Diederich NJ, Raman R, Leurgans S, Goetz CG. Progressive Worsening of Spatial and Chromatic Processing Deficits in Parkinson Disease. Arch Neurol. 2002;59(8):1249-1252. doi:10.1001/archneur.59.8.1249
Impairments of color discrimination (CD) and contrast sensitivity are established signs of Parkinson disease (PD), but their temporal evolution has not been studied.
To determine whether there is progressive, longitudinal deterioration of color discrimination (CD) and contrast sensitivity (CS) in PD.
A prospective study.
Tertiary care center–based sample of PD patients without dementia with normal visual acuity (Snellen fraction >0.6 in the best eye).
Main Outcome Measures
With a mean ± SD interval of 19.8 ± 2.8 months, the following tests were applied twice in 28 patients: the Lanthony D15 test and the Farnsworth Munsell 100 Hue test as tests of CD and the monocular and binocular Pelli-Robson test and the binocular Vistech tables as tests of CS.
There was deterioration of both CD (Farnsworth Munsell 100 hue test: P = .002) and CS (binocular Vistech test at a spatial frequency of 6 cycles per degree, P<.001). Both deficits correlated with age, and the chromatic deficit additionally correlated with higher impairment of motor function (Unified Parkinson's Disease Rating Scale motor section, P = .04) and activities of daily life (Unified Parkinson's Disease Rating Scale activities of daily living section, P = .006). Patients with the highest pathologic psychiatric rating score (Brief Psychiatric Rating Scale) performed worse on both CS (P = .02) and CD (P = .01) at the second examination.
Impairments of CD and CS in PD are progressive over time. Visual deficits may influence overall motor function and lead to enhanced motor impairment.
IN PARKINSON DISEASE (PD) visual pathways can be affected at different levels. In autopsy studies of unmedicated PD patients, retinal dopamine concentration is reduced.1 MPTP (1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine) intoxication in patients and monkeys produces a pathologic electroretinogram, implicating a role of dopamine in the center-surround organization of the receptive field.2 Pathologic changes beyond the retina are suggested by abnormal responses of the visual P300 waves and by reduced cerebral metabolism in the occipital area.3 In the clinical setting, visual processing deficits have been established in several studies. Patients with PD have reduced contrast sensitivity (CS), with maximal deficits in horizontal gratings and medium spatial frequencies. These changes are influenced by stage and duration of the disease and circadian variability.4- 6 The deficits of color discrimination (CD) are detected early in the disease7; the tritan axis is most affected, and dopaminergic treatment can improve the deficits. To our knowledge, there has been no longitudinal study of visual impairment with correlation to other parkinsonian features. We therefore designed a study to determine whether there is deterioration of CS and CD in PD and to monitor such changes in the context of clinical motor function.
Thirty-one patients with idiopathic PD who had previously participated in visual tests volunteered for this follow-up study. Three were excluded because of low visual acuity as defined by a Snellen fraction of less than 0.6 in the best eye (2 patients) and dementia as defined as a score of less than 24 on the Mini-Mental State Examination (1 patient). Thus, 28 patients were examined twice and their visual test results were compared. None of the study subjects had florid psychosis as determined by the Brief Psychiatric Rating Scale (BPRS), inborn color blindness as defined by the Ishihara plates, or other coexistent neurologic disease potentially worsening visual perception. Each patient was examined in the same way as in the first session with the same investigator, room, and lighting conditions. The same battery of ophthalmologic and clinical examinations was applied in the same sequence, with the patients being "on" (mobile) and taking their usual medications. The examiner did not have access to the individual test results from the first session. In addition to the tests already mentioned, these examinations included the Lanthony D15 test (LD) and the Farnsworth Munsell 100 Hue test (FM) as tests of CD and the monocular and binocular Pelli-Robson test (PR) and the binocular Vistech tables (VT) as tests of CS. These test pairs are each time complementary: the LD is a succinct CD screening test, and the FM is a largely accepted and user-friendly CD test that separately explores different color axes. The PR compares monocular vs binocular CS, whereas the VT explores different spatial frequencies (cycles per degree [cpd]).8 All these tests have already been applied to PD patients; they have easy instructions and require only short spans of attention. We further applied the Unified Parkinson's Disease Rating Scale (UPDRS), including the section of activities of daily living (ADL), and the Schwab and England test. Both test series were performed between 8 AM and 5 PM. All technical prerequisites were identical and have been described in detail.9 The mean ± SD time interval between the examinations was 19.8 ± 2.8 months. The statistical analysis used standard descriptive statistics, Wilcoxon signed rank tests, and Spearman rank correlation coefficients. Continuous data are summarized as means ± SDs. Statistical significance is recognized at P<.05.
