Hostname: page-component-cd9895bd7-jkksz Total loading time: 0 Render date: 2024-12-23T07:23:07.656Z Has data issue: false hasContentIssue false
  • English
  • Français

Instrumented insoles for assessment of gait in patients with vestibular schwannoma

Part of: Emerging Wearable Technologies to Promote Wellness and Healthy Aging in Older Adults

Published online by Cambridge University Press:  10 May 2023

Stephen Leong Open the ORCID record for Stephen Leong [Opens in a new window] ,
Bing M. Teh ,
Ton Duong ,
Diane Hu ,
Alexander Chui ,
Jocelyn S. Chen ,
Michael B. Sisti ,
Tony J.C. Wang ,
Damiano Zanotto Open the ORCID record for Damiano Zanotto [Opens in a new window]  and
Anil K. Lalwani
Stephen Leong
Affiliation:
Vagelos College of Physicians & Surgeons, Columbia University Irving Medical Center, New York, NY, USA
Bing M. Teh
Affiliation:
Department of Otolaryngology—Head & Neck Surgery, NewYork-Presbyterian/Columbia University Irving Medical Center, New York, NY, USA Department of Otolaryngology—Head & Neck Surgery, Monash Health; Faculty of Medicine, Nursing and Health Sciences, Monash University, Clayton, VC, Australia
Ton Duong
Affiliation:
Department of Mechanical Engineering, Stevens Institute of Technology, Hoboken, NJ, USA
Diane Hu
Affiliation:
Vagelos College of Physicians & Surgeons, Columbia University Irving Medical Center, New York, NY, USA
Alexander Chui
Affiliation:
Department of Biological Sciences, Columbia University, New York, NY, USA
Jocelyn S. Chen
Affiliation:
Department of Biomedical Engineering, Columbia University, New York, NY, USA
Michael B. Sisti
Affiliation:
Department of Otolaryngology—Head & Neck Surgery, NewYork-Presbyterian/Columbia University Irving Medical Center, New York, NY, USA Department of Neurological Surgery, NewYork-Presbyterian/Columbia University Irving Medical Center, New York, NY, USA Department of Radiation Oncology, NewYork-Presbyterian/Columbia University Irving Medical Center, New York, NY, USA
Tony J.C. Wang
Affiliation:
Department of Neurological Surgery, NewYork-Presbyterian/Columbia University Irving Medical Center, New York, NY, USA Department of Radiation Oncology, NewYork-Presbyterian/Columbia University Irving Medical Center, New York, NY, USA
Damiano Zanotto
Affiliation:
Department of Mechanical Engineering, Stevens Institute of Technology, Hoboken, NJ, USA
Anil K. Lalwani*
Affiliation:
Vagelos College of Physicians & Surgeons, Columbia University Irving Medical Center, New York, NY, USA Department of Otolaryngology—Head & Neck Surgery, NewYork-Presbyterian/Columbia University Irving Medical Center, New York, NY, USA Department of Mechanical Engineering, Columbia University, New York, NY, USA
*
Corresponding author: Anil K. Lalwani; Email: [email protected]

Abstract

Background

Imbalance and gait disturbances are common in patients with vestibular schwannoma (VS) and can result in significant morbidity. Current methods for quantitative gait analysis are cumbersome and difficult to implement. Here, we use custom-engineered instrumented insoles to evaluate the gait of patients diagnosed with VS.

Methods

Twenty patients with VS were recruited from otology, neurosurgery, and radiation oncology clinics at a tertiary referral center. Functional gait assessment (FGA), 2-minute walk test (2MWT), and uneven surface walk test (USWT) were performed. Custom-engineered instrumented insoles, equipped with an 8-cell force sensitive resistor (FSR) and a 9-degree-of-freedom inertial measurement unit (IMU), were used to collect stride-by-stride spatiotemporal gait parameters, from which mean values and coefficients of variation (CV) were determined for each patient.

Results

FGA scores were significantly correlated with gait metrics obtained from the 2MWT and USWT, including stride length, stride velocity, normalized stride length, normalized stride velocity, stride length CV, and stride velocity CV. Tumor diameter was negatively associated with stride time and swing time on the 2MWT; no such association existed between tumor diameter and FGA or DHI.

Conclusions

Instrumented insoles may unveil associations between VS tumor size and gait dysfunction that cannot be captured by standardized clinical assessments and self-reported questionnaires.

Keywords

Sensorssoft wearable roboticsvestibular schwannomavestibular dysfunction
Type
Research Article
Information
Wearable Technologies , Volume 4 , 2023 , e14
Check for updates
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0), which permits unrestricted re-use, distribution and reproduction, provided the original article is properly cited.
Copyright
© The Author(s), 2023. Published by Cambridge University Press

