Flash Electroretinography as a Measure of Retinal Function in Myopia and Hyperopia: A Systematic Review

Refractive errors (myopia and hyperopia) are the most common visual disorders and are severe risk factors for secondary ocular pathologies. The development of refractive errors has been shown to be associated with changes in ocular axial length, suggested to be induced by outer retinal elements. Thus, the present study systematically reviewed the literature examining retinal function as assessed using global flash electroretinograms (gfERGs) in human clinical refractive error populations. Electronic database searching via Medline, PubMed, Web of Science, Embase, Psych INFO, and CINAHL retrieved 981 unique records (last searched on the 29 May 2022). Single case studies, samples with ocular comorbidities, drug trials, and reviews were excluded. Demographic characteristics, refractive state, gfERG protocol details, and waveform characteristics were extracted for the eight studies that met the inclusion criteria for the review and were judged to have acceptable risk of bias using the OHAT tool (total N = 552 participants; age 7 to 50). Study synthesis suggests that myopia in humans involves attenuation of gfERG photoreceptor (a-wave) and bipolar cell (b-wave) function, consistent with the animal literature. Meaningful interpretation of the overall findings for hyperopia was limited by inconsistent reporting, highlighting the need for future studies to report key aspects of gfERG research design and outcomes more consistently for myopic and hyperopic refractive errors.


Introduction
Refractive errors (myopia and hyperopia) are the most common visual disorders [1] and pose a growing socioeconomic and public health problem. They occur when the length of the eye and refractive power of optical components prevents accurate focusing of light on the neural retina. A myopic (short-sighted) eye is excessively large, while hyperopia occurs when the eye is too small. Myopia affects >1.5 billion people globally, with prevalence estimated to increase dramatically to~50% of the global population by 2050, a figure purported to be driven by increased educational and near-work demands [2][3][4]. High myopia of more than six diopters (D) affects a significant proportion of myopes, placing them at very high risk of developing severe vision threatening secondary pathologies such as retinal detachment, glaucoma, and choroidal neovascularization later in life [3][4][5]. Low myopes are also at risk of pathological complications such as maculopathy [6], further emphasizing the clinical significance of this condition and the need to better understand its etiology. Thus, to allow exploration of the site of the retinal elements contributing to myopia development, this paper aimed to systematically review the current literature associated with electrophysiological measurement of functioning of all cell types in the retina using global flash electroretinograms (gfERGs) rather than the more common pattern ERG that primarily provides information about the functioning of neurons in the inner retina ( [7]) of clinically defined myopic and hyperopic humans.

Electrophysiology as a Technique for Understanding the Role of Retinal Cells in Human Refractive Errors
Electroretinography can be used to non-invasively assess phototransduction and retinal processing of light at the cellular level in humans and animals [17]. The Electroretinogram (ERG) is an established electrophysiological diagnostic technique that is widely used in clinical and laboratory settings [30,31]. ERGs utilize external electrodes to measure the electrical activity of the retina following a light stimulus, such as a bright flash. Variations in the stimulus and recording setup allow multiple types of ERGs to be recorded, including Pattern (PERG), multifocal (mfERG), and global flash (gfERG), each differing in the specific information that they provide about retinal function [30,31]. In contrast to other ERG types (such as inner retinally focused PERGs), the waveform recorded in the gfERG predominantly reflects the global function of outer retinal photoreceptors, bipolar, RPE and Muller cells that have been theorized to play a key role in driving ocular growth changes and are also associated with the later development of secondary pathologies in myopia [32].

Using the gfERG to Functionally Dissect Retinal Activity
The gfERG is a measure of the mass response of the retina, elicited by a brief flash of light, that has been used to assess generalized retinal function in a broad range of ophthalmic conditions including refractive errors [33]. The gfERG response can be produced under various conditions including dark adaptation (to isolate the scotopic rod response) and light-adaptation (to isolate the photopic cone response) [31]. The gfERG waveform ( Figure 1) reflects a series of current loops that redistribute ions within the extracellular spaces of the retina after the onset of the light flash. These currents result primarily from changes to photoreceptor and bipolar cell polarity and subsequent interactions with Muller glial and RPE cells. Clinical gfERGs typically capture two major components in the recorded waveform; the a-wave primarily reflects the activity of the rod and cone photoreceptors, and the b-wave primarily reflects the activity of the bipolar cells to light onset [33,34]. Further analysis can uncover small rhythmic wavelets during the ascending phase of the b-wave called oscillatory potentials, which primarily reflect inhibitory feedback by inner retinal amacrine cells [35]. The function of each cell type can be inferred by measuring the size of each wave (amplitude) and the difference in time between onset of the response and maximum response reached (implicit time). Consistent with the animal model pharmacological research outlined above, attenuation of the gfERG a-wave and b-wave amplitude has been demonstrated in studies of optically and pharmacologically induced myopia in chicks [36][37][38][39]. However, although gfERG has been used extensively in clinical settings to examine human myopia e.g., [40][41][42][43][44]; the findings have been mixed, and current knowledge has not been systematically reviewed.
Vision 2023, 7, x FOR PEER REVIEW 3 of 15 changes and are also associated with the later development of secondary pathologies in myopia [32].

Using the gfERG to Functionally Dissect Retinal Activity
The gfERG is a measure of the mass response of the retina, elicited by a brief flash of light, that has been used to assess generalized retinal function in a broad range of ophthalmic conditions including refractive errors [33]. The gfERG response can be produced under various conditions including dark adaptation (to isolate the scotopic rod response) and light-adaptation (to isolate the photopic cone response) [31]. The gfERG waveform ( Figure 1) reflects a series of current loops that redistribute ions within the extracellular spaces of the retina after the onset of the light flash. These currents result primarily from changes to photoreceptor and bipolar cell polarity and subsequent interactions with Muller glial and RPE cells. Clinical gfERGs typically capture two major components in the recorded waveform; the a-wave primarily reflects the activity of the rod and cone photoreceptors, and the b-wave primarily reflects the activity of the bipolar cells to light onset [33,34]. Further analysis can uncover small rhythmic wavelets during the ascending phase of the b-wave called oscillatory potentials, which primarily reflect inhibitory feedback by inner retinal amacrine cells [35]. The function of each cell type can be inferred by measuring the size of each wave (amplitude) and the difference in time between onset of the response and maximum response reached (implicit time). Consistent with the animal model pharmacological research outlined above, attenuation of the gfERG a-wave and bwave amplitude has been demonstrated in studies of optically and pharmacologically induced myopia in chicks [36][37][38][39]. However, although gfERG has been used extensively in clinical settings to examine human myopia e.g., [40,[41][42][43][44]; the findings have been mixed, and current knowledge has not been systematically reviewed.  [31]). The onset of the light flash is indicated (thick orange arrow) alongside the a-wave and b-wave amplitude and implicit time components that can be measured in the resulting retinal response.  [31]). The onset of the light flash is indicated (thick orange arrow) alongside the a-wave and bwave amplitude and implicit time components that can be measured in the resulting retinal response.

Rationale and Aim of the Current Systematic Review
Although it is well accepted that the excessive ocular growth observed in animal models of myopia is locally controlled within the retina, the evidence and theories regarding the particular cell types involved in the control of ocular growth and the progression to secondary pathology in humans and animal models are still under debate [16,17]. Furthermore, to develop effective treatments and to facilitate the identification of those at higher risk of progressive myopic eye growth and later development of sight-threatening pathology, a better understanding of cellular contributions to myopia is necessary. As the gfERG measures the function of key cell types theorized to be involved in the onset and progression of myopia, synthesizing the evidence for gfERG changes in refractive error may provide insight for future research regarding the functional state of the retina over the course of the condition and the efficacy of interventions in preserving retinal integrity. Thus, the present paper aimed to systematically review the gfERG literature assessing retinal cell function in myopic and hyperopic humans. In accordance with PRISMA guidelines, our literature search and study selection strategies will be described, followed by our data extraction and risk of bias procedures. Following consideration of the study data extracted, results will examine data pertaining to gfERG waveform characteristics observed in refractive error in general and then for myopia and hyperopia, respectively. The discussion will examine these findings in context with the previous literature.

Literature Search
This review followed the PRISMA guidelines for systematic reviews [45]. PubMed, MEDLINE, Web of science, Embase, CINAHL, and PsycINFO databases were searched to identify all potentially eligible studies. This combination of databases was chosen for the coverage of concepts relevant to this review, and because it has been shown to retrieve more than 95% of all possible relevant references across a large selection of systematic reviews [46]. Google Scholar was not included due lack of specificity, accessibility, and search accuracy [47]. The final database search was conducted on 29 May 2022. Reference list searching of included studies was also conducted. This review was not registered prior to completion. Table 1 outlines the search strategy and provides example results from PubMed.

Study Selection
Studies comparing gfERG measures across different refractive states (e.g., myopia, hyperopia, high myopia, emmetropia) in human subjects were included in the review. Intervention studies (e.g., drug studies) were only included if baseline data were available. Studies were excluded if they investigated refractive error as a peripheral measure secondary to other conditions that are expected to be associated with functional deficits (e.g., congenital stationary night blindness and single gene mutations associated with multifaceted phenotypes). Single-case studies and studies of non-human animals were also excluded from the review.
The screening process was performed using Covidence Systematic Review Data Management Software. The title and abstract screening were performed independently by two reviewers (SZ and NR), where each reviewer decided either to reject or accept each record based on the inclusion and exclusion criteria outlined above. Conflicts were resolved through discussion with a third reviewer (MM). The same process was applied to full-text screening to identify relevant studies for data extraction. Decisions on whether to reject a study were based on the seven-step hierarchy presented in Table 2. Intervention study with no baseline data 6 Wrong study design (e.g., case studies) 7 Wrong outcomes variables

Data Extraction
A data extraction table was constructed in Microsoft Excel. Data were extracted independently by two reviewers (SZ and NR). Table 3 lists the types of information extracted from each included study. Due to inconsistency between studies in the reporting of results, both quantitative and qualitative data were extracted.

