Hand-held optical coherence tomography: advancements in detection and assessment of optic nerve abnormalities and disease progression monitoring

ABSTRACT

Introduction

Developmental abnormalities of the optic nerve (ON) and pediatric optic neuropathies, such as glaucoma, are leading causes of childhood blindness. The recent development of hand-held spectral domain optical coherence tomography (HH-OCT) has enabled noncontact, high-resolution scanning in non-sedated newborns, infants, and young children and has provided for the first time, in vivo visualization of the retina and ON in these patient groups.

Areas covered

This review will address the applications, recent advances, and future potential of HH-OCT in diagnosis and monitoring of pediatric optic neuropathies. We will provide an update on the use of HH-OCT in pediatric glaucoma, congenital optic disc anomalies, optic pathway gliomas, optic atrophy, and papilledema.

Expert opinion

HH-OCT could offer particular utility in children with optic neuropathies, by providing noninvasive, high-resolution characterization of the optic nerve head. Optic nerve parameters, such as retinal nerve fiber layer thickness, could serve as biomarkers to assess the severity of optic nerve disease qualitatively and quantitatively. Hence, HH-OCT is emerging as a powerful imaging tool to facilitate early diagnosis, identify prognostic biomarkers, monitor disease progression, and assess response to treatment.

KEYWORDS:

• Hand-held optical coherence tomography
• optic atrophy
• optic disc coloboma
• optic disc drusen
• optic nerve
• optic nerve hypoplasia
• optic pathway glioma
• pediatric glaucoma
• pediatric intracranial hypertension

1. Introduction

Developmental abnormalities of the optic nerve (ON) and pediatric optic neuropathies, such as glaucoma, are leading causes of childhood blindness [1]. However, diagnosis and monitoring is challenging in preverbal children due to limited cooperation, often leading to poorly reliable assessments of visual acuity (VA) and visual fields.

The recent development of hand-held spectral domain optical coherence tomography (HH-OCT) has enabled noncontact, high-resolution scanning in non-sedated newborns, infants, and young children and has provided for the first time in vivo visualization of the retina and ON in these patient groups [2,3]. HH-OCT has been shown to be reliable, can be obtained with 94% success in children without anesthesia and has a great potential as an initial diagnostic procedure and for monitoring of disease progression by providing additional morphologic information that is not normally clinically discernible [2,4]. Over the last years, it has been used successfully to characterize normal retina and ON development, to establish normative data of the retina and ON in infancy and childhood [3,5], and as an adjunctive diagnostic tool in retinopathy of prematurity (ROP) [6]. Additionally, the role of HH-OCT in investigation of infantile nystagmus and retinal dystrophies has been established [2,7,8]. HH-OCT can also help differentiate between pediatric intraocular tumors, for example, hamartomas and retinoblastoma, monitor tumor progression, and monitor treatment response [9,10].

HH-OCT could offer particular utility in children with optic neuropathies, by providing high-resolution characterization of the ON head, peripapillary retinal nerve fiber layer (RNFL), and cellular layers of the macula, all of which can be used to assess the severity of ON disease qualitatively and quantitatively. For this reason, HH-OCT is being increasingly used as a noninvasive monitoring tool in children with ON pathologies such as glaucoma, ON atrophy and optic nerve hypoplasia (ONH), optic pathway glioma, optic disc drusen and pseudotumor cerebri [11–15].

2. Methods of literature review

This review will address the applications, recent advances, and future potential of HH-OCT in diagnosis and monitoring of pediatric optic neuropathies. A PubMed search of all articles published from January 1980 to July 2021 on the use of HH-OCT in pediatric optic neuropathies was performed. Searches included a combination of the following terms: hand-held, pediatric, childhood, optical coherence tomography, optic atrophy, optic disc coloboma, optic disc drusen, optic neuritis, optic nerve hypoplasia, optic pathway glioma, pediatric glaucoma, pediatric intracranial hypertension, retinal nerve fiber layer. The resulting references were then reviewed for pertinent articles. No language restrictions were applied.

3. Normal optic nerve development and optic nerve development in prematurity

Before the development of OCT, our knowledge on the perinatal development of the ON was based on cadaveric, histologic, and fundus photography studies [16–19]. Preliminary histopathologic studies had shown that the peak count of in utero ON axons was reached at approximately 16–17 weeks gestational age, then declined and stabilized by about week 29 of gestation [16]. Additionally, it was demonstrated that 75% of the growth of the optic disc and retrobulbar nerve occurs by birth, and 95% of the growth occurs before the age of 1 year [17]. Fundus photography studies suggested that birth weight and sex does not influence the size of the optic disc in full-term infants, and provided normative data on cup-to-disc ratios (CDR) and arteriole-to-venule (A/V) ratios in school-aged children [18,20].

The advent of OCT added significant information for the development of ON in children old enough to cooperate on table-mounted devices. The reliability and reproducibility of ON measurements by means of OCT in children was established, normative pediatric data for ON variables were recorded, and it was shown that black children had larger CDR and thicker RNFL [21–25]. Further research with conventional OCT in children between 4 and 10 years old, who were born prematurely, showed thinning of the peripapillary RNFL, which was associated with reduced visual function [26–28]. However, these studies did not include infants and toddlers, and did not provide information during the first years of life, which is the period with the most rapid changes regarding ON development according to histologic studies [16,17].

With the development of HH-OCT, Lee et al. examined 38 preterm neonates and showed that HH-OCT could detect ON abnormalities, which were not visible during routine funduscopy, such as hyaloid artery remnants and large optic cup [6]. Later, Rothman et al. reported RNFL thickness measurements for 50 full-term infants [29], and Tong et al. assessed ON parameters in 44 preterm and 52 term infants using HH-OCT [30]. The authors showed that by age of term birth vertical CDR is larger in preterm than in term infants and that RNFL is thinner for very preterm vs. term infants at the papillomacular bundle ([mean ± standard deviation] 61 ± 17 vs. 72 ± 13 μm, p < 0.001) [30,31]. Shen et al. later suggested that RNFL is 11.2 μm thinner in extremely low birth weight infants than in very low birth weight infants (55.5 ± 8.3 μm vs. 66.7 ± 10.2 μm; p < 0.001), and the main factor that influenced RNFL thickness at 36 weeks postmenstrual age was birth weight [32]. Interestingly, gestational age and ROP stage were not significant predictors of RNFL thickness after adjustment for birth weight [32].

ON parameters in preterm infants may also be predictive of central nervous system (CNS) pathology and future cognitive development, with larger vertical CDR and thinner RNFL being associated with lower Bayley Scales of Infant Development scores [30,31]. HH-OCT has also shown potential in undilated infants with relative contraindication to pharmacological pupil dilation, such as preterm babies and those with hypoxic injury or hydrocephalus, in order to evaluate the eye–brain connection and the ocular manifestations of neurologic diseases [33,34].

