https://orcid.org/0000-0003-1336-3276
Figures
Abstract
Patients with older-onset Parkinson’s disease (PD) have more severe motor symptoms, faster progression, and a worse prognosis. The thinning of the cerebral cortex is one of the causes of these issues. Patients with older-onset PD manifest more extended neurodegeneration associated with α-synuclein deposition in the cerebral cortex; however, the cortical regions that undergo thinning are unclear. We aimed to identify cortical regions with different thinning depending on the age of onset in patients with PD. Sixty-two patients with PD were included in this study. Patients with PD onset at <63 years old were included in the early or middle-onset PD group, and those with PD onset at >63 years old were included in the late-onset PD (LOPD) group. Brain magnetic resonance imaging data of these patients were processed using FreeSurfer to measure their cortical thickness. The LOPD group displayed less cortical thickness in the superior frontal gyrus, middle frontal gyrus, precentral gyrus, postcentral gyrus, superior temporal gyrus, temporal pole, paracentral lobule, superior parietal lobule, precuneus, and occipital lobe than the early or middle-onset PD group. Compared with patients with early and middle-onset PD, elderly patients displayed extended cortical thinning with disease progression. Differences in the clinical manifestations of PD according to the age of onset were partly due to variations in the morphological changes in the brain.
Citation: Seo K, Matunari I, Yamamoto T (2023) Cerebral cortical thinning in Parkinson’s disease depends on the age of onset. PLoS ONE 18(2): e0281987. https://doi.org/10.1371/journal.pone.0281987
Editor: Niels Bergsland, University at Buffalo, UNITED STATES
Received: October 17, 2021; Accepted: February 6, 2023; Published: February 21, 2023
Copyright: © 2023 Seo et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting Information files.
Funding: The authors received no specific funding for this work.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Parkinson’s disease (PD) is a neurodegenerative disorder caused by the loss of dopaminergic neurons in the substantia nigra; however, its clinical phenotype is complex and heterogeneous [1,2], with differing motor and non-motor symptoms, age of onset, and rate of progression in individual patients. Among these factors, the age of onset may determine the clinical phenotype [3–6]. The older the patient, the more severe the motor symptoms are, particularly bradykinesia, gait disturbance, postural instability [7,8], and disease progression. Furthermore, older patients are more likely to develop dementia, hallucinations, insomnia, constipation, and olfactory disturbances [8,9] and have a poorer prognosis [10]. These findings suggest a strong relationship between aging and cerebral cortex degeneration. The specific degeneration of dopaminergic neurons in the substantia nigra is a pathological hallmark of PD; moreover, neurodegeneration occurs in cortical areas with disease progression. The appearance of Lewy body pathology in the cerebral cortex begins in the anterior medial temporal cortex, followed by the higher sensory association areas, prefrontal cortex, primary association sensory areas, premotor areas, and primary sensory and motor areas [11]. Additionally, neurodegeneration due to cortical α-synuclein deposition is more common in elderly patients with PD [12]. The cortical regions that are more severely affected in such patients have not yet been identified. This necessitates clarifying the relationship between the clinical characteristics and pathophysiology in elderly patients with PD.
FreeSurfer is an open-source software that is used for quantifying and analyzing brain magnetic resonance imaging (MRI) images. A quantitative analysis of brain MRI images was previously performed by manually setting the region of interest for each slice. This requires considerable time and effort, but it cannot analyze the whole brain. With the development of computer technology, researchers have created methods and software for quantitatively analyzing brain MRI. Voxel-based morphometry (VBM) [13] was established in 2000 and is a semi-automatic method that can analyze the whole brain in voxel units. It has been used in brain MRI studies instead of manual methods. FreeSurfer is equipped with surface-based morphometry (SBM), which is specialized for the cortical analysis of brain MRI quantitative analysis methods [14]. SBM is used in brain MRI research and can measure the cortical thickness, cortical surface area, and gyrus in the whole brain unlike VBM.
The purpose of this study was to identify cortical regions with reduced thickness in patients with PD depending on the age of onset by using FreeSurfer as a tool for SBM.
