Presented is a case series (3 patients) with LHON and a likely past history of amblyopia during childhood and who later developed LHON on the good eye, whilst the changes in amblyopic eye were subjectively unnoticed.
3.4. Segmentation Analysis and pRNFL
In all eyes of Patients 1-3, segmentation analysis of retinal layers showed significant thinning of GCC in all other segments (
Supplementary Table S3) with discrete preservation of GCC in the central ETDRS field. The amblyopic eyes had an overall thicker retina and GCC in the central ETDRS circle compared to subacutely affected eyes. No differences were observed in other ETDRS rings or for other retinal layers between amblyopic and subacutely affected eyes (
Supplementary Figure S1). The peripapillary RNFL showed a continued trend of atrophy in both the presumed amblyopic eye and the previously good eye (Supplementary
Figures S2–S4) both in the nasal and temporal segments. In the Patient 3, who has improved on the good eye, pattern of GCC thinning showed progressive decline at the disease onset and later stabilization during the follow-up in both, the good and presumably amblyopic eye (
Supplementary Table S3). Improvement of visual acuity was not associated with thickness change.
We describe three patients with monocular low vision in one eye since early childhood, believed to be due to amblyopia, that presented many years after with subacute vision loss on the previously good eye. The question arises whether low vision on the first eye was actually due to an early LHON episode, or was there indeed true amblyopia at that time, with superimposed LHON affecting both eyes later in life, especially as none of the patients reported any loss or change of vision in their amblyopic eye. In the presented patients, unfortunately, no report on optic disc pallor was available in their childhood notes and no OCT or other imaging was available at that time. Patients 1 and 3 already had some pallor of the ONH in the amblyopic eye at the time of the LHON episode in the previously healthy eye. This is an unusual finding in amblyopia and may suggest earlier episode of LHON. Their different scenarios were therefore clinically challenging and was the reason for detailed morphological and functional follow-up described in this paper.
Patient 1 lost vision on the good eye at the age of 17 and there was no strabismus, anisometropia or media opacities or other amblyogenic factors present. Testing for three common genetic pathological variants of LHON was negative. Extended mitochondrial genome testing revealed rare, previously published [
26]. MT-ND1:m.3700G>A pathogenic variant in homoplasmy. As on presentation, partial disc pallor was already present on amblyopic eye it may be possible that LHON affected the amblyopic eye earlier. However, it is not possible to determine whether LHON was already the cause of “amblyopia”. with insidious onset in childhood or whether the true amblyopic eye was additionally affected in subacute fashion later, that remained unnoticed due to poor vision until the good eye was affected. As he had a slight progression of the pRNFL thinning during the disease course and electrophysiology typical for LHON on both eyes, he may correspond to bilateral sequential onset which had initially started on the amblyopic eye, however, is also possible that it could have been caused by a childhood-onset LHON since the patient did not respond to standard amblyopia treatment.
Three different patterns of visual acuity loss were described in childhood LHON: Classical acute in 63%, slowly progressive in 15%, and insidious or subclinical in 22% [
10]. Recent paper by Barboni et al. [
11] proposed a different classification of childhood-onset LHON: Subacute bilateral (66.7%), insidious bilateral (17.3%), unilateral (11.1%), and subclinical bilateral (4.9%). According to these studies, different patterns could affect either eye. The specific pattern of insidious unilateral onset in early infancy is quite peculiar, and asymmetric involvement may arise as a consequence of the subtle anatomical differences between the eyes, such as differences in the architecture and size of the optic nerve head, and the number of axons that are known to vary by up to 20% between the eyes [
27,
28]. In our patient, OCT showed some advancement of atrophy during follow-up also on the amblyopic eye which could suggest insidious onset and slowly progressive pattern.
