- Email: [email protected]
Comparison of posture and balance in cancer survivors and age-matched controls
Clinical Biomechanics 50 (2017) 1–6
Contents lists available at ScienceDirect
Clinical Biomechanics journal homepage: www.elsevier.com/locate/clinbiomech
Comparison of posture and balance in cancer survivors and age-matched controls
MARK
Abigail C. Schmitta,⁎, Chris P. Repkab, Gary D. Heisec, John H. Challisd, Jeremy D. Smithc a
Michael W. Krzyzewski Human Performance Lab, Department of Orthopaedic Surgery, Duke University, 3475 Erwin Rd, Durham, NC 27705, USA Department of Health Sciences, Northern Arizona University, 1100 South Beaver St. #15095, Flagstaff, AZ 86011, USA c School of Sport & Exercise Science, University of Northern Colorado, Gunter Hall 2760, Greeley, CO 80639, USA d Biomechanics Lab, The Pennsylvania State University, 29K Recreation Building, University Park, PA 16802, USA b
A R T I C L E I N F O
A B S T R A C T
Keywords: Cancer Balance Posture Center of pressure Compliant surface Vision
Background: The combination of peripheral neuropathy and other treatment-associated side effects is likely related to an increased incidence of falls in cancer survivors. The purpose of this study was to quantify differences in postural stability between healthy age-matched controls and cancer survivors. Methods: Quiet standing under four conditions (eyes open/closed, rigid/compliant surface) was assessed in 34 cancer survivors (2 males, 32 females; age: 54(13) yrs., height: 1.62(0.07) m; mass: 78.5(19.5) kg) and 34 agematched controls (5 males, 29 females; age: 54(15) yrs.; height: 1.62(0.08) m; mass: 72.8(21.1) kg). Center of pressure data were collected for 30 s and the trajectories were analyzed (100 Hz). Three-factor (group*surface*vision) mixed model MANOVAs with repeated measures were used to determine the effect of vision and surface on postural steadiness between groups. Findings: Cancer survivors exhibited larger mediolateral root-mean square distance and velocity of the center of pressure, as well as increased 95% confidence ellipse area (P < 0.01) when compared with their age-matched counterparts. For example, when removing visual input, cancer survivors had an average increase in 95% confidence ellipse area of 91.8 mm2 while standing on a rigid surface compared to a 68.6 mm2 increase for the control group. No frequency-based center of pressure measures differed between groups. Interpretation: Cancer survivors exhibit decreased postural steadiness when compared with age-matched controls. For cancer survivors undergoing rehabilitation focused on existing balance deficits, a small subset of the center of pressure measures presented here can be used to track progress throughout the intervention and potentially mitigate fall risk.
1. Introduction Understanding postural steadiness as purposeful movements to maintain equilibrium is essential to understanding human movement (Riley et al., 1990), particularly for clinical populations who exhibit postural deficits due to disease and treatments. Researchers have attempted to elucidate the underlying mechanisms of age related postural unsteadiness by comparing elderly individuals and healthy young adults (Hageman et al., 1995; Laughton et al., 2003; Manchester et al., 1989; Prieto et al., 1996). Prieto et al. (1996) showed that removal of visual input during quiet standing resulted in greater instability in elderly individuals, suggesting elderly adults (66–70 years) were less able to control their balance compared with young adults (21–35 years). Thus, when elderly adults relied on proprioceptive and vestibular input they were less able to control their balance (Prieto et al., 1996).
⁎
Postural stability in patients with diabetic neuropathy has also received considerable attention in the literature. Patients with moderate to severe peripheral diabetic neuropathy demonstrate less stability than patients without diabetic neuropathy or control groups suggesting that neuropathy, rather than the disease, is linked to instability (Oppenheim et al., 1999; Simoneau et al., 1994). Given that postural control requires visual, somatosensory and vestibular (Lord et al., 1993; Oppenheim et al., 1999; Prieto et al., 1996; Slobounov et al., 1997) inputs, it is unsurprising that neuropathy is detrimental to postural steadiness. Neuropathy is often found in patients with cancer as well, as a side effect of cancer treatments. In the United States, nearly 1.69 million people were expected to be diagnosed with cancer in 2017 (American Cancer Society, 2017; Siegel et al., 2017). With overall survival rates approaching 70%, there are approximately 15.5 million cancer survivors in the United States
Corresponding author at: 1864 Stadium Road, University of Florida, Gainesville, FL 32611, USA. E-mail address: a.schmitt@ufl.edu (A.C. Schmitt).
http://dx.doi.org/10.1016/j.clinbiomech.2017.09.010 Received 22 April 2016; Accepted 4 September 2017 0268-0033/ © 2017 Elsevier Ltd. All rights reserved.
Clinical Biomechanics 50 (2017) 1–6
A.C. Schmitt et al.
Participants included 34 cancer survivors (2 males, 32 females; age: mean 54 (SD 13) yrs., height: mean 1.62 (SD 0.07) m; mass: mean 77.5 (SD 19.5) kg; body mass index: mean 29.9 (SD 7.3) kg/m2) in various stages of treatment and recovery. The participants in the study were selected consecutively over the course of 1 year from a larger patient population, provided they were able to safely stand unassisted for longer than 1 min. The group included breast (n = 20), colon (n = 4), skin (n = 1), brain (n = 1), ovarian (n = 1), prostate (n = 1), nonHodgkin's lymphoma (n = 2), multiple myeloma (n = 1), leukemia (n = 1), lung (n = 1), and kidney/liver (n = 1) cancers. Overall, 94% of the subjects underwent surgery, 32% had radiation treatment, and 71% had chemotherapy treatment. At the time of assessment, 21% of patients were still undergoing radiation or chemotherapy treatment. A table provided as supplemental material provides more subject specific information on the group of cancer survivors (Supplementary Table 1). Thirty-four healthy adults (5 males, 29 females; age: mean 54 (SD 15) yrs.; height: mean 1.62 (SD 0.08) m; mass: mean 72.8 (SD 21.1) kg; body mass index: mean 27.7 (SD 6.6) kg/m2) free from neurologic and vestibular impairments were recruited to serve as an age-matched control group. Body mass index (BMI) was similar between groups. No significant differences were found between groups for any group characteristic described above. Each participant provided informed written consent prior to participation in the balance assessment. The University's Institutional Review Board approved the protocol.
(American Cancer Society, 2017; Siegel et al., 2017). Cancer treatments have been associated with detrimental side effects including pain, fatigue, depression, weakness, peripheral neuropathy, mobility limitations, balance impairments, and falls (Delanian et al., 2012; Silver and Gilchrist, 2011; Winters-Stone et al., 2011). Many neurotoxic chemotherapy drugs produce side-effects such as peripheral neuropathy (Wilkes, 2007), and vestibular dysfunction which may lead to decreased postural stability (Silver and Gilchrist, 2011). In particular, chemotherapy agents often result in axonal degeneration which may cause issues with both the sensory and motor neurons required to continuously maintain postural stability (Visovsky, 2003). Several investigations have identified significantly higher fall rates among community-dwelling cancer survivors (Chen and Janke, 2014; Spoelstra et al., 2013; Wildes et al., 2015; Winters-Stone et al., 2011) compared to the average fall rate for older adults. While the annual fall rate for adults 65 years and older is ~30%, several studies have documented 56–58% of cancer survivors have fallen at least once within the past 12 months (Hornbrook et al., 1994; Huang et al., 2015; Spoelstra et al., 2013; Winters-Stone et al., 2011). The combination of peripheral neuropathy and other treatment-associated side effects is likely related to the increased fall risk often reported for cancer survivors. Current research in this area (Cianfrocca et al., 2006; Silver and Gilchrist, 2011; Tofthagen, 2010; Wilkes, 2007; Winters-Stone et al., 2011) tends to be limited to descriptions of balance deficits and dysfunction in cancer survivors without quantitative assessments to support these descriptions. One investigation by Wampler et al. (2007) assessed balance in breast cancer survivors using center of pressure (CoP) velocities and a pair of clinical tests (i.e., Timed Up and Go and the Fullerton Advanced Balance Scale). Their findings indicated decreased postural control in cancer survivors compared with healthy controls for both CoP velocity and the clinical measures. Although Wampler et al. found differences in both clinical and quantitative techniques, simple clinical tests, such as single limb stance or assessments that use pass/fail criteria, may not be sensitive enough to accurately detect differences in postural steadiness between healthy controls and cancer survivors (Balasubramanian, 2015; Battaglini et al., 2011; Pardasaney et al., 2012). While there are a variety of tests to measure balance, from simple clinical tests to complex techniques (Faraldo-Garcia et al., 2012; Wampler et al., 2007), simple clinical tests have obvious limitations in measurement sensitivity, thus more sophisticated analysis techniques should be relied upon to characterize postural control when available. While it is redundant to use multiple highly-related measures to typify postural steadiness, it may also be necessary to use a small number of measures in order to adequately describe balance performance (Prieto et al., 1996). Quantitative assessments of postural steadiness in cancer survivors need to be performed to better understand the underlying causes of postural unsteadiness in this population. The purpose of this study was to quantify differences in postural stability between healthy age-matched controls and cancer survivors during quiet standing. Moreover, this study attempts to quantify these differences under modified visual and surface conditions using measures of CoP trajectory. Our focus was on CoP based measures that were previously shown to be sensitive to changes in vision conditions and indicators of balance deficits, including fall risk, in older adults without cancer (Kurz et al., 2013; Prieto et al., 1996). Finally, this study sought to identify a subset of the investigated CoP based measures that best characterize postural stability in cancer survivors.