Materials and methods
Subjects
We enrolled 62 patients who met the clinical diagnosis criteria of PD by the Movement Disorder Society [6] and who visited Saitama Medical University Hospital between January 2000 and September 2019. This study was a retrospective study and was approved by the ethical review committee of Saitama Medical University (Approval No. 19074.01). Patients with a history of heavy alcohol consumption; anticancer drug use; psychiatric disorders; neurodegenerative diseases not including PD; and intracranial lesions, such as cerebral infarction, cerebral hemorrhage, and head trauma with the loss of consciousness, were excluded from the study. Patients had one or more MRI examinations under the imaging conditions described in the MRI acquisition. Patients with PD onset at <63 years old were included in the early or middle-onset PD (E-MOPD) group. Patients with PD onset at ≥63 years old were included in the late-onset PD (LOPD) group. The standard age of 63 years old was based on the average age of PD onset in Japanese patients [15]. The age of onset was defined as the time when the patient noticed motor symptoms. The disease duration (years) and levodopa equivalent daily dose (LEDD) [16] were used as indicators of disease progression.
MRI acquisition.
We performed MRI examinations by using a 1.5T magnetic resonance system (Magnetom ESSENZA Siemens Healthcare, Erlangen, Germany). The acquisition parameters were as follows: 3D magnetization-prepared rapid acquisition gradient echo, head matrix coil, 800 ms inversion time, 1700 ms repetition time, 4 ms time to echo, 12° fractional anisotropy, 230 × 230 field of view, 256 × 256 matrix size, and 1.2 mm slice thickness. MRI examinations were performed 119 times (minimum: once; maximum: five times per patient).
MRI processing.
We used MacOS Mojave version 10.14.6 for MRI image processing. FreeSurfer (version 6.0; Massachusetts General Hospital, Harvard Medical School) was used to estimate and analyze the cortical thickness from 3D T1 images. First, we created an automated cortical model as a preprocessing step. Cortical model creation processing includes motion correction, averaging of multiple T1 images, removal of non-brain tissue, automated Talairach transformation, segmentation of subcortical and deep gray matter volume structures, signal value normalization, gray matter boundary tessellation, automated geometric correction, maximum signal value turning defined as the transition to other tissues, and surface deformation based on signal value gradients that optimally locate gray and white matter boundaries and gray matter cerebrospinal fluid boundaries. The process for estimating the cortical thickness from a complete cortical model includes the registration of the cortical model to a sphere, cortical segmentation and labeling, and the estimation of the cortical thickness from soft and white matter surfaces. The obtained data were subjected to quality control according to the ENIGMA Cortical Quality Control Protocol 2.0 (http://enigma.ini.usc.edu/protocols/imaging-protocols/). No participants were excluded because of excessive superficial or subcortical segmentation errors.
Statistical analyses
Demographic variables.
We performed the demographic and clinical statistical analyses using JMP 13.2.0 (SAS Institute Inc., Cary, NC, USA). We tested for group differences in the demographic and clinical variables between the E-MOPD and LOPD groups by using Mann–Whitney pairwise comparisons or Pearson’s chi-squared test. Statistical significance was set at p < 0.05.
Cortical thickness analyses.
Statistical analysis of the cortical thickness was performed using a general linear model in Qdec (http://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/QdecGroupAnalysisV6.0) embedded in FreeSurfer. We examined the areas of cerebral cortical thinning in relation to disease progression. Additionally, we investigated between-group differences in the slope of regression lines between the cortical thickness and disease progression. These represented areas had more rapid thinning with disease progression in LOPD than in E-MOPD. S1 Fig provides a schematic of this concept. To remove the effect of age, we used the age of onset as a covariate in the correlation analysis between the clinical course and cortical thickness in each group. The age during MRI imaging was used as a covariate in the intergroup comparison of the correlation between the cortical thickness and clinical course in both groups. Statistical results were corrected for multiple comparisons per cluster by using the Monte Carlo method. All resulting cortical areas were significantly different (p < 0.05).
Results
Demographic variables and clinical information during MRI imaging
There were no significant differences in the demographic variables or clinical information between the E-MOPD and LOPD groups except for age. Table 1 summarizes the demographic variables and clinical information during MRI.
