Subject
Accepted refractive surgery patients included 130 eyes (67 patients). Twelve patients underwent surgery in only one eye, with the contralateral eye untreated due to monocular vision. The study population comprised 30 males (40.54%) and 44 females (59.46%), with a mean age of 24.01 years (range: 18-34 years). Moreover, in this experiment, an additional 40 pieces of information from normal human eyes were collected as a control group for comparison with the experimental group.
All patients in this study underwent SMILE refractive surgery. SMILE was performed using the VisuMax femtosecond laser (Carl Zeiss Meditec, Jena, Germany). The following parameters were used: 100-µm cap thickness, 7.5-mm cap diameters, 6.0- to 6.5-mm optical zone, 500-kHz repetition rate, 35 to 37 (130 nJ) energy cut index, 2-mm side-cut incision, and 4.5-µm spot and track distance. The patient was positioned under the curved contact glass of the femtosecond laser and asked to fixate on a blinking target. When appropriate centration was achieved, suction was applied and the laser started. Following creation of the lenticule, the incision was opened and the two planes of the lenticule were identified. A thin blunt spatula was used to dissect the superficial and deep planes of the lenticule and break the remaining tissue bridges, thus separating the lenticule from the surrounding stroma. This lenticule was grasped with a pair of forceps and extracted through the 2-mm incision. The corneal interface was then flushed with balanced salt solution. The postoperative regimen included prednisolone acetate, ofloxacin, and lubricating eye drops.
All participants were recruited from patients who underwent refractive surgery between 2020 and 2022. Patients were randomly selected from all consultations conducted between September 2020 and March 2021. Selection for analysis was not based on the presence or absence of reported visual symptoms.
This study followed the tenets of the Declaration of Helsinki and was approved by the Ethics Committee of West China Hospital of Sichuan University (SCU-2020-886).
Instruments
OQAS (OQAS II, Visiometrics, SL) and iTrace analyzer (iTrace, Tracey Technologies) were used to measure intraocular scatter index (OSI) and root mean square (RMS) of HOAs at 5mm pupil diameter respectively in this study. Both instruments are widely used for objective optical quality inspection and have been adequately verified in previous work[21, 34].
Based on the double-pass technique, OQAS is known as the device that can quantitatively provide the data of intraocular scatter. The most dominant parameters detected by OQAS include: (1) SR that indicates the convergence ratio of light intensity in the image field of an optical system with aberrations, and the high the value is, the better the optical quality is; (2) MTF cutoff that characterizes the spatial frequency corresponding to the minimum resolution of human eyes in the modulation transfer function curve, and the higher the value is, the better the optical quality becomes; (3) OSI that objectively reflects the scatter of the refractive medium, and the higher the value is, the muddier the refractive media is[35].
Another evaluation instrument, i.e. iTrace analyzer, is composed of a corneal topographer and an aberrometer whose working principle is ray tracing. It is applicable for the measurement of corneal, internal and total intraocular wavefront aberrations. Wavefront aberrations are distortions in the phase of light entering the eye, which leads to the defects in image-forming and thereby decreasing the quality of vision. They could be caused by the non-optimal surface shapes, irregularities and misalignments in the eye’s optical elements. It’s generally accepted that as the wavefront aberrations were corrected, the visual sensitivity, night vision, incidence of glare and halo would significantly improve. They could be mathematically represented as the sum of a series of polynomial functions of different orders, and the higher the wavefront aberration is, the more the visual quality is affected. In particular, the HOAs have a greater impact on the visual quality. This study applied iTrace analyzer to measure the total intraocular wavefront aberrations before and after the operation, including TWA, TLOA, THOA, etc[36].
The SPSS statistical software (Version 25.0; IBM SPSS Inc., Chicago, Illinois, USA)[37] and Microsoft Excel were utilized for statistical analysis.
Method
The purpose of this study is to explore whether there is a causal relationship between ocular scatter and higher-order aberrations. We used the commonly used criteria for inferring causal relationships in the biomedical field[38-43]. There are 8 criteria for inferring causal relationships, among which the temporality and correlation of the association are necessary conditions; the reproducibility of the association, ruling out other possibilities, the strength of the association, and empirical evidence are also of great significance; other criteria can be used as references in determining causality.
