FP386 : Corneal Changes in Parkinson’s Disease Patients on Long Term Amantadine Therapy

Dr. Satya Sudha Daggumilli, D18494, Dr. Vanathi M, Dr. Vinay Goyal, Dr. Radhika Tandon, Dr. Noopur Gupta Parakh

Abstract:

Purpose : To evaluate corneal changes and their progression in patients with Parkinson’s disease on long term oral amantadine therapy.

Materials and methods: Prospective longitudinal Study of 90 patients (180 eyes) of Parkinson’s disease (PD) on more than six months of oral amantadine therapy, 30 amantadine naïve PD patients and 30 age and gender matched healthy controls. Central corneal thickness (CCT), corneal endothelial cell parameters and corneal sub basal nerve fibre layer (SBNFL) changes were studied over one year follow up.

Results: The amantadine PD patients group had statistically significant decrease of endothelial cell density(ECD) (1.51% vs 0.94% vs 0.55%), decrease of hexagonality (4.98% vs 3.56% vs 2.31%) and increase of coefficient of variation (6.12% vs 4.80% vs 3.30%) compared to amantadine naïve PD patients controls respectively. PDgroup on 400 mg of amantadine had greater change in endothelial parameters.

Conclusions: Long term amantadine has a toxic effect on corneal endothelium

Introduction                                                                                                                             Amantadine  hydrochloride, introduced as an antiviral medication^1,2 for preventing and treating type A influenza is being  widely used in the management of Parkinson’s  disease and drug (levodopa)  induced dyskinesia^3-6. Several case reports  have documented amantadine induced corneal toxicity after it’s widespread use as anti parkinson’s drug such as superficial punctate keratitis, epithelial edema and stromal edema.

There are several case reports^7-14 of amantadine induced transient corneal toxicity. In some cases corneal toxicity of amantadine was noted several years after long term oral  amantadine therapy. The mechanism underlying corneal toxicity^11,14-17 secondary to amantadine use is unclear. Fraunfelder FT et al⁹ suggested that drug in the tear film can create superficial corneal deposits , corneal edema and superficial punctate keratitis. Jeng  et al¹⁶ had  hypothesized that drug levels in the aqueous may contribute to corneal endothelial toxicity as he noted corneal edema several years after the initiation of  amantadine therapy. Chang et al¹⁵ suggested dose dependent toxicity of amantadine on corneal endothelium in PD patients. Studies conducted by Bahrani et al²⁵ and Lee et al²⁶ also showed dose dependent toxicity of amantadine on corneal endothelium.

There is paucity of studies on the effect of amantadine on corneal endothelium and sub basal nerve fibre layer on a long term basis. Most of the available literature on the corneal toxicity of amantadine are cross sectional studies.

Hence we undertook this study to evaluate  for the changes in corneal  parameters including sub basal nerve fibre layer in PD patients following long term oral amantadine therapy.

Subjects and methods

The study was  designed as a prospective longitudinal study and was conducted in a tertiary care centre.120 patients with  Parkinson’s disease fulfilling the inclusion and exclusion criteria (table 1) were recruited  from  neurology outpatient clinic and 30 age and gender matched healthy controls were recruited from ophthalmology out patient clinic  between October 2015 and December 2015 and followed up for a period of one year till February 2017. The study subjects were divided into three groups,  90 PD patients on amantadine (group I), 30 PD patients amantadine naïve (group II) and 30 controls (group III). Patients on amantadine  were sub divided based on daily dosage of amantadine as PD patients on 200 mg, 300 mg and  400 mg  as subgroup I, subgroup II and sub group III respectively. Out of the 90 PD patients on amantadine therapy enrolled  in our study 30 were excluded owing to various reasons . Final analysis was conducted on 60 patients (120 eyes) of amantadine group with 20 patients in each subgroup, 30 patients of amantadine naïve group and 30 controls.

Parkinson’s disease  parameters evaluated included duration of Parkinson’s disease, daily dosage of amantadine (milligrams) , duration of  amantadine use (years) and H and Y staging²² of PD  .