1. Introduction

Vestibular schwannomas (VS) are benign tumors of the vestibulocochlear nerve that may present with unilateral hearing loss, tinnitus, episodic vertigo, and imbalance (Kentala and Pyykkö, Reference Kentala and Pyykkö2001). Additionally, patients with VS often have some level of gait disturbance, although most patients achieve a level of vestibular compensation such that their disequilibrium is tolerable on a daily basis (Ishikawa et al., Reference Ishikawa, Cao, Wang, Wong, Tanaka, Miyazaki and Toyoshima2001, Reference Ishikawa, Wang, Wong, Shibata and Itasaka2004; Kentala and Pyykkö, Reference Kentala and Pyykkö2001; Angunsri et al., Reference Angunsri, Ishikawa, Yin, Omi, Shibata, Saito and Itasaka2011; Wang et al., Reference Wang, Jiang, Guan, Jiang, Kazuo and Zhou2011; Yin et al., Reference Yin, Ishikawa, Omi, Saito, Itasaka and Angunsuri2011; Nam et al., Reference Nam, Jung, Kim and Son2018). Nonetheless, gait disturbance is a significant risk factor for falls, especially in elderly patients, and the link between gait disturbance and fall risk has been identified for multiple domains of gait (Hausdorff et al., Reference Hausdorff, Rios and Edelberg2001; Cesari et al., Reference Cesari, Kritchevsky, Penninx, Nicklas, Simonsick, Newman, Tylavsky, Brach, Satterfield, Bauer, Visser, Rubin, Harris and Pahor2005; Springer et al., Reference Springer, Giladi, Peretz, Yogev, Simon and Hausdorff2006; Abellan van Kan et al., Reference Abellan van Kan, Rolland, Andrieu, Bauer, Beauchet, Bonnefoy, Cesari, Donini, Gillette Guyonnet, Inzitari, Nourhashemi, Onder, Ritz, Salva, Visser and Vellas2009; Nordin et al., Reference Nordin, Moe-Nilssen, Ramnemark and Lundin-Olsson2010; Oh-Park et al., Reference Oh-Park, Wang and Verghese2011; Studenski et al., Reference Studenski, Perera, Patel, Rosano, Faulkner, Inzitari, Brach, Chandler, Cawthon, Connor, Nevitt, Visser, Kritchevsky, Badinelli, Harris, Newman, Cauley, Ferrucci and Guralnik2011; Pamoukdjian et al., Reference Pamoukdjian, Paillaud, Zelek, Laurent, Lévy, Landre and Sebbane2015). Specifically, gait speed, stride length, and gait variability have been shown to be correlated with frailty and overall survival in elderly patients (Hausdorff et al., Reference Hausdorff, Rios and Edelberg2001; Springer et al., Reference Springer, Giladi, Peretz, Yogev, Simon and Hausdorff2006; Abellan van Kan et al., Reference Abellan van Kan, Rolland, Andrieu, Bauer, Beauchet, Bonnefoy, Cesari, Donini, Gillette Guyonnet, Inzitari, Nourhashemi, Onder, Ritz, Salva, Visser and Vellas2009; Nordin et al., Reference Nordin, Moe-Nilssen, Ramnemark and Lundin-Olsson2010; Studenski et al., Reference Studenski, Perera, Patel, Rosano, Faulkner, Inzitari, Brach, Chandler, Cawthon, Connor, Nevitt, Visser, Kritchevsky, Badinelli, Harris, Newman, Cauley, Ferrucci and Guralnik2011; Pamoukdjian et al., Reference Pamoukdjian, Paillaud, Zelek, Laurent, Lévy, Landre and Sebbane2015). In patients with VS, increased gait variability and greater foot pressure on the side of the lesion have been demonstrated, particularly with visual deprivation (Ishikawa et al., Reference Ishikawa, Cao, Wang, Wong, Tanaka, Miyazaki and Toyoshima2001, Reference Ishikawa, Wang, Wong, Shibata and Itasaka2004; Angunsri et al., Reference Angunsri, Ishikawa, Yin, Omi, Shibata, Saito and Itasaka2011; Wang et al., Reference Wang, Jiang, Guan, Jiang, Kazuo and Zhou2011; Yin et al., Reference Yin, Ishikawa, Omi, Saito, Itasaka and Angunsuri2011).

Previous studies have utilized clinical assessments and standardized surveys to assess gait in patients with VS or vestibular disorders; in particular, the Functional Gait Assessment (FGA) and Dizziness Handicap Index (DHI) have been used as strong predictors of quality-of-life in patients with VS receiving surgical treatment (Said et al., Reference Said, Lee, Moshtaghi, Saliba, Richardson, Ngo, Mehranpour, Schwartz and Friedman2021; Zobeiri et al., Reference Zobeiri, Mischler, King, Lewis and Cullen2021). The DHI is a 25-item self-reported questionnaire that quantifies patients’ dizziness disability in three domains: functional, emotional, and physical (Jacobson and Newman, Reference Jacobson and Newman1990). In contrast, the FGA is a 10-part series of ambulation tasks, where patients walk with eyes closed, over an obstacle, over a set of stairs, etc. (Wrisley et al., Reference Wrisley, Marchetti, Kuharsky and Whitney2004). Though both assessments have had utility in the VS population, the DHI is more commonly used due to its ease of administration (Mutlu and Serbetcioglu, Reference Mutlu and Serbetcioglu2013; Zobeiri et al., Reference Zobeiri, Mischler, King, Lewis and Cullen2021). The FGA has been validated against other measures of dizziness, balance, and fall risk (Wrisley and Kumar, Reference Wrisley and Kumar2010) and provides more comprehensive data on patients’ walking capacity (Marchetti et al., Reference Marchetti, Lin, Alghadir and Whitney2014).

Recently, the assessment of gait in patients with VS has involved a number of gait parameters, including the trajectory of the center of force (TCOF), foot pressure, stance time, swing time, and double support time, along with the coefficient of variation (CV) associated with each of these metrics (Ishikawa et al., Reference Ishikawa, Wang, Wong, Shibata and Itasaka2004; Angunsri et al., Reference Angunsri, Ishikawa, Yin, Omi, Shibata, Saito and Itasaka2011; Yin et al., Reference Yin, Ishikawa, Omi, Saito, Itasaka and Angunsuri2011). Notably, studies on these parameters have demonstrated that patients with VS do not have significant gait phase changes, but do have a higher CV for stance and swing when walking with eyes closed (Wang et al., Reference Wang, Jiang, Guan, Jiang, Kazuo and Zhou2011; Yin et al., Reference Yin, Ishikawa, Omi, Saito, Itasaka and Angunsuri2011). Additionally, patients with VS exhibit a gait shift to the side of their lesion when walking with their eyes closed, demonstrating the role of visual feedback in producing compensation (Ishikawa et al., Reference Ishikawa, Wang, Wong, Shibata and Itasaka2004).