The longitudinal analysis of 26 preterm eyes at 31–36 weeks and 37–42 weeks postmenstrual age showed an increase in vertical disc diameter and decrease in vertical CDR [30]. Patel et al. performed a HH-OCT prospective study with 352 healthy children aged between 1 day and 13 years and established normative data of the retina and ON in infancy and childhood [3]. The most dramatic changes in ocular development occurred within the first 2 years of life [3]. The size of the optic disc and cup diameter at term was 76.9% and 71.4% of that observed in children aged 7–13 years [3]. The disc and cup diameter increased rapidly between birth and 2 years, following age-related changes in globe diameter [3]. However, when expressed as a visual angle, disc and cup diameters and CDR did not change significantly with age, indicating that those parameters increase proportionally with increasing axial length [3]. Additionally, the peripapillary temporal RNFL demonstrated a marked initial decrease of nearly 35% between birth and 18 months of age, which was followed by a slow increase up to 12 years of age [3]. Lim et al. also provided HH-OCT data for ganglion cell complex (GCC) and RNFL in 67 healthy children from 3.4 to 70.9 months under sedation or anesthesia and found that average GCC volume (0.28 ± 0.04 mm³) and RNFL thickness (38.2 ± 9.5 μm) were stable from 6 months to 5 years of age [35].

Recent research efforts have focused on optimizing scanning acquisition protocols for HH-OCT use in children. Shah et al. examined 180 healthy children (mean age, 7.09 ± 4.56 years) to investigate feasibility of obtaining three-dimensional (3D) full circumpapillary RNFL analysis by HH-OCT. 3D imaging by means of HH-OCT was possible in both eyes of 360 controls, and where scanning was possible, success at obtaining full circumpapillary RNFL was 89% for controls [36]. Consequently, 3D HH-OCT volumes in non-sedated children is both feasible and reliable and is optimal at 6° from ON center [36].

Describing the trajectory of normal ON development and establishing normative databases for ON and retinal parameters for children of different ages is important, because it provides a comparison for future studies of children with ON pathology. The interpretation of pediatric HH-OCT should take into account the normal foveal and optic nerve development, the changes in axial length with age and the higher refractive error [21,37,38]. Further HH-OCT longitudinal studies of healthy children should also describe ON development across children of different race and ethnicity.

3.1. Retinopathy of prematurity

HH-OCT is being increasingly used for diagnosis, follow-up and assessment of treatment response in ROP and a large body of evidence supports its clinical utility in ROP screening. Since this review focuses on the use of HH-OCT in optic nerve abnormalities, this section will only briefly assess the recent developments in ROP.

Preliminary work with HH-OCT during routine ROP examinations identified macular cystoid structures in 39% of examinations and epiretinal membrane in 32% of examinations, which were not visualized by indirect ophthalmoscopy [6]. Persistence of inner retinal layers in premature infants regardless of maximal ROP stage, subclinical cystoid macular edema and vitreous bands have been also documented [39,40]. It has been suggested that punctate hyperreflective vitreous opacities and tractional vitreous bands predict the presence and severity of ROP [41]. In the study by Legocki et al., the presence of hyperreflective vitreous opacities at least once was associated with a diagnosis of ROP (62% vs. 29% without opacities; P = 0.003), maximum ROP stage (P = 0.001), preplus or plus disease (24% vs. 5%; P = 0.005), and type 1 disease (14% vs. 2%; P = 0.03) [41]. HH-OCT of the temporal retina of preterm neonates has enabled imaging of sequential structural changes of abnormal extraretinal neovascularization, including buds, bridging networks, and placoid lesions [42,43]. Additionally, in infants with stage 4 ROP HH-OCT is useful in the differentiation of retinal detachment and retinoschisis [44].

Further studies showed increased central foveal thickness in stage 1 and 2 ROP eyes compared to normal eyes (156.9 ± 28.3 μm, 206.5 ± 98.7 μm, and 135.9 ± 17.6 μm, respectively, p < 0.001) [45]. With an increase in ROP stage from 0 to 2, the mean ± standard deviation foveal retinal thickness increased from 227 ± 124 μm to 297 ± 99 μm (p < 0.001) [46]. Additionally, it has been shown that the most premature infants have the thickest inner retinal layer (IRL) and shallowest pits at all postmenstrual ages, suggesting arrest of foveal development and explaining the foveal hypoplasia seen in children with a history of ROP [47]. Interestingly, in the study by Anwar et al. ROP was significantly correlated with only foveal width in either gestational age or birth weight adjusted statistical models, recommending foveal width as a potential early indicator of ROP [48]. Foveal width decreased when ROP was absent, at a mean (±SEM) rate of −11.18 ± 4.46 µm per week but increased when ROP was present at a rate of +24.96 ± 6.92 µm per week [48].

Finally, the recent use of HH-OCT angiography (OCTA) has enabled dye-free visualization of vascular development and showed that large- and small-vessel parameters were associated with prematurity and ROP severity, respectively [49].

4. Pediatric glaucoma

Pediatric glaucoma is a potentially blinding disease, with 47% of affected children having a VA ≤ 6/60 (Kargi). Primary congential glaucoma (PCG) has the most favorable outcome [1], while secondary glaucoma, i.e. related to aphakia, uveitis, anterior segment dysgenesis, have a more guarded prognosis [50]. Traditionally, the diagnosis of pediatric glaucoma and monitoring of disease progression in infants and young children require frequent examinations under anesthesia, with potential long-term adverse neurodevelopmental effects [51]. Because of the challenge of obtaining reliable visual fields, measure intraocular pressure and quantify ON damage in children, OCT is particularly well suited for diagnosis and monitoring of pediatric glaucoma.

