Graefe's Archive for Clinical and Experimental Ophthalmology (2022) 260:1153–1160 https://doi.org/10.1007/s00417-021-05424-5 RETINAL DISORDERS Retinal thinning in phenylketonuria and Gaucher disease type 3 Susanne Hopf1  · Alexander K. Schuster1 · Julia B. Hennermann2 · Norbert Pfeiffer1 · Susanne Pitz1,3 Received: 3 August 2021 / Revised: 14 September 2021 / Accepted: 20 September 2021 / Published online: 12 October 2021 © The Author(s) 2021 Abstract Purpose Retinal alterations in inherited metabolic diseases associated with neurodegeneration are poorly studied. The objec- tive was to study retinal thickness, specifically the components of the ganglion cell complex (GCC)—nerve fiber layer (NFL), ganglion cell layer (GCL), and inner plexiform layer (IPL)—using spectral-domain optical coherence tomography (SD-OCT) in two different diseases with potential dopaminergic depletion, phenylketonuria (PKU) and Gaucher disease type 3 (GD3). Methods Retinal layers in 19 patients with PKU, 15 patients with GD3, and 93 healthy individuals were measured using peripapillary ring scan and macular SD-OCT. Linear mixed models were computed including an adjustment for age, sex, and spherical equivalent. We calculated Spearman’s rank correlations between retinal layer measurements and clinical and/ or laboratory parameters. Results Thinning of total retinal thickness was found in the macular inner ring (p = 0.002), and outer ring (p = 0.012), spar- ing the fovea (p = 0.12) in PKU, while in GD3, all subfields were thinned (fovea p < 0.001, inner ring p = 0.047, outer ring 0.07). In both conditions, thinning was most evident in the NFL, GCL, and IPL, while OPL (outer plexiform layer) was thickened. Peripapillary retinal nerve fiber layer measurements remained normal. GCL and IPL in PKU correlated with tyrosine serum concentration. Conclusion Thinning of the NFL, GCL, and IPL, with thickened OPL, are both found in PKU and in GD3. Low dopamine concentrations in the retina might promote these effects. However, these data do not give evidence that retinal measurements can be used as a biomarker for disease severity in patients with GD3. Key messages Patients with phenylketonuria and Gaucher disease type 3 are known to present total retinal thinning in the macula and reduced retinal layer measurements of the retinal ganglion cell complex. Each of the individual components of the ganglion cell complex -NFL, GCL, and IPL -are reduced in thickness, while OPL is thickened in patients with phenylketonuria. A similar pattern of retinal alteration is evident in patients with Gaucher disease type 3. Keywords Ganglion cell layer · Ganglion cell complex · Retinal thinning · PKU · Gaucher · Lysosomal storage disease * Susanne Hopf hopf101@gmail.com; susanne.hopf@unimedizin-mainz.de Introduction 1 Department of Ophthalmology, University Medical Center of the Johannes Gutenberg-University Mainz, Optical coherence tomography (OCT), a non-invasive Langenbeckstraße 1, 55131 Mainz, Germany technique with a high resolution, enables to describe 2 Villa Metabolica, Department of Pediatric and Adolescent retinal structures almost at a cellular level in vivo. OCT Medicine, University Medical Center Mainz, Mainz, layers correlate well with histological findings of the Germany retina [1, 2]. The use of this imaging technique to dem- 3 Orbital Center, Ophthalmic Clinic, Bürgerhospital Frankfurt, onstrate neurodegeneration of the retina or the optic disc Frankfurt, Germany Vol.:(012 3456789) 1154 Graefe's Archive for Clinical and Experimental Ophthalmology (2022) 260:1153–1160 in neurological and neuro-ophthalmological conditions is Methods growing [3]. Retinal correlates have been proved primarily in Parkinson’s disease, Alzheimer’s disease, and multiple Study population sclerosis, but also in rare conditions such as amyotrophic lateral sclerosis, and Huntington’s disease [3–5]. They This study included 19 patients aged 6 to 46 years with resemble changes found in glaucoma by thinning of the mild, or classical, phenylketonuria or tetrahydrobiopterin- macular ganglion cell complex [6], defined by the three deficient hyperphenylalaninemia, 15 patients aged 6 to innermost retinal layers (nerve fiber layer, NFL; ganglion 44 years with Gaucher disease type 3, and 93 controls cell layer, GCL; and inner plexiform layer, IPL) [7]. On aged 6 to 75 years. The patients’ inclusion criteria were the one hand, thinning of the combined ganglion cell-inner genetical and/or biochemical prove of their disease. Partic- plexiform layer (GCIPL) and thinning of the peripapillary ipants under 6 years of age were excluded. Controls were