Quantitative gait analysis for patients with VS may allow for early detection of imbalance and disequilibrium and early initiation of physical and vestibular therapy. However, traditional equipment for gait analysis is expensive and/or cumbersome and cannot easily be applied to the clinical setting. Examples of devices used in previous studies include non-reusable tactile sensors attached to patients’ feet with adhesive tape (Ishikawa et al., Reference Ishikawa, Cao, Wang, Wong, Tanaka, Miyazaki and Toyoshima2001; Angunsri et al., Reference Angunsri, Ishikawa, Yin, Omi, Shibata, Saito and Itasaka2011), instrumented walkways (Ohara et al., Reference Ohara, Yasuhiro, Mochizuki, Shinohe, Sasano, Suzuki, Mikami and Koizuka2021), and optical motion analysis systems (Anson et al., Reference Anson, Pineault, Bair, Studenski and Agrawal2019), none of which can easily be introduced into office spaces. Additionally, these systems offer only a limited working distance (8–10 m) over which patients can navigate, when in fact dozens to hundreds of strides may be required to reliably evaluate gait variability (Hollman et al., Reference Hollman, Childs, McNeil, Mueller, Quilter and Youdas2010; Lord et al., Reference Lord, Howe, Greenland, Simpson and Rochester2011).

Our team has previously developed various iterations of instrumented footwear for the assessment of gait and balance, all of which consist of highly portable sets of inertial sensors embedded in footwear (Zanotto et al., Reference Zanotto, Turchet, Boggs and Agrawal2014; Minto et al., Reference Minto, Zanotto, Boggs, Rosati and Agrawal2016; Zhang et al., Reference Zhang, Duong, Rao, Mazzoni, Agrawal, Guo and Zanotto2022). We have used this footwear to explore associations between gait disturbances and DHI scores (Zanotto et al., Reference Zanotto, Mamuyac, Chambers, Nemer, Stafford, Agrawal and Lalwani2017), and to demonstrate that hearing loss is highly correlated with gait variability, which highlights the importance of auditory feedback for walking and balance (Szeto et al., Reference Szeto, Zanotto, Lopez, Stafford, Nemer, Chambers, Agrawal and Lalwani2021). Despite the usability of our technology, patient comfort may have been a limiting factor—our technology thus far has required patients to wear instrumented sandals in place of their own footwear to complete gait analysis, which may have affected their natural walking patterns. In this study, we use new, minimally obtrusive instrumented insoles that may be fitted into the patients’ own footwear, which increases both portability and comfort during walking tasks (Duong et al., Reference Duong, Uher, Montes and Zanotto2022; Zhang et al., Reference Zhang, Zanotto and Agrawal2017, Reference Zhang, Guo and Zanotto2020). Specifically, we use instrumented insoles to evaluate gait in patients with VS; we perform the FGA as a standardized assessment, then collect insole data during a 2-minute walk test (2MWT) and an uneven surface walk test (USWT). The 2MWT is a validated test commonly used to evaluate functional endurance in adults (Bohannon et al., Reference Bohannon, Wang and Gershon2015); variations of the USWT have frequently been used to assess gait dysfunction in the elderly (Bogen et al., Reference Bogen, Aaslund, Ranhoff and Moe-Nilssen2019; Osoba et al., Reference Osoba, Rao, Agrawal and Lalwani2019). Used in combination with the instrumented insoles, the 2MWT allows a large number of steady-state footsteps to be recorded; the USWT test may emphasize gait and balance dysfunction in patients with VS due to the challenging walking surface.

2. Materials and methods

2.1. Patient recruitment

Patients presenting to otology, neurosurgery, and radiation oncology clinics in a tertiary care setting were evaluated via chart review for an active diagnosis of VS. Patients with a significant neurologic disorder, diagnosis of neurofibromatosis II, and age > 80 years were excluded from the study. Patients fulfilling both inclusion and exclusion criteria were contacted via phone and asked for voluntary participation in the study; a confirmation email was sent to all patients agreeing to participate. All study procedures and details were approved by the Institutional Review Board (IRB) of Columbia University.

2.2. Instrumented insoles

The instrumented insole system was developed to capture spatiotemporal gait parameters in real-life environments (Duong et al., Reference Duong, Uher, Montes and Zanotto2022). The system consists of a pair of instrumented insoles, a pair of logic units clipped to the posterolateral side of the patient’s footwear, and an Android smartphone (Figure 1). Each insole consists of an 8-cell force-sensitive resistor (FSR) and a 9-degree-of-freedom inertial measurement unit (IMU). The logic unit consists of a Linux-based single-board computer with integrated Wi-Fi connectivity and on-board data storage. Each logic unit is powered by a 3.7 V 2000mAh Li-Po battery through a 5 V voltage booster with built-in charger circuit. Data from the instrumented insoles are acquired by the logic units at 333 Hz. The system is controlled by the Android smartphone through a custom application. In this study, four insole sizes were used to cover most of the common US shoe sizes (US W5.5 to M11). The overall weight of an instrumented insole and a logic unit is less than 130 grams. The hardware and software require approximately 5 minutes to equip onto patients and initiate data collection. This technology has been validated against gold-standard gait analysis systems in young healthy individuals (Zhang et al., Reference Zhang, Guo and Zanotto2020), older adults (Zhang et al., Reference Zhang, Duong, Rao, Mazzoni, Agrawal, Guo and Zanotto2022), and patients with neuromuscular (Duong et al., Reference Duong, Uher, Young, Duong, Sangco, Cornett, Montes and Zanotto2021) or neurodevelopmental (Duong et al., Reference Duong, Goldman, Zhang, Salazar, Beenders, Cornett, Bain, Montes and Zanotto2020) disorders.

Figure 1. Components of the instrumented insole system, including (a) an 8-cell force sensitive resistor (FSR) and 9-degree-of-freedom inertial measurement unit (IMU), (b) a smartphone control application, and (c) the fully assembled insole and logic unit.

