Ophthalmological involvement in wild‐type transthyretin amyloidosis: A multimodal imaging study

Ophthalmological abnormalities have been reported in hereditary transthyretin‐related amyloidosis (ATTRv, v for variant) but not in wild‐type transthyretin‐related amyloidosis (ATTRwt).


| INTRODUCTION
Amyloidosis includes a group of rare diseases characterized by the extracellular deposition of amyloid, derived from different proteins that are destabilized, modified, and deposited in different tissues and organs. 1 Light chain amyloidosis (AL) is the most common form of systemic amyloidosis and is formed due to the deposition of the fibrillar protein resulting from immunoglobulin light chain misfolding.
Transthyretin (TTR)-related gene mutation (ATTRv, v for variant) is the most common form of hereditary amyloidosis. 1,23][4] ATTRwt is characterized by the extracellular deposition of ATTR in its native conformation in tissues and organs, mainly in the heart, leading to debilitating physical symptoms, decreased functional capacity, diminished quality of life, and eventually death. 5e ophthalmological involvement in ATTRwt has only been suggested, and no study has analyzed the possible ocular manifestations in these patients. 6,7Currently, the true prevalence of ATTRwt is unknown, but it seems to be relatively high compared with ATTRv. 5,8e advancement in the diagnostic techniques and therapies and the greater awareness for the clinical characteristics of ATTR, particularly for ATTRwt, have led to a wider disease clinical spectrum and an increased rate of new diagnoses. 2,9e aim of the present study was to analyze the ocular involvement in ATTRwt and to compare the findings with those detected in ATTRv and AL.Moreover, we used new noninvasive imaging techniques to identify signs of damage at the level of corneal nerve fibers and the retinal vascularization, which are more difficult to disclose at a systemic level.

| Population
This was a cross-sectional, 6-month, non-interventional study.Consecutive patients diagnosed with ATTRwt, confirmed by negative molecular testing, and available to undergo ophthalmological evaluation were included.Patients with other types of amyloidosis (ATTRv and immunoglobulin AL) were enrolled in the same period as controls.
Inclusion criteria were the following: age ≥18 years and definite diagnosis of AL or ATTR (including ATTR mutations' carriers), according to the current guidelines. 1 Exclusion criteria were the following: poor collaboration or the presence of disabilities that will make the procedures difficult to perform or prevent adequate patient's collaboration; allergy to mydriatics; refractive errors ≥6D; any congenital, infant, or inherited alteration of the cornea, retina, or choroid (e.g., hereditary corneal, retinal, or choroidal dystrophy, retinopathy of prematurity, and ocular malformations); any infectious eye disease that could modify the analysis; ocular media opacities precluding the quality of retinal imaging; and any type of previous intraocular surgery/laser treatment in the 12 months prior the enrollment.Moreover, patients' records were carefully reviewed to assess the presence of systemic comorbidities that might be potential confounders in the planned analysis: previous or ongoing chemotherapy, uncontrolled hypertension, cardiovascular and cerebrovascular events within the previous 12 months, and systemic diseases, not amyloidosis-related, which might cause peripheral neuropathy or any other neurological and vascular alteration potentially influencing the analysis, were excluded, except for diabetes, which was taken into account as a potential confounder in the statistical analyses.
The presence of mutation of the TTR gene and systemic involvement of amyloidosis were also assessed from the patients' records.In particular, the neurological involvement was considered when abnormalities were reported in previous neurological examinations and/or neurophysiological/electromyography assessments performed within 1 month from the ophthalmological examination.
An age-matched control group was also studied in the same period, including subjects without history of known significant systemic (e.g., uncontrolled hypertension and diabetes) and ocular diseases.
Approval was obtained from the Ethics Committee of the Azienda Ospedale Università di Padova (code: 357NAO23).Informed written consent was obtained from all individual participants included in the study.

| Procedures
All patients underwent complete eye examination, including bestcorrected visual acuity assessment (BCVA) using the Early Treatment Diabetic Retinopathy Study (ETDRS) charts, anterior segment examination, intraocular pressure, fundus examination, optical coherence tomography (OCT) and OCT angiography (OCTA), and in vivo corneal confocal microscopy (CCM).All examinations were carried out in both eyes if they met the inclusion criteria.