4.1. Table-mounted OCT in children with glaucoma

Imaging of the ON is an essential part of diagnosing and monitoring adult glaucoma [52,53]. Similar to adults, spectral-domain OCT is reliable in glaucoma and glaucoma-suspect eyes of older children for objective measurement of the RNFL [38,54]. Peripapillary RNFL and macular thickness have been shown to correlate well with disease severity and to have reasonable reproducibility [38,54,55]. After glaucoma surgery, OCT has demonstrated RNFL and ganglion cell layer (GCL) thinning, and enlarged cupping with thinner rim areas in children aged 10.1 ± 3.6 years [56]. Additionally, recent OCT studies have questioned the value of postoperative cupping reversal as a marker of successful glaucoma surgery in PCG and juvenile open-angle glaucoma. In eyes with postoperative cupping reversal, OCT has shown continued thinning of the peripapillary RNFL, despite the IOP reduction and the healthier funduscopic appearance of the optic head [57]. Similarly, a recent series of four children with secondary glaucoma did not find a correlation between OCT RNFL and cupping reversal after successful control of IOP with medical or surgical treatment [58]. Hence, it was suggested that preoperative RNFL thickness is a better predictor of visual outcome and OCT may be a more sensitive method of characterizing the severity of glaucomatous damage [57,58]. Recently, outer retinal changes have been described in 13.2% of pediatric glaucoma eyes associated with uveitis or prior intraocular surgery, including pigment epithelial detachment, schisis of the nerve fiber and/or ganglion cell layers, choroidal folds, disruption of the photoreceptor inner and outer segments, aatrophy,and cystoid macular edema [11]. Interestingly, OCT was helpful in differentiating between glaucomatous and nonglaucomatous optic atrophy in children, with the rate of isolated inner nuclear layer (INL) cysts being 0.8% in glaucoma versus 30% in nonglaucomatous optic atrophy [11]. Anterior segment OCT has been also used to assess Haab striae in congenital glaucoma and provide differential diagnostic clues between acute and chronic corneal changes [59].

4.2. HH-OCT in children with glaucoma

These findings have led to the use of HH-OCT in pediatric glaucoma, in order to assess its diagnostic value in infants and young children who cannot cooperate on table-mounted devices. The first observational study by Pilat et al. included 20 children with PCG with a mean age of 4.64 ± 2.79 years, who underwent HH-OCT of the fovea and ON [60]. HH-OCT was successful in 83.3% eyes of patients with PCG without sedation or anesthesia. Children with PCG had significant increase in cup depth (165% increased from control children, 781.8 μm vs. 473.5 μm, p < 0.000), increased cup diameter (159% increased from control children, 5.59° vs. 6.01°, p < 0.000) and reduction in rim area (36.4% reduced from control children, 215.0 μm vs. 590.3 μm, p < 0.000) as compared with 20 controls with high sensitivity and specificity [60]. Additionally, the foveal pit width was significantly reduced in children with PCG and the external limiting membrane (ELM) was absent in all patients with PCG, suggesting early photoreceptor damage and reduced blood supply to the retina (Figure 1). Although many HH-OCT findings in PCG were similar to those of 10 adults with primary open-angle glaucoma (thinner RNFL, larger cupping, smaller rim area), the increase in cup depth, the reduction of the foveal pit width and a non-detectable external limiting membrane were novel findings in PCG [60].

Figure 1. HH-OCT images of the optic nerve (ON) (A) and fovea (B): (A) horizontal B-scan through the center (deepest excavation) of the optic disc; the disc diameter was defined as an interval between the edges of Bruch’s membrane (red line), the cup diameter as the distance between the nasal and temporal internal limiting membrane (green dotted line) 150 μm anterior to the plane of the disc (blue line), the rim area consisted of the area anterior to the same plane (white dotted lines) within the disc edges (white vertical lines) and the internal limiting membrane (green dotted lines); maximal cup depth (vertical yellow line) was measured using a line perpendicular to the line between the cup diameter (blue line) and the deepest point of the cup; RNFL thickness was measured at 6° from disc margins (red dotted lines). (B) Horizontal B-scan of the fovea with labeled individual retinal layers (BM, Bruch’s membrane; ILM, inner limiting membrane; RNFL, retinal nerve fiber layer; GCL, ganglion cell layer; INL, inner nuclear layer; IPL, inner plexiform layer; IS, inner segment; ONL, outer nuclear layer; ELM, external limiting membrane; OPL, outer plexiform layer; OS, outer segment; RPE, retinal pigment epithelium). Retinal layer thickness was measured in the center of the fovea, in the paracentral area (from 1° nasally to 1° temporally) and nasally and temporally (from 2° to 6°). The green line connecting the most prominent positions of the ILM nasally and temporally was used to define the foveal width; the yellow line indicates the foveal depth (the axial distance from the green line to the deepest point of the foveal pit); the area in blue indicates the foveal pit area. (C) Horizontal spectral domain–optical coherence tomography B-scan images of the optic nerve (top) and fovea (bottom) of a patient with primary congenital glaucoma in the left eye (PCG, middle column), an unaffected right eye (left column) and an eye of a healthy age-, gender- and ethnicity-matched control child (right column). On the ON scan, a larger and deeper cup is seen in PCG. On the foveal scan, the ELM is not visible in PCG while it is distinctly seen in the unaffected eye and in the healthy control. Reprinted with permission from [60], licensed under https://creativecommons.org/licenses/by/4.0/.

A limitation of previous studies in pediatric glaucoma is that single horizontal B-scans through the center of the optic disc may not be so sensitive in detecting glaucomatous damage, because RNFL thinning begins in the inferior and superior quadrants, as suggested by studies on adult glaucoma. Hence, the recent improvement of OCT acquisition protocols has overcome difficulties in fixation stability of infants and young children, allowed the volumetric peripapillary RNFL thickness analysis and recommended protocol settings for pediatric ON imaging using a specific HH-OCT device [36]. Shah et al. showed that 3D imaging by means of HH-OCT was possible in 64 out of 90 children with glaucoma (mean age, 6.98 ± 4.42 years), and where 3D imaging was possible, success at obtaining full circumpapillary RNFL was 67% in children with glaucoma and 89% for controls [36].

In addition to the examination of the ON, HH-OCT of the anterior segment (AS) structures in patients with PCG has also provided novel insights into the structural changes of the anterior segment, which cannot be visualized by means of the slit-lamp examination [61]. The main findings of Pilat et al. included Haab’s striae, uneven internal contour, and epithelial thickening of the cornea, flat iris with a thin collarette zone and anterior iris insertion [61] (Figures 2 and 3). Abdeen et al. also performed AS HH-OCT in 26 eyes with PCG and 22 normal eyes of infants younger than 24 months [62]. Distinct anatomical feature were identified in PCG-eyes: an abnormal structure occluding the angle in 26.9% of eyes, a hyper-reflective membrane in 19.2% of eyes and an anterior iris insertion in all eyes [62]. The nasal and temporal anterior chamber angle width in PCG infants was found significantly larger (39.3 ± 6.6° vs. 30.4 ± 5.6, and 40.1 ± 5.3° vs. 32.5 ± 6.2 respectively) (p < 0.001), and iris thickness was significantly reduced (121.7 ± 43.9 μm in PCG-infants, vs. 160.3 ± 38.6 μm in normal-eyes) (p < 0.01) [62].