2.2. Data acquisition Participants completed four quiet standing tests without shoes in the following order: 1) rigid surface with eyes open (RSEO), 2) rigid surface with eyes closed (RSEC), 3) compliant surface with eyes open (CSEO), and 4) compliant surface with eyes closed (CSEC). A single trial for each condition was completed in the standardized order. No familiarization trials were performed. Prior to assessments, feet were placed at a selfselected width, symmetrical about the mid-line with medial malleoli aligned with markings on the plate running from medial-to-lateral. The positions of the feet were measured to maintain position when switching to the compliant surface. The compliant surface consisted of a 3 in medium density foam block used to alter the proprioceptive feedback from the feet. The compliant surface also served to disrupt postural control by decreasing the effectiveness of adjustments made at the foot and ankle to maintain static balance (Patel et al., 2008). During assessments, participants focused on a visual target located at eye-level 1 m away. For conditions with the eyes closed, participants began by staring at the visual target and closed their eyes on a verbal cue. A harness was used during all testing conditions for safety. Force and CoP data (1000 Hz) were collected using Bertec's BalanceCheck™ system (Bertec Corp, Columbus, OH, USA) for 30 s for each condition.
2.3. Data analysis Data were exported and further processed using custom software (MATLAB r2010a, MathWorks, Lowell, MA, USA). Based on previous literature (Prieto et al., 1996), data were resampled at 100 Hz for analysis. CoP data were filtered using a digital Butterworth fourthorder, zero lag, low pass filter with a cutoff frequency of 5 Hz. Only the middle 28 s of data from each trial were used in calculations to avoid any perturbations caused by initiation and conclusion of the data collection trial. The following time domain measures of CoP motion were computed using algorithms previously defined in the literature (Prieto et al., 1996): root-mean square distance (RMS) of the anterior-posterior (AP), and medial-lateral (ML) CoP; mean velocities of AP and ML CoP; and 95% confidence ellipse (CE) area. In addition, the frequency domain measures computed were: mean AP and ML CoP frequencies, and 95% power frequency of the AP and ML CoP.
2. Methods 2.1. Participants A convenience sample of cancer survivors entering a cancer rehabilitation program designed to help mitigate the detrimental side effects of their cancer treatments was recruited for this study. 2
Clinical Biomechanics 50 (2017) 1–6
A.C. Schmitt et al.
2.4. Statistical analysis Three-factor (group*surface*vision) mixed model multivariate analyses of variance (MANOVA) (α = 0.05) with repeated measures were used to determine the effect of vision and surface on postural steadiness between the two groups. One MANOVA was used to identify significant changes in time domain measures, and the other for frequency measures. If multivariate tests resulted in significant differences, univariate analysis of variance (ANOVAs) were used to identify where differences occurred (P < 0.05). Although multiple tests were performed, which could increase the possibility of a type I error, a Bonferroni adjustment was not made given the exploratory nature of this study. Finally, Pearson product moment correlations were used to determine the relationships between the CoP based measures (time domain and frequency domain independently). Statistical analyses were completed using SPSS 21 (SPSS Inc., IBM, Chicago, Illinois). 3. Results Stabilograms qualitatively illustrate increased dispersion of the CoP trajectory when visual input was removed and when proprioceptive input was altered via the compliant surface (Fig. 1). The increased distribution of the CoP trajectory as conditions were altered was apparent regardless of group. Multivariate tests revealed significant main effects for group (Λ⁎ = 0.81, F(5,62) = 3.01, P = 0.017), and an interaction between the surface and vision conditions (Λ⁎ = 0.74, F(5,62) = 4.32, P = 0.002) for time domain based CoP measures (Fig. 2). Follow-up univariate tests revealed that the cancer survivors displayed increased magnitudes of the ML components of both the RMS (P < 0.001) and mean velocity (P = 0.012) of the CoP, as well as increased 95% confidence ellipse area (P = 0.002) when compared with their age-matched counterparts. The interaction between the surface and vision conditions suggests participants became less stable when moving from the rigid surface to the compliant surface, to a greater extent when visual feedback was also removed. In both groups, the magnitudes of all time-domain measures increased significantly when participants stood on the compliant surface compared with the rigid surface (P < 0.001). Furthermore, all measures increased significantly (P < 0.001), with the exception of the ML RMS (P = 0.169), when participants closed
Fig. 2. Group differences in time domain based center of pressure (CoP) measures during the rigid surface, eyes open (RSEO), rigid surface, eyes closed (RSEC), compliant surface, eyes open (CSEO), and compliant surface, eyes closed (CSEC) conditions for cancer survivors and matched controls. Medial-lateral (ML) root mean square (RMS); top panel, anterior-posterior (AP) RMS; second panel from the top, ML Velocity; third panel from the top, AP Velocity; fourth panel from the top, and 95% confidence ellipse area; bottom panel. The significant main effect of group is denoted (—*—) and the vision by surface interaction is represented (†) (P < 0.05). Vision and surface effects are shown collapsed across conditions. RS = rigid surface, CS = compliant surface, EO = eyes open, EC = eyes closed.
Fig. 1. Stabilograms of the rigid surface, eyes open (RSEO; top, left panel), rigid surface, eyes closed (RSEC; top, right panel), compliant surface, eyes open (CSEO; bottom, left panel), and compliant surface, eyes closed (CSEC; bottom, right panel) conditions. These data are from one cancer survivor that was chosen at random to illustrate the characteristic differences in center of pressure (COP) trajectories that was evident in both groups during the quiet standing assessments. Anterior-posterior (AP) and medial-lateral (ML) COP in millimeters.
their eyes (Fig. 2). For frequency based CoP measures, significant main effects were observed for surface (Λ⁎ = 0.57, F(4,63) = 11.83, P < 0.001) and vision (Λ⁎ = 0.61, F(4,63) = 9.93, P = 0.002), but no interaction was 3
Clinical Biomechanics 50 (2017) 1–6
A.C. Schmitt et al.
Table 1 Correlations between time-domain based center of pressure measures in the anteriorposterior (AP) and medial-lateral (ML) directions for the four experimental conditions; rigid surface eyes open, rigid surface eyes closed, compliant surface eyes open, and compliant surface eyes closed. Bolded values indicate strongest and weakest correlations between measures within each condition. Measure
Root mean square AP
Rigid surface eyes open Root mean square 0.537 ML Mean velocity AP 0.418 Mean velocity ML 0.536 95% confidence 0.819 ellipse area Rigid surface eyes closed Root mean square 0.627 ML Mean velocity AP 0.622 Mean velocity ML 0.68 95% confidence 0.813 ellipse area Compliant surface eyes open Root mean square 0.493 ML Mean velocity AP 0.24 Mean velocity ML 0.267 95% confidence 0.809 ellipse area Compliant surface eyes closed Root mean square 0.382 ML Mean velocity AP 0.545 Mean velocity ML 0.392 95% confidence 0.766 ellipse area
Root mean square ML
Mean velocity AP
Mean velocity ML
– 0.776 0.408
– 0.587
– 0.869 0.623
– 0.796
– 0.715 0.355
– 0.482
– 0.596 0.533
– 0.707
– 0.36 0.522 0.871
– 0.508 0.703 0.916
– 0.43 0.63 0.848
– 0.317 0.734 0.837
Table 2 Correlations between frequency-domain based center of pressure measures in the anterior-posterior (AP) and medial-lateral (ML) directions for the four experimental conditions; rigid surface eyes open, rigid surface eyes closed, compliant surface eyes open, and compliant surface eyes closed. Bolded values indicate strongest and weakest correlations between measures within each condition. Measure Fig. 3. Frequency domain based center of pressure (CoP) measures during the rigid surface, eyes open (RSEO), rigid surface, eyes closed (RSEC), compliant surface, eyes open (CSEO), and compliant surface, eyes closed (CSEC) conditions for cancer survivors and matched controls. Medial-lateral (ML) Frequency; top panel, anterior-posterior (AP) Frequency; second panel from the top, ML 95% Power; third panel from the top, and AP 95% Power; bottom panel. * Main effects for vision and surface are shown collapsed across conditions (P < 0.05). No main effects for group were observed. RS = rigid surface, CS = compliant surface, EO = eyes open, EC = eyes closed.