Patient 2 showed characteristic ONH hyperemia, tortuosity of blood vessels, and pseudoedema simultaneously in both eyes although no additional worsening of vision was noted on the previously amblyopic eye. This patient remembered strabismus management with surgery in childhood, in keeping with true strabismic amblyopia. Clinically, this patient had a bilateral symmetric onset of LHON at the rather advanced age of 61, and symmetrical pseudoedema with atypical visual field loss prompted diagnostics towards other neurological causes until genetic testing confirmed the diagnosis.
In Patient 3, the affected eye developed convergent strabismus and could therefore be associated with strabismic amblyopia, however, strabismus can also develop secondary to visual loss due to LHON or any other cause [
29]. Partial ONH atrophy was probably present before the second eye was affected since changes in the visual field and color vision were already noted at the first presentation in the absence of acute signs of disease. It is possible that the first attack could have occurred before the age of five (convergent strabismus usually occurs if the visual acuity in one eye is low before the age of five, while divergent strabismus develops at a later age) [
29].
In the case of Patient 3 the most probable scenario is insidious unilateral LHON with consequential development of strabismus. Strabismus has been reported in cases of the unilateral childhood LHON and is usually associated with low visual function [
4,
12]. Barboni et al. reported strabismus in all patients with unilateral LHON [
11]. They also described the subgroup of patients with the involvement of the second eye at the later age (≥15 years old) which could correspond to our Patient 1. In insidious unilateral patient group the younger presumed age of onset and the presence of strabismus could influence the final visual outcome due to a mechanism of cerebral suppression [
30,
31].
Electrophysiology was characteristic for LHON in both eyes of all three patients, confirming LHON event at some point. If the fixation is normal, the PERG N95 in amblyopic patients is usually normal [
32], whilst in all three patients, the PERG N95 wave was abnormal to similar extent. In all patients, VEP was also abnormal on both presumably amblyopic and LHON eye. This confirms that the disease on the presumably amblyopic happened prior or at the disease onset on the LHON eye.
All our patients also had low color vision both in LHON and presumably amblyopic eye. In patient with VA improvement, color vision on the presumably amblyopic eye also did not improve. This corroborates with LHON event in both eyes as color vision is not significantly reduced in amblyopia [
33], with some alteration in color perception [
34].
With regards to OCT characteristics either no differences were reported in pRNFL and macular thickness between amblyopic and non-amblyopic eyes [
35], or significantly thicker RNFL was reported in amblyopic eyes [
36]. Moreover, thicker retina in ETDRS center and thinner in the inner and outer ETDRS ring was observed in amblyopic compared to normal eyes [
37], Although in our case series, all eyes were affected and this comparison is not valid, we also noted thicker GCC in the ETDRS center in amblyopic eyes than in subacutely affected eyes (
Supplementary Figure S1 and
Supplementary Table S5). Also, all eyes showed the pattern of better preserved GCC thickness in the central ETDRS field than paracentral ones, as described recently, suggesting that both eyes were also affected by LHON [
38].
Taking into account everything above, it is most likely that Patients 1 and 3 in this case series suffered from a LHON episode in the presumed amblyopic eye earlier in life but had only noticed unilateral visual loss upon visual loss in the other eye. In Patient 2, the clinical picture was suggestive of severe bilateral subacute LHON although noted only at the previously good eye.
Delayed involvement of the second eye has been described as rare in adult-onset LHON (37–40). Despite the fact that 97% of LHON patients suffer from the involvement of the other eye within one year [
2,
4], there are a few reports of unilateral or delayed presentation of LHON. Two studies document unilateral vision loss with subsequent visual recovery in the affected eye [
3,
4]. Several reports describe a delay in fellow eye involvement for 3–8 years follow-up period [
5,
6]. Other reports document cases of unilateral involvement with a follow-up period of 18 months [
7], 10 years [
8], and 16 years [
9]. The longest interval reported in the literature was 18-year difference in disease onset (at the age of 5) between the two eyes [
12]. This time difference between the involvement of the two eyes could be explained by the possible presence of a larger number of mutated mtDNA in one optic nerve or the fact that mitochondrial expression in the other eye is increased after the attack on the first eye, compensating the energy defect [
12].