Comparison of vagus nerve cross-sectional area between brain-first and body-first Parkinson’s disease Authors: Shuangshuang Dong, Bo Shen, Xu Jiang, Jun Zhu, Haiying Zhang, Yang Zhao, Yaning Chen, Dongfeng Li, Yuanyuan Feng, Yi Chen, Yang Pan, Feng Han, Ben Liu, Li Zhang Published: 2024 ABSTRACT AbstractThe vagus nerve (VN) is the main neural pathway linking the gut and brain in Parkinson’s disease (PD). In this study, we utilized high-resolution ultrasound to measure the VN cross-sectional area (CSA) in 96 healthy controls (HCs) and 75 PD patients. The PD group was further categorized into three subgroups: PD-preRBD, PD-postRBD, and PD-nonRBD. PD-preRBD was the body-first subtype, and PD-postRBD and PD-nonRBD were the brain-first subtype. The PD group had a significantly lower VN CSA than HCs. Subgroup analysis revealed that the PD-preRBD group tended to exhibit a smaller VN CSA than both the PD-postRBD and PD-nonRBD groups. The VN CSA, specifically the right VN, was significantly correlated with the body-first subtype and some components of PD-related assessment scales. Overall, these findings provide evidence of VN atrophy in PD, especially in body-first PD, suggesting that VN ultrasound could serve as an adjunctive diagnostic tool.Subject terms: Parkinson's disease, Diagnostic markers, Clinical trials Introduction Misfolding and aggregation of the protein α-synuclein (α-syn) are hallmarks of Parkinson’s disease (PD) and various other neurodegenerative conditions collectively referred to as synucleinopathies1. Progressive degeneration of the substantia nigra (SN), coupled with accumulation of α-syn within Lewy bodies, is responsible for the motor symptoms observed in PD1. However, this mechanism cannot fully account for the non-motor symptoms also prevalent in these conditions. Notably, some PD patients manifest non-motor symptoms, such as autonomic dysfunction, 20 years or more before motor symptom onset2,3. Studies have revealed that α-syn is not restricted to the central nervous system and SN. For example, autopsy-based studies by Braak et al. detected α-syn in these areas as well as in the gastrointestinal tract, dorsal motor nucleus of the vagus nerve (DMN), and nerves supplying the enteric nervous system. Braak et al. hypothesized that α-syn originates in the DMN or olfactory bulb, possibly due to insult to both the enteric and olfactory epithelia, and subsequently propagates to the rostral brainstem and supratentorial systems4,5. However, some studies challenge the notion that PD invariably originates in the enteric nervous system and follows the Braak staging pattern of α-syn pathology. Autopsy studies have demonstrated that some PD cases lack the characteristic pathological α-syn deposition in DMN6. Building on these findings, it has been proposed that PD can be categorized into two distinct subtypes differentiated by the initial site of α-syn aggregation: brain-first PD versus body-first PD. The brain-first subtype is characterized by initial pathological α-syn aggregates in the brain that spread through connected neuronal structures, ultimately involving the autonomic nervous system. In contrast, the body-first subtype is defined by α-syn pathology originating in the enteric nervous system that subsequently propagates through the vagus nerve (VN) to the DMN, ultimately extending to the lower brainstem and rest of the brain7. Rapid eye movement (REM) sleep behavior disorder (RBD) is thought to result from damage to pontine structures, including the magnocellular reticular formation and locus subcoeruleus8,9. Recent studies have hypothesized that the appearance of isolated RBD well before parkinsonism could serve as a robust indicator of the body-first subtype and propose pre-motor RBD as a criterion for distinguishing between the body-first and brain-first subtypes10,11. The VN is the main neural pathway linking the gut and brain in two distinct subtypes of PD12. The VN is a mixed nerve comprising ~20% motor efferent fibers and 80% sensory afferent fibers13–15. It serves as the primary pathway connecting the gastrointestinal tract and brain, facilitating bidirectional communication between these regions12. Studies utilizing animal models support the hypothesis that the transmission of pathological α-syn through the VN plays a causative role in the development of PD16,17. A-syn aggregates can induce degeneration and loss of VN neurons, leading to vagal atrophy18. High-resolution ultrasound (HRUS), a non-invasive and radiation-free method, can be used to assess vagal degeneration by measuring the cross-sectional area (CSA) of the VN. Ultrasound studies of the VN in PD patients have yielded mixed results. Six studies18–23 have reported significant atrophy, indicated by a reduction in CSA, in both the left and right VN of PD patients, whereas three studies24–26 found no discernible differences in VN CSA between PD patients and healthy controls. Notably, Horsager et al.27 reported a difference in CSA in the right VN, but not in the left. However, no study to date has aimed to identify CSA disparities in the VN between the brain-first and body-first PD subtypes. To address this research gap, we measured the VN CSA of PD patients and controls and further compared the VN CSA between the two PD subtypes. Results Study population A total of 171 