Mean frequency AP
Rigid surface eyes open Mean frequency ML 0.209 95 power frequency 0.952 AP 95 power frequency 0.236 ML Rigid surface eyes closed Mean frequency ML 0.303 95 power frequency 0.947 AP 95 power frequency 0.358 ML
detected (Λ⁎ = 0.92, F(4,63) = 1.63, P = 0.27) (Fig. 3). Interestingly, there was no group effect detected in the frequency based CoP measures (Λ⁎ = 0.87, F(4,63) = 2.42, P = 0.058). All frequency measures increased significantly when participants closed their eyes regardless of surface condition (P < 0.002). Additionally, both the mean frequency and 95% power of the AP CoP increased significantly (P < 0.001) when participants stood on the compliant surface compared with the rigid surface. In contrast, the mean frequency and 95% power of ML CoP decreased significantly (P < 0.001) when participants stood on the compliant surface (Fig. 3). The Pearson product moment correlations indicated several of the time domain based CoP measures were highly correlated within each testing condition (Table 1). In particular, ML RMS and 95% confidence ellipse area demonstrated the greatest relationship across conditions (RSEO: r = 0.87; RSEC: r = 0.92; CESO: r = 0.85; CSEC: r = 0.84). 4
Mean frequency ML
95 power frequency AP
– 0.183
–
0.928
0.231
– 0.223
–
0.926
0.279
Compliant surface eyes Mean frequency ML 95 power frequency AP 95 power frequency ML
open 0.25 0.972
– 0.266
–
0.248
0.929
0.266
Compliant surface eyes Mean frequency ML 95 power frequency AP 95 power frequency ML
closed 0.478 0.95
– 0.452
–
0.476
0.912
0.453
Clinical Biomechanics 50 (2017) 1–6
A.C. Schmitt et al.
Wampler et al. (2007), who suggested women with breast cancer exhibit increased postural deficits when compared with healthy controls. Both Wampler et al. and our study identified differences in CoP velocity between cancer survivors and controls, and although Wampler et al. did not separate ML and AP components of the CoP velocity, their findings clearly indicated decreased postural control in cancer survivors as evidenced by both CoP velocity and the clinical tests employed. Our findings of increased CoP velocity in the ML direction also suggest decreased postural stability in cancer survivors and support the notion that cancer treatments may be detrimental to postural stability. These findings are consistent with previous literature suggesting ML components of postural sway are indicators of decreased stability and related to falls (Lord et al., 1999; Melzer et al., 2010; Piirtola and Era, 2006). Comparing our results with those of Prieto et al. (1996), who focused on comparisons between young and elderly adults, our cancer group appeared more similar to the elderly group and our healthy agematched control appeared more similar to the young adults. However, without vision, postural instability increased to a greater extent in the cancer survivors compared with the elderly group in Prieto et al. For example, RMS and mean velocity of the AP CoP increased more dramatically in cancer survivors compared with Prieto's elderly group (47% vs. 12%; 86% vs. 37%, respectively). The control group in our study had smaller mean velocity of the ML CoP than the young adult group in Prieto et al., (3.1 mm/s vs. 3.8 mm/s with eyes open; 3.9 mm/ s vs. 4.4 mm/s, respectively). Overall, this suggests instability increased more in cancer survivors than their healthy counterparts when reliance on proprioceptive and vestibular inputs was required. Based on these comparisons with elderly adults, cancer survivors (55 years) responded similarly to a group of individuals ~15 years older. Thus, cancer survivors displayed a postural stability profile that reflected that of an older population. One interpretation could be that cancer treatments appear to have effects on postural steadiness similar to aging.
Similarly, the frequency domain measures were also highly correlated within directions across conditions (r > 0.9, Table 2). 4. Discussion The aim of this study was to assess differences in CoP based measures of postural steadiness between matched controls and cancer survivors. Varied testing conditions were used to investigate the effects of removing visual input and decreasing proprioceptive input on postural control measures during quiet standing. Previous research on balance in cancer survivors has failed to make clinical recommendations to quantitatively assess postural control, although there are both anecdotal and qualitative descriptions that have discussed balance concerns (Cianfrocca et al., 2006; Silver and Gilchrist, 2011; Tofthagen, 2010; Wilkes, 2007; Winters-Stone et al., 2011), few studies have used quantitative measures of balance in their assessments (Wampler et al., 2007). Additionally, this study sought to find a collection of CoP based measures that adequately characterizes postural stability in cancer survivors. 4.1. Time domain measures The results of this study indicate differences exist between cancer survivors and age-matched controls when examining spatiotemporal parameters of the CoP. Specifically, increased CoP velocity in the ML direction, increased RMS in the ML direction, and increased 95% confidence ellipse area in the cancer survivors suggest this group demonstrates decreased stability in the mediolateral direction. In addition to the group differences, the significant surface by vision interaction suggests that all participants exhibit decreased stability as they move from the rigid surface to the compliant surface, and further decreases in stability are seen when visual input is removed. Unsurprisingly, all of the CoP based measures assessed, with the exception of RMS in the ML direction, suggest decreased postural control when reliance on vestibular and proprioceptive inputs was required. When comparing the two groups, it appears that the cancer survivor group is more unstable than the age-matched control group across conditions. The measures of postural stability investigated suggest the cancer survivors are less stable than the control group when removing vision and/or changing to a compliant surface. This difference may be best illustrated using the 95% confidence ellipse area (Fig. 2). When removing visual input, the cancer survivor group had an average increase in CE area of 91.8 mm2 while standing on a rigid surface with eyes open, compared to 68.6 mm2 increase for the age-matched control group. While there may be a host of factors contributing to the differences between the cancer survivors and the control group, the groups were closely matched for age and BMI, suggesting neither of those factors known to be detrimental to balance is likely a contributor. Chemotherapy or radiation-induced peripheral neuropathy (Delanian et al., 2012; Winters-Stone et al., 2011) may be a potential catalyst for this negative impact on postural stability seen in the cancer group. There is no single underlying mechanism responsible for treatmentassociated peripheral neuropathy, but many possible mechanisms have been identified, and may depend on treatment type. Chemotherapy can result in, among other things, inflammation, mitochondrial changes, and oxidative stress, all which may precipitate neural dysfunction and pain (Jaggi and Singh, 2012). Many chemotherapies have also been linked to impaired motor functions, specifically deep tendon reflexes and sensitivity thresholds (Wampler et al., 2007), which likely contribute to postural instability in cancer survivors. Radiation-induced peripheral neuropathy, which is less common, is likely associated with nerve compression caused by radiation-induced fibrosis, peripheral ischemia due to oxidative stress-associated capillary damage, and/or direct damage to peripheral nerves (Delanian et al., 2012). The results of this study are consistent with the general findings of
4.2. Frequency domain measures The absence of a group effect in the frequency based CoP measures may indicate that, although there are spatial differences between the groups, similar postural control strategies are being employed to achieve postural steadiness. Previous investigations into the effects of compliant foam on postural stability have suggested the compliant surface reduces the ability to detect body orientation and implement accurate corrections; moreover, a lack of visual input exacerbated the destabilizing effect of the compliant foam surfaces (Patel et al., 2008; Strang et al., 2011). Frequency based CoP measures increased when participants closed their eyes, suggesting the removal of visual information resulted in an increase in direction changes of the CoP. This likely indicates participants perceive decreased stability and the postural control system compensates by making changes more rapidly than when visual input is available. Interestingly, both the mean frequency and 95% power of the AP CoP increased significantly when participants stood on the compliant surface compared with the rigid surface. In contrast, the mean frequency and 95% power of ML CoP decreased significantly when participants stood on the compliant surface. This difference in the directional components of these measures may indicate a shift in the strategy participants are using from ML corrections to more AP adjustments as the surface condition became more compliant. 4.3. General discussion This study detected differences between groups in both ML RMS and ML mean velocity, which have been associated with increased fall risk (Melzer et al., 2010; Piirtola and Era, 2006). Consistent with previous reports, in the current study, different CoP based measures were sensitive for each group. Prieto et al. (1996) reported differences in vision conditions in young adults were detected with time domain measures, 5
Clinical Biomechanics 50 (2017) 1–6
A.C. Schmitt et al.
