Elsevier

Gait & Posture

Volume 54, May 2017, Pages 236-241
Gait & Posture

Full length article
Dynamical analysis of balance in vestibular schwannoma patients

https://doi.org/10.1016/j.gaitpost.2017.03.015Get rights and content

Highlights

  • The complexity in CoP time series is lower in vestibular schwannoma patients.

  • The ablation of the vestibular tumor lead to an immediate decrease of complexity.

  • This decrease is followed by a progressive restoration of complexity over time.

  • The SampEn method is able to highlight the postural effects of vestibular pathology.

Abstract

The analysis of the complexity of postural fluctuations is a recent method for assessing postural control. Complexity relates to the irregularity of the center of pressure time series and characterizes the ability of postural control to meet a changing environment. In our study, we used the sample entropy (SampEn) parameter to evaluate the complexity of postural sway velocity time series in patients with vestibular schwannoma (n = 19) compared to healthy controls (n = 20), using the sensory organization test. Patients performed postural assessments three days before surgical ablation of the tumor, then three times after surgery, at eight, thirty, and ninety days. The control group underwent posturographic tests only once. Our results demonstrated that SampEn values distinguished both groups before surgery only in postural tasks where vestibular afferences significantly contribute to maintaining balance. We also found an immediate decrease of complexity after the surgical resection of the tumor. Our results are in line with the theory of complexity loss of physiological systems stating that reducing the number of their structural components or altering their coupling leads to a decrease in complexity. Finally, our findings showed that progressive restoration of complexity over time was such that no difference was found between the two groups ninety days after surgery, due to the implementation of central adaptive mechanisms and the substitution by other sensory afferences. Thus, the SampEn parameter can highlight the postural effects of vestibular pathology, and complexity analysis appears to be a valuable tool for investigating the temporal structure of CoP time series.

Introduction

Postural control relies on feedback from the somatosensory, vestibular, and visual systems [1]. The alteration of one of these systems can lead to increased postural instability. This degradation arises notably in patients with vestibular schwannoma (VS) – a benign tumor affecting Schwann cells surrounding the vestibular nerve – whose slow growth leads to a gradual vestibular dysfunction. This process is progressively compensated by central adaptive mechanisms [2], but the surgical resection of the tumor using a translabyrinthine approach induces unilateral vestibular deafferentation (uVD), leading to a decompensation of the previously compensated situation. Therefore, the uVD results in serious deterioration of balance control, which is progressively restored due to the implementation of central adaptive mechanisms. These could be of vestibular origin and could be the result of learning mechanisms involving neural structures and pathways beyond the vestibular nuclei [3].

In clinical practice, the quantification of center of pressure (CoP) displacement – using classic stabilometric measures such as sway area, sway path or length, mean velocity and variability of CoP fluctuations – is an important outcome to assess balance control [4]. Typically, low values for these parameters are interpreted as indicators of stability [5], [6]. The sensory organization test (SOT), a common protocol used to study balance disorders, gives a good knowledge of the time-course of balance compensation in VS [3]. Yet, previous studies highlighted the limitations of the main calculated variable, called equilibrium score (ES), to produce an accurate assessment of balance control. Among these limitations, it was emphasized that the ES computation is based only on the two extreme sway angle values recorded during trial, thus ignoring all other postural fluctuations, and on a 12.5° theoretical range of sway limit of stability without taking into account individual differences (see [7] for a review). These limitations questioned the validity of this method for analyzing postural fluctuations to distinguish between individuals presenting different health status, and capture accurately the balance compensation in VS patients following the ablation of the tumor.

Moreover, conflicting results were reported in the literature concerning the effectiveness of stabilometric measures to highlight the effects of disease [8], as well as to distinguish between populations of various ages [9], and between various experimental conditions [10]. Thus, the assumption of an association between stability of posture and variability of classical parameters is currently debated [11], [12]. Newell et al. [12] were the first to advocate not to associate these two concepts systematically. Likewise, Woollacott [13] showed that quantity of displacements of CoP is not correlated to the quality of postural control. These discrepancies prevented us from drawing conclusions regarding factors (e.g. aging and disease) that could modulate these measures and suggest that new methods may be necessary to investigate the changes in postural control.

For several decades new methods based on dynamics systems were increasingly used for characterizing the dynamical features of postural sway [14], [15], [16], [17]. Among them, some techniques consist in assessing CoP dynamics through the quantification of the complexity of CoP signals. In this context, complexity is related to regularity, predictability and temporal correlations. Authors developed the theory of complexity loss suggesting that advancing in age and disease − hence the deterioration of physiological systems – seems to be associated with a decrease of complexity of CoP trajectories [15], [16]. The two basic principles behind this theory are that (i) the output of a healthy system reveals a type of complex variability associated with long-range correlations and nonlinear interactions; (ii) and this complexity breaks down with aging and disease, reducing the adaptive capabilities of the individual. Over the years, various algorithms were developed to better estimate the entropy (i.e. complexity) of a system [14], [18], [19]. Recently, Richman and Moorman [17] introduced the sample entropy (SampEn) method to quantify the regularity of time series. The more irregular the signal is, the higher the SampEn is. The SampEn acts as a good measurement of complexity in many applications such as heart rate variability [17], EMG recordings [20], and postural sway [21].

Therefore, the purpose of this study was to use SampEn (i) to evaluate the effect of vestibular dysfunction on the complexity of CoP trajectories and (ii) to assess the time-course of this complexity (pre- and post-uVD). According to the theory of loss complexity, we assumed that SampEn values decrease with VS compared to controls; and we predict a decrease of complexity in VS patients early after uVD, which will be progressively restored over time.