Risk of Bias Assessment
Quality assessment of each included study was conducted using the Office of Health Assessment and Translation (OHAT) risk of bias tool for human and animal studies [48]. The OHAT is recommended by the National Health and Medical Research Council of Australia as a practical and flexible best practice tool for risk of bias assessment [49][50][51][52]. Questions 1, 2, and 5 from the tool (Table 4) were excluded from the present assessment as they were only applicable to either human-control trails or experimental animal studies. Question 11 provides the option for additional questions about other potential threats to internal validity to be added on a project-specific basis and hence was not used in the current review. The risk of bias assessment was independently completed by SZ, and where required all authors were consulted regarding judgements. Were experimental conditions identical across study groups? 6 Were the research personnel and human subjects blinded to the study group during the study? 7 Were outcome data complete without attrition or exclusion from analysis? 8 Can we be confident in the exposure characterization? 9 Can we be confident in the outcome assessment? 10 Were all measured outcomes reported? 11 Were there no other potential threats to internal validity?

Study Selection
As seen in Figure 2, electronic database searching and hand searching of reference lists identified 981 unique records (n = 1118 duplicates). Of the 981 records that entered title and abstract screening, 933 studies were deemed irrelevant based on the inclusion criteria. Full-text screening of the remaining 48 records identified 40 studies that did not meet inclusion criteria and were excluded. Among the excluded studies, 29 did not have full-text available (e.g., primarily conference abstracts); five had the wrong group comparison; three had the wrong study design; one was not in English; one had no baseline data; and one had the wrong outcome variable. Eight studies that met the inclusion were included in the systematic review.
6 study? 7 Were outcome data complete without attrition or exclusion from analysis? 8 Can we be confident in the exposure characterization? 9 Can we be confident in the outcome assessment? 10 Were all measured outcomes reported? 11 Were there no other potential threats to internal validity?

Study Selection
As seen in Figure 2, electronic database searching and hand searching of reference lists identified 981 unique records (n = 1118 duplicates). Of the 981 records that entered title and abstract screening, 933 studies were deemed irrelevant based on the inclusion criteria. Full-text screening of the remaining 48 records identified 40 studies that did not meet inclusion criteria and were excluded. Among the excluded studies, 29 did not have full-text available (e.g., primarily conference abstracts); five had the wrong group comparison; three had the wrong study design; one was not in English; one had no baseline data; and one had the wrong outcome variable. Eight studies that met the inclusion were included in the systematic review.

Risk of Bias Assessment
Overall risk of bias using the OHAT tool was deemed to be either below the critical level or minimal across the studies except for two studies which demonstrated a definitely high risk of bias on question 7 "Were outcome data complete without attrition or exclusion from analysis?" (Table 5). Data from six participants for one study who were part of baseline data for comparison could not be obtained [50]. The other study did not obtain data from some participants for certain components of the ERG [41]. All included studies displayed a "probably high" risk of bias for the question 6 "Were the personnel and human subjects blinded to study group during study?". These studies recruited clinical populations (i.e., referred patients); therefore, blinding to study groups could not be achieved. Although these factors increase the risk of bias, refractive error is an objective measure of ocular biometrics, and the gfERG is an objective measure of retinal function, which together minimize the potential effect on internal validity.  [50] an et al., 1984 [42] idanandam et al., 2017 [52] et al., 2020 [40] all et al., 2001 [41] amoto et al., 1997 [51] definitely low probably low probably high definitely high.

Study Characteristics
Study characteristics are summarized in Table 6. The design of the included studies was either between group (37.50%), correlational (25%), within group (12.50%), or mixed (37.50%). A total of 522 participants were identified across all included studies, and where the information was provided, an approximately proportional number of female and male participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included studies. Fifty percent of studies reported age of participants in M(SD), ranging from 7.1(4.4) to 26.9(2.4) years. In the remaining 50% of the studies not reporting M(SD), age ranged from 10 to 50 years. Fifty percent of the studies assessed gfERGs in younger populations (10-23 years), while the remaining assessed older populations (7-50 years). One study did not report maximum age of the participants [43]. In total, 37.50% of studies reported that the controls were age-matched to the myopic/hyperopic participants.
Myopia was assessed in 62.50% of the studies, with refractive state ranging from +0.5D to −27D, meanwhile refractive error ranged from 0 to +11D in the study assessing hyperopia (12.50%). Both myopia and hyperopia were assessed in 25% of studies, where refractive error ranged from ≤ −6D to ≥+6D. Most participants were identified has having no pathologies secondary to myopia including retinal detachment, retinopathy, or any other ocular disease (62.50%). The remaining participants were predominantly characterized as either having some retinal degeneration (12.50%), reduced vision (12.50%), or posterior vitreous detachment (12.50%).

Study Characteristics
Study characteristics are summarized in Table 6. The design of the inclu was either between group (37.50%), correlational (25%), within group (12.50% (37.50%). A total of 522 participants were identified across all included studies the information was provided, an approximately proportional number of fem participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included studies. Fift studies reported age of participants in M(SD), ranging from 7.1(4.4) to 26.9(2 the remaining 50% of the studies not reporting M(SD), age ranged from 10 Fifty percent of the studies assessed gfERGs in younger populations (10-23 y the remaining assessed older populations (7-50 years). One study did not r mum age of the participants [43]. In total, 37.50% of studies reported that the c age-matched to the myopic/hyperopic participants.
Myopia was assessed in 62.50% of the studies, with refractive state ra +0.5D to −27D, meanwhile refractive error ranged from 0 to +11D in the stud hyperopia (12.50%). Both myopia and hyperopia were assessed in 25% of stu refractive error ranged from ≤ −6D to ≥+6D. Most participants were identif ing no pathologies secondary to myopia including retinal detachment, retinop other ocular disease (62.50%). The remaining participants were predominant ized as either having some retinal degeneration (12.50%), reduced vision (12.5 terior vitreous detachment (12.50%).

Study Characteristics
Study characteristics are summarized in Table 6. The design of the included studies was either between group (37.50%), correlational (25%), within group (12.50%), or mixed (37.50%). A total of 522 participants were identified across all included studies, and where the information was provided, an approximately proportional number of female and male participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included studies. Fifty percent of studies reported age of participants in M(SD), ranging from 7.1(4.4) to 26.9(2.4) years. In the remaining 50% of the studies not reporting M(SD), age ranged from 10 to 50 years. Fifty percent of the studies assessed gfERGs in younger populations (10-23 years), while the remaining assessed older populations (7-50 years). One study did not report maximum age of the participants [43]. In total, 37.50% of studies reported that the controls were age-matched to the myopic/hyperopic participants.
Myopia was assessed in 62.50% of the studies, with refractive state ranging from +0.5D to −27D, meanwhile refractive error ranged from 0 to +11D in the study assessing hyperopia (12.50%). Both myopia and hyperopia were assessed in 25% of studies, where refractive error ranged from ≤ −6D to ≥+6D. Most participants were identified has having no pathologies secondary to myopia including retinal detachment, retinopathy, or any other ocular disease (62.50%). The remaining participants were predominantly characterized as either having some retinal degeneration (12.50%), reduced vision (12.50%), or posterior vitreous detachment (12.50%).

Study Characteristics
Study characteristics are summarized in Table 6. The design of th was either between group (37.50%), correlational (25%), within group (37.50%). A total of 522 participants were identified across all included the information was provided, an approximately proportional number participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included stud studies reported age of participants in M(SD), ranging from 7.1(4.4) to the remaining 50% of the studies not reporting M(SD), age ranged fr Fifty percent of the studies assessed gfERGs in younger populations ( the remaining assessed older populations (7-50 years). One study di mum age of the participants [43]. In total, 37.50% of studies reported tha age-matched to the myopic/hyperopic participants.

Study Characteristics
Study characteristics are summarized was either between group (37.50%), correl (37.50%). A total of 522 participants were id the information was provided, an approxim participants were included (37.50%).
Participant age ranged from 7 to 50 ye studies reported age of participants in M( the remaining 50% of the studies not rep Fifty percent of the studies assessed gfERG the remaining assessed older populations mum age of the participants [43]. In total, 3 age-matched to the myopic/hyperopic par Myopia was assessed in 62.50% of t +0.5D to −27D, meanwhile refractive error hyperopia (12.50%). Both myopia and hyp refractive error ranged from ≤ −6D to ≥+ ing no pathologies secondary to myopia in other ocular disease (62.50%). The remaini ized as either having some retinal degener terior vitreous detachment (12.50%).

Study Characteristics
Study characteristics are summarized in Table 6. The design of the included studies was either between group (37.50%), correlational (25%), within group (12.50%), or mixed (37.50%). A total of 522 participants were identified across all included studies, and where the information was provided, an approximately proportional number of female and male participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included studies. Fifty percent of studies reported age of participants in M(SD), ranging from 7.1(4.4) to 26.9(2.4) years. In the remaining 50% of the studies not reporting M(SD), age ranged from 10 to 50 years. Fifty percent of the studies assessed gfERGs in younger populations (10-23 years), while the remaining assessed older populations (7-50 years). One study did not report maximum age of the participants [43]. In total, 37.50% of studies reported that the controls were age-matched to the myopic/hyperopic participants.
Myopia was assessed in 62.50% of the studies, with refractive state ranging from +0.5D to −27D, meanwhile refractive error ranged from 0 to +11D in the study assessing hyperopia (12.50%). Both myopia and hyperopia were assessed in 25% of studies, where refractive error ranged from ≤ −6D to ≥+6D. Most participants were identified has having no pathologies secondary to myopia including retinal detachment, retinopathy, or any other ocular disease (62.50%). The remaining participants were predominantly characterized as either having some retinal degeneration (12.50%), reduced vision (12.50%), or posterior vitreous detachment (12.50%).