cancer.org/content/dam/cancer-org/research/cancer-facts-and-statistics/annualcancer-facts-and-figures/2017/cancer-facts-and-figures-2017.pdf. Balasubramanian, C.K., 2015. The community balance and mobility scale alleviates the ceiling effects observed in the currently used gait and balance assessments for the community-dwelling older adults. J. Geriatr. Phys. Ther. 38, 78–89. Battaglini, C., Tysinger, D., Mercer, V., Groff, D., McMurray, R., 2011. Examining static and dynamic balance in breast cancer survivors using inexpensive and practically sound tests. Med. Sci. Sports Exerc. 43 (5), 10–11. Chen, T.Y., Janke, M.C., 2014. Predictors of falls among community-dwelling older adults with cancer: results from the health and retirement study. Support Care Cancer 22, 479–485. Cianfrocca, M., Flatters, S.J., Bennett, G.J., McNicol, E., Relias, V., Carr, D., Gillis, T.A., 2006. Peripheral neuropathy in a woman with breast cancer. J. Pain 7, 2–10. Delanian, S., Lefaix, J.L., Pradat, P.F., 2012. Radiation-induced neuropathy in cancer survivors. Radiother. Oncol. 105, 273–282. Faraldo-Garcia, A., Santos-Perez, S., Crujeiras-Casais, R., Labella-Caballero, T., SotoVarela, A., 2012. Influence of age and gender in the sensory analysis of balance control. Eur. Arch. Otorhinolaryngol. 269, 673–677. Hageman, P.A., Leibowitz, J.M., Blanke, D., 1995. Age and gender effects on postural control measures. Arch. Phys. Med. Rehabil. 76, 961–965. Hornbrook, M.C., Stevens, V.J., Wingfield, D.J., Hollis, J.F., Greenlick, M.R., Ory, M.G., 1994. Preventing falls among community-dwelling older persons: results from a randomized trial. Gerontologist 34, 16–23. Huang, M.H., Shilling, T., Miller, K.A., Smith, K., LaVictoire, K., 2015. History of falls, gait, balance, and fall risks in older cancer survivors living in the community. Clin. Interv. Aging 10, 1497–1503. Jaggi, A.S., Singh, N., 2012. Mechanisms in cancer-chemotherapeutic drugs-induced peripheral neuropathy. Toxicology 291, 1–9. Kurz, I., Oddsson, L., Melzer, I., 2013. Characteristics of balance control in older persons who fall with injury—a prospective study. J. Electromyogr. Kinesiol. 23, 814–819. Laughton, C.A., Slavin, M., Katdare, K., Nolan, L., Bean, J.F., Kerrigan, D.C., Phillips, E., Lipsitz, L.A., Collins, J.J., 2003. Aging, muscle activity, and balance control: physiologic changes associated with balance impairment. Gait Posture 18, 101–108. Lord, S.R., Caplan, G.A., Colagiuri, R., Colagiuri, S., Ward, J.A., 1993. Sensori-motor function in older persons with diabetes. Diabet. Med. 10, 614–618. Lord, S.R., Rogers, M.W., Howland, A., Fitzpatrick, R., 1999. Lateral stability, sensorimotor function and falls in older people. J. Am. Geriatr. Soc. 47, 1077–1081. Manchester, D., Woollacott, M., Zederbauer-Hylton, N., Marin, O., 1989. Visual, vestibular and somatosensory contributions to balance control in the older adult. J. Gerontol. 44, M118–127. Melzer, I., Kurz, I., Oddsson, L.I., 2010. A retrospective analysis of balance control parameters in elderly fallers and non-fallers. Clin. Biomech. 25, 984–988. Oppenheim, U., Kohen-Raz, R., Alex, D., Kohen-Raz, A., Azarya, M., 1999. Postural characteristics of diabetic neuropathy. Diabetes Care 22, 328–332. Pardasaney, P.K., Latham, N.K., Jette, A.M., Wagenaar, R.C., Ni, P., Slavin, M.D., Bean, J.F., 2012. Sensitivity to change and responsiveness of four balance measures for community-dwelling older adults. Phys. Ther. 92, 388–397. Patel, M., Fransson, P.A., Lush, D., Gomez, S., 2008. The effect of foam surface properties on postural stability assessment while standing. Gait Posture 28, 649–656. Piirtola, M., Era, P., 2006. Force platform measurements as predictors of falls among older people - a review. Gerontology 52, 1–16. Prieto, T.E., Myklebust, J.B., Hoffmann, R.G., Lovett, E.G., Myklebust, B.M., 1996. Measures of postural steadiness: differences between healthy young and elderly adults. IEEE Trans. Biomed. Eng. 43, 956–966. Riley, P.O., Mann, R.W., Hodge, W.A., 1990. Modelling of the biomechanics of posture and balance. J. Biomech. 23, 503–506. Siegel, R.L., Miller, K.D., Jemal, A., 2017. Cancer statistics, 2017. CA Cancer J. Clin. 67, 7–30. http://dx.doi.org/10.3322/caac.21387. Silver, J.K., Gilchrist, L.S., 2011. Cancer rehabilitation with a focus on evidence-based outpatient physical and occupational therapy interventions. Am. J. Phys. Med. Rehabil. 90, S5–15. Simoneau, G.G., Ulbrecht, J.S., Derr, J.A., Becker, M.B., Cavanagh, P.R., 1994. Postural instability in patients with diabetic sensory neuropathy. Diabetes Care 17, 1411–1421. Slobounov, S.M., Slobounova, E.S., Newell, K.M., 1997. Virtual time-to-collision and human postural control. J. Mot. Behav. 29, 263–281. Spoelstra, S.L., Given, B.A., Schutte, D.L., Sikorskii, A., You, M., Given, C.W., 2013. Do older adults with cancer fall more often? A comparative analysis of falls in those with and without cancer. Oncol. Nurs. Forum 40, E69–78. Strang, A.J., Haworth, J., Hieronymus, M., Walsh, M., Smart Jr., L.J., 2011. Structural changes in postural sway lend insight into effects of balance training, vision, and support surface on postural control in a healthy population. Eur. J. Appl. Physiol. 111, 1485–1495. Tofthagen, C., 2010. Patient perceptions associated with chemotherapy-induced peripheral neuropathy. Clin. J. Oncol. Nurs. 14, E22–28. Visovsky, C., 2003. Chemotherapy-induced peripheral neuropathy. Cancer Investig. 21, 439–451. Wampler, M.A., Topp, K.S., Miaskowski, C., Byl, N.N., Rugo, H.S., Hamel, K., 2007. Quantitative and clinical description of postural instability in women with breast cancer treated with taxane chemotherapy. Arch. Phys. Med. Rehabil. 88, 1002–1008. Wildes, T.M., Dua, P., Fowler, S.A., Miller, J.P., Carpenter, C.R., Avidan, M.S., Stark, S., 2015. Systematic review of falls in older adults with cancer. J. Geriatr. Oncol. 6, 70–83. Wilkes, G., 2007. Peripheral neuropathy related to chemotherapy. Semin. Oncol. Nurs. 23, 162–173. Winters-Stone, K.M., Torgrimson, B., Horak, F., Eisner, A., Nail, L., Leo, M.C., Chui, S., Luoh, S.W., 2011. Identifying factors associated with falls in postmenopausal breast cancer survivors: a multi-disciplinary approach. Arch. Phys. Med. Rehabil. 92, 646–652.
but differences were seen in older adults with frequency based measures. They postulated that this difference may be indicative of an altered postural control strategy to compensate for the removal of visual input, and that this change in strategy may be age dependent, thus typified by different measures. It is possible that the cancer survivor group may be employing a compensation strategy that is being identified by the time domain measures. It is also worth noting that measures investigated in our study were not independent of each other, which is consistent with Prieto et al. (1996). For example, we found the RMS of the CoP and 95% confidence ellipse area were found to be highly related (range r = 0.76 to 0.92), as were mean frequencies to 95% power frequencies (r > 0.90) when comparing ML with ML measures and AP with AP measures. In general, these outcomes suggest two things: 1) fewer CoP measures can be used to assess postural steadiness and reach the same conclusions and 2) ML and AP CoP trajectories should be assessed separately when possible. This information may be beneficial when designing balance interventions for cancer survivors in an attempt to overcome the diminished postural steadiness observed in this study. Additionally, given that these CoP based measures identified a postural difference between cancer survivors and age-matched controls, these same measures can be used throughout rehabilitation to assess an intervention's effectiveness at improving posture and balance in cancer survivors. This study has several limitations. First, the sample of cancer survivors had diverse diagnoses and were in various stages of treatment when they were enrolled in the study. Further, fall history information was not collected for this study, limiting our ability to assess the risk of falling. Future research should incorporate fall history questionnaires and indicators of peripheral neuropathy and vestibular function to further our understanding of underlying mechanisms associated with balance deficits and fall risk in cancer survivors. 5. Conclusions Cancer survivors exhibit decreased postural steadiness when compared with healthy controls. Both time domain and frequency based CoP measures were sensitive to changes in vision and surface for all participants. Given the strong correlations between many of the measures analyzed, a small subset of the CoP metrics used here would effectively track progress of programs designed to improve postural steadiness in this population (1) CoP mean velocity, 2) either RMS of the CoP trajectories or 95% confidence ellipse area, and 3) either mean or 95% power frequency). It appears that time-domain measures may be more useful clinically when attempting to quantify postural deficits within this group compared to a control group. Future research should seek to stratify balance outcomes based on treatment and cancer type, as well as assess balance interventions for clinical efficacy and fall risk. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.clinbiomech.2017.09.010. Acknowledgements We would like to acknowledge Chris Silvernale and the staff at the University of Northern Colorado Cancer Rehabilitation Institute (UNCCRI) for their assistance with this project. UNCCRI is an entity of the College of Natural and Health Sciences at the University of Northern Colorado. UNCCRI's mission is to relieve suffering, promote self-sufficiency, improve quality of life, and eliminate secondary cancers and cancer recurrence for cancer survivors through prescriptive exercise and nutrition evidence-based interventions. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. References American Cancer Society, 2017. In: Atlanta (Ed.), Cancer Facts & Figures, . https://www.