Ophthalmic examination included visual acuity  assessment , comprehensive slit lamp bio microscopy, assessment of intraocular pressure and corneal pachymetry by non contact tonometer (NT- 530P Non contact tonometer, Nidek Tech). Non contact  specular microscopy (Topcon)  was used to record endothelial cell parameters ; endothelial cell density (ECD), % hexagonality and coefficient of variation (Cov) and average of three readings was taken for each eye. In vivo  confocal microscopy (confoscan4)  was used to evaluate endothelial cell parameters using  Navis endo cell analysis software and neuron j plugin of image J software was used to analyse corneal sub basal nerve fibre layer .

Statistical analysis:

The STATA version 14  SPSS 24.0 statistical software was used to anlyaze all data. Pearsons correlation coefficient was used to establish correlation between these quantitative variables. Only data from the left eye were assessed in each patient to avoid possible errors in statistical analysis caused by inter-ocular correlation.

To determine the significant differences in corneal endothelial cell morphology between the Parkinson’s patients on amantadine , amantadine naïve Parkinson’s patients and subgroups of PD patients on amantadine an independent t test  and one way ANOVA was used followed by post hoc analysis to see the assosciation between the groups and the subgroups of Parkinson’s patients on amantadine according to daily dosage.

Intraclass correlation and cronbachs alpha  was applied to establish agreement between the  two methods (specular and corneal confocal microscopy ) of corneal endothelial cell evaluation. P values less than 0.05 was considered statistically significant.

Results:

The mean age and % of males and females were comparable in the three groups. The demographic characters and Parkinson’s disease related parameters of study groups and subgroups of PD patients on amantadine were comparable.

The ocular parameters of the three study groups are shown in table 2.The corneal endothelial cell parameters obtained by specular and confocal microscopy at initial presentation and  at one year follow up along with sub basal nerve fibre density among the three study groups and sub groups of PD patients on amantadine  are shown in tables 3-7.Corneal endothelial cell parameters  showed a  significant change in PD patients on amantadine therapy  compared to amantadine naïve PD patients and  the controls. The decrease of ECD (p-0.047) , % hexagonality (p-0.01) and increase of Cov (p-0.03) was more in Parkinson’s disease patients on amantadine compared to amantadine naïve and controls. Similar change of endothelial cell parameters was also noted in  amantadine patient subgroups with greater decrease of ECD,% hexagonality and increase of Cov in patients on higher daily dosage of amantadine in accordance with previous studies.

Our study also  showed that PD patients had  less corneal SBNFD compared to controls and the change of SBNFD was more in PD patients with higher daily dosage of amantadine compared to amantadine naïve PD patients and controls .

Discussion:

Parkinson’s disease is a  chronic progressive neurodegenerative disease resulting from loss of dopaminergic neurons  in the substantia nigra. Amantadine, an antiparkinsonian drug  is being commonly used for parkinson’s patients for its efficacy in treating levodopa induced dyskinesia.

There are several case reports of amantadine induced transient corneal toxicity. However the mechanism underlying corneal toxicity ^11,14,15,17secondary to amantadine use is not known. Proposed mechanisms being dose dependent  corneal endothelial  toxicity and idiosyncratic hypersensitivity to amantadine.

Corneal endothelium is a monolayer of hexagonal cells, derived from neural crest¹⁸  and plays a pivotal role in the homeostasis of cornea. Endothelial cell count at birth is 5000-6000 cells/mm³. Age related loss of endothelial cell decline has two components: Rapid component wherein there is an exponential loss to 3500cells/mm² by age of 5 years¹⁹ and to 3000cells/mm² by 14-20 years of age. Slow component, where there is a linear steady rate loss of 0.3 -0.6%²¹ per year. Percentage decrease of ECD in our study was 1.51% in PD patients on amantadine therapy , 0.94% in amantadine naïve PD patients  and  0.55% in controls(p-0.47). Percentage decrease of ECD among sub groups of amantadine group was 1.36% in PD patients on 200 mg amantadine, 1.53% in PD patients on 300 mg amantadine and 2.03% in PD patients on 400 mg of amantadine.

