Effect of Parkinson’s Disease on Cardio-postural Coupling During Orthostatic Challenge

Cardiac baroreflex and leg muscles activation are two important mechanisms for blood pressure regulation, failure of which could result in syncope and falls. Parkinson’s disease is known to be associated with cardiac baroreflex impairment and skeletal muscle dysfunction contributing to falls. However, the mechanical effect of leg muscles contractions on blood pressure (muscle-pump) and the baroreflex-like responses of leg muscles to blood pressure changes is yet to be comprehensively investigated. In this study, we examined the involvement of the cardiac baroreflex and this hypothesized reflex muscle-pump function (cardio-postural coupling) to maintain blood pressure in Parkinson’s patients and healthy controls during an orthostatic challenge induced via a head-up tilt test. We also studied the mechanical effect of the heart and leg muscles contractions on blood pressure. We recorded electrocardiogram, blood pressure and electromyogram from 21 patients with Parkinson’s disease and 18 age-matched healthy controls during supine, head-up tilt at 70°, and standing positions with eyes open. The interaction and bidirectional causalities between the cardiovascular and musculoskeletal signals were studied using wavelet transform coherence and convergent cross mapping techniques, respectively. Parkinson’s patients displayed an impaired cardiac baroreflex and a reduced mechanical effect of the heart on blood pressure during supine, tilt and standing positions. However, the effectiveness of the cardiac baroreflex decreased in both Parkinson’s patients and healthy controls during standing as compared to supine. In addition, Parkinson’s patients demonstrated cardio-postural coupling impairment along with a mechanical muscle pump dysfunction which both could lead to dizziness and falls. Moreover, the cardiac baroreflex had a limited effect on blood pressure during standing while lower limb muscles continued to contract and maintain blood pressure via the muscle-pump mechanism. The study findings highlighted altered bidirectional coupling between heart rate and blood pressure, as well as between muscle activity and blood pressure in Parkinson’s disease. The outcomes of this study could assist in the development of appropriate physical exercise programs to reduce falls in Parkinson’s disease by monitoring the cardiac baroreflex and cardio-postural coupling effect on maintaining blood pressure.

The continuous wavelet transform of a time series of length with values " ( = 1, … , ) sampled from a continuous signal at a time step of ∆ is defined as: where s is the stretch parameter used to change the scale, is the translation parameter used to slide the wavelet function in time, and ⁎ indicates the complex conjugate (Grinsted et al. 2004). Large scales correlate with the low-frequency components of the signal, while small scales are associated with the high-frequency components. In analogy to Fourier analysis, a wavelet power spectrum (Tian et al. 2016) of a time series with values " can be defined as follows: Given two time series and with values " and " and wavelet transforms " # ( ) and " $ ( ), the cross wavelet transform (XWT) of and is defined as: Where * denotes the complex conjugate.
The cross-wavelet power between and is defined as | " #$ ( )| and reveals areas with high common power, while the complex argument of " #$ ( ) represents the relative phase between and ( Grinsted et al. 2004).
The gain between two time series X and Y can be expressed as follows: The squared cross-wavelet coherence R % & (s) measures the localized correlation coefficient between two time series X and Y in the time-frequency domain and ranges between 0 and 1. The squared cross-wavelet coherence wavelet coherence is defined as follow.
where 〈 . 〉 is a smoothing operator in both time and scale dimensions. Smoothing is required to remove the singularities in wavelet power spectra, and enhance regions of significant power, which can be accomplished using a weighted running average in both the time and scale directions, as described by (Torrence and Compo 1998).
The statistical significance threshold of R % & (s) can be estimated using a Monte Carlo simulation with a large ensemble of surrogate data set pairs having the same coefficients as the real input data pair based on the first-order autoregressive (AR1) model (Grinsted et al. 2004).

Appendix B: Convergent cross mapping
Convergent cross mapping is a technique used to calculate the bidirectional causal relationship between two time series ( ' , = 1, … , ) and ( ' , = 1, … , ) where is the length of the time series. CCM relies on state-space reconstruction to infer causality by measuring the extent to which historical values of X can be used to accurately estimate the sates of (cross-mapping) (Sugihara and May 1990;Sugihara et al. 2012). To do so, the lagged coordinates of variables and , are used to construct the shadow manifold of ( # ) and ( $ ) respectively. The lagged coordinates of ( = ' ) and ( = ' ) are formed (Tsonis et al. 2018;Barraquand et al. 2021) as follows: > 2 = ( 2 , 2"3 , 2")3 , … , 2"(5"$)3 ) ) is the Euclidean distance between the two vectors = ' , and = . . Predicting by # is equivalent to causing , and the strength of causality flowing from to is quantified by calculating the Pearson correlation coefficient between the original time series and the estimated N | # . Similarly, to know if is causing (cross mapping of by using $ : N | $ ), we can calculate the Pearson correlation coefficient between and N | $ .