6
Contents lists available at ScienceDirect
Clinical Biomechanics journal homepage: www.elsevier.com/locate/clinbiomech
Comparison of posture and balance in cancer survivors and age-matched controls
MARK
Abigail C. Schmitta,⁎, Chris P. Repkab, Gary D. Heisec, John H. Challisd, Jeremy D. Smithc a
Michael W. Krzyzewski Human Performance Lab, Department of Orthopaedic Surgery, Duke University, 3475 Erwin Rd, Durham, NC 27705, USA Department of Health Sciences, Northern Arizona University, 1100 South Beaver St. #15095, Flagstaff, AZ 86011, USA c School of Sport & Exercise Science, University of Northern Colorado, Gunter Hall 2760, Greeley, CO 80639, USA d Biomechanics Lab, The Pennsylvania State University, 29K Recreation Building, University Park, PA 16802, USA b
A R T I C L E I N F O
A B S T R A C T
Keywords: Cancer Balance Posture Center of pressure Compliant surface Vision
Background: The combination of peripheral neuropathy and other treatment-associated side effects is likely related to an increased incidence of falls in cancer survivors. The purpose of this study was to quantify differences in postural stability between healthy age-matched controls and cancer survivors. Methods: Quiet standing under four conditions (eyes open/closed, rigid/compliant surface) was assessed in 34 cancer survivors (2 males, 32 females; age: 54(13) yrs., height: 1.62(0.07) m; mass: 78.5(19.5) kg) and 34 agematched controls (5 males, 29 females; age: 54(15) yrs.; height: 1.62(0.08) m; mass: 72.8(21.1) kg). Center of pressure data were collected for 30 s and the trajectories were analyzed (100 Hz). Three-factor (group*surface*vision) mixed model MANOVAs with repeated measures were used to determine the effect of vision and surface on postural steadiness between groups. Findings: Cancer survivors exhibited larger mediolateral root-mean square distance and velocity of the center of pressure, as well as increased 95% confidence ellipse area (P < 0.01) when compared with their age-matched counterparts. For example, when removing visual input, cancer survivors had an average increase in 95% confidence ellipse area of 91.8 mm2 while standing on a rigid surface compared to a 68.6 mm2 increase for the control group. No frequency-based center of pressure measures differed between groups. Interpretation: Cancer survivors exhibit decreased postural steadiness when compared with age-matched controls. For cancer survivors undergoing rehabilitation focused on existing balance deficits, a small subset of the center of pressure measures presented here can be used to track progress throughout the intervention and potentially mitigate fall risk.
1. Introduction Understanding postural steadiness as purposeful movements to maintain equilibrium is essential to understanding human movement (Riley et al., 1990), particularly for clinical populations who exhibit postural deficits due to disease and treatments. Researchers have attempted to elucidate the underlying mechanisms of age related postural unsteadiness by comparing elderly individuals and healthy young adults (Hageman et al., 1995; Laughton et al., 2003; Manchester et al., 1989; Prieto et al., 1996). Prieto et al. (1996) showed that removal of visual input during quiet standing resulted in greater instability in elderly individuals, suggesting elderly adults (66–70 years) were less able to control their balance compared with young adults (21–35 years). Thus, when elderly adults relied on proprioceptive and vestibular input they were less able to control their balance (Prieto et al., 1996).
⁎
Postural stability in patients with diabetic neuropathy has also received considerable attention in the literature. Patients with moderate to severe peripheral diabetic neuropathy demonstrate less stability than patients without diabetic neuropathy or control groups suggesting that neuropathy, rather than the disease, is linked to instability (Oppenheim et al., 1999; Simoneau et al., 1994). Given that postural control requires visual, somatosensory and vestibular (Lord et al., 1993; Oppenheim et al., 1999; Prieto et al., 1996; Slobounov et al., 1997) inputs, it is unsurprising that neuropathy is detrimental to postural steadiness. Neuropathy is often found in patients with cancer as well, as a side effect of cancer treatments. In the United States, nearly 1.69 million people were expected to be diagnosed with cancer in 2017 (American Cancer Society, 2017; Siegel et al., 2017). With overall survival rates approaching 70%, there are approximately 15.5 million cancer survivors in the United States
Corresponding author at: 1864 Stadium Road, University of Florida, Gainesville, FL 32611, USA. E-mail address: a.schmitt@ufl.edu (A.C. Schmitt).
http://dx.doi.org/10.1016/j.clinbiomech.2017.09.010 Received 22 April 2016; Accepted 4 September 2017 0268-0033/ © 2017 Elsevier Ltd. All rights reserved.
Clinical Biomechanics 50 (2017) 1–6
A.C. Schmitt et al.
Participants included 34 cancer survivors (2 males, 32 females; age: mean 54 (SD 13) yrs., height: mean 1.62 (SD 0.07) m; mass: mean 77.5 (SD 19.5) kg; body mass index: mean 29.9 (SD 7.3) kg/m2) in various stages of treatment and recovery. The participants in the study were selected consecutively over the course of 1 year from a larger patient population, provided they were able to safely stand unassisted for longer than 1 min. The group included breast (n = 20), colon (n = 4), skin (n = 1), brain (n = 1), ovarian (n = 1), prostate (n = 1), nonHodgkin's lymphoma (n = 2), multiple myeloma (n = 1), leukemia (n = 1), lung (n = 1), and kidney/liver (n = 1) cancers. Overall, 94% of the subjects underwent surgery, 32% had radiation treatment, and 71% had chemotherapy treatment. At the time of assessment, 21% of patients were still undergoing radiation or chemotherapy treatment. A table provided as supplemental material provides more subject specific information on the group of cancer survivors (Supplementary Table 1). Thirty-four healthy adults (5 males, 29 females; age: mean 54 (SD 15) yrs.; height: mean 1.62 (SD 0.08) m; mass: mean 72.8 (SD 21.1) kg; body mass index: mean 27.7 (SD 6.6) kg/m2) free from neurologic and vestibular impairments were recruited to serve as an age-matched control group. Body mass index (BMI) was similar between groups. No significant differences were found between groups for any group characteristic described above. Each participant provided informed written consent prior to participation in the balance assessment. The University's Institutional Review Board approved the protocol.
(American Cancer Society, 2017; Siegel et al., 2017). Cancer treatments have been associated with detrimental side effects including pain, fatigue, depression, weakness, peripheral neuropathy, mobility limitations, balance impairments, and falls (Delanian et al., 2012; Silver and Gilchrist, 2011; Winters-Stone et al., 2011). Many neurotoxic chemotherapy drugs produce side-effects such as peripheral neuropathy (Wilkes, 2007), and vestibular dysfunction which may lead to decreased postural stability (Silver and Gilchrist, 2011). In particular, chemotherapy agents often result in axonal degeneration which may cause issues with both the sensory and motor neurons required to continuously maintain postural stability (Visovsky, 2003). Several investigations have identified significantly higher fall rates among community-dwelling cancer survivors (Chen and Janke, 2014; Spoelstra et al., 2013; Wildes et al., 2015; Winters-Stone et al., 2011) compared to the average fall rate for older adults. While the annual fall rate for adults 65 years and older is ~30%, several studies have documented 56–58% of cancer survivors have fallen at least once within the past 12 months (Hornbrook et al., 1994; Huang et al., 2015; Spoelstra et al., 2013; Winters-Stone et al., 2011). The combination of peripheral neuropathy and other treatment-associated side effects is likely related to the increased fall risk often reported for cancer survivors. Current research in this area (Cianfrocca et al., 2006; Silver and Gilchrist, 2011; Tofthagen, 2010; Wilkes, 2007; Winters-Stone et al., 2011) tends to be limited to descriptions of balance deficits and dysfunction in cancer survivors without quantitative assessments to support these descriptions. One investigation by Wampler et al. (2007) assessed balance in breast cancer survivors using center of pressure (CoP) velocities and a pair of clinical tests (i.e., Timed Up and Go and the Fullerton Advanced Balance Scale). Their findings indicated decreased postural control in cancer survivors compared with healthy controls for both CoP velocity and the clinical measures. Although Wampler et al. found differences in both clinical and quantitative techniques, simple clinical tests, such as single limb stance or assessments that use pass/fail criteria, may not be sensitive enough to accurately detect differences in postural steadiness between healthy controls and cancer survivors (Balasubramanian, 2015; Battaglini et al., 2011; Pardasaney et al., 2012). While there are a variety of tests to measure balance, from simple clinical tests to complex techniques (Faraldo-Garcia et al., 2012; Wampler et al., 2007), simple clinical tests have obvious limitations in measurement sensitivity, thus more sophisticated analysis techniques should be relied upon to characterize postural control when available. While it is redundant to use multiple highly-related measures to typify postural steadiness, it may also be necessary to use a small number of measures in order to adequately describe balance performance (Prieto et al., 1996). Quantitative assessments of postural steadiness in cancer survivors need to be performed to better understand the underlying causes of postural unsteadiness in this population. The purpose of this study was to quantify differences in postural stability between healthy age-matched controls and cancer survivors during quiet standing. Moreover, this study attempts to quantify these differences under modified visual and surface conditions using measures of CoP trajectory. Our focus was on CoP based measures that were previously shown to be sensitive to changes in vision conditions and indicators of balance deficits, including fall risk, in older adults without cancer (Kurz et al., 2013; Prieto et al., 1996). Finally, this study sought to identify a subset of the investigated CoP based measures that best characterize postural stability in cancer survivors.