Figure 2. Corneal horizontal HH-OCT images of patients with PCG and healthy age-, gender- and ethnicity-matched controls. Patients with PCG showed a variety of changes, including epithelial thickening with a tortuous contour of the Bowman membrane (ID1, 6 and 13), uneven internal contour and the presence of Haab’s striae (ID3, 8). Reprinted with permission from [61], licensed under https://creativecommons.org/licenses/by/4.0/.

Figure 3. Horizontal high-resolution HH-OCT images of the temporal and nasal irido-corneal angles in a patient with PCG in the right eye and a healthy age-, gender- and ethnicity-matched control. The image of the right affected eye of the patient shows abnormal anterior iris insertion in the nasal angle with the iris rout inserting at Schwalbe line (SL) covering the trabecular meshwork (TM). Normal configuration with a visible trabecular meshwork in the temporal angle of the affected eye, in the non-affected eye as well as in the control subject. Reprinted with permission from [61], licensed under https://creativecommons.org/licenses/by/4.0/.

Additionally, Pilat et al. provided HH-OCT data of the anterior and posterior segments in a series of children with anterior segment dysgenesis (Axenfeld’s anomaly, Peter’s anomaly, iridocorneal endothelial syndrome (ICE) syndrome, PAX6 mutation with aniridia, iris, and choroidal coloboma and PCG) [63]. A variety of pathological corneal, angle, and iris findings were reported, which were not clinically visible in 4 of 6 patients [63]. Interestingly, subclinical maldevelopment has been found in retinal structures in clinically unaffected eyes [63] (Figure 4). The emerging value of HH-OCT has been also recently shown in a study examining the angle structures during surgery in children with and without glaucoma. Intraoperative AS HH-OCT showed absence of Schlemm’s canal (SC) in 6 of 13 glaucomatous eyes, and abnormal tissue over the angle and SC in 8 of 13 glaucomatous eyes [64].

Figure 4. Anterior segment photography (A) and spectral domain optical coherence tomography (B–J) of case 3 with right Peter’s anomaly (case 3; A–C = affected eye, D–G = unaffected eye) and a healthy control (H–J). Slit-lamp photography (A) shows central corneal opacity with iridocorneal adhesion. The white arrows on figure B show iridocorneal adhesion on anterior OCT. On figure C, the dotted arrow indicates the scleral spur and the white arrow shows thinning of the iris root. Due to lens opacity in the right (affected) eye, posterior OCT was not possible. The horizontal lines on figures G and J connecting the edges of retinal pigment epithelium show the disc diameter. Optical coherence tomography (OCT) of the left (unaffected) eye of the patient with Peter’s anomaly showed normal anterior segment structures (D, E) and macula (F) with small optic nerve cup (G) as compared to the control subject (J). Reprinted with permission from [63], licensed under https://creativecommons.org/licenses/by/4.0/.

The high sensitivity and specificity of ON changes in pediatric glaucoma, suggest that HH-OCT is a powerful tool to detect and monitor glaucoma and should be used as part of the clinical assessment to quantitatively assess particular ON parameters. Further longitudinal studies should describe the optimal HH-OCT parameters for monitoring pediatric glaucoma, address the correlation between structural and functional measures and also include children with other glaucoma subtypes, such as aphakic, uveitic, traumatic and associated with anterior segment dysgenesis. Although the initial diagnosis of PCG is mainly clinical based on the striking symptoms of the disease (epiphora, corneal clouding, buphthalmos, photophobia), HH-OCT enables detailed monitoring of disease progression by assessing specific ON and retinal biomarkers, which are difficult to quantify by means of ophthalmoscopy, that is, cup depth. Finally, HH-OCT can be helpful in differentiating ocular hypertension from other types of pediatric glaucoma with relatively insidious clinical course, that is, in aphakia and pseudophakia after congenital cataract surgery [60].

In conclusion, HH-OCT has the potential to provide in vivo high-resolution images of the cornea, iridocorneal angle, iris, and ON in awake infants under outpatient clinical settings. Hence, it is an extremely useful tool to detect the underlying anatomical abnormalities in PCG, correlate structure with function, determine genotype–phenotype relationships, accurately plan surgical procedures, monitor disease progression and response to treatment.

5. Congenital optic disc anomalies

Congenital optic disc anomalies refer to structural malformations of the ON head and are a common cause of congenital visual impairment. The prevalence of congenital optic disc anomalies has been estimated at 1.1% in the adult South Indian population [65]. Recent advances in imaging and genetics have led to earlier recognition and hence increasing prevalence of these entities [66]. Accurate and timely diagnosis of congenital optic disc anomalies during the first days of life is of outmost importance, due to the possibility of lifelong visual impairment and associated neurological, endocrinological, and systemic comorbidities [66]. However, still little is known about the detailed morphological alterations, structure-function correlation and the possible developmental changes of the optic disc and retina in affected individuals.

5.1. Optic nerve hypoplasia

ONH is a common congenital malformation of the optic disc, with an estimated prevalence between 10.9 and 17.3/100.000 children per year [67,68]. ONH can be also associated with neurological abnormalities, such as septo-optic-dysplasia (known as de Morsier syndrome, which is characterized by midline brain abnormalities, including absent septum pellucidum, hypothalamic dysfunction, and endocrine abnormalities). VA in affected eyes is variable ranging from logMAR 0.0 to no light perception [69,70]. The diagnosis of ONH in newborns can be difficult, especially in cases where the disc changes are subtle. Hence, many cases might go unrecognized or misdiagnosed as congenital optic atrophy. Additionally, optic disc morphologic predictors of VA are still under investigation. Previous studies using ophthalmoscopy, fundus photography, and electrophysiology have found that optic disc size, optic disc pallor, visual evoked potentials (VEPs) and pattern electroretinogram (PERG) correlate with visual outcome later in life [70–72]. By means of OCT, it has been also suggested that disc diameter is correlated with VA (r = 0.32), VEPs (r = 0.66) and average RNFL thickness in the temporal area (r = 0.61) that corresponds to the position of the papillomacular bundle in a group of subjects from 2.4 to 20 years [73].

Pilat et al. have investigated ON and retinal morphology in 16 patients with ONH (mean age, 17.2 years; SD ± 16.22) and 32 healthy controls by HH-OCT [13]. Patients had significantly smaller discs (p < 0.03 and p < 0.001 compared with unaffected eye and healthy controls, respectively), horizontal cup diameter (p < 0.02 for both), and cup depth (p < 0.02 and p < 0.01, respectively) (Figure 5) [13].

Figure 5. Fundus images (top) and SD-OCT horizontal B-scans (bottom) through the center of the disc in (A) a patient with right eye optic nerve hypoplasia (ONH) and (B) bilateral optic nerve head drusen (ONHD) with ONH in the left eye. Arrows indicate the position of the drusen above the edges of the RPE. The disc size of the left eye with both ONHD and ONH is considerably smaller than in the unaffected eye. Reprinted with permission from [13], licensed under https://creativecommons.org/licenses/by-nc-nd/4.0/.