Study Characteristics
Study characteristics are summarized in Table 6. The design of the inclu was either between group (37.50%), correlational (25%), within group (12.50% (37.50%). A total of 522 participants were identified across all included studies the information was provided, an approximately proportional number of fem participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included studies. Fift studies reported age of participants in M(SD), ranging from 7.1(4.4) to 26.9(2 the remaining 50% of the studies not reporting M(SD), age ranged from 10 Fifty percent of the studies assessed gfERGs in younger populations (10-23 y the remaining assessed older populations (7-50 years). One study did not r mum age of the participants [43]. In total, 37.50% of studies reported that the c age-matched to the myopic/hyperopic participants.
Myopia was assessed in 62.50% of the studies, with refractive state ra +0.5D to −27D, meanwhile refractive error ranged from 0 to +11D in the stud hyperopia (12.50%). Both myopia and hyperopia were assessed in 25% of stu refractive error ranged from ≤ −6D to ≥+6D. Most participants were identif ing no pathologies secondary to myopia including retinal detachment, retinop other ocular disease (62.50%). The remaining participants were predominant ized as either having some retinal degeneration (12.50%), reduced vision (12.5 terior vitreous detachment (12.50%).

Study Characteristics
Study characteristics are summarized in Table 6. The design of the included studies was either between group (37.50%), correlational (25%), within group (12.50%), or mixed (37.50%). A total of 522 participants were identified across all included studies, and where the information was provided, an approximately proportional number of female and male participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included studies. Fifty percent of studies reported age of participants in M(SD), ranging from 7.1(4.4) to 26.9(2.4) years. In the remaining 50% of the studies not reporting M(SD), age ranged from 10 to 50 years. Fifty percent of the studies assessed gfERGs in younger populations (10-23 years), while the remaining assessed older populations (7-50 years). One study did not report maximum age of the participants [43]. In total, 37.50% of studies reported that the controls were age-matched to the myopic/hyperopic participants.
Myopia was assessed in 62.50% of the studies, with refractive state ranging from +0.5D to −27D, meanwhile refractive error ranged from 0 to +11D in the study assessing hyperopia (12.50%). Both myopia and hyperopia were assessed in 25% of studies, where refractive error ranged from ≤ −6D to ≥+6D. Most participants were identified has having no pathologies secondary to myopia including retinal detachment, retinopathy, or any other ocular disease (62.50%). The remaining participants were predominantly characterized as either having some retinal degeneration (12.50%), reduced vision (12.50%), or posterior vitreous detachment (12.50%).

Study Characteristics
Study characteristics are summarized in Table 6. The design of th was either between group (37.50%), correlational (25%), within group (37.50%). A total of 522 participants were identified across all included the information was provided, an approximately proportional number participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included stud studies reported age of participants in M(SD), ranging from 7.1(4.4) to the remaining 50% of the studies not reporting M(SD), age ranged fr Fifty percent of the studies assessed gfERGs in younger populations ( the remaining assessed older populations (7-50 years). One study di mum age of the participants [43]. In total, 37.50% of studies reported tha age-matched to the myopic/hyperopic participants.

Study Characteristics
Study characteristics are summarized was either between group (37.50%), correl (37.50%). A total of 522 participants were id the information was provided, an approxim participants were included (37.50%).
Participant age ranged from 7 to 50 ye studies reported age of participants in M( the remaining 50% of the studies not rep Fifty percent of the studies assessed gfERG the remaining assessed older populations mum age of the participants [43]. In total, 3 age-matched to the myopic/hyperopic par Myopia was assessed in 62.50% of t +0.5D to −27D, meanwhile refractive error hyperopia (12.50%). Both myopia and hyp refractive error ranged from ≤ −6D to ≥+ ing no pathologies secondary to myopia in other ocular disease (62.50%). The remaini ized as either having some retinal degener terior vitreous detachment (12.50%).

Study Characteristics
Study characteristics are summarized in Table 6. The design of the included studies was either between group (37.50%), correlational (25%), within group (12.50%), or mixed (37.50%). A total of 522 participants were identified across all included studies, and where the information was provided, an approximately proportional number of female and male participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included studies. Fifty percent of studies reported age of participants in M(SD), ranging from 7.1(4.4) to 26.9(2.4) years. In the remaining 50% of the studies not reporting M(SD), age ranged from 10 to 50 years. Fifty percent of the studies assessed gfERGs in younger populations (10-23 years), while the remaining assessed older populations (7-50 years). One study did not report maximum age of the participants [43]. In total, 37.50% of studies reported that the controls were age-matched to the myopic/hyperopic participants.
Myopia was assessed in 62.50% of the studies, with refractive state ranging from +0.5D to −27D, meanwhile refractive error ranged from 0 to +11D in the study assessing hyperopia (12.50%). Both myopia and hyperopia were assessed in 25% of studies, where refractive error ranged from ≤ −6D to ≥+6D. Most participants were identified has having no pathologies secondary to myopia including retinal detachment, retinopathy, or any other ocular disease (62.50%). The remaining participants were predominantly characterized as either having some retinal degeneration (12.50%), reduced vision (12.50%), or posterior vitreous detachment (12.50%).

Study Characteristics
Study characteristics are summarized in Table 6. The design of the included studies was either between group (37.50%), correlational (25%), within group (12.50%), or mixed (37.50%). A total of 522 participants were identified across all included studies, and where the information was provided, an approximately proportional number of female and male participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included studies. Fifty percent of studies reported age of participants in M(SD), ranging from 7.1(4.4) to 26.9(2.4) years. In the remaining 50% of the studies not reporting M(SD), age ranged from 10 to 50 years. Fifty percent of the studies assessed gfERGs in younger populations (10-23 years), while the remaining assessed older populations (7-50 years). One study did not report maximum age of the participants [43]. In total, 37.50% of studies reported that the controls were age-matched to the myopic/hyperopic participants.
Myopia was assessed in 62.50% of the studies, with refractive state ranging from +0.5D to −27D, meanwhile refractive error ranged from 0 to +11D in the study assessing hyperopia (12.50%). Both myopia and hyperopia were assessed in 25% of studies, where refractive error ranged from ≤ −6D to ≥+6D. Most participants were identified has having no pathologies secondary to myopia including retinal detachment, retinopathy, or any other ocular disease (62.50%). The remaining participants were predominantly characterized as either having some retinal degeneration (12.50%), reduced vision (12.50%), or posterior vitreous detachment (12.50%).

Study Characteristics
Study characteristics are summarized in Table 6. The design of the included studies was either between group (37.50%), correlational (25%), within group (12.50%), or mixed (37.50%). A total of 522 participants were identified across all included studies, and where the information was provided, an approximately proportional number of female and male participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included studies. Fifty percent of studies reported age of participants in M(SD), ranging from 7.1(4.4) to 26.9(2.4) years. In the remaining 50% of the studies not reporting M(SD), age ranged from 10 to 50 years. Fifty percent of the studies assessed gfERGs in younger populations (10-23 years), while the remaining assessed older populations (7-50 years). One study did not report maximum age of the participants [43]. In total, 37.50% of studies reported that the controls were age-matched to the myopic/hyperopic participants.
Myopia was assessed in 62.50% of the studies, with refractive state ranging from +0.5D to −27D, meanwhile refractive error ranged from 0 to +11D in the study assessing hyperopia (12.50%). Both myopia and hyperopia were assessed in 25% of studies, where refractive error ranged from ≤ −6D to ≥+6D. Most participants were identified has having no pathologies secondary to myopia including retinal detachment, retinopathy, or any other ocular disease (62.50%). The remaining participants were predominantly characterized as either having some retinal degeneration (12.50%), reduced vision (12.50%), or posterior vitreous detachment (12.50%).

Study Characteristics
Study characteristics are summarized in Table 6. The design of the included studies was either between group (37.50%), correlational (25%), within group (12.50%), or mixed (37.50%). A total of 522 participants were identified across all included studies, and where the information was provided, an approximately proportional number of female and male participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included studies. Fifty percent of studies reported age of participants in M(SD), ranging from 7.1(4.4) to 26.9(2.4) years. In the remaining 50% of the studies not reporting M(SD), age ranged from 10 to 50 years. Fifty percent of the studies assessed gfERGs in younger populations (10-23 years), while the remaining assessed older populations (7-50 years). One study did not report maximum age of the participants [43]. In total, 37.50% of studies reported that the controls were age-matched to the myopic/hyperopic participants.
Myopia was assessed in 62.50% of the studies, with refractive state ranging from +0.5D to −27D, meanwhile refractive error ranged from 0 to +11D in the study assessing hyperopia (12.50%). Both myopia and hyperopia were assessed in 25% of studies, where refractive error ranged from ≤ −6D to ≥+6D. Most participants were identified has having no pathologies secondary to myopia including retinal detachment, retinopathy, or any other ocular disease (62.50%). The remaining participants were predominantly characterized as either having some retinal degeneration (12.50%), reduced vision (12.50%), or posterior vitreous detachment (12.50%).

Study Characteristics
Study characteristics are summarized was either between group (37.50%), correl (37.50%). A total of 522 participants were id the information was provided, an approxim participants were included (37.50%).
Participant age ranged from 7 to 50 ye studies reported age of participants in M( the remaining 50% of the studies not rep Fifty percent of the studies assessed gfERG the remaining assessed older populations mum age of the participants [43]. In total, 3 age-matched to the myopic/hyperopic par Myopia was assessed in 62.50% of t +0.5D to −27D, meanwhile refractive error hyperopia (12.50%). Both myopia and hyp refractive error ranged from ≤ −6D to ≥+ ing no pathologies secondary to myopia in other ocular disease (62.50%). The remaini ized as either having some retinal degener terior vitreous detachment (12.50%).

Study Characteristics
Study characteristics are s was either between group (37. (37.50%). A total of 522 particip the information was provided, participants were included (37 Participant age ranged fro studies reported age of partici the remaining 50% of the stud Fifty percent of the studies ass the remaining assessed older mum age of the participants [43 age-matched to the myopic/hy Myopia was assessed in +0.5D to −27D, meanwhile refr hyperopia (12.50%). Both myo refractive error ranged from ≤ ing no pathologies secondary t other ocular disease (62.50%). T ized as either having some reti terior vitreous detachment (12  Perlman et al., 1984 [42] gether minimize the potential effect on internal validity.  [50] an et al., 1984 [42] idanandam et al., 2017 [52] et al., 2020 [40] all et al., 2001 [41] amoto et al., 1997 [51] definitely low probably low probably high definitely high.