participants were recruited for this study, including 75 PD patients and 96 controls. The PD group was further categorized into three subgroups: 27 with pre-motor RBD (PD-preRBD), 10 with post-motor RBD (PD-postRBD), and 38 without RBD (PD-nonRBD). A detailed flow diagram of the recruitment process is provided in Fig. 1. Sonographic data were successfully collected from all participants. Participants’ demographic profiles and ultrasonographic data are presented in Tables 1 and 2. Detailed flowchart of participant recruitment and the specific criteria used to classify Parkinson’s disease (PD) patients in this study. Table 1Comparison of clinical characteristics and ultrasound results between patients with Parkinson’s disease and controlsParkinson’s disease (n = 75)Control (n = 96)p-valuesGeneral informationAge (years)66.93 ± 7.2765.48 ± 10.390.304Sex (male, N%)38 (50.67)42 (43.75)0.368BMI (kg/m2)24.23 ± 3.7824.20 ± 3.020.959Hypertension (N%)31 (41.33)43 (44.79)0.651Assessment scalesNMSQ7.31 ± 4.603.18 ± 1.95<0.001RBDSQ4.08 ± 3.411.21 ± 1.07<0.001MMSE27.00 (24.00, 28.00)27.50 (26.00, 30,00)0.011SCOPA-AUT (total)16.19 ± 4.674.28 ± 1.79<0.001High-resolution ultrasoundLeft VN CSA (mm2)1.47 ± 0.341.95 ± 0.51<0.001Right VN CSA (mm2)1.90 ± 0.422.57 ± 0.51<0.001P-values<0.001a<0.001aThe MMSE score is presented as the median and interquartile range. All other values, except for gender and hypertension, are reported as the mean ± standard deviation (SD).BMI Body Mass Index, NMSQ Motor Symptoms Questionnaire, RBDSQ Rapid Eye Movement Sleep Behavior Disorder Screening Questionnaire, MMSE, Mini-Mental State Examination, SCOPA-AUT Scales for Outcomes in Parkinson's disease-autonomic, VN CSA the cross-sectional area of the vagus nerve.aThe p-value was obtained from a t-test comparing VN CSA on both sides within each group. Comparison of clinical characteristics and ultrasound results between patients with Parkinson’s disease and controls The MMSE score is presented as the median and interquartile range. All other values, except for gender and hypertension, are reported as the mean ± standard deviation (SD). BMI Body Mass Index, NMSQ Motor Symptoms Questionnaire, RBDSQ Rapid Eye Movement Sleep Behavior Disorder Screening Questionnaire, MMSE, Mini-Mental State Examination, SCOPA-AUT Scales for Outcomes in Parkinson's disease-autonomic, VN CSA the cross-sectional area of the vagus nerve. aThe p-value was obtained from a t-test comparing VN CSA on both sides within each group. Table 2Comparison of clinical characteristics and ultrasound results between PD subgroupsBody-first (n = 27)Brain-first (n = 48)p-valuesPD-preRBD (n = 27)PD-postRBD (n = 10)PD-nonRBD (n = 38)General informationAge (years)68.67 ± 5.9764.20 ± 7.7166.42 ± 7.860.210Sex (male, N%)17 (62.96)4 (40.0)17(44.74)0.269BMI (kg/m2)23.50 ± 3.4023.09 ± 2.2924.96 ± 4.190.211Hypertension (N%)13 (48.15)3(30.00)15 (39.47)0.577Disease duration (years)2.00 (1.00, 3.00)1.50 (0.88, 3.25)2.00 (0.50, 3.00)0.859Assessment scalesHoehn & Yahr stage2.56 ± 0.472.15 ± 0.412.45 ± 0.600.130UPDRS (total)40.89 ± 14.9636.80 ± 16.5132.21 ± 13.110.059UPDRS-I2.00 (1.00, 3.00)1.00 (0.00, 2.25)1.00 (0.00, 2.00)a0.006UPDRS-II9.70 ± 5.878.50 ± 4.887.89 ± 4.270.357UPDRS-III27.96 ± 8.7426.20 ± 11.2522.61 ± 9.690.084UPDRS-IV0.00 (0.00, 1.00)0.00 (0.00, 0.25)0.00 (0.00, 0.00)0.251AImotor3.30 ± 2.614.00 ± 2.985.58 ± 3.84a0.026HAMD9.26 ± 7.425.60 ± 4.145.53 ± 4.85a0.036HAMA7.22 ± 4.984.30 ± 3.094.53 ± 3.80a0.029NMSQ9.59 ± 5.365.20 ± 4.10a6.24 ± 3.45a0.003PDSS110.93 ± 19.99106.60 ± 13.39123.50 ± 19.72ab0.009RBDSQ7.00 ± 2.027.10 ± 2.511.21 ± 1.30ab<0.001MMSE26.00 (24.00, 28.00)27.00 (24.75, 28.50)27.00 (23.00, 29.00)0.752PDQ3934.07 ± 23.2022.40 ± 10.7018.54 ± 11.88a0.002SCOPA-AUT (total)18.44 ± 4.0714.00 ± 3.77a15.16 ± 4.72a0.004SCOPA-AUT-cardiovascular3.00 ± 1.882.10 ± 1.451.95 ± 1.47a0.038SCOPA-AUT-gastrointestinal6.52 ± 2.594.20 ± 2.15a4.92 ± 2.06a0.006SCOPA-AUT-urinary6.30 ± 2.234.90 ± 1.915.34 ± 2.170.120SCOPA-AUT-thermoregulatory1.93 ± 1.391.50 ± 1.181.87 ± 1.800.762SCOPA-AUT-pupillary0.00 (0.00, 0.50)0.00 (0.00, 1.00)0.00 (0.00, 1.00)0.220SCOPA-AUT-sexuality00.00 (0.00, 1.00)1.00 (0.00, 1.25)1.00 (0.00, 1.00)0.100High-resolution ultrasoundLeft VN CSA (mm2)1.30 ± 0.251.44 ± 0.191.60 ± 0.370.001Right VN CSA (mm2)1.64 ± 0.331.93 ± 0.252.07 ± 0.43<0.001Note: Disease duration, UPDRS-I, UPDRS-IV, and MMSE scores are presented as the median and interquartile range. All other values, except for gender and hypertension, are reported as the mean ± standard deviation (SD).PD Parkinson’s disease, BMI Body Mass Index, UPDRS Unified Parkinson Disease Rating Scale, HAMD Hamilton Depression Scale, HAMA Hamilton Anxiety Scale, NMSQ Motor Symptoms Questionnaire, PDSS Parkinson's Disease Sleep Scale, RBDSQ Rapid Eye Movement Sleep Behavior Disorder Screening Questionnaire, MMSE Mini-Mental State Examination, PDQ-39 Parkinson's Disease Questionnaire-39, SCOPA-AUT Scales for Outcomes in Parkinson's disease-autonomic, VN CSA the cross-sectional area of the vagus nerve.ap < 0.05 for a pairwise difference with PD-preRBD group.bp < 0.05 for a pairwise difference with PD-postRBD group. Comparison of clinical characteristics and ultrasound results between PD subgroups Note: Disease duration, UPDRS-I, UPDRS-IV, and MMSE scores are presented as the median and