retinal nerve fiber layer (pRNFL) correlate with neurode- included if they had no ocular disease and age-appropriate generative processes of the entire CNS, such as in multiple visual function, as well as no relevant systemic disease sclerosis [3, 4], or with nigrostriatal dopaminergic degen- (e.g., neurodegeneration). Controls were recruited stratify- eration related to Parkinson’s disease [8]. On the other ing for age (seven subgroups with at least 10 participants hand, retinal layer measurements may provide structural were built for an even age distribution). evidence for dysfunction in the fovea and parafovea and The study was approved by the Medical Ethical Commit- retinal dopamine loss, as is assumed in Parkinson’s disease tee of the State Chamber of Medicine of Rhineland Palati- [8–10]. The retina is one of the tissues in the body with nate in Mainz, Germany (reference number 837.373.14). highest dopamine concentration [11]. Dopamine recep- All persons or their parents/guardians gave their writ- tors are expressed in retinal cells throughout the retina ten informed consent prior to inclusion in the study. The with varying functions depending on the receptor subtype research adhered to the tenets of the Declaration of Helsinki. and the cell type [12]. The retinal dopaminergic system is involved in eye growth, light adaptation, circadian rhyth- micity, and cell survival [13]. Dopamine has an important Ophthalmic examination procedure role in uncoupling horizontal and amacrine cell junctions [12]. Dopaminergic amacrine cells are located in the inner The examination included non-cycloplegic auto-refraction nuclear layer (INL) [14, 15]. measurements (NIDEK AR-360A, Nidek Co., Japan), best- Ganglion cell complex (GCC) thinning has recently corrected visual acuity testing, slit lamp biomicroscopy, been reported for the first time in the inherited metabolic and fundus examination, as well as orthoptic examination, diseases phenylketonuria[16] and Gaucher disease type 3 which were published elsewhere [21–23]. Spherical equiv- [17]. In both conditions, a dopaminergic depletion can be alents defined as the sum of the spherical power and half speculated. In PKU, dopamine depletion has been related of the cylindric power were used in the statistic models. to reduced tyrosine uptake into the brain and reduced Imaging of the optic nerve head and the macula was tyrosine-hydroxylase activity [18]. In Gaucher type 1, carried out using spectral-domain (SD) optical coherence GCC thinning was demonstrated primarily in the pres- tomography (OCT) (Spectralis, Heidelberg Engineering ence of parkinsonian features or other clinical markers GmbH, Heidelberg, Germany) with automatic real-time of early neurodegeneration (hyposmia, cognitive impair- function for image averaging. We acquired a peripapillary ment, parkinsonian motor signs) [17, 19, 20]. Thus, in OCT and a macular OCT. The peripapillary retinal nerve both diseases, changes in neurotransmission may be one fiber layer (pRNFL) was imaged with a diameter of 12° (cor- mechanism leading to functional and morphological retina responding to 3.47 mm in the standard eye), and a stand- alteration. ard corneal curvature of 7.7 mm. For the macular OCT, 49 Our aim was to study segmental retinal layers, specifically horizontal single scans were acquired. After semi-automated the components of the ganglion cell complex (GCC)—nerve segmentation of the retinal layers as provided by the OCT fiber layer (NFL), ganglion cell layer (GCL), and inner plexi- software (Heidelberg Eye Explorer version 1.10.2.0, viewing form layer (IPL)—by means of spectral-domain (SD) optical module 6.9.5.0; HEYEX, Heidelberg, Germany), all scans coherence tomography (OCT) in two different diseases with were assessed regarding their quality by a board-certified potential dopaminergic depletion, phenylketonuria (PKU) ophthalmologist (SH). Those with segmentation errors and Gaucher disease type 3 (GD3). The second aim was to were corrected, or excluded in cases of poor image qual- identify whether retinal layer measurements correlate with ity. In cases of poor data in only one sector, this sector was established disease features and if they may be used as a excluded prior to analysis; if more sectors were affected, biomarker for disease severity. This should be of special then the complete OCT dataset of this eye was excluded. interest for patients with GD3. 