Study Characteristics
Study characteristics are summarized in Table 6. The design of the included studies was either between group (37.50%), correlational (25%), within group (12.50%), or mixed (37.50%). A total of 522 participants were identified across all included studies, and where the information was provided, an approximately proportional number of female and male participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included studies. Fifty percent of studies reported age of participants in M(SD), ranging from 7.1(4.4) to 26.9(2.4) years. In the remaining 50% of the studies not reporting M(SD), age ranged from 10 to 50 years. Fifty percent of the studies assessed gfERGs in younger populations (10-23 years), while the remaining assessed older populations (7-50 years). One study did not report maximum age of the participants [43]. In total, 37.50% of studies reported that the controls were age-matched to the myopic/hyperopic participants.
Myopia was assessed in 62.50% of the studies, with refractive state ranging from +0.5D to −27D, meanwhile refractive error ranged from 0 to +11D in the study assessing hyperopia (12.50%). Both myopia and hyperopia were assessed in 25% of studies, where refractive error ranged from ≤ −6D to ≥+6D. Most participants were identified has having no pathologies secondary to myopia including retinal detachment, retinopathy, or any other ocular disease (62.50%). The remaining participants were predominantly characterized as either having some retinal degeneration (12.50%), reduced vision (12.50%), or posterior vitreous detachment (12.50%).
gether minimize the potential effect on internal validity.

Study Characteristics
Study characteristics are summarized in Table 6. The design of the inclu was either between group (37.50%), correlational (25%), within group (12.50% (37.50%). A total of 522 participants were identified across all included studies the information was provided, an approximately proportional number of fem participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included studies. Fift studies reported age of participants in M(SD), ranging from 7.1(4.4) to 26.9(2 the remaining 50% of the studies not reporting M(SD), age ranged from 10 Fifty percent of the studies assessed gfERGs in younger populations (10-23 y the remaining assessed older populations (7-50 years). One study did not r mum age of the participants [43]. In total, 37.50% of studies reported that the c age-matched to the myopic/hyperopic participants.
Myopia was assessed in 62.50% of the studies, with refractive state ra +0.5D to −27D, meanwhile refractive error ranged from 0 to +11D in the stud hyperopia (12.50%). Both myopia and hyperopia were assessed in 25% of stu refractive error ranged from ≤ −6D to ≥+6D. Most participants were identif ing no pathologies secondary to myopia including retinal detachment, retinop other ocular disease (62.50%). The remaining participants were predominant ized as either having some retinal degeneration (12.50%), reduced vision (12.5 terior vitreous detachment (12.50%).
gether minimize the potential effect on internal validity.

Study Characteristics
Study characteristics are summarized in Table 6. The design of the included studies was either between group (37.50%), correlational (25%), within group (12.50%), or mixed (37.50%). A total of 522 participants were identified across all included studies, and where the information was provided, an approximately proportional number of female and male participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included studies. Fifty percent of studies reported age of participants in M(SD), ranging from 7.1(4.4) to 26.9(2.4) years. In the remaining 50% of the studies not reporting M(SD), age ranged from 10 to 50 years. Fifty percent of the studies assessed gfERGs in younger populations (10-23 years), while the remaining assessed older populations (7-50 years). One study did not report maximum age of the participants [43]. In total, 37.50% of studies reported that the controls were age-matched to the myopic/hyperopic participants.
Myopia was assessed in 62.50% of the studies, with refractive state ranging from +0.5D to −27D, meanwhile refractive error ranged from 0 to +11D in the study assessing hyperopia (12.50%). Both myopia and hyperopia were assessed in 25% of studies, where refractive error ranged from ≤ −6D to ≥+6D. Most participants were identified has having no pathologies secondary to myopia including retinal detachment, retinopathy, or any other ocular disease (62.50%). The remaining participants were predominantly characterized as either having some retinal degeneration (12.50%), reduced vision (12.50%), or posterior vitreous detachment (12.50%).
gether minimize the potential effect on internal validity.

Study Characteristics
Study characteristics are summarized in Table 6. The design of the included stud was either between group (37.50%), correlational (25%), within group (12.50%), or mix (37.50%). A total of 522 participants were identified across all included studies, and wh the information was provided, an approximately proportional number of female and m participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included studies. Fifty percen studies reported age of participants in M(SD), ranging from 7.1(4.4) to 26.9(2.4) years the remaining 50% of the studies not reporting M(SD), age ranged from 10 to 50 yea Fifty percent of the studies assessed gfERGs in younger populations (10-23 years), wh the remaining assessed older populations (7-50 years). One study did not report ma mum age of the participants [43]. In total, 37.50% of studies reported that the controls w age-matched to the myopic/hyperopic participants.
Myopia was assessed in 62.50% of the studies, with refractive state ranging fr +0.5D to −27D, meanwhile refractive error ranged from 0 to +11D in the study assess hyperopia (12.50%). Both myopia and hyperopia were assessed in 25% of studies, wh refractive error ranged from ≤ −6D to ≥+6D. Most participants were identified has h ing no pathologies secondary to myopia including retinal detachment, retinopathy, or a other ocular disease (62.50%). The remaining participants were predominantly charact ized as either having some retinal degeneration (12.50%), reduced vision (12.50%), or p terior vitreous detachment (12.50%).
gether minimize the potential effect on int

Study Characteristics
Study characteristics are summarized was either between group (37.50%), correl (37.50%). A total of 522 participants were id the information was provided, an approxim participants were included (37.50%).
Participant age ranged from 7 to 50 ye studies reported age of participants in M( the remaining 50% of the studies not rep Fifty percent of the studies assessed gfERG the remaining assessed older populations mum age of the participants [43]. In total, 3 age-matched to the myopic/hyperopic par Myopia was assessed in 62.50% of t +0.5D to −27D, meanwhile refractive error hyperopia (12.50%). Both myopia and hyp refractive error ranged from ≤ −6D to ≥+ ing no pathologies secondary to myopia in other ocular disease (62.50%). The remaini ized as either having some retinal degener terior vitreous detachment (12.50%).
gether minimize the potential   Sachidanandam et al., 2017 [52] ocular biometrics, and the gfERG is an objective measure of retinal function, which together minimize the potential effect on internal validity.

Study Characteristics
Study characteristics are summarized in Table 6. The design of the included studies was either between group (37.50%), correlational (25%), within group (12.50%), or mixed (37.50%). A total of 522 participants were identified across all included studies, and where the information was provided, an approximately proportional number of female and male participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included studies. Fifty percent of studies reported age of participants in M(SD), ranging from 7.1(4.4) to 26.9(2.4) years. In the remaining 50% of the studies not reporting M(SD), age ranged from 10 to 50 years. Fifty percent of the studies assessed gfERGs in younger populations (10-23 years), while the remaining assessed older populations (7-50 years). One study did not report maximum age of the participants [43]. In total, 37.50% of studies reported that the controls were age-matched to the myopic/hyperopic participants.
Myopia was assessed in 62.50% of the studies, with refractive state ranging from +0.5D to −27D, meanwhile refractive error ranged from 0 to +11D in the study assessing hyperopia (12.50%). Both myopia and hyperopia were assessed in 25% of studies, where refractive error ranged from ≤ −6D to ≥+6D. Most participants were identified has having no pathologies secondary to myopia including retinal detachment, retinopathy, or any other ocular disease (62.50%). The remaining participants were predominantly characterized as either having some retinal degeneration (12.50%), reduced vision (12.50%), or posterior vitreous detachment (12.50%). ocular biometrics, and the gfERG is an objective measure of retinal functio gether minimize the potential effect on internal validity.

Study Characteristics
Study characteristics are summarized in Table 6. The design of the inclu was either between group (37.50%), correlational (25%), within group (12.50% (37.50%). A total of 522 participants were identified across all included studies the information was provided, an approximately proportional number of fem participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included studies. Fift studies reported age of participants in M(SD), ranging from 7.1(4.4) to 26.9(2 the remaining 50% of the studies not reporting M(SD), age ranged from 10 Fifty percent of the studies assessed gfERGs in younger populations (10-23 y the remaining assessed older populations (7-50 years). One study did not r mum age of the participants [43]. In total, 37.50% of studies reported that the c age-matched to the myopic/hyperopic participants.
Myopia was assessed in 62.50% of the studies, with refractive state ra +0.5D to −27D, meanwhile refractive error ranged from 0 to +11D in the stud hyperopia (12.50%). Both myopia and hyperopia were assessed in 25% of stu refractive error ranged from ≤ −6D to ≥+6D. Most participants were identif ing no pathologies secondary to myopia including retinal detachment, retinop other ocular disease (62.50%). The remaining participants were predominant ized as either having some retinal degeneration (12.50%), reduced vision (12.5 terior vitreous detachment (12.50%). ocular biometrics, and the gfERG is an objective measure of retinal function, which together minimize the potential effect on internal validity.

Study Characteristics
Study characteristics are summarized in Table 6. The design of the included studies was either between group (37.50%), correlational (25%), within group (12.50%), or mixed (37.50%). A total of 522 participants were identified across all included studies, and where the information was provided, an approximately proportional number of female and male participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included studies. Fifty percent of studies reported age of participants in M(SD), ranging from 7.1(4.4) to 26.9(2.4) years. In the remaining 50% of the studies not reporting M(SD), age ranged from 10 to 50 years. Fifty percent of the studies assessed gfERGs in younger populations (10-23 years), while the remaining assessed older populations (7-50 years). One study did not report maximum age of the participants [43]. In total, 37.50% of studies reported that the controls were age-matched to the myopic/hyperopic participants.
Myopia was assessed in 62.50% of the studies, with refractive state ranging from +0.5D to −27D, meanwhile refractive error ranged from 0 to +11D in the study assessing hyperopia (12.50%). Both myopia and hyperopia were assessed in 25% of studies, where refractive error ranged from ≤ −6D to ≥+6D. Most participants were identified has having no pathologies secondary to myopia including retinal detachment, retinopathy, or any other ocular disease (62.50%). The remaining participants were predominantly characterized as either having some retinal degeneration (12.50%), reduced vision (12.50%), or posterior vitreous detachment (12.50%). ocular biometrics, and the gfERG is an objective measure of retinal f gether minimize the potential effect on internal validity.