2.3. Gait and balance assessments

Patients participating in the study completed an online Qualtrics survey (Qualtrics XM, Seattle, Washington) consisting of demographic questions (age, sex, race, height, medical history) and the standard 25-question DHI prior to in-person testing. For all gait assessments, patients wore instrumented insoles for the duration of testing. Patients completed three standardized gait assessments, including the 10-part FGA (Wrisley et al., Reference Wrisley, Marchetti, Kuharsky and Whitney2004), the 2-minute walk test (Bohannon et al., Reference Bohannon, Wang and Gershon2015), and an 80-meter (10-meter walkway, 8 laps) uneven surface walk test performed on two 8 feet x 4 feet x 2 inch soft exercise mats (Bogen et al., Reference Bogen, Aaslund, Ranhoff and Moe-Nilssen2019). Gait parameters obtained from insoles included stride length, stride time, stride velocity, stance percent, and swing percent. Gait parameters were obtained from the USWT and 2MWT due to the extended duration of walking required for both tests. Normalization for stride length and stride velocity were calculated as previously described (Hof, Reference Hof1996). Following testing, the mean and CV for each metric was calculated for all patients.

2.4. Statistical analysis

All statistical analyses were performed in R (R Core Team, Vienna, Austria) and Prism (GraphPad, San Diego, California). Patient data were approximately normal in distribution. Pearson correlation analyses were performed to compare scores on FGA and DHI versus insole-derived gait parameters. Linear regression was used to analyze the relationship between tumor size and DHI, FGA, and gait parameters. Paired t-tests were performed to compare gait parameters obtained during the 2MWT and the USWT.

3. Results

In total, 20 patients with a diagnosis of VS completed gait testing with instrumented insoles (Table 1). Of these patients, 20 completed USWT, 16 completed 2MWT, and 16 completed the online survey, including the DHI. The mean age among participants was 63.0 ± 10.1; participants were predominantly female (60.0%). Of the patients completing the online survey, 75.0% reported hearing loss, 63.5% reported tinnitus, and 50.0% reported dizziness (Table 1). The mean DHI score among participants was 14.0 ± 16.1; the mean FGA score was 25.1 ± 3.9; and the mean tumor diameter was 13.5 mm ± 6.8 mm (Table 1).

Table 1. Demographics and characteristics of the study population

Note. In total, 20 patients completed some amount of gait testing; of these 20 patients, 16 completed the online survey. Symptomatology and past medical history are based on the 16-patient group; other demographics and scores are based on the 20-patient group, with the exception of the DHI (based on the 16-patient group).

DHI score was significantly correlated with the following 2MWT metrics: swing time (r = −0.592, p = 0.03) and stride time CV (r = 0.719, p = 0.006) (Supplementary Tables S1 and S2). FGA score was significantly correlated with the following 2MWT and USWT metrics: stride length (2MWT: r = 0.81, p = 0.0001; USWT: r = 0.78, p = 0.00004), stride velocity (2MWT: r = 0.78, p = 0.0003; USWT: r = 0.72, p = 0.0004), normalized stride length (2MWT: r = 0.71, p = 0.002; USWT: r = 0.73, p = 0.0003), normalized stride velocity (2MWT: r = 0.73, p = 0.001; USWT: r = 0.67, p = 0.001), stride length CV (2MWT: r = −0.63, p = 0.008; USWT: r = −0.57, p = 0.009), and stride velocity CV (2MWT: r = −0.61, p = 0.01, USWT: r = −0.56, p = 0.01) (Figure 2, Supplementary Tables S1 and S2).

Figure 2. Pearson correlation analyses for FGA versus selected gait parameters obtained from the 2MWT. All selected parameters are significantly correlated with FGA, and include stride length (a), stride velocity (b), normalized stride length (c), normalized stride velocity (d), stride length CV (e), and stride velocity CV (f). Correlation coefficients and corresponding p-values are displayed.

Figure 3. Pearson correlation analyses for FGA versus selected gait parameters obtained from the USWT. All selected parameters are significantly correlated with FGA, and include stride length (a), stride velocity (b), normalized stride length (c), normalized stride velocity (d), stride length CV (e), and stride velocity CV (f). Correlation coefficients and corresponding p-values are displayed.

No linear relationship existed between tumor diameter and FGA score (R 2 = 0.05, p = 0.3) or DHI score (R 2 = 0.19, p = 0.09) (Figure 4). Linear relationships existed between tumor diameter and the following 2MWT metrics: stride time (R 2 = 0.28, p = 0.04) and swing time (R 2 = 0.39, p = 0.01) (Figure 5; Supplementary Table S3). No linear relationships existed between tumor diameter and USWT metrics, although the regression analysis for swing time approached significance (R 2 = 0.190, p = 0.055) (Figure 5; Supplementary Table S3).

Figure 4. Linear regression analyses for tumor diameter versus FGA (a) and DHI (b) scores. Neither FGA nor DHI are significantly associated with tumor size. R 2 values and corresponding p-values are displayed.

Figure 5. Linear regression analyses for tumor diameter versus 2MWT metrics: stride time (a) and swing time (b), and USWT metrics: stride time (c) and swing time (d). Neither FGA nor DHI are significantly associated with tumor size. R 2 values and corresponding p-values are displayed.

4. Discussion

In this study, we used instrumented insoles to assess gait in patients with an active diagnosis of VS. The insoles are highly portable and can be used with the patient’s own footwear to maximize comfort. Additionally, the system is relatively straightforward to initiate, requiring only 5 minutes to equip onto patients for data collection, including software start-up time. Thus, instrumented insoles may easily be integrated into the routine management of patients with VS, in the clinical setting and potentially in the home setting as well.