1 3 Graefe's Archive for Clinical and Experimental Ophthalmology (2022) 260:1153–1160 1155 Mean retinal thickness of 9 macular layers was used for control for the inclusion of one and two eyes of a study par- the analysis: total retinal layer thickness of the macula, nerve ticipant (as random effect). A further adjustment for age, fiber layer (NFL), ganglion cell layer (GCL), inner plexi- sex, and spherical equivalent was included in the statistical form layer (IPL), inner nuclear layer (INL), outer plexiform analysis. Full thickness measurements are more susceptible layer (OPL), outer nuclear layer (ONL), outer retinal layer to these parameters, than single layer measurements, which (ORL) (being limited by the external limiting membrane and is why we focused on single layer correlation analysis as Bruch’s membrane, this layer corresponds to the photorecep- follows. Spearman’s rank correlation was conducted to tors), and retinal pigment epithelium (RPE). correlate thinned or thickened layers (e.g., GCL or OPL, The 6-mm macular scan measurements were classified respectively) with disease-specific variables of PKU (cur- according to the ETDRS segments (“Early Treatment Dia- rent phenylalanine and tyrosine serum concentration, early- betic Retinopathy Study” subfields). Central zone, inner vs. late-treated PKU, and disease type) and GD3 (mSST, ring, and outer ring with diameters of 1, 3, and 6 mm, phenotype severity, horizontal and vertical eye movements). respectively, were included in the analysis. The average The other retinal layers, which were not different to con- of all points within the central zone (1 mm diameter) was trols, or which were underrepresented in the specific subfield defined as foveal thickness, the inner ring (1 to 3 mm) as (fovea), were not further analyzed. Correlation coefficient parafoveal thickness, and the outer ring (3 to 6 mm) as peri- rho of ≥ 0.5 was regarded a moderate correlation, correla- foveal thickness. tion coefficient of ≥ 0.3 was considered a weak correlation, and < 0.3 was considered a no correlation. Statistical analy- Clinical data sis was performed using R version 4.0.4. All p-values should be regarded as continuous parameters that reflect the level Typical variables described for disease stage in PKU of evidence from our explorative analysis and are therefore and GD3 were determined and obtained from the reported exactly. patient’s records. For PKU, we used first, current phe- nylalanine serum concentrations of each individual (mean 693 µmol/l ± 384 µmol/l); second, current tyros- Results ine serum concentrations of each individual (mean 105  µmol/l ± 60  µmol/l); third, disease treatment (16 All patients examined were included in the study. From early-treated vs. three late-treated individuals with PKU); the 19 PKU patients (mean age 20 ± 12 years), 15 GD3 and fourth, disease type (mild phenylketonuria with patients (mean age 20 ± 10 years), and 93 controls (mean untreated blood phenylalanine concentrations of less than age 32 ± 17 years), we excluded 6, 8, and 8 eyes respectively 1000–1200 µmol/l) [24], classical phenylketonuria, and for the pRNFL analysis due to poor image quality, and 0, 2, tetrahydrobiopterin-deficient hyperphenylalaninemia). and 6 eyes respectively for the macular OCT analysis. For GD3, we used first, the modified severity scoring tool (mSST), which is based on twelve domains including hori- Macular full thickness and macular layers zontal gaze palsy, cranial nerve palsy, seizures and age at first seizures, cognitive ability, ataxia, tremor, spasticity, Phenylketonuria rigidity, dysphagia, dysarthria, and spinal alignment [25]. Second, we considered the phenotype severity (mild, inter- The OCT measurements indicate a significant thinner total mediate, severe), of which intermediate phenotype was retinal thickness in the inner and outer ring, but not in the associated with homozygous L444P mutation [22]. Third/ fovea. This pattern affected the NFL, GCL, IPL, and ONL, fourth, we included horizontal/vertical peak velocity of while OPL was thickened. The differences were most evi- reflexive saccades in GD3 (69°/s ± 58°/s and 192°/s ± 92°/s, dent in the NFL, GCL, and IPL (see Table 1) and more evi- respectively). dent than in GD3. The remaining layers (INL, and ORL) did not differ from controls. RPE was thinned in the inner ring Statistical analysis segment, but the difference was not significant after adjust- ment for sex, age, and spherical equivalent. Medians, interquartile ranges, minimums, and maximums GCL correlated with current tyrosine serum concen- were calculated for all continuous variables. For variables tration (outer ring: rho = 0.70, p = 0.0008 (< 0.001); inner distributed normally, means and standard deviations were ring: rho = 0.51, p = 0.025), as well as IPL did (inner ring: computed. For dichotomous