Study Characteristics
Study characteristics are summarized in Table 6. The design of th was either between group (37.50%), correlational (25%), within group (37.50%). A total of 522 participants were identified across all included the information was provided, an approximately proportional number participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included stud studies reported age of participants in M(SD), ranging from 7.1(4.4) to the remaining 50% of the studies not reporting M(SD), age ranged fr Fifty percent of the studies assessed gfERGs in younger populations ( the remaining assessed older populations (7-50 years). One study di mum age of the participants [43]. In total, 37.50% of studies reported tha age-matched to the myopic/hyperopic participants.
Myopia was assessed in 62.50% of the studies, with refractive +0.5D to −27D, meanwhile refractive error ranged from 0 to +11D in t hyperopia (12.50%). Both myopia and hyperopia were assessed in 25% refractive error ranged from ≤ −6D to ≥+6D. Most participants were ing no pathologies secondary to myopia including retinal detachment, other ocular disease (62.50%). The remaining participants were predom ized as either having some retinal degeneration (12.50%), reduced visio terior vitreous detachment (12.50%). ocular biometrics, and the gfERG is an ob gether minimize the potential effect on int

Study Characteristics
Study characteristics are summarized was either between group (37.50%), correl (37.50%). A total of 522 participants were id the information was provided, an approxim participants were included (37.50%).
Participant age ranged from 7 to 50 ye studies reported age of participants in M( the remaining 50% of the studies not rep Fifty percent of the studies assessed gfERG the remaining assessed older populations mum age of the participants [43]. In total, 3 age-matched to the myopic/hyperopic par Myopia was assessed in 62.50% of t +0.5D to −27D, meanwhile refractive error hyperopia (12.50%). Both myopia and hyp refractive error ranged from ≤ −6D to ≥+ ing no pathologies secondary to myopia in other ocular disease (62.50%). The remaini ized as either having some retinal degener terior vitreous detachment (12.50%). ocular biometrics, and the gfE gether minimize the potential

Study Characteristics
Study characteristics are s was either between group (37. (37.50%). A total of 522 particip the information was provided, participants were included (37 Participant age ranged fro studies reported age of partici the remaining 50% of the stud Fifty percent of the studies ass the remaining assessed older mum age of the participants [43 age-matched to the myopic/hy Myopia was assessed in +0.5D to −27D, meanwhile refr hyperopia (12.50%). Both myo refractive error ranged from ≤ ing no pathologies secondary t other ocular disease (62.50%). T ized as either having some reti terior vitreous detachment (12 ocular biometrics gether minimize t Although these factors increase the risk of bias, refractive error is an objective measure of ocular biometrics, and the gfERG is an objective measure of retinal function, which together minimize the potential effect on internal validity.

Study Characteristics
Study characteristics are summarized in Table 6. The design of the included studies was either between group (37.50%), correlational (25%), within group (12.50%), or mixed (37.50%). A total of 522 participants were identified across all included studies, and where the information was provided, an approximately proportional number of female and male participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included studies. Fifty percent of studies reported age of participants in M(SD), ranging from 7.1(4.4) to 26.9(2.4) years. In the remaining 50% of the studies not reporting M(SD), age ranged from 10 to 50 years. Fifty percent of the studies assessed gfERGs in younger populations (10-23 years), while the remaining assessed older populations (7-50 years). One study did not report maximum age of the participants [43]. In total, 37.50% of studies reported that the controls were age-matched to the myopic/hyperopic participants.
Myopia was assessed in 62.50% of the studies, with refractive state ranging from +0.5D to −27D, meanwhile refractive error ranged from 0 to +11D in the study assessing hyperopia (12.50%). Both myopia and hyperopia were assessed in 25% of studies, where refractive error ranged from ≤ −6D to ≥+6D. Most participants were identified has having no pathologies secondary to myopia including retinal detachment, retinopathy, or any other ocular disease (62.50%). The remaining participants were predominantly characterized as either having some retinal degeneration (12.50%), reduced vision (12.50%), or posterior vitreous detachment (12.50%).
Although these factors increase the risk of bias, refractive error is an objective ocular biometrics, and the gfERG is an objective measure of retinal functio gether minimize the potential effect on internal validity.

Study Characteristics
Study characteristics are summarized in Table 6. The design of the inclu was either between group (37.50%), correlational (25%), within group (12.50% (37.50%). A total of 522 participants were identified across all included studies the information was provided, an approximately proportional number of fem participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included studies. Fift studies reported age of participants in M(SD), ranging from 7.1(4.4) to 26.9(2 the remaining 50% of the studies not reporting M(SD), age ranged from 10 Fifty percent of the studies assessed gfERGs in younger populations (10-23 y the remaining assessed older populations (7-50 years). One study did not r mum age of the participants [43]. In total, 37.50% of studies reported that the c age-matched to the myopic/hyperopic participants.
Myopia was assessed in 62.50% of the studies, with refractive state ra +0.5D to −27D, meanwhile refractive error ranged from 0 to +11D in the stud hyperopia (12.50%). Both myopia and hyperopia were assessed in 25% of stu refractive error ranged from ≤ −6D to ≥+6D. Most participants were identif ing no pathologies secondary to myopia including retinal detachment, retinop other ocular disease (62.50%). The remaining participants were predominant ized as either having some retinal degeneration (12.50%), reduced vision (12.5 terior vitreous detachment (12.50%).
Although these factors increase the risk of bias, refractive error is an objective measure of ocular biometrics, and the gfERG is an objective measure of retinal function, which together minimize the potential effect on internal validity.

Study Characteristics
Study characteristics are summarized in Table 6. The design of the included studies was either between group (37.50%), correlational (25%), within group (12.50%), or mixed (37.50%). A total of 522 participants were identified across all included studies, and where the information was provided, an approximately proportional number of female and male participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included studies. Fifty percent of studies reported age of participants in M(SD), ranging from 7.1(4.4) to 26.9(2.4) years. In the remaining 50% of the studies not reporting M(SD), age ranged from 10 to 50 years. Fifty percent of the studies assessed gfERGs in younger populations (10-23 years), while the remaining assessed older populations (7-50 years). One study did not report maximum age of the participants [43]. In total, 37.50% of studies reported that the controls were age-matched to the myopic/hyperopic participants.
Myopia was assessed in 62.50% of the studies, with refractive state ranging from +0.5D to −27D, meanwhile refractive error ranged from 0 to +11D in the study assessing hyperopia (12.50%). Both myopia and hyperopia were assessed in 25% of studies, where refractive error ranged from ≤ −6D to ≥+6D. Most participants were identified has having no pathologies secondary to myopia including retinal detachment, retinopathy, or any other ocular disease (62.50%). The remaining participants were predominantly characterized as either having some retinal degeneration (12.50%), reduced vision (12.50%), or posterior vitreous detachment (12.50%).
Although these factors increase the risk of bias, refractive error is an ob ocular biometrics, and the gfERG is an objective measure of retinal f gether minimize the potential effect on internal validity.

Study Characteristics
Study characteristics are summarized in Table 6. The design of th was either between group (37.50%), correlational (25%), within group (37.50%). A total of 522 participants were identified across all included the information was provided, an approximately proportional number participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included stud studies reported age of participants in M(SD), ranging from 7.1(4.4) to the remaining 50% of the studies not reporting M(SD), age ranged fr Fifty percent of the studies assessed gfERGs in younger populations ( the remaining assessed older populations (7-50 years). One study di mum age of the participants [43]. In total, 37.50% of studies reported tha age-matched to the myopic/hyperopic participants.
Although these factors increase the risk of ocular biometrics, and the gfERG is an ob gether minimize the potential effect on int

Study Characteristics
Study characteristics are summarized was either between group (37.50%), correl (37.50%). A total of 522 participants were id the information was provided, an approxim participants were included (37.50%).
Participant age ranged from 7 to 50 ye studies reported age of participants in M( the remaining 50% of the studies not rep Fifty percent of the studies assessed gfERG the remaining assessed older populations mum age of the participants [43]. In total, 3 age-matched to the myopic/hyperopic par Myopia was assessed in 62.50% of t +0.5D to −27D, meanwhile refractive error hyperopia (12.50%). Both myopia and hyp refractive error ranged from ≤ −6D to ≥+ ing no pathologies secondary to myopia in other ocular disease (62.50%). The remaini ized as either having some retinal degener terior vitreous detachment (12.50%).
Although these factors increas ocular biometrics, and the gfE gether minimize the potential

Study Characteristics
Study characteristics are s was either between group (37. (37.50%). A total of 522 particip the information was provided, participants were included (37 Participant age ranged fro studies reported age of partici the remaining 50% of the stud Fifty percent of the studies ass the remaining assessed older mum age of the participants [43 age-matched to the myopic/hy Myopia was assessed in +0.5D to −27D, meanwhile refr hyperopia (12.50%). Both myo refractive error ranged from ≤ ing no pathologies secondary t other ocular disease (62.50%). T ized as either having some reti terior vitreous detachment (12 Although these fa ocular biometrics gether minimize t  Westall et al., 2001 [41] tions (i.e., referred patients); therefore, blinding to study groups could not be achieved. Although these factors increase the risk of bias, refractive error is an objective measure of ocular biometrics, and the gfERG is an objective measure of retinal function, which together minimize the potential effect on internal validity.  [50] an et al., 1984 [42] idanandam et al., 2017 [52] et al., 2020 [40] all et al., 2001 [41] amoto et al., 1997 [51] definitely low probably low probably high definitely high.