| In vivo corneal confocal microscopy
All enrolled subjects underwent CCM using Heidelberg Retina Tomography with the Rostock Cornea Module (HRTIII/RCM, Heidelberg Engineering, Germany).The HRTIII employs a 670 nm wavelength diode laser source and provides cross-sectional images of 400 Â 400 μm with a lateral resolution of 1 μm.For CCM imaging, a disposable sterile polymethylmethacrylate cap (TomoCap, Heidelberg Engineering) filled with hydroxypropyl methylcellulose 2.5% (GenTeal gel, Novartis Ophthalmics, East Hanover, New Jersey, USA) was placed on the objective lens of the Cornea Module, to improve optical coupling.After instillation of topical anesthesia, the Cornea Module was advanced manually until obtaining an appropriate cap contact with the corneal surface.Using the section mode of the CCM, layerper-layer images of the central cornea full thickness were obtained.
The Cornea Module was manually moved following a left-right and up-down pattern in order to cover the whole surface of the central cornea, as well as in anterior to posterior direction in order to check the entire thickness of the corneal layers from the epithelium to the endothelium passing through the sub-basal nerve plexus and the stroma.The presence/absence of hyper-reflective deposits was recorded after evaluation of all corneal images by an expert operator.
The three best focused not overlapping images of the sub-basal nerve plexus were selected for each of the examined eyes, and the average of the derived measures was used for further analyses.Firstly, a quantitative automated image analysis software (ACCMetrics, University of Manchester, Manchester, UK, courtesy of Prof. Rayaz A. Malik) was used to calculate six parameters: corneal nerve fiber density (CNFD), the number of nerve fibers/mm 2 ; corneal nerve branch density (CNBD), the number of primary branch points on the main nerve fibers/mm 2 ; corneal nerve fiber length (CNFL), the total length of nerves mm/mm 2 ; corneal nerve fiber total branch density (CTBD), the total number of branch points/mm 2 ; corneal nerve fiber area (CNFA), the total nerve fiber area mm 2 /mm 2 ; and corneal nerve fiber width (CNFW), the average nerve fiber width mm/mm 2 . 10en, two other parameters were obtained by a manual quantitative analysis performed by a blinded experienced operator on the best-quality image of the sub-basal nerve plexus: number of beadings (NBe) and fiber tortuosity (FT), according to previously reported methods. 11NBe was defined as the number of hyper-reflective points per unit of length (100 μm) of the best focused fiber, randomly chosen by the operator from all the nerve fibers seen in the corneal sub-basal nerve plexus image.Nerve beadings represent accumulation of mitochondria along the nerve, thus documenting the metabolic activity of corneal fiber sub-basal nerve plexus. 11FT was classified using the grading system proposed by Oliveira-Soto, which simultaneously considers the frequency and amplitude of changes in nerve fiber direction and provides a score ranging from 0 to 4, where 0 represents almost straight nerve fibers, 1 slightly tortuous fibers, 2 moderately tortuous fibers, 3 tortuous fibers with a quite severe amplitude of changes in fiber direction, and 4 very tortuous nerve fibers, with abrupt and frequent changes in direction.It is considered a morphologic marker of nerve degeneration and an attempt at fiber repair. 11nally, analyzing the whole thickness of the corneal stroma in the whole central cornea surface, we searched for stromal nerves, usually represented as thick, almost always stretched, highly reflective structures, recording the presence of fibers with an anomalous tortuous pattern, defined as at least a grade 3 of the Oliveira-Soto and Efron score and reporting such data as a dichotomous variable (presence/ absence). 12,13l CCM images were captured by an expert technician, while images' elaboration and analysis were conducted by a single operator, masked to any clinical condition of the subjects (including control vs amyloidosis).

