Retinal nerve changes in patients with tremor dominant and akinetic rigid Parkinson’s disease
- First Online:
- Cite this article as:
- Rohani, M., Langroodi, A.S., Ghourchian, S. et al. Neurol Sci (2013) 34: 689. doi:10.1007/s10072-012-1125-7
- 282 Views
Parkinson disease is a multisystem neurodegenerative disease which involves not only basal ganglia and extrapyramidal system but also many other neurologic systems such as retinal ganglion cells. Optical coherence tomography (OCT) is a non-invasive method for assessment of retinal nerve fiber layer (RNFL) thickness and its changes in different diseases. To evaluate the RNFL thickness in patients with Parkinson disease (PD), we performed OCT in patients with PD and compared it with a control group. From October 2010 to July 2011, 27 PD patients (54 eyes) and 25 healthy persons (50 eyes) were entered to this analytical cross-sectional study according to the defined criteria. PD patients were categorized into two groups “akinetic rigid (AR) and tremor dominant (TD)”. RNFL was divided into four quadrants and was assessed by OCT. Afterwards; the data were analyzed by bivariate and multivariate models. The RNFL thickness in PD was significantly lower than the control group. Also, the thicknesses of inferior and nasal quadrants of RNFL in TD group were significantly more than AR group. According to these findings, OCT can be used as a sensitive and objective marker for assessment of early neurodegenerative changes of PD and early initiation of neuroprotective treatments. Future studies with adequate sample sizes are recommended to investigate interactions between age, distribution of the disease and type of PD as well as the effects of individual factors.
KeywordsRetinal nerve fiber layerAkinetic rigid (AR) Parkinson’s diseaseTremor dominant (TD) Parkinson’s diseaseParkinson disease (PD)Optic coherence tomography (OCT)
In the recent decade, it has been cleared that PD is a multisystem disease with the involvement of many non-motor systems and neurons in many parts of the central and peripheral nervous system in addition to neurons in basal ganglia . Furthermore, in 2001 some evidences declared that the severity of visual impairment was in association with the severity of motor impairment [4, 5].
Although the main symptoms of PD are motor, non-motor symptoms due to dysfunction in neurons other than striatonigral system can be seen. These non-motor symptoms sometimes occur prior to motor symptoms and can help in early diagnosis of the disease .
Retinal dopaminergic amacrine cells in ganglion layer are involved in PD and it has been shown that the dopamine content of the retina is less in PD patients in comparison with healthy controls .
Different neurodegenerative diseases cause pathologic changes in nerve fiber layers of retina; therefore, the retina is a good marker for quantitative assessment of neurodegeneration .
Optical coherence tomography (OCT) is a non-invasive, reproducible, easy and rapid instrument without any complications for measurement of axonal degeneration that can be a valuable measure for assessment of probable neuroprotective treatments in clinical trials and a scale for understanding the progression rate of neurodegeneration [8, 9].
Since retinal nerve fiber layer is a good quantitative marker for assessment of neurodegeneration, we used OCT in PD patients and compared it with a healthy control group. In addition, the RNFL thickness was assessed in two major subtypes of PD [tremor dominant (TR) and akinetic rigid (AR)].
Methods and materials
The protocol of this study was approved by the ethical committee of Tehran University of Medical Sciences (TUMS).
The authors were obligated to respect Helsinki declaration. All patients were explained about the aim and method prior to the study and a written consent form was signed by them or their own family.
From October 2010 to July 2011, patients who referred to the neurology department of the hospital and agreed to enter to the study were considered according to the defined inclusion and exclusion criteria.
Patients with any signs and symptoms or a history of neurodegenerative diseases such as Alzheimer and multiple sclerosis (MS), and patients who were suspected for optic neuropathy disorders such as glaucoma and ischemic optic neuropathy were excluded from this investigation [10, 11].
In addition, patients with active ophthalmologic disorders or diseases associated with media opacities that interfered with imaging processes were not included in the study.
The presence of hereditary retinal impairments, diabetic retinopathy, macula edema and other pathologic changes around the optic disc were considered as other exclusion criteria.
This study was designed as an analytical cross-sectional investigation. The sampling method was non-probability, convenience of haphazard sampling. All patients were examined by a neurologist and ophthalmologist to exclude patients with confounding disorders. The diagnosis of PD in patients group was confirmed by UK brain bank criteria . Patients who were referred during the “off” period of the disease were categorized into two groups “AR and TD” based on the motor examination UPDRS(part III). Scoring was based on the method that was used by Lewis et al.  in 2005.