2.2. Data acquisition Participants completed four quiet standing tests without shoes in the following order: 1) rigid surface with eyes open (RSEO), 2) rigid surface with eyes closed (RSEC), 3) compliant surface with eyes open (CSEO), and 4) compliant surface with eyes closed (CSEC). A single trial for each condition was completed in the standardized order. No familiarization trials were performed. Prior to assessments, feet were placed at a selfselected width, symmetrical about the mid-line with medial malleoli aligned with markings on the plate running from medial-to-lateral. The positions of the feet were measured to maintain position when switching to the compliant surface. The compliant surface consisted of a 3 in medium density foam block used to alter the proprioceptive feedback from the feet. The compliant surface also served to disrupt postural control by decreasing the effectiveness of adjustments made at the foot and ankle to maintain static balance (Patel et al., 2008). During assessments, participants focused on a visual target located at eye-level 1 m away. For conditions with the eyes closed, participants began by staring at the visual target and closed their eyes on a verbal cue. A harness was used during all testing conditions for safety. Force and CoP data (1000 Hz) were collected using Bertec's BalanceCheck™ system (Bertec Corp, Columbus, OH, USA) for 30 s for each condition.
2.3. Data analysis Data were exported and further processed using custom software (MATLAB r2010a, MathWorks, Lowell, MA, USA). Based on previous literature (Prieto et al., 1996), data were resampled at 100 Hz for analysis. CoP data were filtered using a digital Butterworth fourthorder, zero lag, low pass filter with a cutoff frequency of 5 Hz. Only the middle 28 s of data from each trial were used in calculations to avoid any perturbations caused by initiation and conclusion of the data collection trial. The following time domain measures of CoP motion were computed using algorithms previously defined in the literature (Prieto et al., 1996): root-mean square distance (RMS) of the anterior-posterior (AP), and medial-lateral (ML) CoP; mean velocities of AP and ML CoP; and 95% confidence ellipse (CE) area. In addition, the frequency domain measures computed were: mean AP and ML CoP frequencies, and 95% power frequency of the AP and ML CoP.
2. Methods 2.1. Participants A convenience sample of cancer survivors entering a cancer rehabilitation program designed to help mitigate the detrimental side effects of their cancer treatments was recruited for this study. 2
Clinical Biomechanics 50 (2017) 1–6
A.C. Schmitt et al.
2.4. Statistical analysis Three-factor (group*surface*vision) mixed model multivariate analyses of variance (MANOVA) (α = 0.05) with repeated measures were used to determine the effect of vision and surface on postural steadiness between the two groups. One MANOVA was used to identify significant changes in time domain measures, and the other for frequency measures. If multivariate tests resulted in significant differences, univariate analysis of variance (ANOVAs) were used to identify where differences occurred (P < 0.05). Although multiple tests were performed, which could increase the possibility of a type I error, a Bonferroni adjustment was not made given the exploratory nature of this study. Finally, Pearson product moment correlations were used to determine the relationships between the CoP based measures (time domain and frequency domain independently). Statistical analyses were completed using SPSS 21 (SPSS Inc., IBM, Chicago, Illinois). 3. Results Stabilograms qualitatively illustrate increased dispersion of the CoP trajectory when visual input was removed and when proprioceptive input was altered via the compliant surface (Fig. 1). The increased distribution of the CoP trajectory as conditions were altered was apparent regardless of group. Multivariate tests revealed significant main effects for group (Λ⁎ = 0.81, F(5,62) = 3.01, P = 0.017), and an interaction between the surface and vision conditions (Λ⁎ = 0.74, F(5,62) = 4.32, P = 0.002) for time domain based CoP measures (Fig. 2). Follow-up univariate tests revealed that the cancer survivors displayed increased magnitudes of the ML components of both the RMS (P < 0.001) and mean velocity (P = 0.012) of the CoP, as well as increased 95% confidence ellipse area (P = 0.002) when compared with their age-matched counterparts. The interaction between the surface and vision conditions suggests participants became less stable when moving from the rigid surface to the compliant surface, to a greater extent when visual feedback was also removed. In both groups, the magnitudes of all time-domain measures increased significantly when participants stood on the compliant surface compared with the rigid surface (P < 0.001). Furthermore, all measures increased significantly (P < 0.001), with the exception of the ML RMS (P = 0.169), when participants closed
Fig. 2. Group differences in time domain based center of pressure (CoP) measures during the rigid surface, eyes open (RSEO), rigid surface, eyes closed (RSEC), compliant surface, eyes open (CSEO), and compliant surface, eyes closed (CSEC) conditions for cancer survivors and matched controls. Medial-lateral (ML) root mean square (RMS); top panel, anterior-posterior (AP) RMS; second panel from the top, ML Velocity; third panel from the top, AP Velocity; fourth panel from the top, and 95% confidence ellipse area; bottom panel. The significant main effect of group is denoted (—*—) and the vision by surface interaction is represented (†) (P < 0.05). Vision and surface effects are shown collapsed across conditions. RS = rigid surface, CS = compliant surface, EO = eyes open, EC = eyes closed.
Fig. 1. Stabilograms of the rigid surface, eyes open (RSEO; top, left panel), rigid surface, eyes closed (RSEC; top, right panel), compliant surface, eyes open (CSEO; bottom, left panel), and compliant surface, eyes closed (CSEC; bottom, right panel) conditions. These data are from one cancer survivor that was chosen at random to illustrate the characteristic differences in center of pressure (COP) trajectories that was evident in both groups during the quiet standing assessments. Anterior-posterior (AP) and medial-lateral (ML) COP in millimeters.
their eyes (Fig. 2). For frequency based CoP measures, significant main effects were observed for surface (Λ⁎ = 0.57, F(4,63) = 11.83, P < 0.001) and vision (Λ⁎ = 0.61, F(4,63) = 9.93, P = 0.002), but no interaction was 3
Clinical Biomechanics 50 (2017) 1–6
A.C. Schmitt et al.
Table 1 Correlations between time-domain based center of pressure measures in the anteriorposterior (AP) and medial-lateral (ML) directions for the four experimental conditions; rigid surface eyes open, rigid surface eyes closed, compliant surface eyes open, and compliant surface eyes closed. Bolded values indicate strongest and weakest correlations between measures within each condition. Measure
Root mean square AP
Rigid surface eyes open Root mean square 0.537 ML Mean velocity AP 0.418 Mean velocity ML 0.536 95% confidence 0.819 ellipse area Rigid surface eyes closed Root mean square 0.627 ML Mean velocity AP 0.622 Mean velocity ML 0.68 95% confidence 0.813 ellipse area Compliant surface eyes open Root mean square 0.493 ML Mean velocity AP 0.24 Mean velocity ML 0.267 95% confidence 0.809 ellipse area Compliant surface eyes closed Root mean square 0.382 ML Mean velocity AP 0.545 Mean velocity ML 0.392 95% confidence 0.766 ellipse area
Root mean square ML
Mean velocity AP
Mean velocity ML
– 0.776 0.408
– 0.587
– 0.869 0.623
– 0.796
– 0.715 0.355
– 0.482
– 0.596 0.533
– 0.707
– 0.36 0.522 0.871
– 0.508 0.703 0.916
– 0.43 0.63 0.848
– 0.317 0.734 0.837
Table 2 Correlations between frequency-domain based center of pressure measures in the anterior-posterior (AP) and medial-lateral (ML) directions for the four experimental conditions; rigid surface eyes open, rigid surface eyes closed, compliant surface eyes open, and compliant surface eyes closed. Bolded values indicate strongest and weakest correlations between measures within each condition. Measure Fig. 3. Frequency domain based center of pressure (CoP) measures during the rigid surface, eyes open (RSEO), rigid surface, eyes closed (RSEC), compliant surface, eyes open (CSEO), and compliant surface, eyes closed (CSEC) conditions for cancer survivors and matched controls. Medial-lateral (ML) Frequency; top panel, anterior-posterior (AP) Frequency; second panel from the top, ML 95% Power; third panel from the top, and AP 95% Power; bottom panel. * Main effects for vision and surface are shown collapsed across conditions (P < 0.05). No main effects for group were observed. RS = rigid surface, CS = compliant surface, EO = eyes open, EC = eyes closed.
