Structural changes in the retina and serum HMGB1 levels are associated with decreased cognitive function in patients with Parkinson's disease

Neurobiol Dis. 2024 Jan:190:106379. doi: 10.1016/j.nbd.2023.106379. Epub 2023 Dec 15.

Abstract

Background: Cognitive impairment is a serious nonmotor symptom in patients with Parkinson's disease (PD). Currently, there are few studies investigating the relationship of serum markers and retinal structural changes with cognitive function in PD.

Objective: To investigate the relationship between retinal structural changes, serum high mobility group box-1 (HMGB1) levels and cognitive function and motor symptoms in PD patients.

Methods: Eighty-nine participants, including 47 PD patients and 42 healthy subjects, were enrolled. PD patients were divided into Parkinson's disease with normal cognitive (PD-NC), Parkinson's disease with mild cognitive impairment (PD-MCI), and Parkinson's disease with dementia (PDD) groups. The motor and nonmotor symptoms of PD patients were evaluated with clinical scale. Serum HMGB1 levels were detected by enzyme-linked immunosorbent assay (ELISA), and ganglion cell-inner plexiform layer complex (GCIPL) thickness changes in the macula were quantitatively analyzed by swept source optical coherence tomography (SS-OCT) in all patients.

Results: Compared with the control group, the macular GCIPL (t = -2.308, P = 0.023) was thinner and serum HMGB1 (z = -2.285, P = 0.022) was increased in PD patients. Macular GCIPL thickness in patients with PD-MCI and PDD were significantly lower than that in PD-NC patients, but there were no significant difference between the PD-MCI and PDD groups. Serum HMGB1 levels in patients with PD-MCI and PDD were significantly higher than those in PD-NC patients, and serum HMGB1 levels in PDD patients were higher than those in PD-MCI patients. Correlation analysis showed that serum HMGB1 levels in PD patients were positively correlated with disease duration, HY stage, UPDRS-I score, UPDRS-III score, and UPDRS total score and negatively correlated with MOCA score. Macular GCIPL thickness was negatively correlated with HY stage and positively correlated with MOCA score, and macular GCIPL thickness was negatively correlated with serum HMGB1 level. Logistic regression analysis showed that elevated serum HMGB1 level, thinner macular GCIPL thickness, and higher HY stage were independent risk factors for Parkinson's disease with cognitive impairment (PD-CI). The areas under the receiver operating characteristic curve (AUC) for the serum HMGB1 level and macular GCIPL thickness-based diagnosis of PD-MCI, PDD and PD-CI based on in patients with PD were 0.786 and 0.825, 0.915 and 0.856, 0.852 and 0.841, respectively. The AUC for the diagnosis of PD-MCI, PDD and PD-CI with serum HMGB1 level and GCIPL thickness combined were 0.869, 0.967 and 0.916, respectively.

Conclusion: The macular GCIPL thickness and serum HMGB1 level are potential markers of cognitive impairment in PD patients, and their combination can significantly improve the accuracy of the diagnosis of cognitive impairment in PD.

Keywords: Biomarkers; Cognitive function; GCIPL; HMGB1; Nonmotor symptoms; Parkinson's disease.

MeSH terms

  • Alzheimer Disease*
  • Cognition
  • Cognitive Dysfunction*
  • HMGB1 Protein*
  • Humans
  • Parkinson Disease*
  • Retina

Substances

  • HMGB1 Protein
  • HMGB1 protein, human