Patients in the control group were chosen parallel to the case group. As a patient with PD was taken to the OCT room, another patient of the same age and gender from other clinics was sent to the room if he/she was in accordance with inclusion and exclusion criteria of the study.
To cause pupil dilation, Tropicamide was administered to all patients 20 min prior to the OCT. The RNFL thickness was measured by 3D-OCT 1000 Mark II made by Japan with cooperation with Topcon in November 2009. The frequency of the instrument was 50–60 Hz with the power of normal 160 VA (Maximum 400 VA). The serial number and PC software edition were 208081 and versions 3.3X, respectively.
In each patient, RNFL was divided into four parts (superior, inferior, nasal and temporal) and they were measured separately.
All measurements were entered to a self-designed checklist included the age of the patients, code of the groups (case: 1 and control: 2), gender, type of the PD (AR, TD) in the case group, duration of the disease, RNFL thickness in superior, inferior, temporal and nasal quadrants in each eye.
The data extracted from the checklists were entered to the SPSS software version 16. To describe qualitative characteristic, frequency index was used. In addition, the continuous data were presented by the mean of the group and standard deviation was considered to be reported as related dispersion index. To specify the distribution style of each data, one-sample Kolmogorov–Smirnov (KS) test was used. Independent sample t test or Mann–Whitney U test (in non-parametric situations) was used to compare the means in two groups of qualitative variables such as subtypes of PD. To understand the correlation between the duration of the disease, Spearman test was used. To evaluate the differences between parametric quantitative variables in the two groups, paired t test (or Wilcoxon signed rank test in non-parametric situations) was performed. To evaluate the effect of confounding variables—duration of the disease—in PD patients, linear regression was used. The data were entered to the model if there was a significant relationship between the dependent variable and covariate/factor using bivariate analysis prior to multivariate analysis. Beyond statistical significance, the data selection was basically performed considering the probable clinical effect of some factors on the dependent variable. The P values <0.05 were considered as significant.
Demographic characteristics in two groups and the P value of the differences
PD group (54 cases)
Control group (50 cases)
54.55 ± 10.37
55 ± 8
40 (74.1 %)
34 (68 %)
The presence of DM
4 (7.4 %)
4 (8 %)
The presence of HTN
4 (7.4 %)
8 (16 %)
The statistical differences of demographic characteristics between two subtypes of patients with PD
TD group (42 cases)
AR group (12 cases)
54.9 ± 9.8
53.33 ± 12.35
34 (80.95 %)
6 (50 %)
The presence of DM
4 (9.52 %)
The presence of HTN
5 (11.90 %)
3 (25 %)
Duration of the disease
6.85 ± 4 y
4.5 ± 1.4 y
The mean RNFL thickness in each quadrant in the case and control group
Temporal quadrant (nm)
Nasal quadrant (nm)
Inf. quadrant (nm)
Sup. quadrant (nm)
Total RNFL thickness (nm)
73.27 ± 12.04
15.79 ± 94.24
12.56 ± 100.29
96.90 ± 18.53
91.18 ± 10.50
80.32 ± 12.25
102.54 ± 18.06
115.24 ± 24.99
112.18 ± 21.94
102.57 ± 13.46
The mean RNFL thickness in each quadrant in the TD and AR group
Temporal quadrant (nm)
Nasal quadrant (nm)
Inf. quadrant (nm)
Sup. quadrant (nm)
Total RNFL thickness (nm)
73.19 ± 13.22
96.5 ± 16.79
101.54 ± 13.69
97.83 ± 20.22
92.26 ± 11.40
73.58 ± 6.89
86.33 ± 7.92
95.91 ± 5.93
93.66 ± 10.79
87.37 ± 5.12
The correlation test between age or duration of the disease and RNFL thicknesses in each quadrant resulted in no significant P value in either Parkinsonism patients (PP) or non-parkinsonism group.
The results of bivariate tests between case and control group are shown in Table 3. The results of independent sample t test for comparing the RNFL changes in TD and AR subtypes are also shown in Table 4.