interquartile range. All other values, except for gender and hypertension, are reported as the mean ± standard deviation (SD). PD Parkinson’s disease, BMI Body Mass Index, UPDRS Unified Parkinson Disease Rating Scale, HAMD Hamilton Depression Scale, HAMA Hamilton Anxiety Scale, NMSQ Motor Symptoms Questionnaire, PDSS Parkinson's Disease Sleep Scale, RBDSQ Rapid Eye Movement Sleep Behavior Disorder Screening Questionnaire, MMSE Mini-Mental State Examination, PDQ-39 Parkinson's Disease Questionnaire-39, SCOPA-AUT Scales for Outcomes in Parkinson's disease-autonomic, VN CSA the cross-sectional area of the vagus nerve. ap < 0.05 for a pairwise difference with PD-preRBD group. bp < 0.05 for a pairwise difference with PD-postRBD group. The PD and control groups were well-matched in terms of demographic characteristics. Gender, age, BMI, and hypertension did not differ significantly between groups. Regarding the duration of PD, we found no significant differences between the PD-preRBD, PD-postRBD, and PD-nonRBD patients. The respective median durations were 2.00 years (IQR: 1.00, 3.00 years), 1.5 years (IQR: 0.88, 3.25 years), and 2.00 years (IQR: 0.50, 3.00 years). Comparison of PD patients and controls The PD group differed significantly from the control group in several assessment scores. As shown in Table 1, the PD group had higher Non-Motor Symptoms Questionnaire (NMSQ) score (p < 0.001), REM Sleep Behavior Disorder Screening Questionnaire (RBDSQ) score (p < 0.001), and Scales for Outcomes in Parkinson’s Disease-autonomic (SCOPA-AUT) score (p < 0.001), indicating greater impairment or symptom severity, but lower Mini-Mental State Examination (MMSE) score (p = 0.011), reflecting greater cognitive dysfunction. The right VN CSA was significantly larger than the left VN CSA in both the PD group (p < 0.001) and control group (p < 0.001). As shown in Fig. 2, the VN CSA on both sides was significantly reduced in the PD group compared to the control group (left VN: p < 0.001, right VN: p < 0.001). Characteristic sonographic images of the vagus nerve. a, b Left and right vagus nerves (indicated by arrows) in a healthy control. c, d Left and right vagus nerves (indicated by arrows) in a patient with Parkinson’s disease. The cross-sectional area of the vagus nerve on both sides was significantly reduced in the patient with Parkinson’s disease. ICA Internal carotid artery, IJV Internal jugular vein. Comparison between body-first and brain-first PD patients As shown in Table 2, the PD-preRBD group had a significantly higher Unified Parkinson’s Disease Rating Scale (UPDRS) Part I (UPDRS-I) score (p = 0.007), Hamilton Depression Scale (HAMD) score (p = 0.040), Hamilton Anxiety Scale (HAMA) score (p = 0.038), Parkinson’s Disease Questionnaire-39 (PDQ-39) score (p = 0.002), and cardiovascular component of the SCOPA-AUT score (p = 0.037) than the PD-nonRBD group. The PD-preRBD group also scored significantly higher on the NMSQ (PD-preRBD vs.PD-postRBD p = 0.022, PD-preRBD vs. PD-nonRBD p = 0.008), SCOPA-AUT (PD-preRBD vs.PD-postRBD p = 0.023, PD-preRBD vs. PD-nonRBD p = 0.012), and gastrointestinal component of the SCOPA-AUT (PD-preRBD vs.PD-postRBD p = 0.023, PD-preRBD vs. PD-nonRBD p = 0.020) than both the PD-postRBD and PD-nonRBD groups. In contrast, there were no significant differences between the PD-postRBD and PD-nonRBD groups on the aforementioned scales (all p > 0.05). The PD-preRBD and PD-postRBD groups both differed significantly from the PD-nonRBD group in terms of the Parkinson’s Disease Sleep Scale (PDSS) score (PD-preRBD vs. PD-nonRBD p = 0.033, PD-postRBD vs. PD-nonRBD p = 0.046) and RBDSQ score (PD-preRBD vs. PD-nonRBD p < 0.001, PD-postRBD vs. PD-nonRBD p < 0.001). The PD-preRBD group exhibited a lower motor asymmetry index (AImotor) than PD-nonRBD patients (p = 0.025). However, AImotor did not differ significantly between the PD-postRBD and PD-nonRBD groups. No significant differences were observed in UPDRS total score, UPDRS Part II (UPDRS-II), UPDRS Part III (UPDRS-III), UPDRS Part IV (UPDRS-IV), or the Hoehn & Yahr (H-Y) stage scores (all p > 0.05). Furthermore, there were no significant group differences in the results of the MMSE or urinary, thermoregulatory, pupillary, and sexuality components of the SCOPA-AUT (all p > 0.05). The right VN CSA was significantly reduced in the PD-preRBD group compared to both the PD-postRBD and PD-nonRBD groups (PD-preRBD vs.PD-postRBD p = 0.025, PD-preRBD vs. PD-nonRBD p < 0.001), indicating that greater vagal atrophy is associated with the body-first PD subtype. Similarly, the left VN CSA was significantly reduced in the PD-preRBD group compared to the PD-nonRBD group (p < 0.001). In contrast, no significant differences in VN CSA were observed between the PD-postRBD and PD-nonRBD groups (p > 0.05). Correlations between VN CSA, clinical assessment scores, and subgroups For the correlation analyses, the PD-postRBD and PD-nonRBD groups were consolidated under the category of brain-first PD subtype, while PD-preRBD indicates the body-first PD subtype. The right VN CSA demonstrated a moderate negative correlation with the body-first PD subtype (r = -0.461, p < 0.001) and RBDSQ (r = −0.360, p = 0.002), and a weak negative correlation with the H-Y stage (r = −0.264, p = 0.002), UPDRS total score (r = -0.268, p = 