Biol] 1933;
112:428450. Available at: http://rspb.royalsocietypublishing.
org/content/112/778/428.full.pdf. Accessed November 15, 2011
15. Olsen T. On the Stiles-Crawford effect and ocular imagery. Acta
Ophthalmol (Copenh) 1993; 71:8588
16. International Organization for Standardization. Ophthalmic Implants Intraocular Lenses Part 2: Mechanical Properties and
Test Methods. Geneva, Switzerland, ISO, 2006; (ISO 119792)

17. Holladay JT, Long SA, Lewis JW, Prager TC, Koester CJ,
Bourgeois KA, Winn TL. Determining intraocular lens power
within the eye. Am Intraocular Implant Soc J 1985; 11:353363.
Available at: http://www.docholladay.com/publications/paper16.
pdf. Accessed November 15, 2011
18. Olsen T. Measuring the power of an in situ intraocular lens with
the keratometer. J Cataract Refract Surg 1988; 14:6467

J CATARACT REFRACT SURG - VOL 38, APRIL 2012

First author:
Thomas Olsen, MD, PhD
University Eye Clinic, Aarhus Hospital
NBG, Aarhus, Denmark