On the follow-up examination, the mean ± SD age was 65.2 ± 12.3 years; PD duration, 13.53 ± 9.14 years; and levodopa treatment duration, 8.57 ± 7.04 years. At the follow-up examination, patients performed worse on all the visual tests with the exception of the LD, the results of which showed a nonsignificant improvement. The deteriorations concerned the total color spectrum and monocular and binocular CS, especially at the middle spatial frequencies. Statistically significant declines were seen with the FM, for 3 of 5 VT scores, and at the monocular PR (P = .04 to P<.001) (Table 1).
The comparison of the most significant visual findings (the VT score at 6 cpd for CS and the total FM score for CD) with covariables showed that the deterioration in CS (follow-up value minus baseline value) correlated with age (P<.001), duration of PD (P = .02), and follow-up scores of Hoehn and Yahr stage (P = .04), UPDRS motor (P = .04), and ADL (P = .006) assessments. There was no significant correlation with levodopa dosage or UPDRS mental score. The decline in CD (follow-up value minus baseline value) showed only a mild correlation with age (P = .04). The comparison with psychiatric rating scores revealed that the patients with the highest BPRS scores at the follow-up examination performed worse on both CS (P = .02) and CD (P = .01) at this examination. However, there was no statistically significant group decline in motor, behavioral, or global function, as tested by the UPDRS and Schwab and England tests, and the medication remained the same (Table 2).
In PD, CS and CD have been shown to be independent of each other; their impairment is worse in PD patients than in age-matched controls.6 In contrast to prior cross-sectional studies, ours is the first longitudinal study and documents the progressive deficits in both CD and CS. We applied a strict study protocol to ensure that the 2 testing sessions were identical. With the exception of the LD, the deterioration of CS and CD was evident throughout the test battery. The most powerful changes were seen in the tests that are also the most sensitive ones in cross-sectional studies, namely, the total score of the FM for CD and the changes at the medium spatial frequencies for CS.10 These changes cannot be owing to a decline in visual acuity, because we excluded patients with low Snellen scores from the follow-up examination. The decline of contrast vision was directly correlated with follow-up motor function scores and age. Nevertheless, deterioration of CS and CD cannot be seen as an exclusive aging phenomenon for several reasons: (1) cross-sectional studies have convincingly demonstrated that, in terms of decline of CD and CS, PD patients are "in advance" (doing worse) compared with age-matched peers; (2) influence of visual aging of CS is most prominent at high spatial frequencies,11 a phenomenon not seen in this study; and (3) aging of CD acts through blue-yellow confusions,12 and again, there was no significant change of this section of the FM in our group. Additive or overlapping effects of aging and PD are, however, possible. It has been proposed that motion sensitivity (spatial frequencies of 1 and 4 cpd, seen at temporal frequencies of 1 and 3 Hz) can better discriminate than static sensitivity between PD progression and aging.10 However, motion detection is not only luminance dependent but also color specific,13 which brings us back to test CD; furthermore, there are no confirming longitudinal studies for this hypothesis.
No specific item in the UPDRS focuses on visual function, so the clinical impact of our observations cannot be fully delineated. However, the correlation of visual deficits and UPDRS and ADL follow-up scores suggests that the progressive character of color and contrast deficits may affect patient function. The link with the psychiatric rating (BPRS) does not tell us which comes first, visual impairment or psychiatric deterioration; although, in a previous study,9 we demonstrated that patients with visual impairment are at higher risk for developing visual hallucinations. Because of the insidious character of deterioration of contrast and color perception, the patients may not be able to compensate in the same way as they might for a suddenly apparent but nonprogressive visual deficit. Lack of correct visual feedback may subjectively "rattle" the patient while walking. Errors of contrast and color perception may raise the potential for overinterpretations or misinterpretations of the visual input and thus cause or contribute to various clinical signs, such as frequent falls, misperceptions of distances and contours, illusions, and even visual hallucinations.9,14 A larger follow-up study using multivariate techniques should clarify possible links to these frequent signs of advanced PD.
Accepted for publication May 7, 2002.
Author contributions: Study concept and design (Drs Diederich and Goetz); acquisition of data (Dr Diederich); analysis and interpretation of data (Drs Diederich, Leurgans, and Goetz and Ms Raman); drafting of the manuscript (Drs Diederich and Goetz and Ms Raman); critical revision of the manuscript for important intellectual content (Drs Diederich, Leurgans, and Goetz); statistical expertise (Drs Diederich and Leurgans and Ms Raman); obtained funding (Dr Diederich); administrative, technical, and material support (Dr Goetz); study supervision (Dr Goetz).
This study was supported by Foundation Think, Luxembourg, and the Parkinson Disease Foundation, New York, NY.
This study was presented as a poster at the Annual Meeting of the American Academy of Neurology, San Diego, Calif, May 1, 2000.
Corresponding author and reprints: Nico J. Diederich, MD, Department of Neuroscience, Centre Hospitalier de Luxembourg, 4, rue Barblé, L-1210, Luxembourg (e-mail: firstname.lastname@example.org).