0.020), UPDRS-III (r = −0.299, p = 0.009), NMSQ (r = −0.236, p = 0.041), PDQ39 (r = −0.287, p = 0.013), SCOPA-AUT total score (r = −0.294, p = 0.010), and the cardiovascular (r = −0.257, p = 0.026) and gastrointestinal (r = −0.263, p = 0.023) components of the SCOPA-AUT. The right VN CSA also showed a weak positive correlation with PDSS score (r = 0.234, p = 0.043). Similarly, the left VN CSA showed a moderate negative correlation with the body-first PD subtype (r = −0.375, p < 0.001) and RBDSQ (r = −0.398, p < 0.001), and a weak negative correlation with H-Y stage (r = −0.238, p = 0.040), PDQ39 (r = −0.264, p = 0.023), SCOPA-AUT total score (r = −0.293, p = 0.011), and the gastrointestinal (r = −0.269, p = 0.020) component of the SCOPA-AUT. No significant correlations were observed between the VN CSA and UPDRS-I, UPDRS-II, UPDRS-IV, HAMD, HAMA, MMSE, disease duration, age, or gender (all p > 0.05; Table 3). Table 3Correlation of VN CSA with clinical assessment data and subgroupsLeft VN CSARight VN CSAr-valuesp-valuesr-valuesp-valuesLeft VN CSA0.513<0.001Right VN CSA0.513<0.001Body-first PD−0.375<0.001−0.461<0.001Hoehn & Yahr stage−0.2380.040−0.2640.022UPDRS (total)−0.2260.051−0.2680.020UPDRS-III−0.2120.068−0.2990.009NMSQ−0.1720.140−0.2360.041PDSS0.1430.2220.2340.043RBDSQ−0.398<0.001−0.3600.002PDQ39−0.2640.023−0.2870.013SCOPA-AUT (total)−0.2930.011−0.2940.010SCOPA-AUT-cardiovascular−0.1830.116−0.2570.026SCOPA-AUT-gastrointestinal−0.2690.020−0.2630.023VN CSA the cross-sectional area of the vagus nerve, PD Parkinson’s disease, UPDRS Unified Parkinson Disease Rating Scale, NMSQ Motor Symptoms Questionnaire, PDSS Parkinson's Disease Sleep Scale, RBDSQ Rapid Eye Movement Sleep Behavior Disorder Screening Questionnaire, PDQ-39 Parkinson's Disease Questionnaire-39, SCOPA-AUT Scales for Outcomes in Parkinson's disease-autonomic. Correlation of VN CSA with clinical assessment data and subgroups VN CSA the cross-sectional area of the vagus nerve, PD Parkinson’s disease, UPDRS Unified Parkinson Disease Rating Scale, NMSQ Motor Symptoms Questionnaire, PDSS Parkinson's Disease Sleep Scale, RBDSQ Rapid Eye Movement Sleep Behavior Disorder Screening Questionnaire, PDQ-39 Parkinson's Disease Questionnaire-39, SCOPA-AUT Scales for Outcomes in Parkinson's disease-autonomic. Box plots Box plots were used to illustrate the distribution of VN CSA among control participants, PD patients, and their subgroups (Fig. 3). The PD group exhibited a narrower box, suggesting a more concentrated data distribution and less variability in VN CSA than the control group, which showed a wider box suggesting greater data dispersion. Furthermore, the PD group had a lower median VN CSA than the control group. As shown, there was no significant data overlap between the two groups, indicating a distinct group difference in VN CSA. Among the three subgroups, the median for the PD-preRBD group was positioned in the lower part of the box, suggesting a lower median VN CSA than the other two groups. Notably, this differentiation was more pronounced for the right VN than the left VN. The PD-postRBD group had a median similar to that of the PD-nonRBD group but with shorter whiskers, indicating fewer extreme values; thus, there is a more consistent distribution of VN CSA within these two subgroups. Despite a small amount of data overlap among the three subgroups, significant differences between the PD-preRBD group and the other two groups were still discernible, particularly for the right VN CSA. Box plots. Box plots showing the distribution of the cross-sectional area (CSA) of the vagus nerve (VN) among healthy controls (a, b), Parkinson’s disease (PD) patients (a, b), and their subgroups (c, d). Each box in the plot represents the interquartile range (IQR) of the dataset, including the first quartile (Q1), median (Q2), and third quartile (Q3). The bottom and top of the box denote Q1 and Q3, respectively, while the horizontal line inside the box indicates the median (Q2). The whiskers extend from Q1 to the minimum value and from Q3 to the maximum value, excluding outliers. *p < 0.05, **p < 0.001. Overall, the box plots reveal differences and variability in the distribution of VN CSA among different groups. These findings provide evidence of VN atrophy in PD, especially in the PD-preRBD group. ROC curve Figure 4 presents the receiver operating characteristic (ROC) curve for VN CSA in diagnosing PD and the body-first PD subtype. PD-preRBD indicates the body-first PD subtype while the PD-postRBD and PD-nonRBD groups were consolidated as the brain-first PD subtype. For the diagnosis of PD, the area under the curve (AUC) for the right VN CSA was 0.840 (0.781–0.898, p < 0.001), and for the left VN CSA it was 0.778 (0.708–0.849, p < 0.001). The cut-off values derived from the ROC curve at a sensitivity of 0.87 for the right VN CSA were 2.35 mm², while at a sensitivity of 0.83 for the left VN CSA they were 1.75 mm². For the diagnosis of the body-first PD versus brain-first PD subtype, the AUC for the right VN CSA was 0.776 (0.669–0.883, p < 0.001), and for the left VN CSA it was 0.723 (0.604–0.842, p = 0.001). The cut-off values derived from the ROC curve with a sensitivity of 0.74 for the right VN CSA were 1.82 mm², while at a sensitivity of 0.70 for the left VN CSA they were 1.45 mm². Receiver operating