In our patient population, the average FGA score was 25.1 ± 3.9, comparable to the national average for the age group (Walker et al., Reference Walker, Austin, Banke, Foxx, Gaetano, Gardner, McElhiney, Morris and Penn2007). The average DHI score was 14.0 ± 16.1, corresponding to mild levels of dizziness (Formeister et al., Reference Formeister, Krauter, Kirk, Zhu, Rizk and Sharon2020). Most patients reported hearing loss and/or tinnitus. These results are consistent with the most common presentation of VS, where hearing symptoms predominate and vestibular symptoms are relatively mild (Kentala and Pyykkö, Reference Kentala and Pyykkö2001). We found that FGA scores were strongly positively correlated with a number of gait parameters, namely, stride length, stride velocity, normalized stride length, and normalized stride velocity, for both 2MWT and USWT. These results are consistent with the literature, as patients with vestibular gait dysfunction have worse FGA performance, with worse balance, dizziness, and fall risk (Marchetti et al., Reference Marchetti, Lin, Alghadir and Whitney2014), and in general take shorter strides (Yamamoto et al., Reference Yamamoto, Mamoto, Imai, Hirasaki and Kubo2002; Agrawal et al., Reference Agrawal, Davalos-Bichara, Zuniga and Carey2013; Chae et al., Reference Chae, Song and Kim2021). Thus, we demonstrate that our instrumented insoles can reliably be used to detect walking dysfunction. The moderate negative correlation between FGA and the coefficient of variability for stride length and stride velocity adds additional evidence to this notion, since increased variability in gait is expected in patients with vestibular gait dysfunction (Hausdorff et al., Reference Hausdorff, Rios and Edelberg2001; Lord et al., Reference Lord, Howe, Greenland, Simpson and Rochester2011; Szeto et al., Reference Szeto, Zanotto, Lopez, Stafford, Nemer, Chambers, Agrawal and Lalwani2021). Interestingly, we did not find comparable correlations between the DHI and insole-derived gait parameters, which may be due to the highly subjective nature of the DHI. The subjectivity of the DHI is a particular concern for patients with VS, who may not necessarily characterize their vestibular dysfunction as dizziness (Kentala and Pyykkö, Reference Kentala and Pyykkö2001).

Among our patients, the average VS tumor diameter was 13.5 mm ± 6.8 mm, suggesting that most patients presented early in the course of their disease (Sughrue et al., Reference Sughrue, Yang, Aranda, Lobo, Pitts, Cheung and Parsa2010). We found no linear associations between tumor diameter and FGA or DHI scores, likely due to ceiling effects for both assessments. Among gait parameters, we found that tumor diameter was negatively associated with two 2MWT gait parameters, stride time and swing time. In other words, patients with larger tumors had shorter swing periods, producing shorter stride times. This association was weaker during the USWT, though in general patients took slower steps on the uneven surface. These results indicate that tumor size may be related to gait dysfunction, as is consistent with the literature (Wagner et al., Reference Wagner, Glaser, Wowra, Muacevic, Goldbrunner, Cnyrim, Tonn and Strupp2011; Yin et al., Reference Yin, Ishikawa, Omi, Saito, Itasaka and Angunsuri2011), though the exact type of dysfunction is unclear. Thus, instrumented insoles may be used to characterize subtle changes in gait associated with tumor growth that the FGA and DHI cannot detect.

The major strength of our study was the use of innovative technology in a patient population with a specific medical diagnosis. By doing so, we accomplished two goals: we provided further evidence for the utility of our technology, and we drew conclusions about gait dysfunction in patients with VS. Major limitations in our study included non-responses on the online survey and inability to complete gait testing for some patients. Additionally, our study lacked an age-matched control group for comparison, which would have allowed for more robust conclusions to be drawn from the data. Given the usability and portability of our instrumented insoles, assembling a reference group will likely be straightforward, especially if devices may be distributed to patients for testing in the home setting.

5. Conclusion

In this study, we used instrumented insoles to assess gait in patients with an active diagnosis of VS. We found significant correlations between insole-derived gait parameters and FGA scores, indicating that insole data was consistent with validated tests routinely used in the clinical setting. Though tumor size was not a reliable predictor of FGA or DHI scores, tumor size was negatively associated with stride time and swing time on the 2MWT test. Thus, instrumented insoles may unveil associations between tumor growth and gait dysfunction that cannot be captured by standardized clinical assessments and self-reported questionnaires.

Supplementary material

The supplementary material for this article can be found at https://doi.org/10.1017/wtc.2023.11.

Acknowledgments

We would like to acknowledge Elsie Rivera-Negron, RN, Joanna Miegoc, RN, and Maryellen Horan, MSN, FNP-BC for their contributions to patient care and coordination.

Data availability statement

Full datasets have been made available in the supplementary materials. Protocols may be requested through the corresponding author.

Authorship contribution

S.L., B.M.T., T.D., D.Z., and A.K.L. conceived and designed the study. S.L., B.M.T., T.D., D.H., A.C., and J.S.C. conducted data gathering. M.B.S. and T.J.C.W. assisted with patient recruitment and data review. S.L., B.M.T., and T.D. performed statistical analyses. S.L., B.M.T., T.D., D.Z., and A.K.L. wrote the article.

Funding statement

This work was supported by the American Neurotological Society Research Grant Award (awarded to B.M.T.).

Competing interest

Stephen Leong, BA: None. Bing M. Teh, MBBS, PhD, FRACS: None. Ton Duong, MEng: None. Diane Hu, BA: None. Michael B. Sisti, MD: None. Tony J. C. Wang, MD: Support from AbbVie, Cancer Panels, Doximity, Elekta, Merck, Novocure, RTOG Foundation, Rutgers, University of Iowa, Wolters Kluwer, Genentech, Varian, Iylon Precision Oncology. Damiano Zanotto, PhD: None. Anil K. Lalwani, MD: Medical/Surgical Advisory Board of Spiral Therapeutics.

Ethical standard

The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committees on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008.