variables, absolute and relative rho = 0.61, p = 0.006). ONL correlated inversely with current frequencies were computed. phenylalanine serum concentration (outer ring: rho = − 0.59, To analyze the differences of retinal thickness with respect to PKU and GD3, we used linear mixed models to 1 3 1156 Graefe's Archive for Clinical and Experimental Ophthalmology (2022) 260:1153–1160 Table 1 Distribution of retinal layer measurements from SD-OCT in phenylketonuria, Gaucher disease type 3, and controls. A linear mixed model was used for statistical analysis. In the adjusted model, age, sex, and spherical equivalent were included Macular OCT layers and Control eyes (n = 180) PKU eyes (n = 36) GD3 eyes (n = 30) p-value p-value ETDRS segments PKU vs. control GD3 vs. control Mean (SD) [µm] crude model crude model (adj. model) (adj. model) Total retinal thickness   Fovea 278.3 (19.2) 271.0 (16.7) 260.2 (10.3) 0.12 (0.38) < 0.001 (< 0.001)   Inner ring 343.8 (12.8) 331.9 (18.3) 336.1 (15.0) 0.002 (0.001) 0.047 (0.009)   Outer ring 306.9 (20.7) 295.1 (16.9) 298.1 (18.9) 0.012 (< 0.001) 0.07 (0.014) 1) NFL   Fovea 12.7 (2.1) 11.7 (1.6) 11.8 (1.6) 0.05 (0.24) 0.13 (0.10)   Inner ring 21.7 (1.8) 20.2 (1.8) 21.3 (1.6) < 0.001 (0.006) 0.33 (0.40)   Outer ring 36.3 (4.0) 33.0 (3.9) 34.3 (5.4) 0.004 (0.001) 0.08 (0.019) 2) GCL   Fovea 16.1 (4.4) 14.1 (2.7) 13.9 (3.0) 0.07 (0.09) 0.06 (0.008)   Inner ring 52.3 (4.2) 46.5 (6.9) 50.0 (5.9) < 0.0001 (< 0.001) 0.10 (0.004)   Outer ring 35.9 (3.1) 33.7 (3.6) 34.5 (4.0) 0.019 (0.001) 0.16 (0.014) 3) IPL   Fovea 22.0 (3.7) 20.4 (2.3) 19.7 (2.2) 0.07 (0.14) 0.014 (0.002)   Inner ring 42.7 (2.8) 39.4 (3.3) 41.3 (3.5) < 0.0001 (< 0.001) 0.12 (0.013)   Outer ring 29.5 (2.5) 28.4 (2.6) 29.2 (2.8) 0.15 (0.021) 0.73 (0.22) 4) INL   Fovea 18.7 (4.6) 17.2 (3.1) 15.5 (2.8) 0.16 (0.82) 0.006 (0.019)   Inner ring 40.3 (2.7) 41.1 (3.6) 40.2 (2.7) 0.13 (0.33) 0.88 (0.50)   Outer ring 33.3 (2.4) 34.4 (2.4) 33.9 (2.5) 0.05 (0.78) 0.44 (0.47) 5) OPL   Fovea 26.2 (5.3) 27.0 (4.3) 22.6 (3.4) 0.47 (0.76) 0.005 (0.002)   Inner ring 33.5 (3.8) 35.8 (4.6) 33.6 (3.7) 0.008 (0.06) 0.95 (0.92)   Outer ring 27.1 (1.7) 28.2 (2.2) 28.1 (2.3) 0.005 (0.005) 0.030 (0.014) 6) ONL   Fovea 92.5 (10.3) 91.5 (9.7) 86.5 (5.8) 0.58 (0.93) 0.024 (0.11)   Inner ring 72.2 (8.6) 67.9 (9.5) 69.6 (5.0) 0.03 (0.12) 0.26 (0.37)   Outer ring 59.8 (6.5) 57.0 (7.4) 57.4 (4.6) 0.07 (0.043) 0.19 (0.09) 7) ORL (PR)   Fovea 90.6 (4.1) 89.9 (4.4) 91.0 (4.9) 0.49 (0.30) 0.74 (0.64)   Inner ring 81.6 (2.3) 80.9 (2.2) 80.7 (2.4) 0.23 (0.11) 0.14 (0.06)   Outer ring 78.3 (2.1) 77.9 (2.5) 77.7 (2.0) 0.68 (0.25) 0.19 (0.11) 8) RPE   Fovea 17.1 (1.8) 16.7 (1.7) 17.2 (2.0) 0.45 (0.43) 0.80 (0.82)   Inner ring 14.8 (1.5) 14.1 (1.2) 14.4 (1.2) 0.029 (0.28) 0.19 (0.68)   Outer ring 13.1 (1.2) 13.1 (1.2) 13.1 (1.1) 0.61 (0.25) 0.94 (0.70) OCT optical coherence tomography, SD standard deviation, GD3 Gaucher disease type 3, PKU phenylketonuria, NFL nerve fiber layer, GCL ganglion cell layer, IPL inner plexiform layer, INL inner nuclear layer, OPL outer plexiform layer, ONL outer nuclear layer, ORL outer retinal layer (being limited by the external limiting membrane and Bruch’s membrane, this layer corresponds to the photoreceptors), RPE retinal pig- ment epithelium, adj. model adjusted model p = 0.01; inner ring: − 0.63, p = 0.004). Early-treated PKU p = 0.014) and thick IPL (inner ring: rho = 0.54, p = 0.02) patients had rather thick GCL (inner ring: rho = 0.55, compared to late-treated patients. 1 3 Graefe's Archive for Clinical and Experimental Ophthalmology (2022) 260:1153–1160 1157 Gaucher disease type 3 Discussion Thinning of total retinal thickness was found in all sub- This study is one of the first, reporting OCT measurements fields (fovea, inner ring, and outer ring) compared to the in phenylketonuria and Gaucher disease type 3, besides the controls, even after adjusting for sex, age, and spherical publications from Serfozo et al. [16, 26] and Tantawy et al. equivalent (Table 1). The retinal layers affected by thin- [17]. Our main finding, GCC reduction, is in line with these ning were NFL (outer ring significantly), GCL (inner ring two studies [16, 17, 26]. We additionally demonstrate that significantly), and IPL (slightly), while OPL (outer ring) each of the individual components (NFL, GCL, and IPL) is was thickened, and the remaining layers (INL, ONL, ORL, reduced in thickness in phenylketonuria. A similar pattern and RPE) showed no difference in thickness to the con- of retinal thinning was evident in GD3. The retinal measure- trols. Individual foveal layers were normal (when only ments found in