Additionally, there were also macular changes including foveal hypoplasia and thinning of the RNFL, GCL, IPL, ONL, and IS compared with the control group (P < 0.05 for all comparisons) [13]. Mild features of underdevelopment were also detected in clinically unaffected fellow eyes of patients with ONH [13]. Interestingly, the authors found no correlations between VA and RNFL thickness in the temporal area. However, changes in macular structure (GCL and IPL) were associated with VA and presence of septo-optic dysplasia, providing a useful predictor of visual function [13].

5.2. Congenital optic disc pits and optic disc colobomas

Congenital optic disc pits, optic disc colobomas peripapillary staphyloma, and morning glory syndrome belong to the group of congenital excavated optic disc anomalies. The development of swept-source OCT (SS-OCT) and enhanced-depth-imaging OCT (EDI-OCT) enabled the visualization of deeper structures in the ON head complex and provided novel information about the anatomical alterations and pathophysiology of the above entities. Given that approximately two-thirds of congenital optic disc pits are complicated with a serous retinal detachment of the macula, OCT can be extremely valuable to determine the pathogenesis of optic disc pit maculopathy. Ohno-Matsui et al. examined 16 eyes with congenital optic disc pits and seven eyes with optic disc colobomas by means of SS-OCT and found displaced lamina cribrosa and herniation of retinal tissue [74]. Lee et al. found distinct OCT features in 14 eyes with congenital excavated optic disc anomalies, and concluded that OCT can be helpful for differential diagnosis and follow-up of these disorders [75].

Grewal et al. first used HH-OCT to describe the characteristics of pediatric choroidal neovascular membranes (CNVs) associated with retinochoroidal and ON coloboma [76]. Vision impairment in coloboma eyes can be caused by optic disc and/or foveal involvement, retinal detachment and CNVs, and OCT has permitted identification of subclinical retinal detachments along the margin of the coloboma. The prevalence of choroidal or retinal detachment in patients with chorioretinal colobomas has been reported between 2.4% and 43% [77–81]. Grewal et al. reported that CNV occurred at the temporal margin of the coloboma closest to the fovea in all cases, and characteristic OCT features were subretinal fluid, intraretinal fluid, cysts and subretinal hyperreflective material [76]. Importantly, serial OCT imaging allows monitoring of treatment response with improvement in subretinal and intraretinal fluid [76]. Pilat et al. have also performed HH-OCT in a 12-year-old boy with iris and ON coloboma, which showed a large ON with RNFL thinning and a larger foveal pit in the unaffected eye [63] (Figure 6).

Figure 6. Anterior (A) and posterior (B) segment photography and OCT (C–J) of a patient with right iris and optic nerve coloboma (A–G) and of the healthy control (H, I). Anterior segment photography (A) shows iris coloboma at 6 o’clock with visible zonula fibers of the lens; fundus photograph (B) demonstrates right optic nerve coloboma without clearly defined optic disc margins, and the fovea was not identified. Dotted lines in figures B and E display the corresponding area of the coloboma on fundus photography and OCT. On OCT, the optic nerve has a large excavation; the fovea was not seen. Figure C shows a normal cornea in the affected eye; figure D shows the anterior tomogram in the area of the coloboma with a flat hypoplastic iris stump. The fovea in the unaffected side (F) had a wider and larger foveal pit as compared to the healthy control (H). Tomograms of the optic nerve in the unaffected side (G) and healthy control (I) appeared similar. Reprinted with permission from [63], licensed under https://creativecommons.org/licenses/by/4.0/.

Given the difficulty of managing retinal detachments in pediatric patients with chorioretinal colobomas, HH-OCT will be extremely helpful in early identification of associated retinal pathology. Additionally, accurate diagnosis will enable prompt referral to other pediatric specialists, as up to 38% of patients with chorioretinal colobomas have associated systemic abnormalities [77]. Finally, little is known about the visual prognosis of eyes with congenital excavated optic disc anomalies based on their funduscopic appearance. Based on echographic images, it has been suggested that increased relative coloboma excavation and the presence of a retrobulbar cyst were associated with increased risk of retinal detachment and severe visual impairment [82]. HH-OCT is a powerful tool to assess in detail possible longitudinal changes over time, response to treatment and prognostic OCT biomarkers for VA.

6. Albinism

Albinism is a group of genetic disorders characterized by disruption of melanin biosynthesis [83]. In ocular albinism (OA) the abnormality is localized to the eyes, and in oculocutaneous albinism (OCA) there is also hypopigmentation of the hair and skin. A spectrum of visual disorders has been described in subjects with albinism, including iris transillumination defects, failure of photoreceptor specialization, foveal hypoplasia, nystagmus, reduced VA, strabismus, astigmatism and other refractive errors, and misrouting of the ON fibers at the chiasm, consisting of an excessive crossing of the fibers in the optic chiasm [84–86].

HH-OCT has been used to describe retinal development in albinism in children between 0 and 6 years [8]. In 44 children in the albinism group, inner retinal layer migration from the fovea was arrested prematurely, resulting in a significantly thicker central macular thickness as compared to the group of 224 controls (p < 0.0001) [8]. Additionally, table-mounted anterior segment OCT has shown that phenotypic features of albinism, such as skin and hair pigmentation, VA, and nystagmus intensity, were significantly correlated to iris thickness measurements, but not to iris transillumination grades [84]. The mean iris thickness was 10.7% thicker in the group of 45 controls (379.3 ± 44.0 μm) compared with the group of 55 albinism participants (342.5 ± 52.6 μm; p > 0.001) [84].

HH-OCT has not been used for investigation of the ON in patients with albinism yet. Currently, our knowledge is based on the information derived from conventional OCT devices. Chong et al. reported abnormal elevation of the ON in four of six subjects with albinism [87]. An OCT study of 56 patients with albinism and 60 control participants has also shown smaller CDR (p < 0.001) and reduced peripapillary RNFL thickness especially in the temporal quadrant (p < 0.001) in patients with albinism compared to controls [12]. Several ON head abnormalities, such as CDR, were related to higher refractive errors in albinism [12]. Hence it remains to be elucidated whether there are distinctive features of the ON head in albinism, and how those are correlated with visual function.