References

Abellan van Kan, G, Rolland, Y, Andrieu, S, Bauer, J, Beauchet, O, Bonnefoy, M, Cesari, M, Donini, LM, Gillette Guyonnet, S, Inzitari, M, Nourhashemi, F, Onder, G, Ritz, P, Salva, A, Visser, M and Vellas, B (2009) Gait speed at usual pace as a predictor of adverse outcomes in community-dwelling older people an international academy on nutrition and aging (IANA) task force. The Journal of Nutrition, Health & Aging 13(10), 881–889. https://doi.org/10.1007/s12603-009-0246-zCrossRefGoogle ScholarPubMed
Agrawal, Y, Davalos-Bichara, M, Zuniga, MG and Carey, JP (2013) Head impulse test abnormalities and influence on gait speed and falls in older individuals. Otology & Neurotology: Official Publication of the American Otological Society, American Neurotology Society [and] European Academy of Otology and Neurotology 34(9), 1729–1735. https://doi.org/10.1097/MAO.0b013e318295313cGoogle ScholarPubMed
Angunsri, N, Ishikawa, K, Yin, M, Omi, E, Shibata, Y, Saito, T and Itasaka, Y (2011) Gait instability caused by vestibular disorders—Analysis by tactile sensor. Auris, Nasus, Larynx 38(4), 462–468. https://doi.org/10.1016/j.anl.2011.01.016CrossRefGoogle ScholarPubMed
Anson, E, Pineault, K, Bair, W, Studenski, S and Agrawal, Y (2019) Reduced vestibular function is associated with longer, slower steps in healthy adults during normal speed walking. Gait & Posture 68, 340–345. https://doi.org/10.1016/j.gaitpost.2018.12.016CrossRefGoogle ScholarPubMed
Bogen, B, Aaslund, MK, Ranhoff, AH and Moe-Nilssen, R (2019) Two-year changes in gait variability in community-living older adults. Gait & Posture 72, 142–147. https://doi.org/10.1016/j.gaitpost.2019.06.005CrossRefGoogle ScholarPubMed
Bohannon, RW, Wang, Y-C and Gershon, RC (2015) Two-minute walk test performance by adults 18 to 85 years: Normative values, reliability, and responsiveness. Archives of Physical Medicine and Rehabilitation 96(3), 472–477. https://doi.org/10.1016/j.apmr.2014.10.006Google ScholarPubMed
Cesari, M, Kritchevsky, SB, Penninx, BWHJ, Nicklas, BJ, Simonsick, EM, Newman, AB, Tylavsky, FA, Brach, JS, Satterfield, S, Bauer, DC, Visser, M, Rubin, SM, Harris, TB and Pahor, M (2005) Prognostic value of usual gait speed in well-functioning older people—Results from the health. Aging and Body Composition Study. Journal of the American Geriatrics Society 53(10), 1675–1680. https://doi.org/10.1111/j.1532-5415.2005.53501.xCrossRefGoogle ScholarPubMed
Chae, S, Song, J-J and Kim, W-S (2021) Change of gait after unilateral vestibular neuritis: A prospective longitudinal observation study. Scientific Reports 11(1), 21579. https://doi.org/10.1038/s41598-021-00665-0CrossRefGoogle ScholarPubMed
Duong, TTH, Goldman, S, Zhang, H, Salazar, R, Beenders, S, Cornett, KM, Bain, JM, Montes, J and Zanotto, D (2020) Validation of insole-based gait analysis system in young children with a neurodevelopmental disorder and autism traits. In 2020 8th IEEE RAS/EMBS International Conference for Biomedical Robotics and Biomechatronics (BioRob), New York, NY, USA, pp. 715–720. https://doi.org/10.1109/BioRob49111.2020.9224273CrossRefGoogle Scholar
Duong, TTH, Uher, D, Young, SD, Duong, T, Sangco, M, Cornett, K, Montes, J and Zanotto, D (2021) Gaussian process regression for COP trajectory estimation in healthy and pathological gait using instrumented insoles. In 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). Prague, Czech Republic: IEEE, pp. 9548–9553. https://doi.org/10.1109/IROS51168.2021.9636562CrossRefGoogle Scholar
Duong, T, Uher, D, Montes, J and Zanotto, D (2022) Ecological validation of machine learning models for spatiotemporal gait analysis in free-living environments using instrumented insoles. IEEE Robotics and Automation Letters 7, 1–8. https://doi.org/10.1109/LRA.2022.3188895CrossRefGoogle Scholar
Formeister, EJ, Krauter, R, Kirk, L, Zhu, TR, Rizk, HG and Sharon, JD (2020) Understanding the dizziness handicap inventory (DHI): A cross sectional analysis of symptom factors that contribute to DHI variance. Otology & Neurotology 41(1), 86–93. https://doi.org/10.1097/MAO.0000000000002438CrossRefGoogle ScholarPubMed
Hausdorff, JM, Rios, DA and Edelberg, HK (2001) Gait variability and fall risk in community-living older adults: A 1-year prospective study. Archives of Physical Medicine and Rehabilitation 82(8), 1050–1056. https://doi.org/10.1053/apmr.2001.24893CrossRefGoogle ScholarPubMed
Hof, AL (1996) Scaling gait data to body size. Gait & Posture 4(3), 222–223. https://doi.org/10.1016/0966-6362(95)01057-2CrossRefGoogle Scholar
Hollman, JH, Childs, KB, McNeil, ML, Mueller, AC, Quilter, CM and Youdas, JW (2010) Number of strides required for reliable measurements of pace, rhythm and variability parameters of gait during normal and dual task walking in older individuals. Gait & Posture 32(1), 23–28. https://doi.org/10.1016/j.gaitpost.2010.02.017CrossRefGoogle ScholarPubMed
Ishikawa, K, Cao, ZW, Wang, Y, Wong, WH, Tanaka, T, Miyazaki, S and Toyoshima, I (2001) Dynamic locomotor function in normals and patients with vertigo. Acta Oto-Laryngologica 121(2), 241–244. https://doi.org/10.1080/000164801300043668Google ScholarPubMed
Ishikawa, K, Wang, Y, Wong, WH, Shibata, Y and Itasaka, Y (2004) Gait instability in patients with acoustic neuroma. Acta Oto-Laryngologica 124(4), 486–489. https://doi.org/10.1080/00016480310000566CrossRefGoogle ScholarPubMed
Jacobson, GP and Newman, CW (1990) The development of the dizziness handicap inventory. Archives of Otolaryngology–Head & Neck Surgery 116(4), 424–427. https://doi.org/10.1001/archotol.1990.01870040046011CrossRefGoogle ScholarPubMed
Kentala, E and Pyykkö, I (2001) Clinical picture of vestibular schwannoma. Auris Nasus Larynx 28(1), 15–22. https://doi.org/10.1016/S0385-8146(00)00093-6CrossRefGoogle ScholarPubMed
Lord, S, Howe, T, Greenland, J, Simpson, L and Rochester, L (2011) Gait variability in older adults: A structured review of testing protocol and clinimetric properties. Gait & Posture 34(4), 443–450. https://doi.org/10.1016/j.gaitpost.2011.07.010CrossRefGoogle ScholarPubMed
Marchetti, GF, Lin, C-C, Alghadir, A and Whitney, SL (2014) Responsiveness and minimal detectable change of the dynamic gait index and functional gait index in persons with balance and vestibular disorders. Journal of Neurologic Physical Therapy: JNPT 38(2), 119–124. https://doi.org/10.1097/NPT.0000000000000015CrossRefGoogle ScholarPubMed
Minto, S, Zanotto, D, Boggs, EM, Rosati, G and Agrawal, SK (2016) Validation of a footwear-based gait analysis system with action-related feedback. IEEE Transactions on Neural Systems and Rehabilitation Engineering: A Publication of the IEEE Engineering in Medicine and Biology Society 24(9), 971–980. https://doi.org/10.1109/TNSRE.2015.2498287CrossRefGoogle ScholarPubMed
Mutlu, B and Serbetcioglu, B (2013) Discussion of the dizziness handicap inventory. Journal of Vestibular Research: Equilibrium & Orientation 23(6), 271–277. https://doi.org/10.3233/VES-130488CrossRefGoogle ScholarPubMed