characteristic (ROC) curve. ROC curve analysis and calculation of the area under the ROC curve (AUC) were used to describe the predictive value of the cross-sectional area (CSA) of the vagus nerve (VN) for diagnosing Parkinson’s disease (a) and the body-first Parkinson’s disease subgroup (b). For diagnosing PD, the AUC for the right VN CSA was 0.840 and for the left VN CSA it was 0.778 (a). For diagnosing the body-first PD subtype versus the brain-first PD subtype, the AUC for the right VN CSA was 0.776 and for the left VN CSA it was 0.723 (b). Discussion In this case-control study, we found that the right VN CSA was significantly larger than the left VN CSA in both the PD group and control group. Furthermore, the PD group had a significantly lower bilateral average VN CSA compared to the control group. Subgroup analysis revealed that the PD-preRBD group tended to exhibit a smaller VN CSA than the other two subgroups. The VN CSA, specifically the right VN, exhibited a statistical correlation with body-first PD subtype and some components of PD-related assessment scales. The PD-preRBD group had more serious non-motor symptoms than the other two subgroups. A systematic review and meta-analysis establishing reference values for the VN CSA in a healthy population reported a right VN average area of 2.53 mm2; and left VN average area of 2.03 mm2 28. Similar measurements were observed for the healthy control group in our study, with the right VN averaging 2.57 mm² and the left VN averaging 1.95 mm². Tsukita et al.18 proposed that α-syn aggregates can induce degeneration and loss of VN neurons, leading to vagal atrophy in PD. This reduction in VN CSA can be measured using HRUS. Studies by Walter et al.20, Sartucci et al.19, Pelz et al.21, Huckemann et al.23, and Höppner et al.22 have all reported a decrease in VN CSA in PD patients. Horsager et al.27 found a difference in right VN CSA, but not in left VN CSA. In contrast, Laucius et al.24, Fedke et al.25, and Sijben et al.26 reported no difference in VN CSA between PD patients and controls. A systematic review and meta-analysis by Abdelnaby concluded that PD patients exhibit a degree of vagal atrophy29. As reported in most studies, the right VN is significantly larger than the left VN. In a correlation study, Höppner et al.22 identified a negative relationship between the right VN CSA and UPDRS score. Walter et al.20 found that left VN CSA correlated negatively with UPDRS III score, while bilateral VN CSA correlated negatively with NMSQ score. In contrast, Pelz et al.21 found no correlation between right or left VN CSA and H-Y stage, disease duration, UPDRS-III, or NMSQ. Similarly, Sartucci et al.21 found no significant correlations between VN CSA and H-Y stage. Huckemann et al.23 reported that VN CSA correlated negatively with heart rate and positively with tilt test parameters representing parasympathetic modulation through vagal activity at rest. Ultrasound studies of the VN in PD have yielded mixed results, and no previous studies focused on CSA disparities in the VN between the brain-first and body-first PD subtypes. Previous studies suggest that diabetes30 and demyelinating diseases31 can lead to degeneration or atrophy of the VN. Furthermore, neuropathy linked to immune or vasculitis issues may present with VN hypertrophy32. In light of these findings, we excluded patients with these conditions from our analysis. The findings of our study are consistent with the majority of previous work finding that the right VN CSA is larger than the left VN CSA in both PD patients and health controls. Our findings are also in line with those of six prior studies18–23 reporting a reduction in VN CSA in PD. Notably, we found greater reduction of VN CSA in the PD-preRBD group than the other two groups. This finding suggests that vagal atrophy is more pronounced in body-first PD, supporting the hypothesis of a distinct pathological process associated with this subtype. For diagnosing PD, the AUC for the right VN CSA was 0.840 and for the left VN CSA it was 0.778, indicating that HRUS is potentially a valuable diagnostic tool for clinical assessment of PD. For diagnosing the body-first PD subtype versus the brain-first PD subtype, the AUC for the right VN CSA was 0.776 and for the left VN CSA it was 0.723. Thus, the VN CSA is a meaningful metric for the body-first PD subtype. By identifying differences in VN CSA between the brain-first and body-first PD subtypes, this study makes an innovative contribution to the field. HRUS, which is non-invasive and radiation-free, could emerge as a paraclinical method for diagnosing the body-first PD subtype, especially when combined with pre-motor RBD. As part of our study, we also performed a comprehensive set of PD assessment scales. By examining the relationship between the VN CSA and these scales, we observed some inconsistencies with the findings from the aforementioned studies. We observed that the VN CSA, specifically the right VN, exhibited a statistical correlation with the severity of both motor and non-motor symptoms. This correlation was less pronounced for the left VN than the right VN. We also identified a moderate negative correlation between the VN CSA and body-first PD subgroup. However, given the moderate or weak nature of these correlations, further research is