Firstly, the necessary temporality for causal relationships stipulates that the cause in a causal relationship must precede the occurrence of the result. In other words, if A is considered the cause of B, then A must occur before B. This is the most important criteria in determining causality. Secondly, the correlation of the association is an important manifestation of the causal relationship and is a necessary prerequisite for the establishment of a causal relationship. Thirdly, the reproducibility of the association refers to obtaining consistent conclusions from the same causal relationship through multiple repetitions, which makes the causal inference more convincing. The reproducibility criterion requires that the same causal relationship should produce consistent results in different contexts. This means that if A is considered the cause of B, then A must produce the same results in different situations. Fourthly, the criterion for ruling out other explanations requires the exclusion of other possible explanations, which means that if A is considered the cause of B, then the influence of all other factors must be ruled out to determine that A is the only reasonable explanation, with no other reasonable alternative. Fifthly, the strength of the association refers to the degree of correlation between two factors, and the RR value is commonly used for quantitative description in prospective cohort studies. The relative ratio (RR) is a commonly used indicator in clinical control studies (prospective studies) to quantify the strength of the association between the factors under study. When RR > 1, it indicates that the subjects in the experimental group are more consistent with the research conclusion compared to the control group. Under the background conditions set by the experimental group, there is a high intensity of causal correlation between the factors analyzed. When RR < 1, compared to the experimental group, the subjects in the control group are more consistent with the research conclusion. Under the background conditions set by the experimental group, the intensity of the causal correlation between the factors analyzed is insufficient. When RR = 1, it means that the background conditions set by the experimental group are irrelevant to the research conclusion. The Confidence Interval (CI) and the Z Test can be used to test the significance of the RR value. If the range of the Confidence Interval (CI) does not include 0, it indicates that the calculated RR value has statistical significance. The Z Test is generally used for the difference test of large sample (i.e., sample size greater than 30) means. It uses the theory of standard normal distribution to infer the probability of difference occurrence, thereby comparing whether the difference between the two means is significant. If the z value is greater than 2.58, it indicates that the degree of difference explored is very significant within the 0.01 confidence interval. The strength of the association can serve as an important basis for determining the establishment of a causal relationship, as a stronger association indicates a greater likelihood of a causal relationship. Sixthly, the last important criterion is empirical evidence, which has a high strength as a criterion for determining causal relationships. Empirical evidence can come from field experiments, clinical trials, or basic medical experiments. In addition, there are other criteria for causal association, biological plausibility, and specificity of the association. Since the determination of causality is complex, not all of these criteria need to be met in assessing causality. However, the more criteria that are met, the stronger the evidence for the establishment of a causal relationship.
The subjects in the experimental group were tested before surgery and at 1 week, 6 and 12 months after surgery. Considering that the measurement of ocular higher-order aberrations varies with different pupil diameters[44], the change in ocular higher-order aberrations measured at too small pupil diameters is not significant, while the pupil size of most human eyes is not too large without mydriasis. Therefore, we manually set the pupil diameter as 5mm for each measurement. To indicate the time difference between data collection and refractive surgery, we assigned a numerical label to each monitoring data, where "0" represents preoperative monitoring data, "1" represents monitoring data at 1 week postoperatively, "2" represents monitoring data at 6 months postoperatively, and "3" represents monitoring data at 1 year postoperatively. The change in each monitoring data relative to the preoperative data is marked as "Δ". We conducted correlation analyses between ΔOSI 1 at 1 week postoperatively and RMS 2 at 6 months postoperatively, ΔOSI 2 at 6 months postoperatively and RMS 3 at 1 year postoperatively, and ΔOSI 1 at 1 week postoperatively and RMS 3 at 1 year postoperatively, verifying the correlation of causal relationships under the condition of meeting the temporality criterion (based on Pearson correlation coefficient and Bonferroni correction, p-values less than 0.025 were considered statistically significant). Through repeated analysis of experimental data at different time points, we confirmed whether the same conclusions would be drawn, thus verifying the criterion of reproducibility of causal relationships. To enhance the reliability of the results, we strictly adhered to the method of controlling variables and additionally collected data from 40 normal eyes as a control group. The data from the control group were collected synchronously with the experimental group under the same external conditions, and the analysis results were compared one by one with those of the experimental group to verify the criterion of ruling out other possibilities for causal relationships. The number of individuals in both the experimental and control groups that matched the research results were counted separately to calculate the RR value. The results were tested for reliability using the confidence interval (CI), thereby verifying the strength of the causal relationship. Furthermore, this study is entirely based on clinical experimental results, meeting the requirements for empirical evidence. The specificity of the association is mostly difficult to establish for non-communicable diseases and is not necessary to discuss further in this study. From the perspective of physiological optics, the cause of ocular scatter is the change in corneal transparency, while the cause of higher-order aberrations is the phase effect caused by changes in corneal shape (possibly refractive index), both of which result from defects in the optical medium. The association between the two does not contradict biological plausibility. In this study, combining the analysis and research results of various core indicators, we can further explore the biological plausibility of this association.