Study Characteristics
Study characteristics are summarized in Table 6. The design of the included studies was either between group (37.50%), correlational (25%), within group (12.50%), or mixed (37.50%). A total of 522 participants were identified across all included studies, and where the information was provided, an approximately proportional number of female and male participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included studies. Fifty percent of studies reported age of participants in M(SD), ranging from 7.1(4.4) to 26.9(2.4) years. In the remaining 50% of the studies not reporting M(SD), age ranged from 10 to 50 years. Fifty percent of the studies assessed gfERGs in younger populations (10-23 years), while the remaining assessed older populations (7-50 years). One study did not report maximum age of the participants [43]. In total, 37.50% of studies reported that the controls were age-matched to the myopic/hyperopic participants.
Myopia was assessed in 62.50% of the studies, with refractive state ranging from +0.5D to −27D, meanwhile refractive error ranged from 0 to +11D in the study assessing hyperopia (12.50%). Both myopia and hyperopia were assessed in 25% of studies, where refractive error ranged from ≤ −6D to ≥+6D. Most participants were identified has having no pathologies secondary to myopia including retinal detachment, retinopathy, or any other ocular disease (62.50%). The remaining participants were predominantly characterized as either having some retinal degeneration (12.50%), reduced vision (12.50%), or posterior vitreous detachment (12.50%). tions (i.e., referred patients); therefore, blinding to study groups could not b Although these factors increase the risk of bias, refractive error is an objective ocular biometrics, and the gfERG is an objective measure of retinal functio gether minimize the potential effect on internal validity.

Study Characteristics
Study characteristics are summarized in Table 6. The design of the inclu was either between group (37.50%), correlational (25%), within group (12.50% (37.50%). A total of 522 participants were identified across all included studies the information was provided, an approximately proportional number of fem participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included studies. Fift studies reported age of participants in M(SD), ranging from 7.1(4.4) to 26.9(2 the remaining 50% of the studies not reporting M(SD), age ranged from 10 Fifty percent of the studies assessed gfERGs in younger populations (10-23 y the remaining assessed older populations (7-50 years). One study did not r mum age of the participants [43]. In total, 37.50% of studies reported that the c age-matched to the myopic/hyperopic participants.
Myopia was assessed in 62.50% of the studies, with refractive state ra +0.5D to −27D, meanwhile refractive error ranged from 0 to +11D in the stud hyperopia (12.50%). Both myopia and hyperopia were assessed in 25% of stu refractive error ranged from ≤ −6D to ≥+6D. Most participants were identif ing no pathologies secondary to myopia including retinal detachment, retinop other ocular disease (62.50%). The remaining participants were predominant ized as either having some retinal degeneration (12.50%), reduced vision (12.5 terior vitreous detachment (12.50%). tions (i.e., referred patients); therefore, blinding to study groups could not be achieved. Although these factors increase the risk of bias, refractive error is an objective measure of ocular biometrics, and the gfERG is an objective measure of retinal function, which together minimize the potential effect on internal validity.

Study Characteristics
Study characteristics are summarized in Table 6. The design of the included studies was either between group (37.50%), correlational (25%), within group (12.50%), or mixed (37.50%). A total of 522 participants were identified across all included studies, and where the information was provided, an approximately proportional number of female and male participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included studies. Fifty percent of studies reported age of participants in M(SD), ranging from 7.1(4.4) to 26.9(2.4) years. In the remaining 50% of the studies not reporting M(SD), age ranged from 10 to 50 years. Fifty percent of the studies assessed gfERGs in younger populations (10-23 years), while the remaining assessed older populations (7-50 years). One study did not report maximum age of the participants [43]. In total, 37.50% of studies reported that the controls were age-matched to the myopic/hyperopic participants.
Myopia was assessed in 62.50% of the studies, with refractive state ranging from +0.5D to −27D, meanwhile refractive error ranged from 0 to +11D in the study assessing hyperopia (12.50%). Both myopia and hyperopia were assessed in 25% of studies, where refractive error ranged from ≤ −6D to ≥+6D. Most participants were identified has having no pathologies secondary to myopia including retinal detachment, retinopathy, or any other ocular disease (62.50%). The remaining participants were predominantly characterized as either having some retinal degeneration (12.50%), reduced vision (12.50%), or posterior vitreous detachment (12.50%). tions (i.e., referred patients); therefore, blinding to study groups could not be achieved. Although these factors increase the risk of bias, refractive error is an objective measure of ocular biometrics, and the gfERG is an objective measure of retinal function, which to-gether minimize the potential effect on internal validity.

Study Characteristics
Study characteristics are summarized in Table 6. The design of the included studies was either between group (37.50%), correlational (25%), within group (12.50%), or mixed (37.50%). A total of 522 participants were identified across all included studies, and where the information was provided, an approximately proportional number of female and male participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included studies. Fifty percent of studies reported age of participants in M(SD), ranging from 7.1(4.4) to 26.9(2.4) years. In the remaining 50% of the studies not reporting M(SD), age ranged from 10 to 50 years. Fifty percent of the studies assessed gfERGs in younger populations (10-23 years), while the remaining assessed older populations (7-50 years). One study did not report maxi-mum age of the participants [43]. In total, 37.50% of studies reported that the controls were age-matched to the myopic/hyperopic participants.
Myopia was assessed in 62.50% of the studies, with refractive state ranging from +0.5D to −27D, meanwhile refractive error ranged from 0 to +11D in the study assessing hyperopia (12.50%). Both myopia and hyperopia were assessed in 25% of studies, where refractive error ranged from ≤ −6D to ≥+6D. Most participants were identified has hav-ing no pathologies secondary to myopia including retinal detachment, retinopathy, or any other ocular disease (62.50%). The remaining participants were predominantly character-ized as either having some retinal degeneration (12.50%), reduced vision (12.50%), or pos-terior vitreous detachment (12.50%). tions (i.e., referred patients); therefore, bli Although these factors increase the risk of ocular biometrics, and the gfERG is an ob gether minimize the potential effect on int

Study Characteristics
Study characteristics are summarized was either between group (37.50%), correl (37.50%). A total of 522 participants were id the information was provided, an approxim participants were included (37.50%).
Participant age ranged from 7 to 50 ye studies reported age of participants in M( the remaining 50% of the studies not rep Fifty percent of the studies assessed gfERG the remaining assessed older populations mum age of the participants [43]. In total, 3 age-matched to the myopic/hyperopic par Myopia was assessed in 62.50% of t +0.5D to −27D, meanwhile refractive error hyperopia (12.50%). Both myopia and hyp refractive error ranged from ≤ −6D to ≥+ ing no pathologies secondary to myopia in other ocular disease (62.50%). The remaini ized as either having some retinal degener terior vitreous detachment (12.50%). tions (i.e., referred patients); th Although these factors increas ocular biometrics, and the gfE gether minimize the potential

Study Characteristics
Study characteristics are s was either between group (37. (37.50%). A total of 522 particip the information was provided, participants were included (37 Participant age ranged fro studies reported age of partici the remaining 50% of the stud Fifty percent of the studies ass the remaining assessed older mum age of the participants [43 age-matched to the myopic/hy Myopia was assessed in +0.5D to −27D, meanwhile refr hyperopia (12.50%). Both myo refractive error ranged from ≤ ing no pathologies secondary t other ocular disease (62.50%). T ized as either having some reti terior vitreous detachment (12 tions (i.e., referre Although these fa ocular biometrics gether minimize t

Study Charact
Study charac was either betwee (37.50%). A total o the information w participants were Participant a studies reported the remaining 50 Fifty percent of th the remaining as mum age of the p age-matched to th Myopia was +0.5D to −27D, m hyperopia (12.50% refractive error ra ing no pathologie other ocular disea ized as either hav terior vitreous de Yamamoto et al., 1997 [51] subjects blinded to study group during study?". These studies recruited clinical populations (i.e., referred patients); therefore, blinding to study groups could not be achieved. Although these factors increase the risk of bias, refractive error is an objective measure of ocular biometrics, and the gfERG is an objective measure of retinal function, which together minimize the potential effect on internal validity.

Study Characteristics
Study characteristics are summarized in Table 6. The design of the included studies was either between group (37.50%), correlational (25%), within group (12.50%), or mixed (37.50%). A total of 522 participants were identified across all included studies, and where the information was provided, an approximately proportional number of female and male participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included studies. Fifty percent of studies reported age of participants in M(SD), ranging from 7.1(4.4) to 26.9(2.4) years. In the remaining 50% of the studies not reporting M(SD), age ranged from 10 to 50 years. Fifty percent of the studies assessed gfERGs in younger populations (10-23 years), while the remaining assessed older populations (7-50 years). One study did not report maximum age of the participants [43]. In total, 37.50% of studies reported that the controls were age-matched to the myopic/hyperopic participants.
Myopia was assessed in 62.50% of the studies, with refractive state ranging from +0.5D to −27D, meanwhile refractive error ranged from 0 to +11D in the study assessing hyperopia (12.50%). Both myopia and hyperopia were assessed in 25% of studies, where refractive error ranged from ≤ −6D to ≥+6D. Most participants were identified has having no pathologies secondary to myopia including retinal detachment, retinopathy, or any other ocular disease (62.50%). The remaining participants were predominantly characterized as either having some retinal degeneration (12.50%), reduced vision (12.50%), or posterior vitreous detachment (12.50%).
subjects blinded to study group during study?". These studies recruited clin tions (i.e., referred patients); therefore, blinding to study groups could not b Although these factors increase the risk of bias, refractive error is an objective ocular biometrics, and the gfERG is an objective measure of retinal functio gether minimize the potential effect on internal validity.

Study Characteristics
Study characteristics are summarized in Table 6. The design of the inclu was either between group (37.50%), correlational (25%), within group (12.50% (37.50%). A total of 522 participants were identified across all included studies the information was provided, an approximately proportional number of fem participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included studies. Fift studies reported age of participants in M(SD), ranging from 7.1(4.4) to 26.9(2 the remaining 50% of the studies not reporting M(SD), age ranged from 10 Fifty percent of the studies assessed gfERGs in younger populations (10-23 y the remaining assessed older populations (7-50 years). One study did not r mum age of the participants [43]. In total, 37.50% of studies reported that the c age-matched to the myopic/hyperopic participants.
Myopia was assessed in 62.50% of the studies, with refractive state ra +0.5D to −27D, meanwhile refractive error ranged from 0 to +11D in the stud hyperopia (12.50%). Both myopia and hyperopia were assessed in 25% of stu refractive error ranged from ≤ −6D to ≥+6D. Most participants were identif ing no pathologies secondary to myopia including retinal detachment, retinop other ocular disease (62.50%). The remaining participants were predominant ized as either having some retinal degeneration (12.50%), reduced vision (12.5 terior vitreous detachment (12.50%).
subjects blinded to study group during study?". These studies recruited clinical populations (i.e., referred patients); therefore, blinding to study groups could not be achieved. Although these factors increase the risk of bias, refractive error is an objective measure of ocular biometrics, and the gfERG is an objective measure of retinal function, which together minimize the potential effect on internal validity.