needed to confirm and further elucidate these preliminary findings. Our finding of a larger right VN CSA compared to the left could be attributed to the differential functional topography of left and right brainstem vagal nuclei in the brainstem and their distinct innervation patterns33. The right VN predominantly innervates the small intestine and colon, influencing the anterior gastric plexus, whereas the left VN serves the stomach, liver, and duodenum34. Due to asymmetric innervation, nerve fiber composition differs between the left and right VNs, with the right VN containing ~20% more nerve fibers and a lower proportion of B- and Aδ-nerve fibers. We hypothesize that this lateral variation may be the primary cause for the larger right VN CSA35. RBD is the second most prevalent sleep disorder in the early stages of PD36,37 and is particularly linked to body-first PD. A-syn pathology starts in the enteric nervous system in the body-first PD, which is thought to lead to earlier damage to the autonomic nervous system and lower brainstem structures compared to the brain-first PD. A recent review presented clinical and imaging evidence for these subtypes of PD, supporting the idea the body-first subtype is associated with a higher incidence of non-motor symptoms, particularly those impacting the autonomic nervous system10. Consistent with this, we observed a higher burden of non-motor symptoms in the PD-preRBD group, as reflected by the comprehensive set of PD assessment scales. The observed variance in non-motor symptoms is likely due to more pronounced VN atrophy in patients with body-first PD. This hypothesis is supported by findings from Walter et al. implicating the VN in the non-motor symptoms of PD. Atrophy of the VN could therefore be a significant factor contributing to the increased non-motor symptomatology observed in our PD-preRBD group. To further substantiate this speculation, we conducted additional analyses to quantify the degree of VN atrophy and its correlation with non-motor symptoms. These analyses revealed an association between the severity of VN atrophy and the presence of non-motor symptoms, which we believe strengthens our argument. Urinary function is governed by a sophisticated network involving the brain, spinal cord, and nerves that regulate the bladder and urethral function38. Dysfunction in any component of this system beyond the peripheral nerves may lead to urinary problems. Prostate enlargement in aging men, which is a common issue, could further exacerbate these issues10. Thus, patients with brain-first and body-first PD may face comparable urinary symptoms. Brain-first PD involves a-syn starting in the brain, possibly leading to earlier cognitive decline or dementia. However, not all studies support this idea. PD patients with RBD often face more severe cognitive impairment and show quicker progression to dementia39, as well as signs of reduced neuromelanin in the locus coeruleus and lower noradrenergic transporter density, which are linked to poorer cognition40. Probable RBD is also associated with decreased cortical acetylcholinesterase activity, indicating worse cognitive function41. These findings suggest that body-first PD might have a greater impact on cognition than brain-first PD. Cognitive function is highly complex and shaped by an array of interconnected systems, including neurotransmitter dynamics and other biological processes. The interplay of these factors may blur the lines distinguishing the brain-first and body-first PD subtypes. In our study, no significant disparities in cognitive performance were observed between the three PD subgroups using the MMSE. At the point of motor symptom onset and clinical diagnosis of PD, it is estimated that dopamine neurons of the SN are nearly 50% reduced42. Recent studies have indicated no significant difference in the severity of motor symptoms between the body-first and brain-first PD subtypes11,43. In line with these results, we found no substantial differences in H-Y stage or UPDRS-III scores between PD subgroups. As such, average motor symptoms are not reliable clinical indicators for differentiating between brain-first and body-first PD. Although autonomic symptoms frequently emerge in the early stages of PD, there is a lack of concrete diagnostic tools to predict their onset. HRUS offers a promising avenue for elucidating the structure of deep autonomic nerves, potentially facilitating the identification of PD prior to the manifestation of motor symptoms. As our study is a relatively large case-control investigation focusing on the VN CSA in PD patients and comparing results between the brain-first and body-first PD subtypes, it makes a meaningful and novel contribution to the field. The present study is subject to several limitations. First, RBD diagnosis relied on the RBDSQ, which is not as precise as polysomnography (PSG). The decision to use the RBDSQ was influenced by the prohibitive costs and technical demands of PSG, as well as potential issues related to patient cooperation. Second, it is possible that not all body-first PD patients exhibit clinical RBD; some may experience REM sleep without atonia, falling short of the criteria for RBD diagnosis. This could result in erroneous classification of body-first patients as brain-first patients. To