| Optical coherence tomography (OCT) and OCT angiography (OCTA)
All enrolled subjects underwent OCT and OCTA, using the Spectralis HRA + OCT2 platform (Heidelberg Engineering, Heidelberg, Germany), as previously described. 14Both eyes of each subject were evaluated.The following scans were obtained: a 20 Â 20 OCT volumetric macular map centered on the foveola, a circumpapillary OCT ring scan with a diameter of 3.5 mm centered on the optic nerve head with a resolution of 100 ART (Q index minimum 30 for both OCT scans), a 10 Â 10 OCTA map centered on the foveola, and a 20 Â 20 OCTA map centered on the optic nerve head (Q index minimum 35 for both OCTA scans).
From the volumetric macular map, the following parameters, automatically provided by the incorporated Spectralis software (Heyex Software, Heidelberg Eye Explorer, Heidelberg Engineering) were collected: the central subfield thickness (CST), represented by the mean retinal thickness of the central circular 1 mm centered on the foveola, and the total macular volume (MV), represented by the total retinal volume of the central circular 6 mm centered on the foveola.
With regard to OCTA, en face images of the superficial vascular plexus (SVP), the intermediate capillary plexus (ICP), the deep capillary plexus (DCP), the choriocapillaris (CC), and the choroid (Ch) were obtained from the macular map and of the radial peripapillary capillary plexus (RPCP) from the optic nerve head map.All images were automatically provided by the in-built segmentation algorithm, as previously described. 15All images were reviewed to confirm consistent segmentation by the automated instrument software.
Quantitative analysis of the vascular plexuses in the OCTA en face images was performed using the open-source, available ImageJ software (National Institutes of Health, Bethesda, MD, USA).For the SVP, ICP, DCP, and RPCP, four parameters were analyzed: vessel area density (VAD), vessel length fraction (VLF), vessel diameter index (VDI), and fractal dimension (FD), as previously described. 15Briefly, OCTA en face images were automatically converted into binarized images to obtain VAD, dividing the number of black pixels counted by the software in the binarized image by the total number of image pixels.After that, images were skeletonized (i.e., elaborated until a single pixel remains for each vessel segment) to obtain VLF, dividing the number of vessel pixels in the skeletonized image by the total number of image pixels.VDI was obtained by processing both the binarized and the skeletonized images to calculate the average vessel caliber.Finally, FD represents an index of complexity of the vessels' ramification, obtained on a skeletonized image by a specific count of differently sized squares containing a vessel fragment.With regard to the RPCP, the analysis was conducted after removing from the images the large vessels, according to a previously described method. 15For the CC and Ch, the vascular/stromal ratio (V/S ratio) was obtained after elaboration of the images via ImageJ software, from the vessel density (V), defined as the proportion between the luminal area represented by white pixels and the total area, the stromal density (S), defined as 1-V. 16l OCT and OCTA images were captured by an expert technician, while images' elaboration and analysis were conducted by a single operator, masked to any clinical condition of the subjects (including control vs amyloidosis).

| Statistical analysis
The analysis of the studied parameters was performed according to usual methods provided by descriptive statistics: mean and standard deviation for quantitative variables and absolute and relative (percentage) frequencies for qualitative variables.Demographic and descriptive parameters were compared among groups by ANOVA test or chi-square test according to the type of variable.The analysis consisted in comparing patients with amyloidosis and controls and patients with different types of amyloidosis.For quantitative parameters, a multiple regression model with ANOVA for repeated measures has been used.Confounder parameters such as diabetes and glaucoma have been included in the model as co-factors.Measurement of the parameters in both eyes of the same patient has been taken into account in all the analysis.For qualitative parameters, chi-square test has been used.In case omnibus test resulted statistically significant, pairwise comparisons have been made by Tukey-Kramer post-hoc test for multiple comparisons and Chi-square test if the parameter was expressed as quantitative or qualitative variable, respectively.For all the analyses, SAS ® v.9.4 (SAS Institute, Cary, NC, USA) has been used.All tests have been interpreted as statistically significant if p < .05.

