Slower saccadic reading in Parkinson's disease

Abstract

Idiopathic Parkinson's Disease (PD) is characterized by degeneration of dopaminergic and other neurons, leading to motor and non-motor deficits. Abnormal eye movements in PD, including fixations, saccades, and convergence, are well described. However, saccadic reading, which requires serial and alternating saccades and fixations, is not well studied, despite its obvious impact on the quality of life. In this study, we assessed saccadic reading using variations of the King-Devick (KD) test, a rapid single digit number naming test, as a way to assess the ability to make serial left-to-right ocular motor movements necessary for reading. We recruited 42 treated PD patients and 80 age-matched controls and compared their reading times with a variety of measures, including age, duration of disease, Unified Parkinson's Disease Rating Scale (UPDRS), the National Eye Institute 25-Item Visual Functioning Questionnaire 25 (VFQ-25), and Montreal Cognitive assessment (MoCA) test. The subjects performed 4 trials of reading 120 single digit numbers aloud as fast as possible without making errors. In each trial, they read 3 pages (KD1, KD2, and KD3), and each page contained 40 numbers per page in 8 lines with 5 numbers/line. We found that PD patients read about 20% slower than controls on all tests (KD1, 2, and 3 tests) (p < 0.02), and both groups read irregularly spaced numbers slower than regularly spaced numbers. Having lines between numbers to guide reading (KD1 tests) did not impact reading time in both PD and controls, but increased visual crowding as a result of decreased spacing between numbers (KD3 tests) was associated with significantly slower reading times in both PD and control groups. Our study revealed that saccadic reading is slower in PD, but controls and PD patients are both impacted by visuospatial planning challenges posed by increased visual crowding and irregularity of number spacing. Reading time did not correlate with UPDRS or MoCA scores in PD patients but significantly correlated with age, duration of disease, and VFQ-25 scores. The presence of convergence insufficiency did not significantly correlate with reading time in PD patients, although on average there was slower reading time in those with convergence insufficiency by 8 s (p = 0.2613). We propose that a simple reading task using 120 single-digit numbers can be used as a screening tool in the clinical setting to assess functional ocular motor difficulties in Parkinson's disease that can have a profound impact on quality of life.

Fig 1

Irregularly (A) and regularly (B) spaced single digit number reading tests. The irregularly spaced tests (KD1 irreg, KD2 irreg, and KD3 irreg) were read left-to-right twice. Then the regularly spaced tests (KD1 reg, KD2 reg, and KD3 reg) were read left-to-right (L-R) and then right-to-left (R-L). KD1 has a horizontal line to help guide reading from one number to the next, although that adds some visual crowding. KD2 has the same spacing as KD1 but has no lines to guide reading. KD3 has less vertical space between the lines compared with KD2 and therefore has the most amount of visual crowding.

Fig 2. PD patients read significantly slower in all reading tasks.

(A) Box-and -whisker plot of sum of KD1, KD2 and KD3 (KD123) reading times (3 pages or 120 numbers per task) for KD123 irreg, KD123 reg L-R, and KD123 reg R-L. (B) Box-and-whisker plot of individual KD reading times (1 page, 40 numbers per task). (C) Significant correlation between age and KD123 irreg reading time in the control group (R² = 0.06853 p = 0.0206). (D) Significant correlation between age and KD123 irreg reading times in PD (R² = 0.02389, p = 0.0010). (E) Significant correlation between disease duration and KD123 irreg reading time in PD group (R² = 0.1905 p = 0.0088). (F) No correlation between UPDRS score and KD123 irreg reading time (R² = 0.01811 p = 0.5034).