Corneal endothelial cells are uniquely regular, usually uniform in size to one another. Hexagon is the most energy efficient geometric shape in order to cover a surface completely without leaving gaps. This  hexagonal layer of cells allow minimum exposure of cornea to aqueous. Osmotic gradient created at the basolateral border of the endothelial cells by ion exchanges, passively drives fluid towards anterior chamber. Endothelium ensures corneal transparency by maintaining electrolyte balance and water fraction at 78%. This is carried out by an active pump²⁰ that moves ions, and draws water osmotically from the stroma into the  aqueous humour. Corneal endothelial cells, have zonula occludens  responsible for weak barrier function allowing nutrients and other molecules  to enter  the stroma. The  combined fluid pump and leaky barrier is sometimes called pump leak mechanism. Distortion of this equilibrium caused by damage of the endothelial cells leads to swelling of the corneal stroma resulting in corneal edema and decrease in visual acuity.

Many factors like intraocular surgery, inflammation, drugs can aggravate endothelial cell loss. The limited regenerative capacity of endothelial cells leads to reduced endothelial cell density which is compensated by thinning and elongation of remaining cells. This leads to change in shape of cells leading to decrease of hexagonal cells and increase of variation of cells. Changes in coefficient of variation and percentage hexagonality are indicative of endothelial stress before decrease of endothelial cell density becomes apparent. PD  patients had greater % decrease of hexagonality (4.98% vs 3.56% vs 2.31%)(p-0.01)and increase of coefficient of variation (6.12% vs 4.80% vs 3.30%)(p-0.03) compared to amantadine naïve PD patients and controls respectively. Sub group with higher daily dosage of amantadine had greater decrease of % hexagonality and increase of Cov compared to other two sub groups. Corneal endothelial cell parameters evaluated by confocal and specular microscopy were comparable  similar to the study by Christiana et al²⁴.

None of the patients in our study developed corneal edema,. Critical endothelial cell count to develop corneal edema is less than 500 cells/mm², endothelial cell densities in our patients was never near critical endothelial cell count. Change in central corneal thicknes (CCT)  was also  not significant .

Our study also showed that SBNFD was less in PD patients²³ on amantadine therapy compared to amantadine naïve and controls. The decrease of SBNFD in PD patients on higher daily dosage of amantadine was more compared to the other two sub groups.

Our study also evaluated the assosciation of corneal endothelial cell parameters and SBNFD of PD patients with H and Y staging and found that patients with higher H and Y stage had greater change of  endothelial cell parameters, however this could be attributed to the fact that PD patients on higher stage of H and Y were on amantadine therapy.

Awareness of amantadine induced endothelial cell loss is important especially when prescribing amantadine in patients with decreased endothelial cell densities like fuchs endothelial dystrophy, glaucomaand uveitis. Careful medication review is important in patients with unknown etiology of corneal edema. We suggest Ophthalmic consultation prior to and during amantadine therapy

Conclusion:

Our results show  that  amantadine causes dose dependent toxicity on corneal endothelial cells.

References:

1. StanicovaJ,Miskovsky P, Sutiak V. Amantadine :an antiviral and antiparkinson agent. Vet Med 2001;46(9-10):244-256.
2. Jackson G, Muldoon R, Akers L. Serological evidence for prevention of influenzal infection in volunteers by anti-influenzal drug adamantamine hydrochloride. Antimicrob Agents Chemother 1963; 161:703–707.
3. Schwab RS, England AC, Poskanzer DC, Young RR. Amantadine in the treatment of Parkinson’s disease. JAMA 1969; 208:1168 –1170.
4. Maugh T. Panel urges wide use of antiviral drug. Science 1979; 206:1058 –1060.
5. Schwab R, England A. Amantadine HCL (Symmetrel) and its relation to levodopa in the treatment of Parkinson’s disease. Trans Am NeurolAssoc 1969; 94:85–90.
6. Fahn S, Isgreen WP. Long-term evaluation of amantadine and levodopa combination in Parkinsonism by double-blind crossover analyses. Neurology 1975; 25:695–700.
7. Pearlman JT, Kadish AH, Ramseyer JC. Vision loss associated with amantadine hydrochloride use. JAMA. 1977; 237:1200,110:96–97.
8. Nogaki H, Morimatu M. Superficial punctate keratitis and corneal abrasion due to amantadine hydrochloride. J Neurol. 1993; 240:388–389.
9. Fraunfelder FT, Meyer SM. Amantadine and corneal deposits. Am J Ophthalmol. 1990;110:96-97
10. Hughes B, Feiz V, Flynn SB, et al. Reversible amantadine-induced corneal edema in an adolescent. Cornea. 2004:23:823–824.
11. Galor A, Jeng BH, Lowder CY. A curious case of corneal edema. Eye Net.2007:41–42
12. Blanchard DL.Amantadine caused corneal edem. Cornea. 1990; 9: 181.
13. Kubo S, Iwatake A, Ebihara N, et al. visual impairement in Parkinson’s disease treated with amantadine: case report and review of literature.Parkinsonism RelatDisord 2008; 14:166-9.
14. French DD, Margo CE. Post-marketing surveillance of cornealedema, Fuchsdystrophy, and amantadine use in the Veterans Health Administration.Cornea. 2007; 26:1087–108
15. Chang KC, Kim MK, Wee WR, Lee JH. Corneal endothelial dysfunction associatedwith amantadine toxicity,Cornea. 2008 Dec;27(10):1182-5
16. Jeng BH, Galor A, Lee MS, et al. Amantadine associated corneal edema: potentially irreversible even after cessation of medication. Ophthalmology 2008; 115:1540-4
17. Chang KC, Jeong JH, Kim MK, Wee WR, Lee JH, Jeon BS. The effect of amantadine on corneal endothelium in subjects with Parkinson’s disease,Ophthalmology. 2010 Jun;117(6):1214-9
18. Bahn eta al, classification of corneal endothelial disorders based on neural crest origin. Ophthalmology 1984;91:558-563
19. Nucci P et al, Normal endothelial cell density range in childhood. Arch ophthalmol vis sci 1986; 27: 44-51.
20. Maurice DM. The cornea and sclera. In : Davson H (ed). The Eye, 3^rd Academic press: orlando;1984, p-85.
21. Bourne et al, central corneal endothelial cell changes over a ten year period. Invest ophthalmol vis Sci 2002; 43: 2152-2159.
22. Joel S. Perlmutter, Assessment of Parkinson disease manifestations. Curr Protoc Neurosci.2009 oct; CHAPTER: unit 10.1
23. Lewis kassilliyya, small fibre neuropathy in Parkinson’s disease, 2015
24. Christiana M.C.Klais, comparision of endothelial cell count using confocal and contact specular microscopy
25. Bahrani k, Goyal v, Shukla G, Vanathi M, Behari M. Adverse effects of amantadine in patients of Parkinson’s disease. A cross sectional study (abstract). Movement Disorders 2015;30 Suppl 1:177
26. Po Yen Lee, Hung Pin Tu et al, Amantadine, use as a risk factor for corneal edema: A nationwide cohort study in Taiwan.Am j ophthalmol 2016;171:122-129.
27. Christiana M.C.Klais, comparision of endothelial cell count using confocal and contact specular microscopy

 

Table 1: inclusion and exclusion criteria of study subjects

Inclusion criteria

Exclusion criteria

1.                    Patients with Parkinson’s disease willing to participate in the study 2.                    Parkinson’s disease  patients on long term oral amantadine therapy (more than six months) 3.                    Age > 40 years

Parkinson’s disease patients with history of: 1.                    Cataract surgery 2.                     Glaucoma 3.                     Laser treatment 4.                    Contact lens use 5.                    Use of topical ocular medications 6.                    Ocular trauma 7.                    Smoking 8.                     Not willing for follow up investigations 9.                     Not consenting to participate in the  study

Table 2: Ocular parameters of  the three study groups at initial presentation and at one year follow up