There were no significant differences in the demographic variables or clinical information between the groups, except that the E-MOPD group was younger than the LOPD group.
Correlation between cortical thickness and disease progression
Correlation between cortical thickness and disease progression in all patients.
By using the disease duration as an index of disease progression, we observed significant negative correlations between the cortical thickness and disease progression in the bilateral superior frontal gyrus, middle frontal gyrus, precentral gyrus, postcentral gyrus, paracentral lobule, and right occipital lobe (p < 0.05) (Fig 1). Similarly, by using LEDD as an index of disease progression, we recorded significant negative correlations between the cortical thickness and disease progression in the bilateral precentral gyrus, paracentral lobule, superior parietal lobule, precuneus, and occipital lobule (p < 0.05) (Fig 1). S1 Table outlines the identified clusters. The cluster information includes coordinates of the Montreal Neurological Institute (MNI) displaying the maximum -log10 p-value in the identified cluster, cortical area (anatomical location of MNI coordinates), cluster size (mm2), and clusterwise p-value (p-value for each cluster).
Color maps indicate areas with significant thinning (p < 0.05). LEDD, levodopa equivalent daily dose.
Correlation between cortical thickness and disease progression in E-MOPD and LOPD.
When using the disease duration as an index of disease progression, there were significant negative correlations between the cortical thickness and disease progression in the superior frontal gyrus, middle frontal gyrus, central precentral gyrus, central posterior gyrus, and bilateral paracentral lobule of the E-MOPD and in the superior frontal gyrus, middle frontal gyrus, central precentral gyrus, central posterior gyrus, paracentral lobule, precuneus, and bilateral occipital lobe bilaterally of the LOPD group (p < 0.05) (Fig 2). When using LEDD as an index of disease progression, there was no significant correlation in the E-MOPD group (Fig 2). By contrast, the LOPD group displayed significant negative correlations between the cortical thickness and disease progression in the superior frontal gyrus, middle frontal gyrus, precentral gyrus, postcentral gyrus, paracentral lobule, superior parietal lobule, precuneus, occipital lobe, and left superior temporal gyrus on both sides (p < 0.05) (Fig 2). S2 and S3 Tables present information on the identified clusters.
(a) Areas of cortical thinning with prolonged disease duration. (b) Areas of cortical thickness thinning with an increase in LEDD. LEDD, levodopa equivalent daily dose; E-MOPD, early or middle-onset Parkinson’s disease; and LOPD, late-onset Parkinson’s disease.
Between-group differences in the slope of regression lines between cortical thickness and disease progression.
When using the disease duration as an index of disease progression, the precuneus displayed more rapid cortical thinning in the LOPD group than in the E-MOPD group (p < 0.05) (Fig 3). When using LEDD as an index of disease progression, the bilateral superior temporal gyrus, paracentral lobule, superior parietal lobule, occipital lobule, right superior frontal gyrus, middle frontal gyrus, temporal pole, left precentral gyrus, left postcentral gyrus, and precuneus revealed more rapid cortical thinning in the LOPD group than in the E-MOPD group (p < 0.05) (Fig 3). There were no areas of rapid cortical thinning in the E-MOPD group compared with the LOPD group. S4 Table presents information on the identified clusters.
LEDD, levodopa equivalent daily dose; E-MOPD, early or middle-onset Parkinson’s disease; and LOPD, late-onset Parkinson’s disease.
Discussion
This is the first study to capture age-dependent differences in cortical thinning during disease progression in PD. The LOPD group displayed more extensive cortical thinning with disease progression than the E-MOPD group.