both conditions resemble changes seen dur- considering the layers ONL outwards, because the layers ing aging: GCL and IPL both thin out, while OPL thickens NFL to OPL are very small and do not yield reliable OCT with age [27]. data in general). However, total retinal thinning was most evident in the fovea. Macular thickness NFL outer ring correlated inversely with mSST (outer ring: rho = − 0.49; p = 0.046; inner ring: rho = − 0.39, Alterations of the ganglion cell complex in various neuro- p = 0.11), and OPL correlated inversely with horizontal peak degenerative diseases overlap and tend to show similarities. velocity (OPL outer ring: rho = − 0.57, p = 0.020; OPL inner We found GCC reduction both in PKU and in GD3. ring: rho = − 0.51, p = 0.040). In GD3, no correlations were found for foveal total retinal thickness, and GCL, and none Phenylketonuria Our data support the findings of Serfozo with vertical peak velocity or phenotype severity (mild/ et al., who reported total retinal thinning in the parafoveal intermediate/severe). and perifoveal region, sparing the fovea [26], in early-treated phenylketonuria, and reduced GCC thickness (average, supe- Peripapillary RNFL rior, and inferior quadrants) [16]. We further demonstrated OPL thickening in the inner and outer ring. Inverse cor- The global peripapillary RNFL did not differ signifi- relation of phenylalanine serum concentration with retinal cantly between the groups with 96 µm in PKU and GD3 measurements was, at most, inconsistently found [26]. We and 96.7 µm in control eyes. Regarding the distinct peri- found tyrosine serum concentrations correlating with GCL papillary RNFL sectors, only the temporal-inferior sec- and IPL, which might indirectly indicate low cerebral tyros- tor was significantly thicker in PKU eyes compared to ine and dopamine concentrations. controls (p = 0.029), but the difference did not remain significant after adjusting for sex, age, and spherical Gaucher disease In line with previous investigations, we equivalent (Table 2). The other sectors were normal in confirm significant retinal thinning in the NFL (outer ring), both diseases. GCL (outer and inner ring), and IPL (inner ring) after adjust- ing for sex, age, and refraction, and thickening of the OPL in GD3. Our data thus attest reduced GCC as reported by Table 2 Distribution of Peripapillary RNFL Control eyes PKU eyes GD3 eyes p-value PKU p-value GD3 peripapillary retinal nerve fiber Mean [µm] (n = 178) (n = 30) (n = 24) crude model crude model layer thickness from SD-OCT [adj. model [adj. model in phenylketonuria, Gaucher disease type 3, and controls. Global (average) 96.7 ± 9.8 96.1 ± 9.5 96.0 ± 5.9 0.61 [0.63] 0.25 [0.19] A linear mixed model was used for statistical analysis. In Temporal 69.9 ± 10.5 66.8 ± 10.2 71.1 ± 13.7 0.50 [0.14] 0.90 [0.36] the adjusted model, age, sex, Temoral-superior 135.0 ± 18.0 134.8 ± 22.3 132.1 ± 20.0 0.71 [0.45] 0.88 [0.50] and spherical equivalent were Temporal-inferior 139.3 ± 19.3 148.5 ± 17.5 135.4 ± 21.0 0.029 [0.07] 0.33 [0.26] included Nasal 73.6 ± 15.6 71.1 ± 9.7 70.6 ± 11.1 0.39 [0.40] 0.47 [0.59] Nasal-superior 106.3 ± 19.8 107.8 ± 19.1 109.3 ± 16.0 0.91 [0.85] 0.38 [0.58] Nasal-inferior 105.8 ± 23.6 102.9 ± 17.1 107.8 ± 34.1 0.63 [0.29] 0.44 [0.98] RNFL retinal nerve fiber layer, ± = standard deviation, GD3 Gaucher disease type 3, PKU phenylketonuria, adj. model adjusted model 1 3 1158 Graefe's Archive for Clinical and Experimental Ophthalmology (2022) 260:1153–1160 Tantawy et al., who divided a cohort of GD patients aged 11 quadrants—in early-treated phenylketonuria [16], we did to 29 years into a group with parkinsonian features (n = 11) not detect reduction in pRNFL thickness. The values of our and a group without (n = 37), independently of the type of pRNFL data, however, are in line with those measured by GD. Their results were that GCC thickness differed between Serfozo et al., namely 96.3 µm (± 9.9 µm) in early-treated young patients with parkinsonian features and those without phenylketonuria. The different outcome might be attributed parkinsonian features. However, between GD1 (n = 14) and to thicker pRNFL in their controls (101.9 ± 7.2 µm) com- GD3 (n = 34), GCC thickness did not differ significantly, pared to ours [16]. although all GD patients together (93.1 (± 7.0) µm) differed from the controls (98.7 (± 9.6) µm) [17]. Similarly, thinning Gaucher disease Peripapillary RNFL reduction as of the retinal GCC was associated with potential clinical detected by Matos et  al. in a GD3 patient is not con- markers of early neurodegeneration in GD1 (n = 11) or GBA firmed by our data. In this