Comparative evaluation of refractive outcomes after implantation of two types of intraocular lenses with different diopter intervals (0.25 diopter versus 0.50 diopter)

This prospective study was approved by the institutional review board (IRB) of Korea University Ansan Hospital (IRB number: 2016AS0020) and was registered as a clinical trial at https://​cris.​nih.​go.​kr (identification number: KCT0002192). All patients provided signed informed consent to participate in a clinical research study. All research and data collection practices adhered to the tenets of the Declaration of Helsinki and good clinical practices.

Eighty eyes from 40 patients with bilateral senile cataract who were scheduled to undergo consecutive phacoemulsification and IOL implantation within a period of one to four weeks at Korea University Ansan Hospital between November 1, 2016 and August 31, 2017 were enrolled in this study. The Softec HD™ (Lenstec Inc., St. Petersburg, FL, USA; 40 eyes) IOL with a 0.25 D interval was implanted in one eye of each subject, and the Akreos AO (Bausch & Lomb, Rochester, NY, USA; 40 eyes) IOL with a 0.50 D interval was implanted in the contralateral eye. The RANDBETWEEN(1,2) function in Microsoft Excel (Microsoft Inc., Redmond, WA, USA) was used to randomly decide which IOL would be used in the first eye. The Softec HD™ and Akreos AO IOLs employ similar material and optic design (hydrophilic acrylic bi-aspheric zero aberration with 1.46 refractive index), although their haptic design and overall length are different (Table 1 and Figure 1). We specifically included patients who were scheduled to undergo implantation of IOL with a power range from 15.5 D to 25.0 D because the Softec HD™ IOL provides a 0.25 D interval in this range. Eyes with best corrected visual acuity (BCVA) of 20/40 or better postoperatively were included in this study. Patients with amblyopia, corneal disease such as keratoconus or corneal dystrophy, traumatic cataracts, or a history of previous ocular surgery (e.g., refractive surgery) were excluded. Also excluded from this study were those patients who had undergone complicated ocular surgery (e.g., anterior capsular tears); those who had had any postoperative complications; and those with noticeable postoperative IOL decentration or tilt.

Preoperative objective refraction measured with an autorefractor/keratometer (KR-8100; Topcon, Tokyo, Japan) was recorded at the screening visit. Subjective refraction was recorded when autorefraction was not available. Preoperative K, anterior chamber depth (ACD), and AL values were measured with optical biometry using an IOLMaster® 500 (Carl Zeiss Meditec, Jena, Germany). IOL power was calculated using the SRK/T and Haigis formulas of the device. The data-adjusted SRK/T A-constant was calculated using a spreadsheet (Microsoft Excel; Microsoft Inc.), and the data-adjusted a₀, a₁, and a₂ constants for the Haigis formula were calculated with linear regression analysis in order to obtain zero mean numerical error in IOL power prediction. [9‐12] The data-adjusted SRK/T A-constant was 118.17 for the Softec HD™ IOL and 118.24 for the Akreos AO IOL. The optimized IOL constants for the Haigis formula (a₀, a₁, and a₂) were 0.565, 0.240, and 0.138, respectively, for the Softec HD™ IOL and 1.706, 0.279, and 0.087, respectively, for the Akreos AO IOL. Postoperative uncorrected distance visual acuity, BCVA, objective refraction measured with an autorefractor/keratometer, and subjective refraction were measured at one month after surgery. The refractionist was masked to type of IOL.

All phacoemulsification and IOL implantations were performed by one experienced surgeon (Y.E.) under topical anesthesia with 0.5% proparacaine hydrochloride (Paracaine; Hanmi Pharm, Seoul, Korea). During surgery, a 2.75-mm clear corneal incision was made, and a continuous curvilinear capsulorrhexis slightly smaller in size than the IOL optic was created with a 26-gauge needle and capsulorrhexis forceps. After performing hydrodissection and hydrodelineation, superficial cortex and epinucleus were aspirated with a phaco probe. Then, the nucleus was held by the phaco probe, and the phaco chopper (Nagahara Chopper; ASICO, LLC, Westmont, IL) was placed on the opposite side of the equator of the nucleus. The phaco chopper was pulled and the phaco probe was pushed toward the phaco chopper to divide the nucleus into halves. Then the divided nucleus halves were rotated 90 degrees and divided into quadrants with the same technique. All pieces of segments of the divided nucleus were emulsified and aspirated. An irrigation/aspiration system was used to remove the epinucleus and cortex. The anterior chamber was filled with sodium hyaluronate 1.65% /chondroitin sodium sulfate 4.0% (DisCoVisc; Alcon Laboratories, Inc., Fort Worth, TX), and each IOL was implanted in a capsular bag using an injector system. After removing the ophthalmic viscosurgical device, the wound was sutured with 10–0 nylon, which was removed at postoperative 1 week.

Preoperative medication 0.5% levofloxacin hydrate (Cravit®, Santen, Osaka, Japan) every 6 h and 0.1% bromfenac sodium hydrate (Bronuck®, Taejoon Pharm., Seoul, Korea) every 12 h were used for 3 days before surgery. At the day of surgery, 0.5% Moxifloxacin (Vigamox®, Alcon) every 2 h, 0.1% fluorometholone (Flucon®, Alcon) every 6 h, and 0.1% bromfenac sodium hydrate (Bronuck®) every 12 h were used. At the day after surgery, 0.5% Moxifloxacin was used every 6 h, along with the other eye drops used on 1 day for the next 1 month.

To evaluate the effect of the dioptric error of the IOL produced during the manufacturing process on refractive outcome, only IOL power was used as a random variable in Monte Carlo simulation, and it was assumed that no error was caused by the preoperative estimation of effective lens position and/or the measurement of K, ACD, and AL in the IOL power calculation. In order to satisfy this assumption, IOL power, which has zero refractive prediction error, was calculated from preoperative biometry and postoperative spherical equivalent values in each patient. The mean and standard deviation (SD) values were used to generate random variables of IOL power for each patient with a normal distribution using the RAND() and NORMINV functions in Excel (Microsoft Inc.). The mean value of IOL power was defined as the IOL power that has zero refractive prediction error in each patient, and SD was ±0.40 D in the Akreos AO group and ± 0.11 D in the Softec HD™ group. [7, 8] Generated IOL powers for each patient were used to calculate the refractive prediction error during Monte Carlo simulation. This study conducted 20 Monte Carlo simulations: 10 using the Haigis formula and 10 using the SRK/T formula.

The refractive prediction error was defined as the difference between postoperative spherical equivalent and preoperative predicted refraction as determined using two IOL calculation formulas (i.e., refractive prediction error = postoperative spherical equivalent – preoperative predicted refraction). The mean absolute error (MAE) was defined as the mean absolute value of the refractive prediction error, and the median absolute error (MedAE) was defined as the median absolute value of the refractive prediction error. The percentages of eyes that achieved a postoperative refractive prediction error within ±0.25 D, ± 0.50 D, and ± 0.75 D from the preoperative predicted refraction were estimated.

Descriptive statistics for all patient data were obtained using the Statistical Package for Social Sciences version 21.0 (IBM Corp., Armonk, NY, USA). Student’s t-tests were used to compare K, ACD, AL, IOL power, and MAE, while Mann–Whitney U tests were used to compare MedAE between the two groups. Chi-square tests were conducted to compare the percentage of eyes that achieved a postoperative refractive prediction error within ±0.25 or ± 0.50 D from the preoperative predicted refraction by the SRK/T and Haigis formulas between the two groups. Results were considered statistically significant if the P-value was less than 0.05.