Tomographic Analysis of Incision Sequence on Anterior and Posterior Surgically Induced Astigmatism in Cataract Surgery

Purpose: To evaluate of effect incision sequence on anterior and posterior astigmatism and SIA at cataract surgery using Scheimpug tomography measurements. Methods: We evaluated 86 patients who had gone cataract surgery between October 2018 to March 2019. All eyes were measured with a Scheimpug tomography system preoperatively and at 1 month after surgery which included anterior and posterior corneal surfaces. The surgery was begin with side incision in group 1 differently the main incision was done rst in group 2. Student t test and Pearson’s correlation were performed. P values < 0.05 were considered statistically signicant. Result: Surgically induced anterior astigmatism (SIA-A) 0.76±0.43, 0.67±0.38 and surgically induced posterior astigmatism(SIA-P) 0.27±0.19 0.23±0.19 respectively in group 1 and group 2. 1D or greater SIA-A was found 24.4 % and 15.6% and 0.5 D or greater SIA-P was found 14.6% and 12.2% respectively in group 1 and group 2. There was a signicant increase in anterior astigmatism (p=0.002, 0.005 respectively in group 1 and 2), but not in the posterior astigmatism (p=0.471 and 0.247 respectively in group 1 and 2). In the subgroup analysis, group 2 has lower SIA-A then group 1 except oblique subgroup and group 2 has lower SIA-P for all subgroups but it wasn't statistically signicant. Conclusion: Making the main incision rst caused less SIA on both anterior and posterior surfaces however this difference was not signicant.


Introduction
Cataract is the most common preventable cause of blindness and phacoemulsi cation is considered the gold-standard procedure for cataract [1,2]. Clear corneal incision without suture is a standard cataract surgery procedure however surgery induced astigmatism (SIA) and postoperative astigmatism are still problem. In the modern cataract surgery, to reduce SIA is one of the main aim of surgery by that to improve patient satisfaction and increase uncorrected visual acuity. There are some potent surgery procedure to reduce corneal astigmatism such as selecting proper size and location of the incision, limbal relaxing incision and toric intraocular lens implantation [3][4][5].
Main incision is primary responsible for surgery induced astigmatism. Most of surgeon prefer to make side incision rstly, then ll to anterior chamber with ophthalmic viscosurgical device (OVD) and make a main incision, because this approach might be safer to avoid to the anterior capsule damage. More or less lling of the anterior chamber with OVD can be effect normal anterior chamber depth, chamber volume corneal shape and wound architecture. Under or over lling of the anterior camera may cause upward or downward shift of the inner edge of the incision when the cornea returns to its normal architecture. This wound architectural shift situation may cause deviations from the linear structure of the incision, and this change may affect surgery-induced astigmatism and postoperative astigmatism.
The aim of this study to evaluate of effect incision sequence on anterior and posterior astigmatism and SIA at cataract surgery using Scheimp ug tomography measurements.

Methods
This study included 120 eyes who underwent phacoemulsi cation in the Konya Education and Research Hospital (new name Konya City Hospital). Surgery procedure was performed by the same surgeon between October 2018 to March 2019 The study protocol was approved by the Institutional Review Board/Ethics committee, and written informed consent was obtained from all participants. The study adhered to the tenets of the Declaration of Helsinki.
All patients underwent a routine ophthalmic examination preoperatively. Fundoscopy was performed to evaluate the retinal disease under a dilated pupil. Anterior corneal surface, steep meridian and posterior corneal surface were measured using the Scheimpfug analysis system (Pentacam HR, Oculus Optikgeräte GmbH, Wetzlar, Germany) before and one months later after the surgery. Measurements obtained by the Scheimpfug analysis system, which included the at and steep central radii in the 3mm zone on the anterior and posterior corneal surfaces and with poor-quality Scheimpfug analysis scans were excluded. Astigmatism change (AC) was de ned as the difference between preoperative and postoperative astigmatism measured by corneal tomography.
After the examination patient who have cataracts with regular astigmatism less than 2.5 D were included in this study. Preoperatively, patients were randomly divided into the groups. Patient who have previously ocular surgery history, pterygium, intraoperative surgical complications, severe dry eye or any corneal pathology were excluded. 120 eyes were included in the study but 34 eyes were excluded because of poor follow-up or insu cient topography quality data. Astigmatism was classi ed according to the steepest meridian. Horizontal (steepest meridian 0 to 29.9° or 150 to 180°), vertical (steepest meridian 60 to 119.9°), and oblique (steepest meridian 30 to 59° or 120 to 149.9°). We didn't use the terms of classical astigmatism in accordance with the rule or against the rule because of negative dioptric power of the posterior cornea may cause confusion in nomenclature.