Several reports have been published on the relationship between the clinical symptoms of PD and cortical thinning. In terms of motor symptom severity, bradykinesia is associated with the thinning of the parietotemporal sensory association area [17]. Postural instability is associated with thinning of the cuneus and precuneus [18], whereas tremor is associated with the thinning of a wide range of cortical regions, including the bilateral middle frontal gyrus, left orbitofrontal cortex, bilateral cingulate gyrus, left middle temporal gyrus, left lateral temporal gyrus, left inferior parietal lobule, and left occipital gyrus [19]. Among non-motor symptoms, cognitive dysfunction and psychiatric symptoms are closely related to cerebral cortex atrophy. PD with mild cognitive impairment (PD-MCI) is characterized by the thinning of the parietal and occipital cortices compared with PD without dementia. When combined with the thinning of the frontal cortex, PD-MCI reportedly progresses to PD with dementia [20–23]. Patients with PD-MCI without the thinning of the parietal and occipital lobes recover normal cognitive function, but those with thinning of the aforementioned areas do not [21]. Cognitive domain–specific reports have demonstrated that attention and executive function deficits were correlated with the thinning of frontal regions [22,24]. Moreover, visuospatial cognitive deficits were correlated with the thinning of the parietal and occipital lobes [20]. Radziunas et al. [25] and Kubera et al. [26] mentioned that impulse control disorder is correlated with the thinning of the inferior frontal gyrus, superior frontal gyrus, anterior cingulate gyrus, and precentral gyrus in PD. A study of the correlation between psychological symptoms and cortical thinning in patients with PD, as assessed by the Caregiver-Administered Neuropsychiatric Inventory, demonstrated that nocturnal behavior and irritability were correlated with frontal cortex thinning, aggression and agitation with temporal lobe thinning, and apathy with insular cortex thinning [27].
In this study, we found that the LOPD group had cortical regions with more rapid thinning than the E-MOPD group. Among all cortical areas that displayed more rapid cortical thinning in the LOPD group, the superior temporal gyrus, temporal pole, paracentral lobule, superior parietal lobule, precuneus, and occipital lobe were reportedly associated with bradykinesia [17], the precuneus and occipital lobe were associated with postural instability [18,19], and extensive cortices from the frontal to the parietal and occipital lobes were associated with cognitive dysfunction [20–23]. The pattern of cortical thinning observed in the current study may be explained by the pathophysiology underlying the clinical features of elderly patients with PD, such as α-synuclein deposition in elderly patients [12]. The serum level of uric acid, which is an antioxidant, is reportedly low in patients with PD; this situation may be attributed to the weakening of the neuroprotective mechanisms of the body [28]. In addition to α-synuclein deposition, amyloid-β deposition is also observed in older patients with PD and may contribute to the difference in thinning depending on the age of onset [29,30]. The cause of rapid PD progression in elderly patients has not yet been clarified. However, this clarification will likely lead to the development of therapies that inhibit disease progression.
A limitation of this study was its retrospective nature because it made the acquisition of motor and non-motor symptoms difficult in a standardized manner. Therefore, we were unable to analyze the relationship between cortical thinning and the characteristics of motor symptoms, such as tremor, bradykinesia, and postural instability, as well as non-motor symptoms, such as cognitive dysfunction and autonomic symptoms. Although it is desirable to standardize the number of MRI scans for patients, this study had a retrospective design and could not be standardized.
Conclusion
Patients with older-onset PD demonstrated more extensive areas of reduced cortical thickness compared with those with early and middle-onset PD. Furthermore, patients with older-onset PD had cortical areas with more rapid thinning. Therefore, brain structural MRI with automated cortical thickness analysis may provide clues for a better understanding of age-related structural changes in patients with PD in an objective manner, which may be related to the underlying pathophysiology.
Supporting information
S1 Fig. A schematic image depicting the difference in the slope of regression lines between the E-MOPD and LOPD groups.
E-MOPD, early or middle-onset Parkinson’s disease; LOPD, late-onset Parkinson’s disease.
https://doi.org/10.1371/journal.pone.0281987.s001
(DOCX)
S1 Table. Information on the identified clusters in Fig 1.
https://doi.org/10.1371/journal.pone.0281987.s002
(DOCX)
S2 Table. Information on the identified clusters in Fig 2(A).
https://doi.org/10.1371/journal.pone.0281987.s003
(DOCX)
S3 Table. Information on the identified clusters in Fig 2(B).
https://doi.org/10.1371/journal.pone.0281987.s004
(DOCX)
S4 Table. Information on the identified clusters in Fig 3.