7. Optic pathway gliomas

Optic pathway gliomas (OPGs) are rare low-grade glial tumors with unstable evolution that demand cautious monitoring and appropriate management. OPGs are sporadic or affect 15–20% of patients with neurofibromatosis type 1 (NF1). It has been suggested that the visual prognosis for sporadic OPGs is worse compared to OPGs associated with NF1 [71]. The mean age at diagnosis is 4 years, with 90% of patients being diagnosed in the first two decades of life [88–90]. Although OPGs have excellent 5-year-survival rates of more than 90%, their anatomic location can lead to visual impairment, which in some cases may progress to complete blindness [91,92]. Currently, there is controversy regarding the optimal management of OPGs and observation of indolent tumors is an important option. Due to their infiltrating nature and involvement of the optic nerves, chiasm, hypothalamus and optic tracts, the role of surgical excision is limited. Hence, the current consensus is to treat children with chemotherapy, and in some cases with biologic agents and early radiation therapy [93–95]. Treatment is generally indicated in children with evidence of visual or neurological deterioration and/or radiological tumor progression.

Most OPGs become symptomatic between 1 and 8 years of age [96]. Unfortunately, in this age group VA and visual field testing are difficult to perform and are not reliable, and collaboration with traditional table-mounted OCT devices is not always possible. As a result, many children may undergo unnecessary chemotherapy. Moreover, the alterations in tumor size as demonstrated with neuroimaging are not proportionate to visual deterioration, and there are reports of visual decline without noticeable tumor growing [14]. Therefore, it is of outmost importance to introduce a novel biomarker, which is not dependent on the child’s cooperation and can reliably detect visual deterioration, in order to allow early treatment and also monitor response to treatment.

Avery et al. studied the use of HH-OCT in sedated children with OPGs prior to MRI imaging, and found that measures of circumpapillary RNFL demonstrate good intra- and inter-visit reproducibility, concluding that HH-OCT has the potential for longitudinal monitoring of circumpapillary RNFL changes [14]. Similarly, highly reproducible measures of retinal thickness and ganglion cell-inner plexiform layer (GC-IPL) thickness were demonstrated in another study of sedated children with OPGs undergoing HH-OCT [14]. It was suggested that a loss of ≥10% in the GC-IPL is probably clinically significant, while a normal or stable thickness of GC-IPL probably indicates that the tumor is not causing progressing damage to the visual pathway [14]. The same authors showed that RNFL thickness by means of HH-OCT during sedation can differentiate between young children with and without vision loss from OPGs (abnormal VA and/or visual field), confirming previous results with table-mounted devices [14,90]. RNFL thickness was decreased in all quadrants in children with vision loss (n = 15 eyes) compared with children without vision loss (n = 49 eyes) from OPGs (average RNFL thickness 75.8 μm vs. 125.1 μm, respectively, p < 0.001) [90]. In addition, there was no significant difference in RNFL thickness between children without vision loss from OPGs and 20 control participants (children with NF1 without OPGs). This study showed that vision loss is proportional to RNFL thickness, which could be used as a surrogate measure of vision and an objective quantitate biomarker [90]. A recent longitudinal cohort study added that children with vision loss from OPGs usually show a ≥ 10% decline of RNFL thickness in one or more quadrants [97].

Hence, RNFL thickness could serve as a reliable, quantitative, clinically relevant biomarker, in order to longitudinally monitor vision in children with OPGs, who are unable to cooperate with VA or visual field testing. HH-OCT could aid clinicians in deciding which child with an OPG requires observation vs treatment, especially in the presence of new or worsening magnetic resonance imaging (MRI) findings. Additionally, early detection of visual decline by means of RNFL thickness could allow treatment prior to massive axonal damage and subsequent visual loss. Therefore, it is of outmost importance to run large, longitudinal multicenter studies in order to investigate HH-OCT quantitative biomarkers, such as RNFL and GC-IPL thickness, in order to establish optimal monitoring protocols for children with OPGs.

8. Optic nerve tumors

8.1. Retinoblastoma

Recently, HH-OCT has almost become a standard tool for evaluation of retinoblastoma, with increasing use in the diagnosis, guidance of treatment, assessment of response to treatment and life-long monitoring [9]. Recently, HH-OCT has been also implemented in the diagnosis of retinoblastoma at the optic nerve head. Fabian et al. have described six children with secondary epipapillary retinoblastoma, didiagnosed,nd monitored with HH-OCT [98]. Children were subsequently treated with intraophthalmic artery chemotherapy and/or intravitreous chemotherapy. The significance of this report is that, in two of six cases the relapse was not evident in ophthalmoscopy, but was detected by the HH-OCT, which showed the presence of a vitreous seed in the ON cup [98]. The invaluable role of HH-OCT in detecting subclinical tumors has been reported by the same authors in an additional series of 16 recurrent subclinical retinoblastoma foci all measuring less than 400 microns [99]. These lesions are translucent and avascular in the initial stages, hence undetectable on ophthalmoscopy and fluorescein angiography [99]. Prompt diagnosis is hence essential for detection of lesions threatening the macula and ON, preservation of VA and optimal survival rates [100].

9. Optic atrophy

To date, no reports have been published on the use of HH-OCT in optic atrophy due to various causes. Studies with table-mounted OCT devices have provided valuable insights into the morphological features of atrophic discs and progression over time.

As mentioned above, OCT can distinguish between glaucomatous and nonglaucomatous optic atrophy in children, with INL cysts, outer retinal and photoreceptor loss, total retinal atrophy, choroidal folds, and inner segment disruption more likely to be present in the latter [11]. Causes of nonglaucomatous optic atrophy in this study included atrophic papilledema, optic neuritis, anterior visual pathway tumors and periventricular leukomalacia. Children with these conditions will inarguably benefit from HH-OCT assessment not only during the initial diagnostic workup, but also during follow-up examinations.

In particular, in patients with autosomal dominant optic atrophy (ADOA) peripapillary RNFL thickness is reduced mainly in the temporal and inferior quadrants as compared to controls (average temporal RNFL thickness 34.33 μm vs. 68.23 μm, average inferior RNFL thickness 72.12 μm vs. 120.30 μm respectively) and also correlates significantly with VA in patients with ADOA (r = −0.845, p = 0.008) [101,102]. Additionally, OCT studies have suggested that patients with ADOA are born with fewer ON axons and subsequent visual loss depends on further age-related loss of fibers [101,103]. HH-OCT allows imaging of suspicious optic discs from infancy and also monitoring for progression, and should be implemented in other types of mitochondrial optic neuropathies and also Leber’s hereditary optic neuropathy with childhood onset.

Traumatic optic neuropathy (TON) is another cause of optic atrophy, which occurs in 0.5% to 5% of patients after head trauma, with motor vehicle collisions and sports injuries being the most common causes [104]. Diagnosing TON in children can be a challenge, because children may not notice the visual deficit, may not report it, or may not be able to perform VA measurements. Previous OCT studies have documented progressive RNFL thinning and macular thinning and have described the progression of morphological changes over time, suggesting that OCT is able to assess and monitor axonal loss after TON [105–107].