Study Characteristics
Study characteristics are summarized in Table 6. The design of the included studies was either between group (37.50%), correlational (25%), within group (12.50%), or mixed (37.50%). A total of 522 participants were identified across all included studies, and where the information was provided, an approximately proportional number of female and male participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included studies. Fifty percent of studies reported age of participants in M(SD), ranging from 7.1(4.4) to 26.9(2.4) years. In the remaining 50% of the studies not reporting M(SD), age ranged from 10 to 50 years. Fifty percent of the studies assessed gfERGs in younger populations (10-23 years), while the remaining assessed older populations (7-50 years). One study did not report maximum age of the participants [43]. In total, 37.50% of studies reported that the controls were age-matched to the myopic/hyperopic participants.
Myopia was assessed in 62.50% of the studies, with refractive state ranging from +0.5D to −27D, meanwhile refractive error ranged from 0 to +11D in the study assessing hyperopia (12.50%). Both myopia and hyperopia were assessed in 25% of studies, where refractive error ranged from ≤ −6D to ≥+6D. Most participants were identified has having no pathologies secondary to myopia including retinal detachment, retinopathy, or any other ocular disease (62.50%). The remaining participants were predominantly characterized as either having some retinal degeneration (12.50%), reduced vision (12.50%), or posterior vitreous detachment (12.50%).
subjects blinded to study group during study?". Thes tions (i.e., referred patients); therefore, blinding to st Although these factors increase the risk of bias, refrac ocular biometrics, and the gfERG is an objective mea gether minimize the potential effect on internal validi

Study Characteristics
Study characteristics are summarized in Table 6. was either between group (37.50%), correlational (25% (37.50%). A total of 522 participants were identified ac the information was provided, an approximately prop participants were included (37.50%).
Participant age ranged from 7 to 50 years across a studies reported age of participants in M(SD), rangin the remaining 50% of the studies not reporting M(SD Fifty percent of the studies assessed gfERGs in young the remaining assessed older populations (7-50 year mum age of the participants [43]. In total, 37.50% of stu age-matched to the myopic/hyperopic participants. Myopia was assessed in 62.50% of the studies, +0.5D to −27D, meanwhile refractive error ranged fro hyperopia (12.50%). Both myopia and hyperopia wer refractive error ranged from ≤ −6D to ≥+6D. Most p ing no pathologies secondary to myopia including reti other ocular disease (62.50%). The remaining participa ized as either having some retinal degeneration (12.50 terior vitreous detachment (12.50%).
subjects blinded to study group during st tions (i.e., referred patients); therefore, bli Although these factors increase the risk of ocular biometrics, and the gfERG is an ob gether minimize the potential effect on int

Study Characteristics
Study characteristics are summarized was either between group (37.50%), correl (37.50%). A total of 522 participants were id the information was provided, an approxim participants were included (37.50%).
Participant age ranged from 7 to 50 ye studies reported age of participants in M( the remaining 50% of the studies not rep Fifty percent of the studies assessed gfERG the remaining assessed older populations mum age of the participants [43]. In total, 3 age-matched to the myopic/hyperopic par Myopia was assessed in 62.50% of t +0.5D to −27D, meanwhile refractive error hyperopia (12.50%). Both myopia and hyp refractive error ranged from ≤ −6D to ≥+ ing no pathologies secondary to myopia in other ocular disease (62.50%). The remaini ized as either having some retinal degener terior vitreous detachment (12.50%).
subjects blinded to study grou tions (i.e., referred patients); th Although these factors increas ocular biometrics, and the gfE gether minimize the potential

Study Characteristics
Study characteristics are s was either between group (37. (37.50%). A total of 522 particip the information was provided, participants were included (37 Participant age ranged fro studies reported age of partici the remaining 50% of the stud Fifty percent of the studies ass the remaining assessed older mum age of the participants [43 age-matched to the myopic/hy Myopia was assessed in +0.5D to −27D, meanwhile refr hyperopia (12.50%). Both myo refractive error ranged from ≤ ing no pathologies secondary t other ocular disease (62.50%). T ized as either having some reti terior vitreous detachment (12 subjects blinded to study group during study?". tions (i.e., referred patients); therefore, blinding t Although these factors increase the risk of bias, re ocular biometrics, and the gfERG is an objective gether minimize the potential effect on internal va

Study Characteristics
Study characteristics are summarized in Tab was either between group (37.50%), correlational (37.50%). A total of 522 participants were identifie the information was provided, an approximately p participants were included (37.50%).
Participant age ranged from 7 to 50 years acr studies reported age of participants in M(SD), ran the remaining 50% of the studies not reporting M Fifty percent of the studies assessed gfERGs in yo the remaining assessed older populations (7-50 y mum age of the participants [43]. In total, 37.50% o age-matched to the myopic/hyperopic participant Myopia was assessed in 62.50% of the stud +0.5D to −27D, meanwhile refractive error ranged hyperopia (12.50%). Both myopia and hyperopia refractive error ranged from ≤ −6D to ≥+6D. Mo ing no pathologies secondary to myopia including other ocular disease (62.50%). The remaining part ized as either having some retinal degeneration (1 terior vitreous detachment (12.50%). from some participants for certain components of the ERG [41]. All included studies displayed a "probably high" risk of bias for the question 6 "Were the personnel and human subjects blinded to study group during study?". These studies recruited clinical populations (i.e., referred patients); therefore, blinding to study groups could not be achieved. Although these factors increase the risk of bias, refractive error is an objective measure of ocular biometrics, and the gfERG is an objective measure of retinal function, which together minimize the potential effect on internal validity.

Study Characteristics
Study characteristics are summarized in Table 6. The design of the included studies was either between group (37.50%), correlational (25%), within group (12.50%), or mixed (37.50%). A total of 522 participants were identified across all included studies, and where the information was provided, an approximately proportional number of female and male participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included studies. Fifty percent of studies reported age of participants in M(SD), ranging from 7.1(4.4) to 26.9(2.4) years. In the remaining 50% of the studies not reporting M(SD), age ranged from 10 to 50 years. Fifty percent of the studies assessed gfERGs in younger populations (10-23 years), while the remaining assessed older populations (7-50 years). One study did not report maximum age of the participants [43]. In total, 37.50% of studies reported that the controls were age-matched to the myopic/hyperopic participants.
Myopia was assessed in 62.50% of the studies, with refractive state ranging from +0.5D to −27D, meanwhile refractive error ranged from 0 to +11D in the study assessing hyperopia (12.50%). Both myopia and hyperopia were assessed in 25% of studies, where refractive error ranged from ≤ −6D to ≥+6D. Most participants were identified has having no pathologies secondary to myopia including retinal detachment, retinopathy, or any other ocular disease (62.50%). The remaining participants were predominantly characterized as either having some retinal degeneration (12.50%), reduced vision (12.50%), or posterior vitreous detachment (12.50%). definitely low from some participants for certain components of the ERG [41]. All included studies displayed a "probably high" risk of bias for the question 6 "Were the personnel and human subjects blinded to study group during study?". These studies recruited clinical populations (i.e., referred patients); therefore, blinding to study groups could not be achieved. Although these factors increase the risk of bias, refractive error is an objective measure of ocular biometrics, and the gfERG is an objective measure of retinal function, which together minimize the potential effect on internal validity.

Study Characteristics
Study characteristics are summarized in Table 6. The design of the included studies was either between group (37.50%), correlational (25%), within group (12.50%), or mixed (37.50%). A total of 522 participants were identified across all included studies, and where the information was provided, an approximately proportional number of female and male participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included studies. Fifty percent of studies reported age of participants in M(SD), ranging from 7.1(4.4) to 26.9(2.4) years. In the remaining 50% of the studies not reporting M(SD), age ranged from 10 to 50 years. Fifty percent of the studies assessed gfERGs in younger populations (10-23 years), while the remaining assessed older populations (7-50 years). One study did not report maximum age of the participants [43]. In total, 37.50% of studies reported that the controls were age-matched to the myopic/hyperopic participants.
Myopia was assessed in 62.50% of the studies, with refractive state ranging from +0.5D to −27D, meanwhile refractive error ranged from 0 to +11D in the study assessing hyperopia (12.50%). Both myopia and hyperopia were assessed in 25% of studies, where refractive error ranged from ≤ −6D to ≥+6D. Most participants were identified has having no pathologies secondary to myopia including retinal detachment, retinopathy, or any other ocular disease (62.50%). The remaining participants were predominantly characterized as either having some retinal degeneration (12.50%), reduced vision (12.50%), or posterior vitreous detachment (12.50%).
probably low from some participants for certain components of the ERG [41]. All included studies displayed a "probably high" risk of bias for the question 6 "Were the personnel and human subjects blinded to study group during study?". These studies recruited clinical populations (i.e., referred patients); therefore, blinding to study groups could not be achieved. Although these factors increase the risk of bias, refractive error is an objective measure of ocular biometrics, and the gfERG is an objective measure of retinal function, which together minimize the potential effect on internal validity.