https://doi.org/10.1371/journal.pone.0281987.s005
(DOCX)
References
- 1. Berg D, Postuma RB, Bloem B, Chan P, Dubois B, Gassere T, et al. Time to redefine PD? Introductory statement of the MDS Task Force on the definition of Parkinson’s disease. Mov Disord. 2014 Apr;29(4):454–462. pmid:24619848
- 2. Fereshtehnejad SM, Zeighami Y, Dagher A, Postuma RB. Clinical criteria for subtyping Parkinson’s disease: biomarkers and longitudinal progression, Brain. 2017 Jul;140(7):1959–1976. pmid:28549077
- 3. Lewis SJ, Foltynie T, Blackwell AD, Robbins TW, Owen AM, Barker RA. Heterogeneity of Parkinson’s disease in the early clinical stages using a data driven approach. J Neurol Neurosurg Psychiatry. 2005 Mar;76(3):343–348. pmid:15716523
- 4. Reijnders JS, Ehrt U, Lousberg R, Aarsland D, Leentjens AF. The association between motor subtypes and psychopathology in Parkinson’s disease. Parkinsonism Relat Disord. 2009 Jun;15(5):379–382. pmid:18977165
- 5. Liu P, Feng T, Wang YJ, Zhang X, Chen B. Clinical heterogeneity in patients with early-stage Parkinson’s disease: a cluster analysis. J Zhejiang Univ Sci B. 2011 Sep;12(9):694–703. pmid:21887844
- 6. Postuma RB, Berg D, Stern M, Poewe W, Olanow CW, Oertel W, et al. MDS clinical diagnostic criteria for Parkinson’s disease. Mov Disord. 2015 Oct;30(12):1591–1601. pmid:26474316
- 7. Wickremaratchi MM, Ben-Shlomo Y, Morris HR. The effect of onset age on the clinical features of Parkinson’s disease. Eur J Neurol. 2009 Apr;16(4):450–456. pmid:19187262
- 8. Keezer MR, Wolfson C, Postuma RB. Age, Gender, Comorbidity, and the MDS-UPDRS: Results from a Population-Based Study. Neuroepidemiology. 2016;46(3):222–227.
- 9. Zhou MZ, Gan J, Wei YR, Ren XY, Chen W, Liu ZG. The association between non-motor symptoms in Parkinson’s disease and age at onset. Clin Neurol Neurosurg. 2013 Oct;115(10):2103–2107.
- 10. Kempster PA O’Sullivan SS, Holton JL, Revesz T, Lees AJ. Relationships between age and late progression of Parkinson’s disease: a clinico-pathological study. Brain. 2010 Jun;133(Pt 6):1755–1762.
- 11. Braak H, Del Tredici K, Rub U, de Vos RA, Jansen Steur EN, Braak E. Staging of brain pathology related to sporadic Parkinson’s disease. Neurobiol Aging. 2003;24: 197–211. pmid:12498954
- 12. Selikhova M, Williams DR, Kempster PA, Holton JL, Revesz T, Lees AJ. A clinico-pathological study of subtypes in Parkinson’s disease. Brain. 2009 Nov;132(Pt 11):2947–2957. pmid:19759203
- 13. Ashburner J, Friston KJ. Voxel-based morphometry—the methods. Neuroimage. 2000 Jun;11(6 Pt 1):805–821.