10. Neurological diseases

The potential of HH-OCT to capture in-vivo images of the human retina in non-sedated patients permits assessment of not only infants and small children, but also of noncooperative adult patients, patients with neurodevelopmental delay and patients who are unable to sit on table-mounted devices [108].

10.1. Microcephaly

Primary microcephaly is a brain development disorder defined as a head circumference more than 3 standard deviations below the mean for age and gender, and affected subjects usually have intellectual disability and language delay, with varying degrees of motor delay [109]. Various ocular abnormalities have been described in microcephaly, such as chorioretinal degeneration, pigmentary changes, optic disc coloboma and atrophy, falciform retinal folds, microphthalmia, hypoplastic fovea, strabismus, and nystagmus. Papageorgiou et al. investigated the morphology of the retina and ON in 27 adults and children with primary microcephaly by means of HH-OCT [110]. The authors found retinal changes on OCT in 85% of patients vs only 33% on funduscopy. Macular OCT alterations included disruption of the ellipsoid zone, persistent inner retinal layers, irregular foveal pits, reduced retinal thickness, and thinning of the ganglion cell layer [110]. Optic nerve OCT also showed reduced peripapillary retinal thickness in patients with microcephaly compared to controls for both temporal (275 vs 318 μm, p < 0.001) and nasal sides (239 vs 268 μm, p = 0.013) (Figure 7) [110].

Figure 7. Left column: Patient 17 (microcephaly lymphedema chorioretinal dysplasia) with smaller disc (bracket) and cup (line) diameter. Patient 22 (possible progressive encephalopathy with edema, hypsarrhythmia, and optic atrophy [PEHO]) with reduced peripapillary retinal nerve fiber layer (RNFL) thickness (arrows) and smaller disc (bracket) and cup (line) diameter. Patient 23 (possible PEHO) with reduced peripapillary RNFL thickness (arrows) and smaller disc (bracket) and cup (line) diameter. Right column: Healthy controls. Lower panel: Cross-sectional schematic diagrams represent mean values of optic nerve head measures of patients with microcephaly (left) and controls (right). Upper horizontal dotted lines represent horizontal offset (150 μm) used to determine cup diameters, and lower horizontal dotted lines indicate disc horizontal diameters. Vertical dotted lines show margins of rim areas. Significant values are marked with an asterisk. CDR = cup-to-disc ratio; n = nasal; t = temporal. Reprinted with permission from [110], licensed under https://creativecommons.org/licenses/by/4.0/.

10.2. Craniosynostosis

A variety of ophthalmological findings has been described in children with craniosynostosis, including strabismus and other oculomotor disorders, refractive errors, increased intracranial pressure and papilledema, proptosis, hypertelorism [111]. Previous studies using table-mounted OCT devices demonstrated that OCT in children with craniosynostosis is feasible and has a high potential as an adjunctive tool to screen for papilledema in a quantitative manner [112]. Peripapillary RNFL thickness could identify optic neuropathy in patients with craniosynostosis and appeared more sensitive at detecting optic atrophy than papilledema [113]. The sensitivity for detecting optic atrophy was 88%, for papilledema 60% and for either form of optic neuropathy 77% [113].

Swanson et al. first used the iVue portable OCT device to assess 40 children with craniosynostosis aged between 0 and 18 years under general anesthesia. The authors found that quantitative HH-OCT parameters of the peripapillary retina were correlated with invasively measured intracranial pressure (ICP) [114]. ICP correlated with maximal RNFL thickness (r = 0.60, p ≤ 0.001) and maximal retinal thickness (r = 0.53, p ≤ 0.001) [114]. HH-OCT has been also used by Rufai et al. to examine 50 children with syndromic and non-syndromic craniosynostosis between 0 and 18 years demonstrating good feasibility and repeatability of the method [115]. Examinations were mainly performed in an outpatient setting without sedation and, in some cases under general anesthesia. Interestingly, HH-OCT identified two cases of intracranial hypertension, one of which could not be detected by fundoscopy [115].

These findings suggest that HH-OCT is a promising tool in diagnosis, susurveillance,nd management of increased ICP and optic neuropathy in children with craniosynostosis.

10.3. Pediatric intracranial hypertension and pseudopapilledema (optic disc drusen)

An important potential application of HH-OCT in pediatric ophthalmology is in differentiating papilledema from pseudopapilledema. Papilledema is swelling of the optic nerve head due to raised intracranial pressure. On the other hand, pseudopapilledema is related to a congenital anomalous appearance that mimics papilledema, such as optic disc drusen. The diagnosis of papilledema often requires MRI of the brain, followed by a lumbar puncture, which are potentially traumatic for children and often need to be performed under general anesthesia. Additionally, the gold standard intraparenchymal ICP measurement is invasive and carries several risks.

Initial studies have shown that OCT is a convenient ancillary test that assists in distinguishing papilledema from pseudopapilledema and have recommended qualitative and quantitative OCT features for recognizing optic disc drusen: hyporeflective boot-shaped area adjacent to the drusen, isolated or clustered hyperreflective bands, signal-poor regions in the core, increased RNFL and angling of Bruch’s membrane opening toward the vitreous cavity [116–119]. Once a diagnosis of true papilledema has been made, OCT can be used to monitor treatment response and is potentially useful as a visual prognostic tool.

Increased RNFL thickness in temporal and superior quadrants and increased central macular thickness has been described in 11 children affected by pseudotumor cerebri compared to 37 normal controls [120]. The average RNFL for pseudotumor cerebri eyes was thicker compared with controls (125.7 vs 106.5 μm, p < 0.0001) and the macular volumes for pseudotumor cerebri versus control eyes were 7.21 versus 6.97 mm³, respectively (p = 0.005) [120]. Additionally, the presence of optic atrophy and photoreceptor damage on OCT has been associated with permanent visual dysfunction in pseudotumor cerebri [15].

HH-OCT has been utilized by Swanson et al. to assess the optic nerve head morphology in five patients with hydrocephalus and intracranial pressure [114]. In this study, noninvasive quantitative measures of the peripapillary retinal structure including maximal RNFL (r = 0.60, p ≤ 0.001), maximal retinal thickness (r = 0.53, p ≤ 0.001), and maximal anterior retinal projection (r = 0.53, p = 0.003), were correlated with invasively measured ICP [114]. In contrast, conventional clinical signs had low sensitivity for detecting intracranial hypertension [114]. The utility of HH-OCT is currently investigated in an ongoing prospective study, the RIO study: ‘Recognizing intracranial hypertension using hand-held OCT in children with craniosynostosis’ [121]. Clearly, HH-OCT has a unique potential to advance the current diagnostic and treatment algorithms in pediatric diseases with elevated ICP.