| Characteristics of the patients
Thirty-five patients were enrolled, 32 men and 3 women.Seventeen patients were affected by ATTRwt (all men), nine (one woman) patients by ATTRv, and two patients were pre-symptomatic carriers of the ATTRv mutation.In addition, seven patients with AL were enrolled (two women).Four patients were also affected by diabetes mellitus (three with ATTRwt and one with AL) (Table 1).
The mean age for the whole patients' group was 70.1 ± 11 years and 75.1 ± 6.7 for the ATTRwt group alone.The control group consisted of 49 healthy subjects with a mean age of 66.2 ± 10.2 years.The whole study sample and the control group did not differ with regard to mean age (p = .0964).There was instead a significant difference in age among groups (p = .0044),which was maintained only between ATTRwt patients and carriers in the single group analysis (p = .0072)(Table 2).
With regard to the main systemic clinical manifestations, 17 patients presented signs of symptomatic neuropathy (in some patients subclinical), 30

| Clinical findings
The mean visual acuity of the entire study group was 77.In addition, the visual acuity of the ATTRwt group was more than 10 letters lower than the other groups, however without statistical significance.(Table 2).
Seventeen (50%) ATTRwt eyes presented with a clinically significant cataract, eight of them requiring surgery; 13 (38%) ATTRwt eyes had already undergone surgery for cataract (Table 2).Mean intraocular pressure was 15 ± 4 mmHg, but three ATTRwt, one ATTRv, and one AL patients were already on topical hypotonic treatment for glaucoma, and two additional ATTRwt patients were diagnosed with glaucoma during our examination by finding glaucoma-related field defects at perimetry and a corresponding reduction in peripapillary fibers at OCT.One of these patients presented with a previously unknown pseudoexfoliatio lentis in one eye, characterized by deposition of a white fibrillar material in the anterior segment of the eye, well visible at slit lamp examination. 17One ATTRv patient showed abnormal conjunctival vessels.
With regard to the alterations at the retinal level, abnormalities in the RPE such as drusen or pigmentary alterations were identified at the ophthalmoscopic examination and confirmed at OCT, in 19 (56%) ATTRwt, 4 (22%) ATTRv, 1 (25%) carrier, and 4 (29%) AL eyes.In In no patient, deposits or further anomalies attributable to amyloidosis at the level of the iris were found.At the corneal level, epithelial and/or endothelial deposits were identified in six ATTRwt patients.These deposits were confirmed at CCM, which was able to identify epithelial and/or endothelial deposits in additional 10 patients (four ATTRwt, two ATTRv, three AL, and one pre-symptomatic carrier) (Tables 2 and 4).

| Corneal confocal microscopy
The CCM examination found a significant reduction in the mean length of the sub-basal corneal plexus fibers (CNFL 9.7 ± 4.3 vs 12.4 ± 4.0, p = .0230),which was also confirmed after correction for diabetic patients ( p = .0270)(Table 3).Moreover, the examination showed tortuous stromal nerves in a greater number of amyloidosis eyes compared to controls ( p = .0094),as well as corneal deposits (30 vs 0 eyes, for amyloidosis and control eyes, respectively, p < .0001)(Figure 1, Supplemental File S1).
The comparison among groups (ATTRwt, ATTRv, AL, and carriers) did not show significant differences, except for CNBD ( p = .0377)and tortuous stromal nerves ( p = .0052)(Table 3).When making a group-to-group comparison, a higher number of tortuous stromal nerves was found in ATTRwt eyes compared to both AL and ATTRv (p = .0330and .0045,respectively).

| Optical coherence tomography
The OCT examination of the macular area showed a reduction in the mean macular volume (in the central macular area with a diameter of 6 mm) with an increase in the central subfield thickness (CST) in all amyloid patients compared to controls (p = .0059and p < .0001,respectively) also after correction for diabetes and glaucoma (p = .0003and p = .0180,respectively) (Supplemental File S2).These results have been influenced by the presence of an epiretinal membrane in seven eyes (four ATTRwt and three ATTRv), leading to a thickening involving the more central area of the macula, despite a generalized thinning of the whole macula, shown by the reduced volume (Supplemental File S2).In addition, a generalized reduction in the thickness of the peripapillary nerve fibers, both average and in each peripapillary sector (p < .05),was identified, which was maintained in the superior temporal sector and in the average even after correction for glaucoma and diabetes (Supplemental File S2).
The comparison among groups (ATTRwt, ATTRv, AL, and carriers) showed a significant difference among groups in the global peripapillary RNFL thickness and in the temporal and temporal inferior sectors ( p = .0413,.0205,and .0009,respectively), where the lowest thickness was found in the ATTRwt patients, reaching statistical significance in the temporal inferior sector vs ATTRv ( p = .0059)and carriers ( p = .0053)(Table 4).