Ocular parameters	Group I (amantadine)	Group II amantadine naive	Group III ( controls)	p- value
Mean UCVA_0 Mean UCVA_12	0.387±0.236 0.477±0.251	0.409±0.181 0.434±0.193	0.416±0.169 0.437±0.175	0.136 0.115
Mean BCVA_0 Mean BCVA_12	0.146±0.153 0.166±0.163	0.080±0.104 0.107±0.110	0.144±0.149 0.155±0.159	0.026 0.116
MeanNCT_0 Mean NCT_12	15.06±2.24 16.03±2.18	15.8±2.69 16.41±2.85	15.56±2.12 15.96±3.58	0.271 0.281
Mean CCT_0 Mean CCT_12	519.86±22.03 523.6±21.23	530.20±17.08 533.11±17.747	523.90±11.51 526.76±10.97	0.055 0.119

UCVA: uncorrected visual acuity, BCVA: best corrected visual acuity, NCT: non contact tonometry, CCT: central corneal thickness 

Table 3: Corneal endothelial cell density changes among the  study subjects over one year follow up by specular microscopy

Groups	Specular ECD_0±SD	Specular ECD_12±SD	Decrease/ Year	p-value
Parkinson’s patients amantadine naïve (n=30)	2736±362.30	2721.93±363.82	0.94%	0.062
Parkinson’s  patients on amantadine   (n=60)	2668.31±247.78	2636.83±247.10	1.51%	0.041
Controls (n=30)	2702.63±292.72	2682.6±289.89	0.55%	0.071

ECD_0: endothelial cell density at initial visit; ECD_12: endothelial cell density at one year follow up; SD: standard deviation 

Table 4: Corneal endothelial cell density  changesamong subgroups of Parkinson’s disease patients on amantadine therapy over one year follow up by specular microscopy

Subgroups ( daily amantadine    dosage)	Specular ECD_0± SD	Specular ECD_12 ± SD	Percentagedecrease/ Year	p value
Patients on 200mg,(n=20)	2732.95±295.68	2696.55±295.83	1.36%	0.0512
Patients on 300mg,(n=20)	2705.9±232.41	2665.7±230.47	1.53%	0.0403
Patients on  400mg, (n=20)	2557.1±174.18	2505.1±180.82	2.03%	0.018

ECD_0: endothelial cell density at initial visit; ECD_12: endothelial cell density at one year follow up; SD: standard deviation 

Table 5: Percentage hexagonality difference of the study groups over one year follow up by specular microscopy

Groups	Hex at presentation	Hex at 12 months	Percentage decrease /year	p-value
Parkinson’s  patients amantadinenaive  (n=30)	52.58±6.31	50.71±6.18	3.56%	0.058
Parkinson’s patients on amantadine (n=60)	50.17±3.66	47.64±3.63	4.98%	0.045
Controls(n=30)	57.35±5.72	55.96±5.72	2.31%	0.071

Hex – hexagonality

Table 6 : Increase of coefficient of variation among the three study groups over one year follow up by specular microscopy.

Groups	Coefficient of variation at presentation	Coefficient of variation at   12 months	Increase/ Year	p value
Parkinson’s  patients amantadinenaive  (n=30)	36.35±5.93	38.35±4.72	4.80%	0.051
Parkinson’s patients on amantadine (n=60)	39.03±3.59	41.42±3.70	6.12%	0.038
Controls(n=30)	36.06±5.28	37.25±5.45	3.30%	0.068

 

Table 7: Change of sub basal nerve fibre density among the three study groups over a span of one year by invivo confocal microscopy.

Group	Sbnfd_0	Sbnfd_12	Percentage decrease /year	p- value
Parkinson’s  patients amantadine naive (n-15)	9959.12±4282.44	9843.59 ±4235.98	1.16%	0.048
Parkinson’s  patients on amantadine(n-34)	8472.58 ±2781.88	8318.38 ±3201.89	1.82%	0.032
Controls ( n-30)	11389.73±2538.8	11275.4 ±2373.2	0.92%	0.087

SBNFD_0:sub basal nerve fibre density at initial visit, SBNFD_12: sub basal nerve fibre density at one year