case, the authors attributed mutation carriers (intermediate level of glucocerebrosidase their findings to either glaucoma of normal pressure or a activity) in a study conducted by McNeill et al. [20]. GD1 nervous degeneration [33]. While no other studies have patients without neurodegenerative symptoms do not show investigated pRNFL in GD3, Weill et. al reported abnor- retinal GCC thinning [19, 20], although subsections of the mal pRNFL scans in one-third of GD1 patients with sig- GCC thickness (e.g., of the outer macular GCC nasally and nificant thinning of the average, superior, and inferior inferiorly) revealed significant thinning [19]. In GD1, no pRNFL [19]. This pattern of damage was supposed to correlation was found between retinal measurements and match a magnocellular type, which is also characteristic disease severity [19]. In GD3, we detected a weak corre- in Alzheimer’s disease [34]. From our data, we cannot lation between modified severity scoring tool, and retinal conclude a damage or a pattern of damage at the level of NFL, and no correlation between saccadic measurements the optic disc primarily due to GD3. and retinal layer measurements. Limitations and perspectives The pathways of retinal damage including that of reti- nal ganglion cells are not fully understood in both diseases. This study has several limitations. First, we did not have Changes in neurotransmitter metabolism are discussed to complete long-term blood biomarker data of the PKU play a role. Retinal dopamine deficiency is discussed to play group (mean phenylalanine serum concentrations over a role in primary retina degeneration and secondary loss the past 5 or 10 years) for the analysis. We also could of dopamine-regulated neurons [26]. Besides disturbances not provide data of parkinsonian motor signs (bradykin- in neurotransmitter metabolism, other changes may lead to esia, rigidity, rest tremor) [17], or prodromal symptoms morphological changes of the retina in both disorders. In (hyposmia, cognitive impairment, hallucinations, depres- PKU, this might be a direct neurotoxic effect of phenylala- sion, sleep disorders, and autonomic dysfunction) [20] nine [28]. In the lysosomal storage disorder GD, the reduced in the GD3 group. Unfortunately, these data were not activity of β-glucocerebrosidase is associated with accumu- available in our cohort. Longitudinal data is still miss- lation of α-synuclein, inhibition of apoptosis, and reduced ing, which could elucidate whether progression occurs mitochondrial function with associated oxidative stress [29, over time. This is even more important, as progression in 30]. Clinically, vitreous fluid of GD3 patients may contain neuro-ophthalmologic diseases is subject to high interin- visible opacities as well as Gaucher cells, and high concen- dividual variation. tration of glucosylceramide [31, 32]. In addition, vascular abnormalities with tortuosity, and occlusion were discussed Conclusion to induce retinal damage [19]. In the present cohort, only three patients presented increased tortuosity of retinal ves- This OCT study with PKU and GD3 patients confirmed that sels bilaterally, one patient presented caliper changes of reti- retinal thickness is reduced, to elucidate, at the level of the nal vessels, and two patients showed peripapillary atrophy NFL, GCL, IPL, and ONL, while OPL is thickened in PKU of the outer retina [22]. patients. The same is held true for GD3 although with fewer As follow-up data are missing, we cannot exclude that in significance. Individual follow-up examinations are required GD3 metabolic imbalances during embryonic development for evaluation and detection of a progression of retinal neu- already influence a normal retina development. rodegeneration. This is important, because current therapies for these conditions might interfere with progressive mor- Peripapillary RNFL phological changes of the retina. Phenylketonuria While Serfozo et al. found that average pRNFL was reduced—without prove of reduction in the 1 3 Graefe's Archive for Clinical and Experimental Ophthalmology (2022) 260:1153–1160 1159 Author contribution All authors contributed to the study conception sclerosis: a review. Mult Scler Relat Disord 22:77–82. https://d oi. and design. Data collection was performed by Susanne Hopf, Julia org/ 10.1 016/j.m sard. 2018. 03. 007 Hennermann, and Susanne Pitz. Data analysis was performed by Alex- 5. Yap TE, Balendra SI, Almonte MT, Cordeiro MF (2019) Reti- ander Schuster. Management and interpretation of data were conducted nal correlates of neurological disorders. Ther Adv Chronic Dis by Susanne Hopf and Alexander Schuster. The first draft of the manu- 10:2040622319882205. https:// doi. org/ 10. 