October 2018


EyeWorld journal club
Review of “A large retrospective database analysis comparing outcomes of intraoperative aberrometry with conventional preoperative planning”

by Asadolah Movahedan, MD, Stefanie Sherman, MD, Nandita Anand, MD, Megan Silas, MD, Blake Williams, MD, Christos Theophanous, MD, Ema Avdagic, MD, Andrea Blitzer, MD, Lincoln Shaw, MD, and Peter Veldman, MD

Peter Veldman, MD
, residency program director, Department of Ophthalmology and Visual Science, University of Chicago

University of Chicago residents, from left: Megan Silas, MD, Lincoln Shaw, MD, Asadolah Movahedan, MD, Stefanie Sherman, MD, Blake Williams, MD, Ema Avdagic, MD, Andrea Blitzer, MD, Nandita Anand, MD; not pictured: Christos Theophanous, MD

Many have debated whether the refractive outcomes justify the additional cost and time of conducting intraoperative aberrometry. I asked the University of Chicago residents to review this study that appears in the October issue of JCRS.

—David F. Chang, MD,
EyeWorld journal club editor

As technology has advanced, the goal of cataract surgery has evolved from restoration of vision to visual enhancement. Many patients expect spectacle independence after cataract surgery, and it has become increasingly important to have reliable and predictable postoperative refractive outcomes. Despite prior advancements, minimizing prediction error remains a major concern. Current preoperative biometric devices with the latest generation theoretical formulas are popular among surgeons around the world, and their accuracy has been compared in different clinical settings in numerous studies. Intraoperative aberrometry (IA) is a relatively new surgical planning modality that measures aphakic refraction at the time of cataract surgery, promising a more accurate IOL power calculation. Prior studies on the use of IA have demonstrated advantages in IOL selection in certain circumstances; however, the applicability and broad adaptation of this technology into routine cataract surgery is not yet well studied.1
The article by Cionni et al. is a retrospective analysis of intraoperative aberrometry calculations performed by the Optiwave Refractive Analysis (ORA) system (Alcon, Fort Worth, Texas) and standard IOL biometry in more than 30,000 eyes undergoing phacoemulsification. The purpose of the study is to compare the absolute prediction error of IA in IOL power calculation to conventional preoperative biometry. The authors conclude that calculations using IA outperform calculations from preoperative biometry in this cohort.
The study is designed to reflect a general population, with the cohort including patients with a wide range of axial lengths, various types of intraocular lens implants, and different surgeons. With a sample size of 32,189 eyes, the study is highly powered to detect subtle differences among subgroups. It serves as a landmark first step in determining the utility of this new technology in today’s ever-changing ophthalmologic surgery toolkit.
Nevertheless, the results of such a large study should be interpreted with caution, and the applicability of this study to routine practice should be vetted. The authors indicate that there is a mean absolute prediction error (MAPE) difference of 0.06 D between IA and standard aberrometry (p<0.001). While statistically significant, the clinical significance of such a small difference may be minimal. Additionally, by reporting superior performance of IA in terms of MAPE, the study fails to mention the final refractive outcome or other clinically relevant endpoints, such as visual acuity and spectacle independence. Of note, the authors do not describe the timing of the measurement of postoperative refractions, which has been shown to alter refractive outcome.2 The authors also cite the statistical significance of a “greater percentage of eyes with an aberrometry APE <0.5 D than eyes with a preoperative calculation APE <0.5 D (81.9% [26,357/32,189] vs. 75.9% [24,437/32,189], P<.0001).” It would be interesting to know how many of the cases that fell outside of this APE <0.5 D metric were actually unhappy with their outcome. This information may also provide a measure of the clinical relevance of the authors’ 0.5 D standard.
The relevance of this small identified advantage of IA with respect to MAPE is further nuanced by limitations in current intraocular lens technology. While the study accounted for variability across manufacturers by limiting the data set to eyes with lenses manufactured by Alcon (the study’s sponsor), there remains variability within a single manufacturer. IOL manufacturing tolerance, the internal variability of lens power within a single manufacturer, is set by the International Organization for Standardization and ranges from 0.3 D to 1.0 D depending on total IOL dioptric power.3 In addition, most manufacturers produce IOLs in 0.5 D increments. As such, the identified slightly greater precision offered by IA likely does not yet translate into a clinically tangible postoperative outcome with the currently available IOL technology.
In critically evaluating IA’s role in routine cataract surgery, the cost-effectiveness of broadly adopting this technology must also be considered. While no studies have yet been conducted to compare cost-effectiveness of IA in cataract surgery to conventional biometry, other technologies, such as femtosecond laser-assisted cataract surgery (FLACS), have not been shown to be cost-effective compared to conventional surgery.4 Moreover, IA adds extra steps to the surgical procedure, thus increasing operating room and anesthesia time and overall healthcare resource utilization.
For situations in which the preoperative and IA measurements differed in the study, the surgeons decided which lens would be chosen for the patient. It was in this area that the IA showed the largest (12.5%) benefit in achieving an APE <0.5 D (81.3% [10,385/12,779] vs. 68.8% [8794/12,779], P<.0001). A subgroup analysis of these cases has the potential to highlight particular patient populations that may benefit from the utilization of the technology. Subgroups might include patients with increased age, extreme axial lengths, relative ptosis, increased density of cataract, irregular astigmatism, etc. In addition, the study design excluded patients with corneal scarring, ectasia, prior transplantation procedures, and those who were post-refractive surgery from the analysis, although these subgroups may benefit the most from this technology given known limitations of preoperative biometry in these complex cases.5,6
The article by Cionni et al. is the largest study of IA to this point, providing insight into the precision of IA in IOL power calculation in cataract surgery. Despite its large sample size and the high degree of statistical significance in the study’s endpoints, there remains uncertainty as to the clinical applicability of the findings. It is evident that the vast majority of cases likely received no additional benefit from the technology relative to standard preoperative biometry. Nonetheless, IA may ultimately prove useful in cases in which further subgroup analysis demonstrates a greater likelihood of benefit. Similar to the way that aspirin is known to be of clear benefit in those with pre-existing cardiovascular disease but is not suggested for prophylactic thromboembolic prevention in the general population, targeting appropriate patients for the use of IA may maximize clinical utility and patient outcomes while more responsibly utilizing healthcare dollars.


1. Hatch KM, et al. Intraocular lens power selection and positioning with and without intraoperative aberrometry. J Refract Surg. 2015;31:237–42.
2. Limburg H, et al. Routine monitoring of visual outcome of cataract surgery. Part 2: Results from eight study centres. Br J Ophthalmol. 2005;89:50–2.
3. www.iso.org/standard/55682.html
4. Abell RG, Vote BJ. Cost-effectiveness of femtosecond laser-assisted cataract surgery versus phacoemulsification cataract surgery. Ophthalmology. 2014;121:10–16.
5. Koch DD. The enigmatic cornea and intraocular lens calculations: The LXXIII Edward Jackson Memorial Lecture. Am J Ophthalmol. 2016;171:xv–xxx.
6. Yesilirmak N, et al. Intraoperative wavefront aberrometry for toric intraocular lens placement in eyes with a history of refractive surgery. J Refract Surg. 2016;32:69–70.

Contact information

: pveldman@bsd.uchicago.edu

Review of “A large retrospective database analysis comparing outcomes of intraoperative aberrometry with conventional preoperative planning” Review of “A large retrospective database analysis comparing outcomes of intraoperative aberrometry
Ophthalmology News - EyeWorld Magazine
283 110
220 120
True, 10