The severity of corneal nerve loss differentiates motor subtypes in patients with Parkinson's disease

Abstract

Background: Parkinson's disease (PD) is a heterogeneous movement disorder with patients manifesting with either tremor-dominant (TD) or postural instability and gait disturbance (PIGD) motor subtypes. Small nerve fiber damage occurs in patients with PD and may predict motor progression, but it is not known whether it differs between patients with different motor subtypes.

Objective: The aim of this study was to explore whether there was an association between the extent of corneal nerve loss and different motor subtypes.

Methods: Patients with PD classified as TD, PIGD, or mixed subtype underwent detailed clinical and neurological evaluation and corneal confocal microscopy (CCM). Corneal nerve fiber density (CNFD), corneal nerve branch density (CNBD), and corneal nerve fiber length (CNFL) were compared between groups, and the association between corneal nerve fiber loss and motor subtypes was investigated.

Results: Of the 73 patients studied, 29 (40%) had TD, 34 (46%) had PIGD, and 10 (14%) had a mixed subtype. CNFD (no./mm², 24.09 ± 4.58 versus 28.66 ± 4.27; p < 0.001), CNBD (no./mm², 28.22 ± 11.11 versus 37.37 ± 12.76; p = 0.015), and CNFL (mm/mm², 13.11 ± 2.79 versus 16.17 ± 2.37; p < 0.001) were significantly lower in the PIGD group compared with the TD group. Multivariate logistic regression showed that higher CNFD (OR = 1.265, p = 0.019) and CNFL (OR = 1.7060, p = 0.003) were significantly associated with the TD motor subtype. The receiver operating characteristic (ROC) analysis demonstrated that combined corneal nerve metrics showed excellent discrimination between TD and PIGD, with an area under the curve (AUC) of 0.832.

Conclusion: Greater corneal nerve loss occurs in patients with PIGD compared with TD, and patients with a higher CNFD or CNFL were more likely to have the TD subtype. CCM may have clinical utility in differentiating different motor subtypes in PD.

Keywords: Parkinson’s disease; corneal nerve fiber; motor subtypes; postural instability and gait disturbance; small nerve fiber; tremor-dominant.

Figure 1.

CCM measurements in PD patients with different motor subtypes. Scatter plots of (a) CNFD, (b) CNBD, and (c) CNFL in TD, PIGD, and mixed groups. Errors bars represent mean ± SD. CCM, corneal confocal microscopy; CNBD, corneal nerve branch density; CNFD, corneal nerve fiber density; CNFL, corneal nerve fiber length; PD, Parkinson’s disease; PIGD, postural instability and gait disturbance; TD, tremor-dominant. *p < 0.05; ***p < 0.001.

Figure 2.

Representative CCM images in (a, e) a healthy control and patient with (b, f) TD, (c, g) PIGD, and (d, h) mixed disease. (a–d) Corneal nerve fibers are linear homogeneous and highly reflective. (e–h) Nerve fiber trunks are highlighted in red, green dots indicate the origin of the branches, and blue and red lines combined indicate the CNFL. Scale bar = 100 µm. CCM, corneal confocal microscopy; PIGD, postural instability and gait disturbance; TD, tremor-dominant.

Figure 3.

Relationship between (a) CNFD and (b) CNFL and the probability of motor subtypes after correction for confounders. CNFD, corneal nerve fiber density; CNFL, corneal nerve fiber length; PIGD, postural instability and gait disturbance; TD, tremor-dominant.

Figure 4.

The ROC for CCM distinguishing TD from PIGD. CCM, corneal confocal microscopy; CNBD, corneal nerve branch density; CNFD, corneal nerve fiber density; CNFL, corneal nerve fiber length; PIGD, postural instability and gait disturbance; ROC, receiver operating characteristic; TD, tremor-dominant.