Mean frequency AP
Rigid surface eyes open Mean frequency ML 0.209 95 power frequency 0.952 AP 95 power frequency 0.236 ML Rigid surface eyes closed Mean frequency ML 0.303 95 power frequency 0.947 AP 95 power frequency 0.358 ML
detected (Λ⁎ = 0.92, F(4,63) = 1.63, P = 0.27) (Fig. 3). Interestingly, there was no group effect detected in the frequency based CoP measures (Λ⁎ = 0.87, F(4,63) = 2.42, P = 0.058). All frequency measures increased significantly when participants closed their eyes regardless of surface condition (P < 0.002). Additionally, both the mean frequency and 95% power of the AP CoP increased significantly (P < 0.001) when participants stood on the compliant surface compared with the rigid surface. In contrast, the mean frequency and 95% power of ML CoP decreased significantly (P < 0.001) when participants stood on the compliant surface (Fig. 3). The Pearson product moment correlations indicated several of the time domain based CoP measures were highly correlated within each testing condition (Table 1). In particular, ML RMS and 95% confidence ellipse area demonstrated the greatest relationship across conditions (RSEO: r = 0.87; RSEC: r = 0.92; CESO: r = 0.85; CSEC: r = 0.84). 4
Mean frequency ML
95 power frequency AP
– 0.183
–
0.928
0.231
– 0.223
–
0.926
0.279
Compliant surface eyes Mean frequency ML 95 power frequency AP 95 power frequency ML
open 0.25 0.972
– 0.266
–
0.248
0.929
0.266
Compliant surface eyes Mean frequency ML 95 power frequency AP 95 power frequency ML
closed 0.478 0.95
– 0.452
–
0.476
0.912
0.453
Clinical Biomechanics 50 (2017) 1–6
A.C. Schmitt et al.
Wampler et al. (2007), who suggested women with breast cancer exhibit increased postural deficits when compared with healthy controls. Both Wampler et al. and our study identified differences in CoP velocity between cancer survivors and controls, and although Wampler et al. did not separate ML and AP components of the CoP velocity, their findings clearly indicated decreased postural control in cancer survivors as evidenced by both CoP velocity and the clinical tests employed. Our findings of increased CoP velocity in the ML direction also suggest decreased postural stability in cancer survivors and support the notion that cancer treatments may be detrimental to postural stability. These findings are consistent with previous literature suggesting ML components of postural sway are indicators of decreased stability and related to falls (Lord et al., 1999; Melzer et al., 2010; Piirtola and Era, 2006). Comparing our results with those of Prieto et al. (1996), who focused on comparisons between young and elderly adults, our cancer group appeared more similar to the elderly group and our healthy agematched control appeared more similar to the young adults. However, without vision, postural instability increased to a greater extent in the cancer survivors compared with the elderly group in Prieto et al. For example, RMS and mean velocity of the AP CoP increased more dramatically in cancer survivors compared with Prieto's elderly group (47% vs. 12%; 86% vs. 37%, respectively). The control group in our study had smaller mean velocity of the ML CoP than the young adult group in Prieto et al., (3.1 mm/s vs. 3.8 mm/s with eyes open; 3.9 mm/ s vs. 4.4 mm/s, respectively). Overall, this suggests instability increased more in cancer survivors than their healthy counterparts when reliance on proprioceptive and vestibular inputs was required. Based on these comparisons with elderly adults, cancer survivors (55 years) responded similarly to a group of individuals ~15 years older. Thus, cancer survivors displayed a postural stability profile that reflected that of an older population. One interpretation could be that cancer treatments appear to have effects on postural steadiness similar to aging.
Similarly, the frequency domain measures were also highly correlated within directions across conditions (r > 0.9, Table 2). 4. Discussion The aim of this study was to assess differences in CoP based measures of postural steadiness between matched controls and cancer survivors. Varied testing conditions were used to investigate the effects of removing visual input and decreasing proprioceptive input on postural control measures during quiet standing. Previous research on balance in cancer survivors has failed to make clinical recommendations to quantitatively assess postural control, although there are both anecdotal and qualitative descriptions that have discussed balance concerns (Cianfrocca et al., 2006; Silver and Gilchrist, 2011; Tofthagen, 2010; Wilkes, 2007; Winters-Stone et al., 2011), few studies have used quantitative measures of balance in their assessments (Wampler et al., 2007). Additionally, this study sought to find a collection of CoP based measures that adequately characterizes postural stability in cancer survivors. 4.1. Time domain measures The results of this study indicate differences exist between cancer survivors and age-matched controls when examining spatiotemporal parameters of the CoP. Specifically, increased CoP velocity in the ML direction, increased RMS in the ML direction, and increased 95% confidence ellipse area in the cancer survivors suggest this group demonstrates decreased stability in the mediolateral direction. In addition to the group differences, the significant surface by vision interaction suggests that all participants exhibit decreased stability as they move from the rigid surface to the compliant surface, and further decreases in stability are seen when visual input is removed. Unsurprisingly, all of the CoP based measures assessed, with the exception of RMS in the ML direction, suggest decreased postural control when reliance on vestibular and proprioceptive inputs was required. When comparing the two groups, it appears that the cancer survivor group is more unstable than the age-matched control group across conditions. The measures of postural stability investigated suggest the cancer survivors are less stable than the control group when removing vision and/or changing to a compliant surface. This difference may be best illustrated using the 95% confidence ellipse area (Fig. 2). When removing visual input, the cancer survivor group had an average increase in CE area of 91.8 mm2 while standing on a rigid surface with eyes open, compared to 68.6 mm2 increase for the age-matched control group. While there may be a host of factors contributing to the differences between the cancer survivors and the control group, the groups were closely matched for age and BMI, suggesting neither of those factors known to be detrimental to balance is likely a contributor. Chemotherapy or radiation-induced peripheral neuropathy (Delanian et al., 2012; Winters-Stone et al., 2011) may be a potential catalyst for this negative impact on postural stability seen in the cancer group. There is no single underlying mechanism responsible for treatmentassociated peripheral neuropathy, but many possible mechanisms have been identified, and may depend on treatment type. Chemotherapy can result in, among other things, inflammation, mitochondrial changes, and oxidative stress, all which may precipitate neural dysfunction and pain (Jaggi and Singh, 2012). Many chemotherapies have also been linked to impaired motor functions, specifically deep tendon reflexes and sensitivity thresholds (Wampler et al., 2007), which likely contribute to postural instability in cancer survivors. Radiation-induced peripheral neuropathy, which is less common, is likely associated with nerve compression caused by radiation-induced fibrosis, peripheral ischemia due to oxidative stress-associated capillary damage, and/or direct damage to peripheral nerves (Delanian et al., 2012). The results of this study are consistent with the general findings of
4.2. Frequency domain measures The absence of a group effect in the frequency based CoP measures may indicate that, although there are spatial differences between the groups, similar postural control strategies are being employed to achieve postural steadiness. Previous investigations into the effects of compliant foam on postural stability have suggested the compliant surface reduces the ability to detect body orientation and implement accurate corrections; moreover, a lack of visual input exacerbated the destabilizing effect of the compliant foam surfaces (Patel et al., 2008; Strang et al., 2011). Frequency based CoP measures increased when participants closed their eyes, suggesting the removal of visual information resulted in an increase in direction changes of the CoP. This likely indicates participants perceive decreased stability and the postural control system compensates by making changes more rapidly than when visual input is available. Interestingly, both the mean frequency and 95% power of the AP CoP increased significantly when participants stood on the compliant surface compared with the rigid surface. In contrast, the mean frequency and 95% power of ML CoP decreased significantly when participants stood on the compliant surface. This difference in the directional components of these measures may indicate a shift in the strategy participants are using from ML corrections to more AP adjustments as the surface condition became more compliant. 4.3. General discussion This study detected differences between groups in both ML RMS and ML mean velocity, which have been associated with increased fall risk (Melzer et al., 2010; Piirtola and Era, 2006). Consistent with previous reports, in the current study, different CoP based measures were sensitive for each group. Prieto et al. (1996) reported differences in vision conditions in young adults were detected with time domain measures, 5
Clinical Biomechanics 50 (2017) 1–6
A.C. Schmitt et al.