Section snippets

Participants

Nineteen patients (Table 1) with unilateral VS who were scheduled for surgical ablation using translabyrinthine approach took part in the protocol. Patients were compared to a healthy control group (n = 20). The patients performed postural assessment three days before surgery (BS) and three times after surgery, at eight (AS8), thirty (AS30), and ninety days (AS90). Participants enrolled in the control group underwent posturographic tests only once. Each participant provided written informed

Healthy profile of complexity

The one-way ANOVA showed significant changes in SampEn of both AP- and ML-direction CoP time series between SOT conditions (AP: F(5,295) = 124.70, p < 0.001, ηp2 = 0.68; ML: F(5,295) = 51.13, p < 0.001, ηp2 = 0.46). Tukey’s post hoc test indicated that the complexity of AP time series declined as the degree of task’s sensory difficulty increased (i.e. C1 > C2 > C3 > C4 > C5 and C6, all p < 0.05), except for the C5-C6 comparison (p = 0.06). In ML direction, the SampEn was lower in C5 and C6 compared to the first four

Discussion

The aim of this study was to use the sample entropy method to explore the temporal structure of postural fluctuations in patients with VS compared to healthy controls. The main results are that (i) SampEn values distinguished both groups before uVD only in postural tasks which require a substantial contribution of vestibular afferences in maintaining balance (i.e. C5 and C6); (ii) the surgical resection of the tumor led to an immediate decrease of complexity in CoP time series which was

Conclusion

As a conclusion, our results revealed changes in complexity in patients with VS as well as healthy controls according to the constraints – information available for postural regulation – set by conditions. Moreover, a decrease in complexity of CoP fluctuations was found in VS patients compared to healthy controls, especially in situations involving vestibular inputs to maintain balance. In addition to this, the surgical resection of the tumor leads to a decrease in complexity of CoP

Funding

No sources of funding were used to assist in the preparation of this study.

Conflict of interest

The authors have no conflicts of interest that are directly relevant to the content of this study.

References (30)

  • M. Roerdink et al.

    Center-of-pressure regularity as a marker for attentional investment in postural control: a comparison between sitting and standing postures

    Hum. Mov. Sci.

    (2011)
  • D.E. Vaillancourt et al.

    Changing complexity in human behavior and physiology through aging and disease

    Neurobiol. Aging

    (2002)
  • L. Ribeyre et al.

    Preoperative balance control compensation in patients with a vestibular schwannoma: does tumor size matter?

    Clin. Neurophysiol.

    (2015)
  • I.S. Curthoys

    Vestibular compensation and substitution

    Curr. Opin. Neurol.

    (2000)
  • C. Parietti-Winkler et al.

    Prediction of balance compensation after vestibular schwannoma surgery

    Neurorehabil. Neural Repair

    (2016)
  • Cited by (14)

    • Postural disorders induced by cochlear implantation in normo-reflexic patients: A potential origin of a transient geocentric perception disorder

      2020, Gait and Posture
      Citation Excerpt :

      Each time series was calculated using the negative natural logarithm of the conditional probability that a time series of length N, having repeated himself for m samples within a tolerance r, will also repeat itself for m + 1 samples, excluding the selected on itself (i.e., self-matches). A(r) and B(r) were the total number of templates matches of length m + 1 and m, respectively, within a tolerance r. Parameters for the analysis were set at m = 3 and r = 0.3 [19]. The PhysioToolkit-PhysioNet SampEn software was used for the estimation of the SampEn values [20].

    • Weighting and reweighting of visual input via head mounted display given unilateral peripheral vestibular dysfunction

      2019, Human Movement Science
      Citation Excerpt :

      Clinically, the relationship between age, vestibular dysfunction and postural control has been primarily studied within tasks that include complete elimination of visual input, e.g., standing with eyes closed (Strang, Haworth, Hieronymus, Walsh, & Smart Jr., 2011; Vitkovic, Le, Lee, & Clark, 2016), or dynamic posturography (Sensory Organization Test, SOT) that includes challenging sway-referenced environments (Hoffmann, Seigle, Frère, & Parietti-Winkler, 2017; Pedalini, Cruz, Bittar, Lorenzi, & Grasel, 2009). It is known that people with vestibular loss increase postural sway when presented with a sensory conflict (e.g., swaying visual surround) (Eysel-Gosepath, McCrum, Epro, Brüggemann, & Karamanidis, 2016; Gazzola, Caovilla, Doná, Ganança, & Ganança, 2019; Hoffmann et al., 2017; Mulavara, Cohen, Peters, Sangi-Haghpeykar, & Bloomberg, 2013; Pedalini et al., 2009; Sprenger, Wojak, Jandl, & Helmchen, 2017). However, increased postural sway in response to visual perturbations is not unique to vestibular loss and has also been observed in aging adults (Cohen, Heaton, Congdon, & Jenkins, 1996; Mancini & Horak, 2010; Prieto, Myklebust, Hoffmann, Lovett, & Myklebust, 1996).

    • Control Mechanisms of Static and Dynamic Balance in Adults With and Without Vestibular Dysfunction in Oculus Virtual Environments

      2018, PM and R
      Citation Excerpt :

      Note that the comparison of values is made possible, although the assessment was done in 2 separate studies, because we processed the signal in the same manner in the 2 studies (down-sampling by 2, no smoothing applied) [26] and used the same parameters to calculate SampEn (r, m) [27]. Our findings are in agreement with previous studies on environmental constraints [7,8]: regularity increases (ie, SampEn decreases) when participants are standing on a compliant surface compared with a stable floor and when visual complexity increases (ie, dark vs light, larger sway of stars compared with minimal sway). In fact, the effect of the support surface on regularity was so dominant that any between-group differences vanished when participants were standing on the stability trainer.

    View all citing articles on Scopus
    View full text