1177/ 20406 22319 script was written by Susanne Hopf, and all authors commented on 882205 previous versions of the manuscript. All authors read and approved 6. Chrysou A, Jansonius NM, van Laar T (2019) Retinal layers in the final manuscript. Parkinson’s disease: a meta-analysis of spectral-domain optical coherence tomography studies. Parkinsonism Relat Disord 64:40– Funding Open Access funding enabled and organized by 49. https:// doi. org/ 10. 1016/j.p arkr eldis. 2019. 04. 023 Projekt DEAL. 7. Tan O, Chopra V, Lu AT-H et al (2009) Detection of macular gan- glion cell loss in glaucoma by Fourier-domain optical coherence The data are included in the manuscript. Participants tomography. Ophthalmology 116:2305-2314.e2. https:// doi. org/ Data availability of the study did not agree for their individual raw dataset to be shared 10. 1016/j.o phtha. 2009. 05.0 25 publicly, so supporting data is not available. 8. Lee J-Y, Ahn J, Shin JY, Jeon B (2021) Parafoveal change and dopamine loss in the retina with Parkinson’s disease. Ann Neurol 89:421–422. https:// doi.o rg/ 10. 1002/ ana.2 5972 Declarations 9. Lee J-Y, Ahn J, Kim TW, Jeon BS (2014) Optical coherence tomography in Parkinson’s disease: is the retina a biomarker? J Ethics approval The study was approved by the Medical Ethical Com- Parkinsons Dis 4:197–204. https:// doi.o rg/ 10.3 233/J PD- 130306 mittee of the State Chamber of Medicine of Rhineland Palatinate in 10. Tian T, Zhu X-H, Liu Y-H (2011) Potential role of retina as a bio- Mainz, Germany (reference number 837.373.14). The research adhered marker for progression of Parkinson’s disease. Int J Ophthalmol to the tenets of the Declaration of Helsinki. 4:433–438. https://d oi. org/1 0. 3980/j.i ssn. 2222-3 959. 2011. 04. 21 11. Brandies R, Yehuda S (2008) The possible role of retinal dopa- Consent to participate Informed consent was obtained from all indi- minergic system in visual performance. Neurosci Biobehav Rev vidual participants included in the study, or from legal guardians. 32:611–656. https://d oi. org/1 0. 1016/j.n eubio rev. 2007.0 9. 004 12. Nguyen-Legros J, Versaux-Botteri C, Vernier P (1999) Dopamine Consent for publication Participants signed informed consent regard- receptor localization in the mammalian retina. Mol Neurobiol ing publishing their data. 19:181–204. https:// doi. org/ 10.1 007/B F0282 1713 1 3. Witkovsky P (2004) Dopamine and retinal function. Doc Oph- thalmol 108:17–40. https://d oi.o rg/1 0.1 023/b:d oop.0 00001 9487. Conflict of interest The authors declare no competing interests. 88486.0 a 14. Frederick JM, Rayborn ME, Laties AM et al (1982) Dopaminergic Open Access This article is licensed under a Creative Commons Attri- neurons in the human retina. J Comp Neurol 210:65–79. https:// bution 4.0 International License, which permits use, sharing, adapta- doi. org/ 10. 1002/ cne. 90210 0108 tion, distribution and reproduction in any medium or format, as long 1 5. Eglen SJ, Raven MA, Tamrazian E, Reese BE (2003) Dopamin- as you give appropriate credit to the original author(s) and the source, ergic amacrine cells in the inner nuclear layer and ganglion cell provide a link to the Creative Commons licence, and indicate if changes layer comprise a single functional retinal mosaic. J Comp Neurol were made. The images or other third party material in this article are 466:343–355. https://d oi. org/ 10.1 002/c ne.1 0891 included in the article's Creative Commons licence, unless indicated 1 6. Serfozo C, Barta AG, Horvath E et al (2020) Altered visual func- otherwise in a credit line to the material. If material is not included in tions, macular ganglion cell and papillary retinal nerve fiber layer the article's Creative Commons licence and your intended use is not thickness in early-treated adult PKU patients. Mol Genet Metab permitted by statutory regulation or exceeds the permitted use, you will Rep 25:100649. https:// doi. org/ 10.1 016/j.y mgmr.2 020. 100649 need to obtain permission directly from the copyright holder. To view a 17. Tantawy AAG, Adly AAM, Hashem NH et al (2020) Ganglion copy of this licence, visit http://c reati vecom mons.o rg/l icens es/b y/4.0 /. cell complex thinning in young Gaucher patients: relation to pro- dromal parkinsonian markers. Mov Disord 35:2211–2219. https:// doi.o rg/ 10.1 002/ mds. 28256 18. Pascucci T, Giacovazzo G, Andolina D et al (2012) In vivo cat- echolaminergic metabolism in the medial prefrontal cortex of References ENU2 mice: an investigation of the cortical dopamine deficit in phenylketonuria. J Inherit Metab Dis 35:1001–1009. https://d oi. 1. Moreira-Neto CA, Bergeron S, Coblentz J et al (2019) Optimizing org/1 0.1 007/ s10545-0 12- 9473-2 optical coherence tomography and histopathology correlation in 1 9. Weill Y, Zimran A, Zadok D et al (2020) Macular ganglion cell retinal imaging. Can J Ophthalmol 54:280–287. https:// doi. org/ complex and peripapillary retinal nerve fiber layer thinning in 10.1 016/j.j cjo.2 018.0 6.0 24 patients with type-1 Gaucher disease. Int J Mol Sci 21. https:// 2. Xie W, Zhao M, Tsai S-H et al (2018) Correlation of spectral doi. org/ 10. 