| Optical coherence tomography angiography
The study of retinal vascularization at OCTA showed a diffuse reduction in vascular density both at the level of the RPCP, located in the peripapillary nerve fiber layer, and at the level of the macular vascular plexuses, as well as at the level of the vascular component of the choriocapillaris (even after correction for diabetes and glaucoma) (Figure 2, Supplemental File S3).
This reduction was particularly pronounced in ATTRwt and ATTRv patients compared to AL and ATTRv carriers; however, in most cases, a statistical significance is not reached (Table 5).
No significant correlation with any major systemic manifestation was found for all the above-reported analyzed variables.

| DISCUSSION
In our cohort of ATTRwt patients, we found a high prevalence of ophthalmological alterations, involving both the anterior and the posterior segment of the eye, significantly modifying the visual function of the affected patients, also when compared to ATTRv and AL patients.
Chronic open angle glaucoma has already been reported to be a main cause of irreversible visual loss in ATTRv patients.It seems to be due to the perivascular amyloid deposition in conjunctival and episcleral tissues and due to the amyloid protein released into the aqueous humor causing trabecular meshwork obstruction; these mechanisms may remain asymptomatic for months and years, until the damage is irreversible, particularly in older patients, like the ATTRwt ones. 18In fact, two new diagnoses of glaucoma (confirmed by pathological defects at both the peripapillary OCT and the visual field) have been made in our ATTRwt patients, during our examination.Moreover, the presence of amyloid deposits causing ocular hypertension may reduce the efficacy of the conventional therapy, as already suggested for ATTRv.[20][21] F I G U R E 1 Corneal confocal microscopy images of the epithelium (A and B), the sub-basal nerve plexus (C and D), and the endothelium (e and f) captured with Heidelberg Retina Tomograph III Rostock Cornea Module (Heidelberg, Germany) from the central cornea of a healthy subject (A-C) and a ATTRwt subject (D-F).While the corneal nerve fibers of the healthy subject are mostly straight and continuous, those of the ATTRwt patient appear fragmented and more tortuous.Moreover, hyper-reflective deposits may be observed at both the epithelium and the endothelium.
With regard to OCT results, we found a reduction in the whole macular volume (MV), even after correction for diabetes and glaucoma and despite the presence of increased retinal central thickness (CST) secondary to the presence of vitreomacular traction.Indeed, studies of immunochemistry showed an intense immunoreactivity particularly at the level of ganglion cells somata and axons of rats eyes. 22though we did not analyze each retinal layer thickness in the macula, we found a generalized reduction, particularly in ATTRwt eyes, in the peripapillary retinal fiber layer, constituted by the axons of the ganglion cells, which might therefore be a direct consequence of the ganglion cells' impairment secondary to ATTR.
Moreover, TTR is a minor component of drusen, extracellular deposits of a material found in the Bruch membrane, which lay in conjunction with RPE and the choroid of the eye and are increased in age-related macular degeneration (AMD). 23We found a very high prevalence of RPE alterations, including drusen, in the macula with a prevalence of 55.9% and 22.2% in patients with ATTRwt and ATTRv, respectively, and even focal alterations in a single eye of a pre-symptomatic carrier.The reported prevalence of the same alterations is about 20% in the age-matched general population. 24Moreover, advanced maculopathy was detected in three of 17 (18%) ATTRwt patients (in two of whom it was unknown) and it required a specific therapeutic pathway of care.It is possible that in ATTRwt, the accumulation of ATTR at Bruch's membrane is enhanced, increasing the agerelated degeneration of the complex constituted by RPE, Bruch's membrane, and choriocapillaris, characteristic of AMD patients.
We also found a prevalence of 12% and 17% in ATTRwt and ATTRv eyes, respectively, of alterations at the vitreomacular interface, mainly represented by a hyper-reflective and thickened inner limiting membrane (ILM) at OCT.Although they may also be found in the general population, no control eyes and AL patients presented these alterations.[27] Carpal tunnel syndrome and spinal stenosis have been observed in ATTRwt several years before the onset of systemic symptoms, and, although rarely, peripheral neuropathy in ATTRwt has also been described. 28We also analyzed the small fibers of the corneal sub-basal nerve plexus, in vivo, using CCM and an automatic (and repeatable) method of quantification of CCM parameters, allowing a reliable evaluation of the fibers. 4,29,30A reduction in sub-basal nerve fiber length (CNFL) and an increase in stromal nerve tortuosity were found in all amyloidosis patients compared with control group.CNFL has already been shown to be the most reproducible and strong parameter for the detection of abnormalities of small fiber morphology in diabetes, also associated with different measures of neuropathy severity. 31Moreover, Rousseau et al. 31 proposed CNFL as a sensitive marker of denervation in ATTRv, correlating with the severity of both sensorimotor and autonomic neuropathies, as well as with walking status and lower limbs intraepidermal nerve fiber density (IENFD).The results of our study not only confirm the relevance of CNFL in ATTRv but also reveal its importance in ATTRwt.An increased number of tortuous stromal nerves was also detected in most ATTRwt patients compared to controls.Even if this has been suggested as a sign of suffering nerves, the study of the stromal nerves with CCM has been less extensively reported than the sub-basal nerve plexus and the meaning of their alterations should be cautiously interpreted. 13,31,32TA allowed the identification of a diffuse reduction of all vascular parameters compared to controls both at the macula plexi and in the peripapillary plexus, particularly in ATTRwt and ATTRv eyes, but without significant difference among groups.Signs of angiopathy have been described in ATTRv patients using invasive imaging modalities. 18,33reover, retinal microangiopathy has been suggested to correlate with cerebral amyloid microangiopathy. 34We confirmed the presence of vascular damage also in our amyloidosis population, using OCTA, that is able to detect the vascular impairment in a completely noninvasive way.
