The Effect of Biometric Parameters on Spot Vision Screener Results

objective: Our aim in this study is to investigate the relationship between biometric parameters and spot vision screener [SVS] Materyal-Metods: 250 eyes of 125 children between the age of 6,77±1,59 included the study. The results of cyloplegic [CSVS] and noncycloplegic [NSVS] spot vision screener [Plusoptix p12,Germany] measurements and autorefractometer with cycloplegia [CA] were compared. The spherical equivalent [SE] differences of between CA and NSVS measurements with CA and CSVS measurements were compared with AL, ACD, CR, Mean K and AL / CR values. The relationship between them was examined. Results: According to the amblyopia risk factors [ARFs] based on the criteria from AAPOS 2013 guidelines , 33 eyes [13%] in the NSVS results and 34 eyes [13.6%] in the CSVS results were detected as amblyopic. According to CA results, NSVS had 67.3% sensitivity, 94.5% specicity, CSVS 69.4% sensitivity and 89.1% specicity in detecting amblyopia. When the SE differences of between CA and NSVS values were comparaed with biometric parameters a negative correlation was observed with ACD , AL and especially AL/CR ratio . Conclussion: Both NSVS and CSVS showed moderate sensitivity and high specicity in detecting AFRs based on the criteria from AAPOS 2013 guidelines. CSVS has no additional clinical advantage. Biometric parameters effect the NSVS results.


Introduction
Amblyopia, unilateral or bilateral, is characterized by low best-corrected visual acuity [BCVA] caused by visual de ciency and/or abnormal binocular interaction without structural defects in the eye or posterior visual pathways [1]. The main causes are strabismus, anisometropia, and deprivation [2]. In the presence of one or more of these causes, the retina cannot form a clear image, resulting in amblyopia in infancy and childhood, when vision develops rapidly [3]. Early diagnosis and treatment can prevent permanent vision loss [4]. Because amblyopia usually causes no symptoms, it can be detected only by screening.
Thus, eye examinations are important, especially in early childhood [5]. However, because the accommodation response in children is high, autorefractometer measurements without cycloplegia may give incorrect results [6]. Therefore, measurements using cycloplegia are necessary, but because cooperation and concentration in children are also low, measurement may be di cult. In addition, the side effects of cycloplegic agents are a limiting circumstance [7]. These limitations may be overcome by using spot vision screener [SVS] devices, which are handheld photo refractors that use infrared technology to acquire data from both eyes simultaneously. The greatest advantages of SVS are that it measures quickly and without cycloplegia. Many studies have compared SVS devices with cycloplegia autorefractometers [CA] and retinoscopy [8][9][10][11]. SVS sensitivity has been reported as 60.9%-89.8% and its speci city as 70.4%-94.9% [8,[12][13][14]. However, few studies have investigated factors that may affect SVS performance [15]. The aim of the present study was to compare SVS measurements made with and without cycloplegia with biometric parameters to identify the factors that may affect SVS performance. The data's compliance with normal distribution was tested using the Shapiro Wilk test. The differences among the measurements were analyzed using the Friedman test. Paired comparisons of the parameters that differed were made using Dunn's test. Agreement between the two raters was examined using the Kappa coe cient. A perfect t was de ned as a Kappa value > 0.76, a good t beyond chance was de ned as a value between 0.40 and 0.75, a weak t beyond chance was de ned as a value between 0.00 and 0.39, and no t as values < zero. The correlation among quantitative variables was examined and interpreted using Spearman's rho correlation coe cient. As descriptive statistics, mean ± standard deviation was used for numerical variables, and median, quarter 1 [Q1], and quarter 3 [Q3] were used for categorical variables, which were given as numbers and values. The SPSS for Windows version 24.0 software program was used for statistical analysis, and p < 0.05 was considered statistically signi cant.

Results
The patients' ages ranged from 5 to 10 years, and the average was 6.77 ± 1.59 years. Of the 250 eyes included in the study, 120 belonged to females and 130 to males [ Table 1]. Mean AL was 22.76 ± 1.1, mean ACD was 3.65 ± 0.3, mean CR was 7.76 ± 0.28, and mean AL/CR was 2.93 ± 0.14 [ Table 5].
When the differences of SE between CA and NSVS values were compared with biometric parameters a negative correlation was observed with ACD , AL and AL/CR ratio [ Table 6, Figures 1, 2, and 3].