Since there were significant differences between the duration of the disease in two subtypes of PD patients, multivariate analysis was performed to understand the probable confounding effect of these factors on the differences of RNFL thickness between TD and AR group. Linear regression between the mean RNFL, superior and temporal quadrant thicknesses and the type of the disease, gender and duration of the disease was performed that revealed no changes in their previous non-significant results. By multivariate analysis, the significant results of independent sample t test between inferior and nasal quadrant thickness and type of PD were influenced by entering duration of the disease into the model, but as the adjusted R square of the model and also the P value of the ANOVA table were not significant for both inferior and nasal quadrant thicknesses, and the model was not reportable.
In this study, we compared the RNFL thickness in a large number of PD and non-PD patients. Also the thickness was evaluated in two subtypes of PD. The RNFL thickness in PP was significantly lower than the patients whose eyes were not involved by previously known causes of RNFL impairment (control group). Regarding further analysis, the thickness of inferior and nasal quadrants of RNFL seemed to be changed in AR group more than TD group.
In 1990, dopamine content was found diminished in the retina of PP’ cadavers .
Subsequently, to evaluate the eye neural function in PD, lots of studies were performed. Visual evoked potential (VEP) and color vision studies used to check the visual function did not afford to distinguish the place of impairment if it originated from retinal dysfunction. They all assessed the pathway of entire retina to cortex. VEP and psychophysical assessment provided some data that were comparable with in vivo retinal histology in animals [15, 16].
In 2004, the retinal nerve inner layer thickness was measured in 10 Parkinson patients and a control group with OCT. The thickness in the patient group was significantly lower than the others . In 2008, Altintas et al. investigated the correlation between the severity of PD and morphologic changes of RNFL. In that study, the thickness of RNFL and macula was measured in 17 patients with PD and 11 cases as the control group with OCT. The thickness in PP was significantly lower than the control group .
Regarding our literature review, the largest sample size for this purpose was gathered in a study in 2009. In that study, the RNFL morphologic pathologies in 24 patients with PD were compared with 17 healthy patients by means of OCT. By their results, no significant differences were found between the thickness of the superior and inferior quadrants of the outer layer of the RNFL in patients with PD and the control group. This result was in contrast with a significant P value between the inner retinal layer thickness in patients with PD and the control group .
Though the effect of dopamine diminution in retinal nerves was previously discussed, we tried to understand the effect of these pathologic changes on RNFL thickness in common subtypes of PD. Also the large number of data in this study and the significant pathologic thinning of retinal nerve diameter in PP provide well-organized evidences for future meta-analysis.
In this study, we compared the RNFL thickness in two subtypes of PD (AR and TD) to understand if the neurodegeneration process in AR group was more than TD group or not. To our knowledge, this comparison has not been previously reported.
Although the total peripapillary RNFL thickness in AR group was less than TD group, it was not significantly different. This might happen due to the small sample size especially in AR group. Also tremor in TD patients is usually associated with fixation during the performance of OCT that can cause inaccurate interpretation. The duration of the disease was another factor that could affect our results.
The significant RNFL thickness reduction in nasal and inferior quadrants of AR group expressed that the neurodegenerative process in AR group was more severe than the TD group. Also the pattern of involvement showed that impairment can be followed by a specific pattern starting from inferior and nasal quadrants. Although this cross-sectional study cannot confirm these hypotheses, it motivates future studies with larger sample sizes to understand if there are any specific patterns for RNFL involvement. If the results will be confirmed, starting neuroprotective treatments in patients with Parkinson disease who do not present with tremor can prevent neurodegeneration process.
Because the visual impairments may occur without any obvious alterations in the general ophthalmological follow-ups, electrophysiological recordings can provide some useful information prior to clinical presentations .
Considering the role of dopamine as an efficient modulator in the retinal neurochemical system, diminution of dopaminergic reservoirs effects on GABA, glycine and glutamate. One of the important anxieties is the over production of glutamate that is usually followed by an atrophic process in axonal fibers .
In our study, lots of patients in the case group and even the control group did not agree to undergo OCT as they understood that it was not associated with any therapeutic effects.
Although the sample size of our study was more than previous studies, it was not enough for multivariate analysis. In this study, we tried to reduce the effect of some probable confounding factors on RNFL thickness such as age and the duration of the disease in PP, but the regression indexes were not worthy to be reported due to the lack of enough sample size. According to this limitation, future studies with adequate sample sizes are highly recommended to investigate interactions between age, distribution of the disease and type of PD as well as the effects of individual factors.