Surgical Technique
All Operator procedures were performed under topical anesthesia with proparacaine HCl %0.5 (Alcaine, Alcon Laboratories Inc ). In grup 1, rstly side incision performed with 20 gauge mvr blade (Alcon Laboratories Inc) on 150 degree after the lling to anterior chamber with OVD (Protectalon 2,0 VSY inc) second side incision and 2,8mm single-step clear corneal incision were performed with double bevel up blade (Alcon Laboratories Inc). Main incision was performed on 90 degree and second side incision was performed on 60 degree away from main incision at cornea. In group 2, rstly 2,8mm single-step clear corneal incision was performed after that the anterior chamber was lled with OVD and side incision were made with 20 gauge mvr blade . After standard cataract surgery by phacoemulsi cation bimanual cortex aspiration was performed and acrylic single pieces non-toric hydrophobic IOL (Acrysof SA60AT; Alcon) was inserted in the capsular bag All incision was carried out on same location on cornea in the both groups and corneal incision place were left without suture .
Patients were treated with a combination of moxi oxacine and dexamethasone (Moxidexa, Abdi Ibrahim Inc.) 8 times for rst week then dexamethasone 1 mg/mL (Maxidex, Alcon Laboratories Inc) eyedrops 4 times a day for 3 weeks postoperatively.

SIA Calculation
We used vector analysis software based on Alpin's method to calculate SIA [6,7].

Statistical Analysis
For statistical analysis, Statistical Package for Social Science (SPSS) program (Worldwide Headquarters SPSS Inc. 21.0 Windows package program) was used. Descriptive ndings are displayed as mean ± standard deviation. Normal distribution of the data was according to assessed Kolmogorov-Smirnov test in both groups. Comparisons between groups were analyzed via student's t test. Chi-square test was used to compare categorical variables. Pearson's correlation was performed to analyze the relation preoperative astigmatism and the SIA. P values < 0.05 were considered statistically signi cant.

Sample Size Calculation
The smallest astigmatic change that might be clinically relevant was de ned to be 0.25 and 0.125 D respectively for anterior and posterior corneal surface. Based on pilot data of 20 patients, the standard deviation of SIA was approximately to be 0.4 and 0.2D respectively for anterior and posterior corneal surface. Setting a to 0.05 and power 0.8, the minimum sample size for each group was calculated to be 41 eyes.   and SIA-P ). 1D or greater SIA-A was found 24.4 % and 15.6% and 0.5 D or greater SIA-P was found 14.6% and 12.2% respectively in group 1 and group 2. Table 2 shows information about postoperative data. Abbreviations: CA: Corneal astigmatism, AC: Astigmatism change, SIA: Surgery induced astigmatism As shown in Table 3, there is a signi cant change in anterior corneal astigmatism values in both groups compared to preoperative values (p = 0.002, 0.005 respectively in group 1 and 2) but the difference between the magnitudes of the preoperative and postoperative posterior corneal astigmatism was insigni cant (p = 0.471 and 0.247 respectively in group 1 and 2). The median posterior astigmatism change is -0.005 and − 0.01 respectively in group 1 and 2.  The correlation was not signi cant between the magnitude of the SIA-A and preoperative anterior corneal astigmatism (r = 0.08 p = 0.639, r=-0.07 p = 0.963 respectively in group 1 and 2) or between the SIA-P and posterior corneal astigmatism (r = 0.249, p = 0.100, r = 0.139 p = 0.391 respectively in group 1 and 2) In the subgroup analyses for anterior corneal surface, anterior corneal astigmatism was signi cantly change in horizontal subgroup in groups 1 however corneal astigmatism was signi cantly different in horizontal and oblique subgroups in group 2 (p = 0.005, 0.017, 0.015 respectively).
In the comparison of subgroups between groups, there was no signi cant difference in according to SIA-A, SIA-P, AC anterior and posterior, however SIA-A were lower in group 2 compared to the group 1 except oblique subgroup and groups 2 has less SIA-P for all subgroups. Table 4 shows SIA and AC between groups to compare subgroups for anterior and posterior corneal surface.