mitigate these concerns, we endeavored to expand the sample size to help offset potential inaccuracies in RBD diagnosis, although this cannot entirely eliminate the possibility of misclassification. Advanced imaging techniques play a crucial role in distinguishing the brain-first and body-first subtypes of PD. This differentiation is supported by recent studies employing various modalities including positron emission tomography (PET)11, 123I-MIBG myocardial scintigraphy44, abdominal CT imaging11, and striatal dopaminergic imaging45. We concur that these imaging methods offer profound insights into the pathological trajectory of PD and are dedicated to ongoing refinement of our research methods. Should future funding and opportunities arise, we hope to consider integration of PSG assessments and add imaging techniques to our study protocol that would allow us to enhance the depth and breadth of our investigation. Third, our study exclusively examined the VN and did not investigate other cranial or peripheral nerves. This limitation prevents us from asserting that the observed reduction of CSA is specific to the VN. A broader investigation measuring a range of peripheral and cranial nerves as internal controls could provide a more conclusive assessment. In conclusion, our findings support the hypothesis of atrophy of the VN in PD and further suggest that this atrophy is more pronounced in the body-first PD as opposed to brain-first PD subtype. The use of VN HRUS could potentially augment the diagnostic toolkit for PD, offering a valuable method for identifying and distinguishing between PD subtypes. Methods Participants All participants were recruited from the outpatient clinics and geriatric neurology ward of Nanjing Brain Hospital, China, between July 2023 and September 2024. We included PD patients that had been diagnosed within 1 year or less. Inclusion was based on a clinical diagnosis of PD confirmed by the treating neurologist, in accordance with the 2015 diagnostic criteria established by the Movement Disorder Society46. The control group comprised individuals without PD. The exclusion criteria, which applied to both groups, were comorbidities such as diabetes mellitus, demyelinating diseases, peripheral neuropathy, schizophrenia, severe cognitive impairment, amyotrophic lateral sclerosis, or secondary Parkinson’s syndrome attributed to vascular or drug-induced etiologies. PSG, which is considered the gold standard for RBD diagnosis, was not feasible due to its limited availability. Instead, we utilized the RBDSQ, which has shown solid diagnostic accuracy47, with a score of 5 or higher indicating probable RBD48,49. We also conducted detailed medical history assessments and inquiries regarding dream-enactment behaviors to exclude secondary causes of RBD, such as alcohol consumption or medication use. In patients with probable RBD, we closely examined the timing of dream-enactment behaviors in relation to the onset of motor symptoms. PD patients were categorized into the following three groups based on the presence of RBD and its temporal relationship to motor symptoms: PD-preRBD, PD-postRBD, and PD-nonRBD. PD-preRBD indicates body-first PD, whereas PD-postRBD and PD-nonRBD indicate brain-first PD. In total, this study included 27 PD-preRBD patients, 10 PD-postRBD patients, 38 PD-nonRBD patients, and 96 healthy controls. A detailed flow diagram of the recruitment process is shown in Fig. 1. This study was approved by the local ethics committee of Nanjing Brain Hospital, China (approval number. 2021-KY007-01) and performed in line with the 2013 Declaration of Helsinki). All participants were adults and provided informed consent. Clinical assessment PD patients underwent a comprehensive evaluation including detailed medical histories, clinical and neurological examinations, and a suite of PD-related measures. Control participants completed some of these assessments, but not all PD-related measures. The assessment tools included the UPDRS, RBDSQ, NMSQ, PDSS, HAMD, HAMA, MMSE, PDQ-39, and SCOPA-AUT. The MMSE was utilized to evaluate cognitive status in all participants. The UPDRS was used to score disease severity, with the UPDRS III specifically assessing motor function. The AImotor distribution was characterized by the absolute difference between the left and right motor scores on the UPDRS-III scale for each extremity, with assessments including bradykinesia, rigidity, and tremor43. Non-motor symptoms were evaluated using the NMSQ. The HAMD and HAMA scales were used to measure depression and anxiety levels, respectively. Sleep status was evaluated using the PDSS. The PDQ-39 assessed participants’ quality of life, while the SCOPA-AUT evaluated functionality of the autonomic nervous system across six domains: gastrointestinal, urinary, cardiovascular, thermoregulatory, pupillary, and sexuality. The 13-item RBDSQ uses a threshold score of 5 or higher to indicate the presence of probable RBD48,49. Body-first patients were identified by either an initial diagnosis of idiopathic RBD that progressed to PD or a verifiable history of RBD symptoms occurring before motor symptoms. In contrast, brain-first patients were defined by the absence of RBD at PD diagnosis or by RBD symptoms emerging distinctly after the onset of motor symptoms10,11. Data collection