3390/i jms21 19702 7 domain optical coherence tomography with histology and elec- 20. McNeill A, Roberti G, Lascaratos G et al (2013) Retinal thinning tron microscopy in the porcine retina. Exp Eye Res 177:181–190. in Gaucher disease patients and carriers: results of a pilot study. https://d oi.o rg/1 0. 1016/j. exer. 2018. 08.0 03 Mol Genet Metab 109:221–223. https://d oi.o rg/1 0.1 016/j.y mgme. 3. Papadopoulou A, Oertel FC, Zimmermann H et al (2018) Optical 2013. 04.0 01 coherence tomography in disorders of the central nervous sys- 2 1. Hopf S, Nowak C, Hennermann JB et al (2020) Saccadic reaction tem. Klin Monbl Augenheilkd 235:1242–1258. https://d oi.o rg/1 0. time and ocular findings in phenylketonuria. Orphanet J Rare Dis 1055/a-0 715- 7961 15:124. https://d oi. org/ 10. 1186/ s13023- 020- 01407-7 4. Alonso R, Gonzalez-Moron D, Garcea O (2018) Optical coher- 22. Hopf S, Pfeiffer N, Liesenfeld M et al (2019) A comprehensive ence tomography as a biomarker of neurodegeneration in multiple monocentric ophthalmic study with Gaucher disease type 3 patients: vitreoretinal lesions, retinal atrophy and characterization 1 3 1160 Graefe's Archive for Clinical and Experimental Ophthalmology (2022) 260:1153–1160 of abnormal saccades. Orphanet J Rare Dis 14:257. https:// doi. damage. Neurochem Int 62:1–7. https://d oi. org/1 0.1 016/j.n euint. org/ 10.1 186/s 13023- 019-1 244-9 2012. 10. 010 23. Hopf S, Liesenfeld M, Schmidtmann I et al (2018) Age dependent 3 0. Klein AD, Mazzulli JR (2018) Is Parkinson’s disease a lysosomal normative data of vertical and horizontal reflexive saccades. PLoS disorder? Brain 141:2255–2262. https:// doi. org/ 10. 1093/ brain/ ONE 13:e0204008. https://d oi.o rg/1 0.1 371/j ourna l.p one.0 20400 8 awy147 2 4. van Wegberg AMJ, MacDonald A, Ahring K et al (2017) The 3 1. Winter AW, Salimi A, Ospina LH, Roos JCP (2019) Ophthalmic complete European guidelines on phenylketonuria: diagnosis and manifestations of Gaucher disease: the most common lysosomal treatment. Orphanet J Rare Dis 12:162. https:// doi. org/ 10. 1186/ storage disorder. Br J Ophthalmol 103:315–326. https:// doi. org/ s13023- 017- 0685-2 10.1 136/b joph thalm ol-2 018- 312846 25. Davies EH, Mengel E, Tylki-Szymanska A et al (2011) Four-year 32. Shrier EM, Barr CC, Grabowski GA (2004) Vitreous opacities follow-up of chronic neuronopathic Gaucher disease in Europe- and retinal vascular abnormalities in Gaucher disease. Arch Oph- ans using a modified severity scoring tool. J Inherit Metab Dis thalmol 122:1395–1398. https://d oi. org/ 10. 1001/ archop ht.1 22.9. 34:1053–1059. https:// doi.o rg/1 0.1 007/s 10545- 011-9 347-z 1395 26. Serfozo C, Barta AG, Horvath E et al (2021) Reduced macular 33. Matos A, Gurgel V, Gonçalves M (2017) Ophthalmologic findings thickness and macular vessel density in early-treated adult patients in Gaucher’s disease type III: case report. Revista Brasileira de with PKU. Mol Genet Metab Rep 27:100767. https://d oi.o rg/ 10. Oftalmologia 76. https://d oi.o rg/ 10. 5935/0 034-7 280. 20170 066 1016/j. ymgmr. 2021. 100767 3 4. La Morgia C, Di Vito L, Carelli V, Carbonelli M (2017) Patterns 27. Nieves-Moreno M, Martínez-de-la-Casa JM, Morales-Fernández of retinal ganglion cell damage in neurodegenerative disorders: L et al (2018) Impacts of age and sex on retinal layer thicknesses parvocellular vs magnocellular degeneration in optical coherence measured by spectral domain optical coherence tomography with tomography studies. Front Neurol 8:710. https://d oi.o rg/1 0.3 389/ Spectralis. PLoS ONE 13:e0194169. https://d oi.o rg/1 0.1 371/j ourn fneur. 2017. 00710 al. pone. 019416 9 28. Campistol Plana J (2019) Early diagnosis of phenylketonuria. Publisher's note Springer Nature remains neutral with regard to Physiopathology of the neuronal damage and therapeutic options. jurisdictional claims in published maps and institutional affiliations. Medicina (B Aires) 79(Suppl 3):2–5 29. Cleeter MWJ, Chau K-Y, Gluck C et al (2013) Glucocerebrosi- dase inhibition causes mitochondrial dysfunction and free radical 1 3