10.4. Pediatric optic neuritis

Optic nerve head swelling in children can be due to inflammation, malignant infiltration, compression by a mass lesion, infection, ischemia, multiple sclerosis (MS) or neuromyelitis optica [122,123]. Previous OCT studies of pediatric patients with MS have reported a significant decrease in mean RNFL thickness among optic neuritis eyes compared with healthy control eyes [124–127]. In the study by Yeh et al., in children with optic neuritis, average RNFL thickness was 97 μm in unaffected (n = 5) versus 89 μm in affected eyes (n = 9), and average RNFL thickness in healthy controls was 107 μm [124]. The differences between children with demyelinating disease and controls and between optic neuritis and non-optic neuritis eyes were statistically significant (p < 0.001) [124]. Eyes of MS patients without a history of optic neuritis were also found to have lower RNFL and GCL volumes than control eyes [128,129].

Future studies with HH-OCT may be of value in detecting RNFL thickening in retrobulbar optic neuritis without clinically apparent optic nerve head swelling. Additionally, longitudinal monitoring will enable assessment of morphological changes of the ON and evaluation of the degree and progression of optic atrophy.

11. Conclusions

HH-OCT is feasible in infants from birth in a variety of pediatric ocular conditions and can be performed reliably without sedation under routine outpatient settings. It is especially indicated in premature infants, nystagmus, albinism, children with unclear vision loss and suspected retinal dystrophies. HH-OCT has been already used to describe normal ON development from birth through adolescence and also to investigate the morphological alterations of the ON in prematurity [130]. In the field of pediatric neuropathies it has been used in primary congenital glaucoma, optic nerve hypoplasia, retinochoroidal and optic disc colobomas, optic pathway gliomas, and papilledema in craniosynostosis and hydrocephalus. Its future potential applications are within a broad range of pediatric optic neuropathies: pediatric pseudotumor cerebri and papilledema from different etiologies, pediatric optic neuritis, optic disc drusen, mitochondrial optic neuropathies, and other causes of nonglaucomatous optic atrophy such as traumatic optic neuropathy. Current research suggests that it is a powerful adjunctive tool for diagnosis, understanding of pathophysiology, monitoring of disease progression and treatment response and for defining structure-function correlations and visual prognostic biomarkers.

12. Expert opinion

HH-OCT has been established in the last few years in pediatric ocular diseases as a noninvasive examination method, which provides high-resolution, in-vivo images of the human retina and ON and can be performed without dilation under routine clinical settings. HH-OCT is especially useful in infants, toddlers and preverbal children, who cannot cooperate for VA and visual field testing. HH-OCT can be also used during sedation or general anesthesia for other procedures, as well as in the clinic or intensive care unit. Numerous studies have validated the reproducibility and reliability of quantitative retinal and ON parameters obtained by HH-OCT for diagnosis and longitudinal monitoring of pediatric ocular conditions. In addition, qualitative HH-OCT characteristics have been described in various pediatric ocular pathologies.

The potential of HH-OCT for correlating anatomic features, such as RNFL thickness and retinal thickness, with functional measures, such as VA and visual field, makes it an important tool for characterizing the severity and progression of ocular diseases, assessing treatment outcomes, determining screening frequency, as well as planning individualized surgical intervention. With recent technological improvements and rapidly expanding research efforts, HH-OCT will possibly revolutionize the diagnosis and management of optic neuropathies in children, while reducing the number of examinations under anesthesia. The first longitudinal studies have demonstrated the value of HH-OCT in monitoring of pediatric glaucoma and optic nerve glioma, and a variety of other optic neuropathies are currently being investigated. Obviously, HH-OCT should not be used as a replacement for other diagnostic modalities, such as MRI, but as an adjunctive tool to improve understanding of disease pathophysiology, enhance earlier diagnosis and facilitate appropriate care pathways, to help avoiding invasive and costly tests under general an aesthetic, to provide an indication of possible prediction of VA, and identify new therapeutic targets.

Optic neuropathies in children are relatively rare, implying difficulties in performing research with large study groups. However, their visual consequences can be devastating and severe irreversible vision loss complicated by amblyopia can occur from the first years of life, posing a lifelong burden on the health and wealth of affected individuals. Hence, the development of tools, such as HH-OCT, which enables early diagnosis, identifies prognostic biomarkers, monitors disease progression, and response to treatment, is of outmost importance. The need to establish an early and accurate diagnosis will eventually be more important in the future due to the development of novel gene therapies.

Article highlights

• Ηand-held spectral domain optical coherence tomography (HH-OCT) enables noncontact, high-resolution imaging of the retina and optic nerve (ON) in non-sedated newborns, infants, and young children.

• In recent years, HH-OCT has been used increasingly in children with optic neuropathies, to provide high-resolution characterization of the ON head, peripapillary retinal nerve fiber layer, and cellular layers of the macula, all of which can be used to assess the severity of ON disease qualitatively and quantitatively.

• HH-OCT has been used successfully to characterize normal retina and ON development and to establish normative data of the retina and optic nerve in infancy and childhood.

• In primary congenital glaucoma, there were novel HH-OCT findings, such as the increase in cup depth, the reduction of the foveal pit width and a non-detectable external limiting membrane, in addition to thinner retinal nerve fiber layer (RNFL), larger cupping and smaller rim area, which were similar to those in primary open-angle glaucoma.

• Because of the difficulty in obtaining reliable visual fields, measuring intraocular pressure, and quantifying ON damage in children with glaucoma, OCT is particularly well suited for the diagnosis and monitoring of pediatric glaucoma.

• HH-OCT is useful for the characterization of congenital optic disc anomalies and has been introduced in patients with optic nerve hypoplasia and children with choroidal neovascular membranes associated with retinochoroidal and ON coloboma.

• In children with optic pathway gliomas HH-OCT is extremely useful in assessing RNFL thickness, in order to longitudinally monitor disease progression in children, who are unable to cooperate with VA or visual field testing.

• In secondary epipapillary retinoblastoma, HH-OCT could detect subclinical tumors not evident on funduscopy.

• HH-OCT has been introduced in assessment of neurological conditions, such as in children with microcephaly, craniosynostosis, and intracranial hypertension.

• HH-OCT is a promising tool in diagnosis, surveillance, and management of pediatric intracranial hypertension, because HH-OCT parameters of the peripapillary retina were correlated with invasively measured intracranial pressure and HH-OCT also identified cases of intracranial hypertension, which could not be detected by fundoscopy.

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.