Discussion
Corneal incision causes tissue damage and healing process changes anterior and posterior corneal curvature. Consequently, these changes make as surgery-induced astigmatism therefore SIA should be evaluated well but the classical spherocylinder format is not suitable for SIA thus different method was de ned to manage this type of data, such as vector analysis technique or polar value analysis [6-8].
Many factors affects SIA such as the lenght, type, widht and location of insition. For this purpose various incision location (like steep axis or temporal incision) or different type of incision (such as single plane, bi-plane corneal incision or scleral incision) were analyzed but the incision sequence was ignored [9,10].
In the literature, no evidence was found about the sequence of the incision on anterior and posterior astigmatism and SIA so we evaluated to sequence of the incision affect on SIA with 2.8 mm single plane clear corneal incision.
Our mean SIA anterior is 0.76D, 0.67 D respectively in group 1 and group 2 with 2.8 mm superior clear corneal incision. Performing the main incision rst caused less SIA however there was no difference between the groups statistically. In a study comparing the temporal incision with the superior incision, average SIA was found 0.77 and 1.29 with 2.8 mm corneal incision respectively [11]. In another study showed that superior incision was cause 48.28 % more SIA than by the temporal incision [12]. Similarly, there is a publication stating that the superotemporal or superionasal incision is better options because it causes less SIA than the superior incision [13]. We prefer superior incision because of the manipulations are made more easy than the temporal position.
In our study, SIA posterior is 0.27 D and 0.23 D respectively in group1, group 2 and these rates are compatible with the literature. Although there was no statistically difference between the groups and subgroup analysis, group 2 has lower SIA-P than group 1 and subgroups of group 2 has less SIA-P than subgroups of group 1 In the current study, posterior corneal astigmatism wasn't different from the preoperative in the both of groups. Similarly Kohnen et all. showed that posterior corneal astigmatism didn't change postoperative in one months period after 2.2 mm temporal clear corneal incisions in femtosecond laser-assisted cataract surgery [14]. In the another study with 2.75 mm temporal corneal incision, posterior astigmatic values didn't change after the operation [15]. Kim et al. found that the mean SIA of posterior cornea was 0.19D with a 2.2 mm temporal incision cataract surgery and stated that it did not make any changes in posterior astigmatism with regard to preoperative posterior cornea curvature [16].
There are also studies that have found a lower posterior SIA score. Hayashi et al. demonstrated mean SIA-P was 0,10D in short temporal CCI group but their study didn't compare postoperative posterior astigmatism with preoperative astigmatism [17]. In these studies concluded that changing posterior astigmatism after cataract surgery are clinically insigni cant and negligible [14][15][16][17]. In our study, AC-P is almost zero. Additionally, a decrease in posterior astigmatism was observed in the ATR-P subgroup of group 1, but it was not signi cant.
Although cataract surgery does not seem to have a signi cant effect on posterior astigmatism in some studies in the literature, there are also publications reporting opposing opinions. Cheng et al. pointed out that the posterior corneal astigmatism is a signi cant contributor to astigmatism after cataract surgery and demonstrated that posterior astigmatism is 0.5D or more in 24% of patients [18]. However, in this study, a single corneal suture was placed on main corneal incision and suture wasn't removed before measured by the Scheimpfug analysis system. In the study of Li et al. the mean SIA-p value was found to be 0.34 D but this study enrolled patient with regular corneal astigmatism ≥ 1.0 D and at the end of the study, a signi cant correlation was found between preoperative astigmatism value and SIA-P [19]. The mean preoperative astigmatism value was 1.58, and this may be the reason for the high SIA-P.
Nemeth at al. found that mean SIA-P was 0,31 D and it was 0.5 D or greater in 25% of eyes with 2.8 mm on axis manual corneal incision [20]. The mean SIA-P and 0.5 D or greater SIA-P were found is slightly higher than the our study. These rates were 14.6% and 12.2% respectively in group 1 and group 2 in the current study. The mean SIA-P and greater SIA-P than 0.5 D ratio were higher than our result whereas 2.8 mm corneal incision was used in both studies. However preoperative rate of posterior astigmatism more than 0.5 D was 11.36% in Nemeth et al.'s study whereas this ratio was just 2.4% in our study. A signi cant correlation was found between preoperative posterior astigmatic values of diopters and SIA-P in the study of Nemeth but in the current study although this correlation was present in both groups but it wasn't signi cant.
The limitations of our study are the short follow-up time and relatively small sample size. Since the data are obtained during a short follow-up period, long-term effect of incision order after surgery could not be analyzed in this study. Sample size has particularly limited the evaluation of subgroup analysis, so subgroup result should be treated carefully. Although SIA-A and SIA-P values were lower in group 2 especially in the horizontal subgroup, the results wasn't different, probably it caused by the small sample size.

Conclusion
Making the main incision rst caused less SIA especially on posterior surfaces however this difference was not signi cant with 2.8 mm superior clear corneal incision cataract surgery. However, further prospective, clinical trials with larger numbers of patients are needed to investigate the long-term effects of the incision sequence on SIA and corneal astigmatism.

Declarations
Authors Contribution: All authors met the ICMJE authorship criteria. All authors made substantial contributions to conception, design, analysis and interpretation of data, contributed to writing the article, provided critical revision of the manuscript, and approved the nal version.
Funding: The authors received no nancial support for this study.
Data availability: Open request