Study Characteristics
Study characteristics are summarized in Table 6. The design of the included studies was either between group (37.50%), correlational (25%), within group (12.50%), or mixed (37.50%). A total of 522 participants were identified across all included studies, and where the information was provided, an approximately proportional number of female and male participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included studies. Fifty percent of studies reported age of participants in M(SD), ranging from 7.1(4.4) to 26.9(2.4) years. In the remaining 50% of the studies not reporting M(SD), age ranged from 10 to 50 years. Fifty percent of the studies assessed gfERGs in younger populations (10-23 years), while the remaining assessed older populations (7-50 years). One study did not report maximum age of the participants [43]. In total, 37.50% of studies reported that the controls were age-matched to the myopic/hyperopic participants.
Myopia was assessed in 62.50% of the studies, with refractive state ranging from +0.5D to −27D, meanwhile refractive error ranged from 0 to +11D in the study assessing hyperopia (12.50%). Both myopia and hyperopia were assessed in 25% of studies, where refractive error ranged from ≤ −6D to ≥+6D. Most participants were identified has having no pathologies secondary to myopia including retinal detachment, retinopathy, or any other ocular disease (62.50%). The remaining participants were predominantly characterized as either having some retinal degeneration (12.50%), reduced vision (12.50%), or posterior vitreous detachment (12.50%).
probably high from some participants for certain components of the ERG [41]. All included studies dis-played a "probably high" risk of bias for the question 6 "Were the personnel and human subjects blinded to study group during study?". These studies recruited clinical popula-tions (i.e., referred patients); therefore, blinding to study groups could not be achieved. Although these factors increase the risk of bias, refractive error is an objective measure of ocular biometrics, and the gfERG is an objective measure of retinal function, which to-gether minimize the potential effect on internal validity.

Study Characteristics
Study characteristics are summarized in Table 6. The design of the included studies was either between group (37.50%), correlational (25%), within group (12.50%), or mixed (37.50%). A total of 522 participants were identified across all included studies, and where the information was provided, an approximately proportional number of female and male participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included studies. Fifty percent of studies reported age of participants in M(SD), ranging from 7.1(4.4) to 26.9(2.4) years. In the remaining 50% of the studies not reporting M(SD), age ranged from 10 to 50 years. Fifty percent of the studies assessed gfERGs in younger populations (10-23 years), while the remaining assessed older populations (7-50 years). One study did not report maxi-mum age of the participants [43]. In total, 37.50% of studies reported that the controls were age-matched to the myopic/hyperopic participants.
Myopia was assessed in 62.50% of the studies, with refractive state ranging from +0.5D to −27D, meanwhile refractive error ranged from 0 to +11D in the study assessing hyperopia (12.50%). Both myopia and hyperopia were assessed in 25% of studies, where refractive error ranged from ≤ −6D to ≥+6D. Most participants were identified has hav-ing no pathologies secondary to myopia including retinal detachment, retinopathy, or any other ocular disease (62.50%). The remaining participants were predominantly character-ized as either having some retinal degeneration (12.50%), reduced vision (12.50%), or pos-terior vitreous detachment (12.50%). definitely high.

Study Characteristics
Study characteristics are summarized in Table 6. The design of the included studies was either between group (37.50%), correlational (25%), within group (12.50%), or mixed (37.50%). A total of 522 participants were identified across all included studies, and where the information was provided, an approximately proportional number of female and male participants were included (37.50%).
Participant age ranged from 7 to 50 years across all included studies. Fifty percent of studies reported age of participants in M(SD), ranging from 7.1(4.4) to 26.9(2.4) years. In the remaining 50% of the studies not reporting M(SD), age ranged from 10 to 50 years. Fifty percent of the studies assessed gfERGs in younger populations (10-23 years), while the remaining assessed older populations (7-50 years). One study did not report maximum age of the participants [43]. In total, 37.50% of studies reported that the controls were age-matched to the myopic/hyperopic participants.
Myopia was assessed in 62.50% of the studies, with refractive state ranging from +0.5D to −27D, meanwhile refractive error ranged from 0 to +11D in the study assessing hyperopia (12.50%). Both myopia and hyperopia were assessed in 25% of studies, where refractive error ranged from ≤−6D to ≥+6D. Most participants were identified has having no pathologies secondary to myopia including retinal detachment, retinopathy, or any other ocular disease (62.50%). The remaining participants were predominantly characterized as either having some retinal degeneration (12.50%), reduced vision (12.50%), or posterior vitreous detachment (12.50%).
Among the ERG stimulus types, 50% of studies reported employing International Society Clinical Electrophysiology of Vision (ISCEV) standards for photopic and scotopic flash in light and dark adapted conditions, respectively, while the remaining 50% of studies were published before ERG standard protocols for measurement were established [42,43,50,51]. These studies used customized photopic and scotopic stimuli for the ERG measurements. In 90% of the studies the participants were dark/light-adapted before ERG flash was delivered. International Society Clinical Electrophysiology of Vision protocols for ERG stimulus specifies ≥20 min for dark-adaptation and ≥10 min for light-adaptation [31]. Dark-adaptation duration across these studies ranged from 5 to 30 min. Light-adaptation duration across these studies ranged from 10 to 20 min. Participants were either dark-adapted (50%), both dark/light-adapted (37.50%), or neither (12.50%).  comparison groups. However, the reviewed studies identified b-wave attenuation more frequently than a-wave attenuation in myopia, and two studies reported an increased a/b ratio associated with high and pathological myopia [42,43], suggesting that transmission of the visual signal from photoreceptor to bipolar cells, or bipolar cell responses, may display proportionally more functional impairment as myopia progresses. Findings regarding implicit time and oscillatory potential components of the gfERG waveform were less frequently reported. Where information was available, implicit time between the onset of the response and maximum a-wave and b-wave amplitude did not differ between groups suggesting that the speed of signal transduction from the photoreceptors to the bipolar cells is not affected in myopia. Oscillatory potentials are generally thought to reflect the activity of amacrine and ganglion cells [35], though their exact origin remains unresolved. For studies reporting OP amplitudes, findings were mixed (with one study reporting an increase under scotopic conditions [40] and another a decrease under photopic conditions [41]).
In contrast to myopia, only three studies assessing gERGs in hyperopic participants were identified. Inconsistencies in the reporting and findings of these studies hindered synthesis and interpretation of the results. Hyperopia is a far less common clinical condition in adults than myopia and is not increasing in worldwide prevalence [55,56], which may account for the lack of clinical gfERG studies of hyperopes.
The gfERG findings in human in human myopes reviewed here build on electrophysiological, structural, and pharmacological evidence from animal models implicating the photoreceptors and ON-and OFF-bipolar cell pathways in the control of eye growth and the development of myopia [18][19][20]57]. The human studies reviewed here consistently concur with the findings of gfERG studies in chick models of early and established myopia identifying decreased a-wave and b-wave amplitudes [36][37][38][39]. Using a quantitative model of phototransduction, Westbrook et al. [39] derived the photoreceptor light response in myopic and emmetropic chick eyes from the leading edge of the a-wave. This model demonstrated that photoreceptors in myopic eyes are significantly more sensitive to lower intensity light stimulus than normally developing eyes. However, at higher intensities, the photoreceptor light response declines faster in myopic than in control eyes, suggesting an increase in negative feedback mechanisms. This reduction in photoreceptor response to intense light is consistent with ultrastructural studies demonstrating that rod outer segments are elongated in both occlusion and lens defocus models of myopia [58][59][60]. Such a reduction in outer segment phagocytosis indicates that a level of photoreceptor inactivity may be involved in both growth paradigms (as disc shedding follows a circadian rhythm stimulated by dark-light transition).
Pharmacological and gene knockout studies in animals have provided further evidence for photoreceptor and bipolar cell involvement in ocular growth regulation. A number of early such studies examining the effects of neurotoxic substances in animal myopia models have indicated that agents which disrupt the majority of amacrine or ganglion cell functioning alone (while they may alter anterior chamber depth) have little effect on vitreous chamber growth [23][24][25][26]. In contrast, agents that also affect bipolar or photoreceptor functioning do alter the rate of postnatal vitreous enlargement and visually induced ocular growth [21,23,24,27,28], and pharmacological or genetic disruption of the balance between bipolar cell ON-and OFF-pathways has been shown to directionally alter ocular growth [20,21,36,57,61].

Limitations
Several limitations were common across the included studies. Not unexpectedly, some early studies did not follow ISCEV standards for testing, as they were conducted before the standards were established [42,43,50,51]. Therefore, methods of gfERG testing varied in these studies considerably. Studies that followed the ISCEV protocol for gfERG stimuli often failed to adhere to the durations outlined for dark-adaptation and light-adaptation. The current ISCEV standard specifies a minimum of 20 min for dark-adaptation and 10 min for light-adaptation. Failure to adhere to the duration of adaptation means the rod systems isolated by dark-adaptation and cone systems isolated by light-adaptation may have been inadequately assessed [31]. In some cases, reporting of the results did not provide clear indications about the direction of change for the outcomes measured. Reporting of the outcomes was too general in these instances to determine which aspects of the ERG waveform were affected and to what degree this effect was detected [42,44,50]. Furthermore, the included studies covered a wide age range, and few studies reported whether the comparison groups were age-matched [40,41,51]. Thus age matching remains a crucial factor to consider, given that age-related changes in the ERG have long been recognized [62,63], and the stability of refractive state and likelihood of secondary associated pathology vary across the life span [64].

Future Directions
The review identified that additional studies that include assessment of hyperopes are required to obtain a greater understanding of relative changes in function associated with signed-directional growth. In order to examine the involvement of the inner retinal contribution to the development of refractive errors, it would be beneficial for future reviews examine PERG waveform characteristics in myopia and hyperopia. Further, it is recommended that future investigations using gfERG should more consistently implement standardized testing protocols through the development of clinical guidelines for assessment and reporting gfERG to enhance the utility of this measure for examination of outer retinal changes associated ametropias in addition to ocular pathologies.

Conclusions
The global flash electroretinogram waveform appears to be altered in the myopic eye with reductions in the a-and b-waves the most frequently reported characteristics. The implicit times of global flash electroretinogram amplitudes remain unchanged between comparison groups, while a/b ratio increases have been associated with high myopia where secondary pathology may be present. While limited to global flash electroretinograms, the current review findings for impaired a-wave and b-wave activity suggest that the function of the rod and cone photoreceptor response to light onset is perturbed, and hence that transmission of the visual signal to the bipolar cells is likely to be reduced in human myopia, consistent with animal model studies. The effects of hyperopia on the global flash electroretinogram waveform are less conclusive given the limited number of studies and inconsistency in their reporting. Higher quality studies consistently and explicitly reporting on global flash electroretinogram outcomes are required to further clarify the evidence.  Data Availability Statement: The data extracted for this paper are available in the original research articles that were systematically reviewed.