Moreover, we found an impaired vascularization in all retinal plexi and choriocapillaris but not in the choroid, which is constituted by larger vessels than those in other plexi.It may be assumed that the damage secondary to amyloid deposition may influence the small vessels first, reducing blood flow.In vitro studies suggested that both ATTRwt and ATTRv proteins regulate endothelial cell functions and oligomers formed during the process of amyloid fibril formation play an important role in damaging the endothelial cells of blood vessels, a process which seems to start before TTR deposition. 27,35Therefore, OCTA seems to be able to detect chorioretinal vascular alteration.
Intraocular amyloid deposition also occurs in the anterior lens capsule, leading to early cataract. 18In our ATTRwt patients, we found a very high percentage of patients with cataract, requiring surgery, or patients who had already undergone it.Even if cataract is a cause of reversible visual loss and may be considered of secondary relevance compared to the systemic manifestations, still it may create a significant discomfort in affected patients.Moreover, the risk of glaucoma post-cataract surgery requires careful follow-up. 36though vitreous opacities are described as a frequent and disabling manifestation in ATTRv, the prevalence of clinically relevant vitreous opacities was not significant in our ATTRwt sample.Literature data are very different even in ATTRv cohorts, also depending on different mutations, and vitreous opacity may also be absent. 4,37Moreover, vitreous opacities are generally described as a late manifestation of the disease, and the definition of opacities is strictly dependent on the examining clinician, making these findings less reliable compared to the other ophthalmologic findings. 37th the advent of the new therapeutic options for ATTR, good results have been obtained in reducing disease progression and obtaining longer survival.Tafamidis has recently been approved also for ATTRwt patients. 38,39However, most of the current treatments are not able to cross the blood-brain and blood-retinal barrier.Therefore, the clinical manifestations involving the central nervous system and eye might become the most prevalent in the near future. 9,40,41reover, the eye examination, eventually also using aqueous humor samples, might provide possible markers of response to treatments. 42nce no plasma or urinary biomarkers are available for the diagnosis of ATTR, recently, the development of specific screening programs for the early identification of ATTR, particularly ATTRwt, has been proposed. 5,43Therefore, the recognition of multiple ocular abnormalities, particularly involving the intraocular pressure and the RPE, associated with other early manifestations of the disease (e.g., carpal tunnel syndrome) may anticipate the suspicious of ATTR, particularly ATTRwt, and suggest more specific evaluations, such as CCM for the study of the sub-basal nerve plexus and OCTA for the evaluation of the chorioretinal vascularization.Following our preliminary results, further studies might better focus on single ocular abnormalities and their correlation with other systemic manifestations to better characterize possible different phenotypes of the diseases, their prevalence, and best management.
One of the main confounding factors in our ATTRwt population may be the older age of the ATTRwt patients, where a high prevalence of alterations usually related to degeneration and aging can be found.However, ATTRwt is, for definition, an age-related disease, whose relevance is increasing rapidly because of the constant increase in the aging population.Therefore, the age of the studied population is a substantial characteristic of these patients. 44A high prevalence of other conditions more represented in the elderly (e.g., carpal tunnel syndrome, rheumatologic disorders involving the spine and limb extremities, polyneuropathy) has been recognized in ATTRwt patients compared to age-matched controls. 2,28,43As suggested by these studies, even if it is possible that some alterations are correlated more to age than to amyloidosis, still it is unlikely that they are not correlated at all since amyloidosis can potentiate the effect of any other degenerative and vascular process, also with regard to the ophthalmological abnormalities. 28,43,45Accordingly, in ATTRwt patients, we found a high prevalence of ophthalmological abnormalities compared to control group and also to the reported prevalence of the same alterations in the age-matched general population (e.g., 55.9% of RPE alterations vs about 20% in the ATTRwt patients vs the general population). 24other limitation of our study was the limited sample size.However, this study represents, at least for ATTRwt, one of the largest cohorts so far described, particularly from an ophthalmological point view.
Finally, the relationship between the amyloid deposition and the clinical manifestations found in this study has not been confirmed by immunohistochemical colorations (Congo red), and no typing of amyloid proteins was performed.7][48] Considering the high invasiveness of the biopsy in the eye and the inaccessibility of some involved ocular tissues in vivo, most studies on ATTR do not perform specific histology of ocular manifestations. 49,50 conclusion, we found an unexpected high prevalence of ocular involvement in ATTRwt that can potentially lead to severe ocular involvement with significant consequences in term of visual acuity and performance when compared to the more studied amyloidosis forms, such as ATTRv and AL.Therefore, in ATTRwt patients, a multidisciplinary management with regular ophthalmological assessments should be recommended from the moment of diagnosis in order to identify, adequately follow up, and eventually treat the alterations that tend to appear over time in different ocular structures.