Nam, G-S, Jung, CM, Kim, JH and Son, EJ (2018) Relationship of vertigo and postural instability in patients with vestibular schwannoma. Clinical and Experimental Otorhinolaryngology 11(2), 102–108. https://doi.org/10.21053/ceo.2017.01277CrossRefGoogle ScholarPubMed
Nordin, E, Moe-Nilssen, R, Ramnemark, A and Lundin-Olsson, L (2010) Changes in step-width during dual-task walking predicts falls. Gait & Posture 32(1), 92–97. https://doi.org/10.1016/j.gaitpost.2010.03.012CrossRefGoogle ScholarPubMed
Ohara, A, Yasuhiro, M, Mochizuki, F, Shinohe, T, Sasano, Y, Suzuki, K, Mikami, K and Koizuka, I (2021) Effects of using cane and vestibular rehabilitation on the walking function in elderly patients with dizziness. Auris Nasus Larynx 48(4), 571–576. https://doi.org/10.1016/j.anl.2020.10.010Google ScholarPubMed
Oh-Park, M, Wang, C and Verghese, J (2011) Stair negotiation time in community dwelling older adults: Normative values and association with functional decline. Archives of Physical Medicine and Rehabilitation 92(12), 2006–2011. https://doi.org/10.1016/j.apmr.2011.07.193Google ScholarPubMed
Osoba, MY, Rao, AK, Agrawal, SK and Lalwani, AK (2019) Balance and gait in the elderly: A contemporary review. Laryngoscope Investigative Otolaryngology 4(1), 143–153. https://doi.org/10.1002/lio2.252CrossRefGoogle ScholarPubMed
Pamoukdjian, F, Paillaud, E, Zelek, L, Laurent, M, Lévy, V, Landre, T and Sebbane, G (2015) Measurement of gait speed in older adults to identify complications associated with frailty: A systematic review. Journal of Geriatric Oncology 6(6), 484–496. https://doi.org/10.1016/j.jgo.2015.08.006Google ScholarPubMed
Said, M, Lee, J, Moshtaghi, O, Saliba, J, Richardson, AJW, Ngo, V, Mehranpour, P, Schwartz, MS and Friedman, RA (2021) The relationship between the functional gait assessment and quality-of-life data in patients undergoing vestibular schwannoma resection. Otology & Neurotology 42(7), 1074–1080. https://doi.org/10.1097/MAO.0000000000003137Google ScholarPubMed
Springer, S, Giladi, N, Peretz, C, Yogev, G, Simon, ES and Hausdorff, JM (2006) Dual-tasking effects on gait variability: The role of aging, falls, and executive function. Movement Disorders: Official Journal of the Movement Disorder Society 21(7), 950–957. https://doi.org/10.1002/mds.20848CrossRefGoogle ScholarPubMed
Studenski, S, Perera, S, Patel, K, Rosano, C, Faulkner, K, Inzitari, M, Brach, J, Chandler, J, Cawthon, P, Connor, EB, Nevitt, M, Visser, M, Kritchevsky, S, Badinelli, S, Harris, T, Newman, AB, Cauley, J, Ferrucci, L and Guralnik, J (2011) Gait speed and survival in older adults. JAMA 305(1), 50–58. https://doi.org/10.1001/jama.2010.1923CrossRefGoogle ScholarPubMed
Sughrue, ME, Yang, I, Aranda, D, Lobo, K, Pitts, LH, Cheung, SW and Parsa, AT (2010) The natural history of untreated sporadic vestibular schwannomas: A comprehensive review of hearing outcomes: Clinical article. Journal of Neurosurgery 112(1), 163–167. https://doi.org/10.3171/2009.4.JNS08895CrossRefGoogle Scholar
Szeto, B, Zanotto, D, Lopez, EM, Stafford, JA, Nemer, JS, Chambers, AR, Agrawal, SK and Lalwani, AK (2021) Hearing loss is associated with increased variability in double support period in the elderly. Sensors (Basel, Switzerland) 21(1), 278. https://doi.org/10.3390/s21010278CrossRefGoogle ScholarPubMed
Wagner, J, Glaser, M, Wowra, B, Muacevic, A, Goldbrunner, R, Cnyrim, C, Tonn, J-C and Strupp, M (2011) Vestibular function and quality of life in vestibular schwannoma: Does size matter? Frontiers in Neurology 2, 55. https://doi.org/10.3389/fneur.2011.00055CrossRefGoogle ScholarPubMed
Walker, ML, Austin, AG, Banke, GM, Foxx, SR, Gaetano, L, Gardner, LA, McElhiney, J, Morris, K and Penn, L (2007) Reference group data for the functional gait assessment. Physical Therapy 87(11), 1468–1477. https://doi.org/10.2522/ptj.20060344CrossRefGoogle Scholar
Wang, Y, Jiang, H-Y, Guan, C, Jiang, X-J, Kazuo, I and Zhou, H-W (2011) Gait instability in patients with small acoustic neuroma. Chinese Medical Journal 124(11), 1735–1738Google ScholarPubMed
Wrisley, DM and Kumar, NA (2010) Functional gait assessment: Concurrent, discriminative, and predictive validity in community-dwelling older adults. Physical Therapy 90(5), 761–773. https://doi.org/10.2522/ptj.20090069CrossRefGoogle ScholarPubMed
Wrisley, DM, Marchetti, GF, Kuharsky, DK and Whitney, SL (2004) Reliability, internal consistency, and validity of data obtained with the functional gait assessment. Physical Therapy 84(10), 906–918. https://doi.org/10.1093/ptj/84.10.906CrossRefGoogle ScholarPubMed
Yamamoto, K, Mamoto, Y, Imai, T, Hirasaki, E and Kubo, T (2002) Effects of caloric vestibular stimulation on head and trunk movements during walking. Gait & Posture 15(3), 274–281. https://doi.org/10.1016/s0966-6362(01)00152-7CrossRefGoogle ScholarPubMed
Yin, M, Ishikawa, K, Omi, E, Saito, T, Itasaka, Y and Angunsuri, N (2011) Small vestibular schwannomas can cause gait instability. Gait & Posture 34(1), 25–28. https://doi.org/10.1016/j.gaitpost.2011.02.026CrossRefGoogle ScholarPubMed
Zanotto, D, Mamuyac, EM, Chambers, AR, Nemer, JS, Stafford, JA, Agrawal, SK and Lalwani, AK (2017) Dizziness handicap inventory score is highly correlated with markers of gait disturbance. Otology & Neurotology 38(10), 1490–1499. https://doi.org/10.1097/MAO.0000000000001586CrossRefGoogle ScholarPubMed
Zanotto, D, Turchet, L, Boggs, EM and Agrawal, SK (2014) SoleSound: Towards a novel portable system for audio-tactile underfoot feedback. In 5th IEEE RAS/EMBS International Conference on Biomedical Robotics and Biomechatronics. New York, NY: Association for Computing Machinery, pp. 193–198. https://doi.org/10.1109/BIOROB.2014.6913775CrossRefGoogle Scholar
Zhang, H, Duong, TTH, Rao, AK, Mazzoni, P, Agrawal, SK, Guo, Y and Zanotto, D (2022) Transductive learning models for accurate ambulatory gait analysis in elderly residents of assisted living facilities. IEEE Transactions on Neural Systems and Rehabilitation Engineering 30, 124–134. https://doi.org/10.1109/TNSRE.2022.3143094CrossRefGoogle ScholarPubMed
Zhang, H, Guo, Y and Zanotto, D (2020) Accurate ambulatory gait analysis in walking and running using machine learning models. IEEE Transactions on Neural Systems and Rehabilitation Engineering 28(1), 191–202. https://doi.org/10.1109/TNSRE.2019.2958679CrossRefGoogle ScholarPubMed
Zhang, H, Zanotto, D and Agrawal, S (2017) Estimating CoP trajectories and kinematic gait parameters in walking and running using instrumented insoles. In IEEE Robotics and Automation Letters. IEEE, pp. 2159–2165. https://doi.org/10.1109/LRA.2017.2721550Google Scholar
Zobeiri, OA, Mischler, GM, King, SA, Lewis, RF and Cullen, KE (2021) Effects of vestibular neurectomy and neural compensation on head movements in patients undergoing vestibular schwannoma resection. Scientific Reports, 11(1), 517. https://doi.org/10.1038/s41598-020-79756-3Google ScholarPubMed
Figure 0