cancer.org/content/dam/cancer-org/research/cancer-facts-and-statistics/annualcancer-facts-and-figures/2017/cancer-facts-and-figures-2017.pdf. Balasubramanian, C.K., 2015. The community balance and mobility scale alleviates the ceiling effects observed in the currently used gait and balance assessments for the community-dwelling older adults. J. Geriatr. Phys. Ther. 38, 78–89. Battaglini, C., Tysinger, D., Mercer, V., Groff, D., McMurray, R., 2011. Examining static and dynamic balance in breast cancer survivors using inexpensive and practically sound tests. Med. Sci. Sports Exerc. 43 (5), 10–11. Chen, T.Y., Janke, M.C., 2014. Predictors of falls among community-dwelling older adults with cancer: results from the health and retirement study. Support Care Cancer 22, 479–485. Cianfrocca, M., Flatters, S.J., Bennett, G.J., McNicol, E., Relias, V., Carr, D., Gillis, T.A., 2006. Peripheral neuropathy in a woman with breast cancer. J. Pain 7, 2–10. Delanian, S., Lefaix, J.L., Pradat, P.F., 2012. Radiation-induced neuropathy in cancer survivors. Radiother. Oncol. 105, 273–282. Faraldo-Garcia, A., Santos-Perez, S., Crujeiras-Casais, R., Labella-Caballero, T., SotoVarela, A., 2012. Influence of age and gender in the sensory analysis of balance control. Eur. Arch. Otorhinolaryngol. 269, 673–677. Hageman, P.A., Leibowitz, J.M., Blanke, D., 1995. Age and gender effects on postural control measures. Arch. Phys. Med. Rehabil. 76, 961–965. Hornbrook, M.C., Stevens, V.J., Wingfield, D.J., Hollis, J.F., Greenlick, M.R., Ory, M.G., 1994. Preventing falls among community-dwelling older persons: results from a randomized trial. Gerontologist 34, 16–23. Huang, M.H., Shilling, T., Miller, K.A., Smith, K., LaVictoire, K., 2015. History of falls, gait, balance, and fall risks in older cancer survivors living in the community. Clin. Interv. Aging 10, 1497–1503. Jaggi, A.S., Singh, N., 2012. Mechanisms in cancer-chemotherapeutic drugs-induced peripheral neuropathy. Toxicology 291, 1–9. Kurz, I., Oddsson, L., Melzer, I., 2013. Characteristics of balance control in older persons who fall with injury—a prospective study. J. Electromyogr. Kinesiol. 23, 814–819. Laughton, C.A., Slavin, M., Katdare, K., Nolan, L., Bean, J.F., Kerrigan, D.C., Phillips, E., Lipsitz, L.A., Collins, J.J., 2003. Aging, muscle activity, and balance control: physiologic changes associated with balance impairment. Gait Posture 18, 101–108. Lord, S.R., Caplan, G.A., Colagiuri, R., Colagiuri, S., Ward, J.A., 1993. Sensori-motor function in older persons with diabetes. Diabet. Med. 10, 614–618. Lord, S.R., Rogers, M.W., Howland, A., Fitzpatrick, R., 1999. Lateral stability, sensorimotor function and falls in older people. J. Am. Geriatr. Soc. 47, 1077–1081. Manchester, D., Woollacott, M., Zederbauer-Hylton, N., Marin, O., 1989. Visual, vestibular and somatosensory contributions to balance control in the older adult. J. Gerontol. 44, M118–127. Melzer, I., Kurz, I., Oddsson, L.I., 2010. A retrospective analysis of balance control parameters in elderly fallers and non-fallers. Clin. Biomech. 25, 984–988. Oppenheim, U., Kohen-Raz, R., Alex, D., Kohen-Raz, A., Azarya, M., 1999. Postural characteristics of diabetic neuropathy. Diabetes Care 22, 328–332. Pardasaney, P.K., Latham, N.K., Jette, A.M., Wagenaar, R.C., Ni, P., Slavin, M.D., Bean, J.F., 2012. Sensitivity to change and responsiveness of four balance measures for community-dwelling older adults. Phys. Ther. 92, 388–397. Patel, M., Fransson, P.A., Lush, D., Gomez, S., 2008. The effect of foam surface properties on postural stability assessment while standing. Gait Posture 28, 649–656. Piirtola, M., Era, P., 2006. Force platform measurements as predictors of falls among older people - a review. Gerontology 52, 1–16. Prieto, T.E., Myklebust, J.B., Hoffmann, R.G., Lovett, E.G., Myklebust, B.M., 1996. Measures of postural steadiness: differences between healthy young and elderly adults. IEEE Trans. Biomed. Eng. 43, 956–966. Riley, P.O., Mann, R.W., Hodge, W.A., 1990. Modelling of the biomechanics of posture and balance. J. Biomech. 23, 503–506. Siegel, R.L., Miller, K.D., Jemal, A., 2017. Cancer statistics, 2017. CA Cancer J. Clin. 67, 7–30. http://dx.doi.org/10.3322/caac.21387. Silver, J.K., Gilchrist, L.S., 2011. Cancer rehabilitation with a focus on evidence-based outpatient physical and occupational therapy interventions. Am. J. Phys. Med. Rehabil. 90, S5–15. Simoneau, G.G., Ulbrecht, J.S., Derr, J.A., Becker, M.B., Cavanagh, P.R., 1994. Postural instability in patients with diabetic sensory neuropathy. Diabetes Care 17, 1411–1421. Slobounov, S.M., Slobounova, E.S., Newell, K.M., 1997. Virtual time-to-collision and human postural control. J. Mot. Behav. 29, 263–281. Spoelstra, S.L., Given, B.A., Schutte, D.L., Sikorskii, A., You, M., Given, C.W., 2013. Do older adults with cancer fall more often? A comparative analysis of falls in those with and without cancer. Oncol. Nurs. Forum 40, E69–78. Strang, A.J., Haworth, J., Hieronymus, M., Walsh, M., Smart Jr., L.J., 2011. Structural changes in postural sway lend insight into effects of balance training, vision, and support surface on postural control in a healthy population. Eur. J. Appl. Physiol. 111, 1485–1495. Tofthagen, C., 2010. Patient perceptions associated with chemotherapy-induced peripheral neuropathy. Clin. J. Oncol. Nurs. 14, E22–28. Visovsky, C., 2003. Chemotherapy-induced peripheral neuropathy. Cancer Investig. 21, 439–451. Wampler, M.A., Topp, K.S., Miaskowski, C., Byl, N.N., Rugo, H.S., Hamel, K., 2007. Quantitative and clinical description of postural instability in women with breast cancer treated with taxane chemotherapy. Arch. Phys. Med. Rehabil. 88, 1002–1008. Wildes, T.M., Dua, P., Fowler, S.A., Miller, J.P., Carpenter, C.R., Avidan, M.S., Stark, S., 2015. Systematic review of falls in older adults with cancer. J. Geriatr. Oncol. 6, 70–83. Wilkes, G., 2007. Peripheral neuropathy related to chemotherapy. Semin. Oncol. Nurs. 23, 162–173. Winters-Stone, K.M., Torgrimson, B., Horak, F., Eisner, A., Nail, L., Leo, M.C., Chui, S., Luoh, S.W., 2011. Identifying factors associated with falls in postmenopausal breast cancer survivors: a multi-disciplinary approach. Arch. Phys. Med. Rehabil. 92, 646–652.
but differences were seen in older adults with frequency based measures. They postulated that this difference may be indicative of an altered postural control strategy to compensate for the removal of visual input, and that this change in strategy may be age dependent, thus typified by different measures. It is possible that the cancer survivor group may be employing a compensation strategy that is being identified by the time domain measures. It is also worth noting that measures investigated in our study were not independent of each other, which is consistent with Prieto et al. (1996). For example, we found the RMS of the CoP and 95% confidence ellipse area were found to be highly related (range r = 0.76 to 0.92), as were mean frequencies to 95% power frequencies (r > 0.90) when comparing ML with ML measures and AP with AP measures. In general, these outcomes suggest two things: 1) fewer CoP measures can be used to assess postural steadiness and reach the same conclusions and 2) ML and AP CoP trajectories should be assessed separately when possible. This information may be beneficial when designing balance interventions for cancer survivors in an attempt to overcome the diminished postural steadiness observed in this study. Additionally, given that these CoP based measures identified a postural difference between cancer survivors and age-matched controls, these same measures can be used throughout rehabilitation to assess an intervention's effectiveness at improving posture and balance in cancer survivors. This study has several limitations. First, the sample of cancer survivors had diverse diagnoses and were in various stages of treatment when they were enrolled in the study. Further, fall history information was not collected for this study, limiting our ability to assess the risk of falling. Future research should incorporate fall history questionnaires and indicators of peripheral neuropathy and vestibular function to further our understanding of underlying mechanisms associated with balance deficits and fall risk in cancer survivors. 5. Conclusions Cancer survivors exhibit decreased postural steadiness when compared with healthy controls. Both time domain and frequency based CoP measures were sensitive to changes in vision and surface for all participants. Given the strong correlations between many of the measures analyzed, a small subset of the CoP metrics used here would effectively track progress of programs designed to improve postural steadiness in this population (1) CoP mean velocity, 2) either RMS of the CoP trajectories or 95% confidence ellipse area, and 3) either mean or 95% power frequency). It appears that time-domain measures may be more useful clinically when attempting to quantify postural deficits within this group compared to a control group. Future research should seek to stratify balance outcomes based on treatment and cancer type, as well as assess balance interventions for clinical efficacy and fall risk. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.clinbiomech.2017.09.010. Acknowledgements We would like to acknowledge Chris Silvernale and the staff at the University of Northern Colorado Cancer Rehabilitation Institute (UNCCRI) for their assistance with this project. UNCCRI is an entity of the College of Natural and Health Sciences at the University of Northern Colorado. UNCCRI's mission is to relieve suffering, promote self-sufficiency, improve quality of life, and eliminate secondary cancers and cancer recurrence for cancer survivors through prescriptive exercise and nutrition evidence-based interventions. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. References American Cancer Society, 2017. In: Atlanta (Ed.), Cancer Facts & Figures, . https://www.
6