4/ 19
of these ATTRwt eyes, there was an advanced maculopathy, two of whom of first diagnosis, requiring a specific diagnostictherapeutic path.Four (12%) ATTRwt, three (17%) ATTRv, and one (7%) AL eyes had vitreoretinal interface alterations, for example, epiretinal membrane, one (AL) already surgically treated at another center fine years before, currently not in follow-up.Finally, vitreous opacities were identified in six (17%) ATTRwt, six (33%) ATTRv, and two (14%) AL eyes, clearly visible at the ophthalmoscopic examination, although not clinically significant.Therefore, no vitreoretinal surgical procedure was deemed necessary, considering the invasiveness of the procedure.One ATTRwt patient also showed signs of previous branch retinal vein occlusion.
patients presented signs of cardiomyopathy, and 19 were affected by carpal tunnel syndrome, at the moment of ocular examination.The prevalence of comorbidities in the different groups of amyloidosis patients is reported in Table 2. Five of nine ATTRv patients were in treatment with tafamidis 20 mg/die, and four patients were on patisiran.One ATTRwt patient was in therapy with tafamidis 61 mg/die.The remaining 16 ATTRwt patients were not treated for amyloidosis.All AL patients were under specific treatment.Clinical and demographic characteristics of enrolled subjects.
T A B L E 2 Clinical and demographic characteristics of different amyloidosis groups.
Corneal confocal microscopy data in amyloidosis type groups.
T A B L E 5 Optical coherence tomography angiography data in amyloidosis type groups.