and evaluation were performed for all participants, especially during each PD patient’s optimal ‘on’ medication phase, by neurologists trained to use these specialized assessment protocols. HRUS examination All imaging was performed using a HRUS system (GE LOGIQ E20, United States) coupled with a model L4-20 linear array transducer. This transducer operates within a frequency range of 4–10 MHz, providing the necessary clarity for detailed assessment. Imaging was performed by a single sonographer who was blinded to participants’ identities and status as a PD patient or control. Participants were scanned in a supine position with their heads gently extended and turned to the side opposite the nerve being scanned. Employing the carotid artery and internal jugular vein as key anatomical references, the sonographer performed an axial scan of the bilateral VN within the neck region. Customized settings, including gain, depth, and focus, were optimized for each participant to ensure high-quality imaging. A depth setting of 4 cm was selected to capture essential details within the scan area. The VN was discerned as a small, rounded structure with hypoechoic or honeycomb-like characteristics, situated deep to the carotid artery and internal jugular vein. The VN is typically positioned laterally adjacent to the common carotid artery and dorsally relative to the internal jugular vein, nestled within the carotid sheath. During the examination, minimal pressure was exerted to avoid compressing the nerve. The sonographer acquired bilateral VN images at the level of thyroid cartilage. The image intensity scale was fine-tuned to highlight the VN’s periphery at its maximum intensity to achieve the best possible contrast between the hypoechoic neural tissue and hyperechoic epineurium. The CSA of the VN was measured in square millimeters with an accuracy of 0.01 mm². Final measurements were derived from the mean of three individual assessments (Fig. 2). Statistical analysis Statistical analyses were performed using the SPSS statistical package (version 25, IBM Corp., Armonk, N.Y., USA) and figures were produced using GraphPad Prism (version 8, GraphPad Software, La Jolla, CA, USA). Inter-group comparisons involved the contrast between the PD group and control group as well as contrasts among the three PD subgroups. The normality of data distribution was evaluated using Q-Q plots and histograms. Normally distributed data are reported using the mean and standard deviation (x̄ ± SD). Comparisons between groups were made using the two-sample t-test or one-way analysis of variance (ANOVA), while differences between the right and left VN CSA within each participant were assessed using the paired samples t-test. Non-normally distributed data are reported as the median and interquartile range [M(Q1, Q3)] and compared between groups using the Mann-Whitney U test or Kruskal-Wallis H test. Categorical data are reported as number of cases and percentage, with comparisons between groups performed using the chi-square test. Results with p < 0.05 were considered statistically significant. Relationships between the VN CSA and participant characteristics were explored using Pearson or Spearman correlation analysis. For dichotomous variables or data that were not normally distributed, Spearman’s rank correlation analysis was used. This approach was applied to parameters such as the body-first subgroup classification, gender, hypertension status, disease duration, and MMSE scale. For normally distributed data, Pearson’s correlation analysis was employed, including variables like age, BMI, and other PD-related scales. ROC curve analysis and calculation of AUC were used to describe the predictive value of VN CSA for diagnosing PD and the body-first PD subtype and to confirm the optimal cut-off values. These cut-off values were established by maximizing Youden’s index on the ROC. PD-preRBD indicates body-first PD, while PD-postRBD and PD-nonRBD indicate brain-first PD. For the correlation and ROC curve analyses, PD-postRBD and PD-nonRBD patients were consolidated under the category of brain-first PD subtype to enable more comprehensive data analysis. Acknowledgements L.Z. received research funding from National Natural Science Foundation of China (NSFC) (No: 82171249). Y.P. received research funding from National Natural Science Foundation of China (NSFC) (No: 82101332). Author contributions S.D.: conception, execution, statistical analysis, and writing of the first draft of the manuscript; B.S.: statistical analysis and critique of the manuscript; X.J., J.Z., H.Z., Y.Z., Y.C., D.L., Y.F., Y.C.: data collection and critique of the manuscript; Y.P.: data collection and funding acquisition; F.H.: review and critique of the manuscript; B.L.: acquisition of ultrasound data; L.Z.: project administration, funding acquisition and review. All authors read and approved the final manuscript. Data availability Data are available upon request from the corresponding author. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. These authors contributed equally: Shuangshuang Dong, Bo Shen. Contributor Information Ben Liu, Email: 914901984@qq.com. Li Zhang, Email: neuro_zhangli@163.com. References Associated Data Data Availability Statement Data are available upon request from the corresponding author.