Discussion
Because amblyopia is a serious condition that can result in permanent vision loss if not treated in childhood, screening programs are recommended [16,17]. Many studies have examined the effectiveness of vision screening tests in preschool children [18][19][20]. Visual screening in children is also important for early diagnosis of strabismus and other eye diseases besides the serious refractive errors that cause amblyopia. Instrument-based vision scanning has been recommended by the American Association for Pediatric Ophthalmology and Strabismus [AAPOS] and the American Association of Certi ed Orthoptists [AACO] [21]. Many studies have reported that SVS effectively detects refractive errors and prevents refractive amblyopia in preschool children [22][23][24][25].
In the current study, Plusoptix p12 was used as an SVS device. The sensitivity of NSVS in predicting amblyopia was 67.3%, and its speci city was 94.5%. The sensitivity of CSVS was 69.4%, and its speci city was 89.1%.
In another study, Yakar stated that SVS without cycloplegia had 59% sensitivity and 94% speci city for detecting the risk factors of amblyopia compared to the autorefractometer with cycloplegia [15].
In their study of the Plusoptix A09, Yan et al. found that it had an overall sensitivity of 94.9% and a speci city of 67.5% for detecting the risk factors of refractive amblyopia [27]. The differences in these results can be attributed to the different devices used and the different age groups tested.
Although SVS is designed to work without cycloplegia, studies have also compared SVS with autorefractometer and/or retinoscopy and with and without cycloplegia [8]. Yakar noted that sensitivity and negative predictive value increased with cycloplegia. Therefore, he concluded that measurement using cycloplegia may be useful in selected cases [8]. In the present study, despite a slight difference in sensitivity, there was no increase in negative predictive value. This result may be due to the devices used. Nevertheless, we believe that SVS measurement using cycloplegia results in no additional bene t and extends the duration of the procedure, which may be more di cult for the child.
It is known that emetropization occurs with balance among the ocular biometric parameters [28]. Changes in AL and CR are important biometric factors that affect refractive errors [29]. In recent years, studies have noted that the AL/CR ratio is a better marker than AL and CR [30,31]. Therefore, in the present study, we wanted to investigate the potential relationships between SVS and AL/CR, which have never been studied. The only related study in the literature is one by Yakar that compared the performance of SVS with those of ACD and AL without cycloplegia [15]. That study found a negative correlation between them. Our study is only the second to investigate this relationship. In addition, our study compared the CR, meanK, and AL/CR values using SVS both with and without cycloplegia. Like Yakar's study, ours found a negative correlation between ACD , AL, AL/CR ratio when using NSVS, but no correlation when using CSVS.
According to these results, NSVS is less reliable in eyes that have narrow anterior chambers, short axial lengths, and especially reduced AL/CR ratios. Studies have shown that the AL/CR ratio is more than 3 in myopia and less than 2.90 in hyperopia [32,33]. Based on these ndings, when measurement using cycloplegia is not possible, SVS values are reliable for myopic children whose AL/CR ratios are 3 or higher.
The strength of the present study is that it compares SVS values both with and without cycloplegia and correlates two conditions with biometric parameters. The study's limitations include its small number of patients and its lack of comparison with retinoscopy.
However, as previously mentioned, studies using SVS have observed different rates of reliability. It is important to know the anatomical factors that may affect NSVS results. To this end, the present study found that NSVS was affected by biometric parameters and had a negative correlation with AL, ACD, and especially the AL/CR ratio.
Based on these results, we conclude that NSVS is more reliable in children who have long axial lengths, deep anterior chambers, and increased AL/CR ratios. We believe that more accurate results will be obtained with larger studies that perform biometry measurements both with and without cycloplegia.

Declarations
To be used for all articles, including articles with biological applications : 'Not applicable' Funding: No funds, grants, or other support was received.        When the differences of SE between CA and NSVS values were compared with biometric parameters a negative correlation was observed with ACD , AL and AL/CR ratio [ Table 6, Figures 1, 2, and 3].

Figure 2
When the differences of SE between CA and NSVS values were compared with biometric parameters a negative correlation was observed with ACD , AL and AL/CR ratio [ Table 6, Figures 1, 2, and 3].