Figure 1. Components of the instrumented insole system, including (a) an 8-cell force sensitive resistor (FSR) and 9-degree-of-freedom inertial measurement unit (IMU), (b) a smartphone control application, and (c) the fully assembled insole and logic unit.

Figure 1

Table 1. Demographics and characteristics of the study population

Figure 2

Figure 2. Pearson correlation analyses for FGA versus selected gait parameters obtained from the 2MWT. All selected parameters are significantly correlated with FGA, and include stride length (a), stride velocity (b), normalized stride length (c), normalized stride velocity (d), stride length CV (e), and stride velocity CV (f). Correlation coefficients and corresponding p-values are displayed.

Figure 3

Figure 3. Pearson correlation analyses for FGA versus selected gait parameters obtained from the USWT. All selected parameters are significantly correlated with FGA, and include stride length (a), stride velocity (b), normalized stride length (c), normalized stride velocity (d), stride length CV (e), and stride velocity CV (f). Correlation coefficients and corresponding p-values are displayed.

Figure 4

Figure 4. Linear regression analyses for tumor diameter versus FGA (a) and DHI (b) scores. Neither FGA nor DHI are significantly associated with tumor size. R2 values and corresponding p-values are displayed.

Figure 5

Figure 5. Linear regression analyses for tumor diameter versus 2MWT metrics: stride time (a) and swing time (b), and USWT metrics: stride time (c) and swing time (d). Neither FGA nor DHI are significantly associated with tumor size. R2 values and corresponding p-values are displayed.

Supplementary material: File

Leong et al. supplementary material

Table S1

Download Leong et al. supplementary material(File)
File 17.1 KB
Supplementary material: File

Leong et al. supplementary material

Table S2

Download Leong et al. supplementary material(File)
File 17.6 KB
Supplementary material: File

Leong et al. supplementary material

Table S3

Download Leong et al. supplementary material(File)
File 17.5 KB