- 14. Fischl B, Dale AM. Measuring the thickness of the human cerebral cortex from magnetic resonance images. Proc Natl Acad Sci U S A. 2000 Sep;97(20):11050–11055. pmid:10984517
- 15. Taniguchi A, Narita Y, Naito Y, Kuzuhara S. Analysis of application form for Parkinson’s disease provided by the specific diseases treatment research program of Ministry of Health, Labour and Welfare of Japan. Rinsho Shinkeigaku. 2008 Feb;48(2):106–113. pmid:18326303
- 16. Tomlinson CL, Stowe R, Patel S, Rick C, Gray R, Clarke CE. Systematic review of levodopa dose equivalency reporting in Parkinson’s disease. Mov Disord. 2010 Nov;25(15):2649–2653. pmid:21069833
- 17. Lyoo CH, Ryu YH, Lee MS. Cerebral cortical areas in which thickness correlates with severity of motor deficits of Parkinson’s disease. J Neurol. 2011 Oct;258(10):1871–1876. pmid:21512741
- 18. Herb JN, Rane S, Isaacs DA, Van Wouwe N, Roman OC, Landman BA, et al. Cortical Implications of Advancing Age and Disease Duration in Parkinson’s Disease Patients with Postural Instability and Gait Dysfunction. J Parkinsons Dis. 2016 May 5;6(2):441–451. pmid:27164041
- 19. Benito-Leon J, Serrano JI, Louis ED, Holobar A, Romero JP, Povalej-Brzan P, et al. Tremor severity in Parkinson’s disease and cortical changes of areas controlling movement sequencing: A preliminary study. J Neurosci Res. 2018 Aug;96(8):1341–1352. pmid:29660812
- 20. Garcia-Diaz AI, Segura B, Baggio HC, Uribe C, Campabadal A, Abos A, et al. Cortical thinning correlates of changes in visuospatial and visuoperceptual performance in Parkinson’s disease: A 4-year follow-up. Parkinsonism Relat Disord. 2018 Jan;46:62–68. pmid:29132765
- 21. Chung SJ, Park YH, Yoo HS, Lee YH, Ye BS, Sohn YH, et al. Mild cognitive impairment reverters have a favorable cognitive prognosis and cortical integrity in Parkinson’s disease. Neurobiol Aging. 2019 Jun;78:168–177. pmid:30947112
- 22. Chung SJ, Yoo HS, Lee YH, Lee HS, Ye BS, Sohn YH, et al. Frontal atrophy as a marker for dementia conversion in Parkinson’s disease with mild cognitive impairment. Hum Brain Mapp. 2019 Sep;40(13):3784–3794. pmid:31090134
- 23. Gasca-Salas C, Garcia-Lorenzo D, Garcia-Garcia D, Clavero P, Obeso JA, Lehericy S, et al. Parkinson’s disease with mild cognitive impairment: severe cortical thinning antedates dementia. Brain Imaging Behav. 2019 Feb;13(1):180–188. pmid:28710667
- 24. Zhang L, Wang M, Sterling NW, Lee EY, Eslinger PJ, Wagner D, et al. Cortical Thinning and Cognitive Impairment in Parkinson’s Disease without Dementia. IEEE/ACM Trans Comput Biol Bioinform. 2018 Mar-Apr;15(2):570–580. pmid:29610105
- 25. Radziunas A, Deltuva VP, Tamasauskas A, Gleizniene R, Pranckeviciene A, Petrikonis K, et al. Brain MRI morphometric analysis in Parkinson’s disease patients with sleep disturbances. BMC Neurol. 2018 Jun;18(1):88. pmid:29925331
- 26. Kubera KM, Schmitgen MM, Nagel S, Hess K, Herweh C, Hirjak D, et al. A search for cortical correlates of trait impulsivity in Parkinsons disease. Behav Brain Res. 2019 Sep;369:111911.
- 27. Ye BS, Jeon S, Yoon S, Kang SW, Baik K, Lee Y, et al. Effects of dopaminergic depletion and brain atrophy on neuropsychiatric symptoms in de novo Parkinson’s disease. J Neurol Neurosurg Psychiatry. 2018 Feb;89(2):197–204. pmid:28951497
- 28. Schirinzi T, Di Lazzaro G, Sancesario GM, Summa S, Petrucci S, Colona VL, et al. Young-onset and late-onset Parkinson’s disease exhibit a different profile of fluid biomarkers and clinical features. Neurobiol Aging. 2010 Jun;90:119–124.
- 29. Jellinger KA, Seppi K, Wenning GK, Poewe W. Impact of coexistent Alzheimer pathology on the natural history of Parkinson’s disease. J N Transm (Vienna). 2002 Mar;109(3):329–339. pmid:11956955
- 30. Mastaglia FL, Johnsen RD, Byrnes ML, Kakulas BA. Prevalence of amyloid-beta deposition in the cerebral cortex in Parkinson’s disease. Mov Disord. 2003 Jan;18(1):81–86. pmid:12518303