IOL power calculations that use surgically induced changes in refraction
or no previous data proved more accurate than methods using keratometry values
prior to corneal refractive surgery and surgically induced changes in
refraction, a study found.
The authors used the American Society of Cataract and Refractive Surgery
IOL power calculator to gauge the accuracy of IOL power prediction methods
after previous LASIK or PRK.
Making accurate IOL power calculations after previous corneal refractive
surgery is difficult and time-consuming, Li Wang, MD, PhD, and colleagues
Warren Hill, MD, and Douglas D. Koch, MD, said in the study.
We found that the method using no prior data at all or using only
part of the previous data was superior to the method using pre-LASIK or pre-PRK
Ks and the change in manifest refraction, Dr. Wang told Ocular
Surgery News. Rely on methods that use no prior data at all or the
method of using changes in manifest refraction only. Pay more attention to
methods in those two categories.
Power calculation methods
The Web-based ASCRS calculator generates IOL power for eyes undergoing
cataract surgery with previous LASIK, PRK or radial keratotomy. It classifies
methods using keratometry values prior to LASIK or PRK and change in manifest
refraction resulting from LASIK or PRK; surgically induced change in
refraction; or no previous refractive data.
IOL power is calculated using the Aramberri double-K method modification
of the Holladay 1 formula, which uses corneal power prior to LASIK or PRK to
estimate effective lens position. A default value of 43.86 D is used when
corneal power prior to LASIK or PRK is not available.
If we have a pre-LASIK corneal power, the Holladay 1 uses that
value to calculate the lens position, Dr. Wang said. If the patient
does not have that value, the formula we are using, the so-called double-K
Holladay 1 formula, uses a default value.
The clinical history method has been considered the standard approach to
calculating IOL power in eyes with previous refractive surgery. However, that
method is reliable only when historical data are accurate; inaccurate
historical data can skew final IOL power calculations, Dr. Wang said.
A 1 D inaccuracy in historical data yields a 1 D error in final
refraction at the corneal plane, she said.
Patients and parameters
The retrospective study, published in the Journal of Cataract and
Refractive Surgery, included 72 eyes of 57 patients with a mean age of
58 years. Refractive data prior to LASIK or PRK were available for 70 eyes, and
keratometry values prior to LASIK or PRK were available for 62 eyes.
All patients underwent implantation of an AcrySof SN60WF aspheric IOL
(Alcon), had no intraoperative or postoperative complications, and had
postoperative corrected distance visual acuity of 20/32 or better.
Outcome measures included mean arithmetic IOL prediction error, mean
absolute IOL prediction error, variance in mean arithmetic IOL prediction error
and percentage of eyes within a certain refractive prediction error (IOL
prediction error of 0.71 D or refractive error at the spectacle plane of 0.5 D,
and IOL prediction error of 1.43 D or refractive error at the spectacle plane
of 1 D, assuming that 1 D of IOL prediction error resulted in 0.7 D of
refractive error at the spectacle plane).
IOL power prediction errors
Study results showed that mean arithmetic IOL prediction errors were as
- 0.66 D to 0.3 D for IOL power calculation methods using
keratometry values prior to LASIK or PRK and surgically induced changes in
- 0.25 D to 0.49 D for methods using surgically induced changes
- 0.24 D to 0.18 D for methods using no previous data
Mean absolute IOL prediction errors were as follows:
- 1.1 D to 1.31 D for methods using values prior to LASIK or PRK and
surgically induced changes in refraction
- 0.62 D to 0.71 D for methods using surgically induced changes in
- 0.57 D to 0.69 D for methods using no previous data
Percentage of eyes in each group achieving ±0.5 D of the
predicted refraction and ±1 D of the predicted refraction were as
- 37% to 44% of eyes within 0.5 D and 60% to 69% of eyes within 1 D in
the group using keratometry values prior to LASIK or PRK and surgically induced
changes in refraction
- 57% to 67% of eyes within 0.5 D and 86% to 91% of eyes within 1 D in
the group using surgically induced changes in refraction
- 58% to 60% of eyes within 0.5 D and 90% to 96% of eyes within 1 D in
the group using no previous data
Further study is devoted to gauging the accuracy of IOL calculation
methods in eyes with previous hyperopic laser vision correction, Dr. Wang said.
We are collecting those data because hyperopic eyes are not as
numerous as the myopic eyes. Also, the change in refraction is normally smaller
in hyperopic eyes. We just have a smaller sample for those groups, she
said. by Matt Hasson
- Li Wang, MD, PhD, can be reached at Cullen Eye Institute, Baylor
College of Medicine, 6565 Fannin St., NC-205, Houston, TX 77030; 713-798-7946;
fax: 713-798-3027; e-mail: firstname.lastname@example.org.
One of the most significant challenges for refractive cataract surgeons
is to provide the best possible uncorrected vision to patients undergoing
cataract surgery following previous LASIK or PRK. These patients are accustomed
to good uncorrected visual acuity, and the refractive results following
cataract surgery have been variable. The authors have reviewed their results
with the ASCRS website that allows physicians to use varied sources of data,
including post-refractive surgery keratometry and historical information. The
authors conclude that the surgically induced variables provide better outcomes
than the historical method. This study documents the usefulness of the ASCRS
website and the importance of using surgically induced changes in the cornea.
Eric D. Donnenfeld, MD
Cornea/External Disease Board Member