- Email: [email protected]
Seat interface pressures of individuals with paraplegia: Influence of dynamic wheelchair locomotion compared with static seated measurements
313
Seat Interface Pressures of Individuals With Paraplegia: Influence of Dynamic Wheelchair Locomotion Compared With Static Seated Measurements Thomas W. Kernozek, PhD, Jeff E. Lewin, MS ABSTRACT. Kernozek TW, Lewin JE. Seat interface pressures of individuals with paraplegia: influence of dynamic wheelchair locomotion compared with static seated measurements. Arch Phys Med Rehabil 1998;79:313-316. Objective: To provide a comparison of the seat interface pressures between static seating and dynamic seating during wheelchair locomotion of individuals with paraplegia. Design: Repeated measures multivariate analysis of variance (MANOVA) comparing two conditions: static seat and dynamic seat interface pressures. Setting: University campus and clinic. Participants: Fifteen participants, each of whom propelled a manual wheelchair for at least 5 hours per week over the previous 6 months and functioned with a spinal cord injury/ disability level of T1 or below. Main Outcome Measures: Peak pressure (PP) and pressure time integral (PTI) as measured by the Novel Pliance System TM, which consists of a flexible 32 × 32 capacitive sensor mat (each sensor 1.5cm 2) interfaced with a PC, was sampled at 10Hz. The participants were measured in their own wheelchair with a new Jay Active seat cushion. Results: The repeated measures MANOVA showed a difference in the PP and PTI between the static and dynamic measurements (Wilk's = .00, p < .05). Follow-up dependent t tests yielded a difference in PP (t = 5.40, p < 0.025) and no difference in the PTI between static and dynamic conditions (t = 1.45, p > 0.025). The PP during static seating (mean = 16.2 -+ 5.0kPa [121 + 37.5mmHgD was less than during dynamic seat interface pressures during wheelchair locomotion (20.03 _+ 6.6kPa [152.3 + 49.5mmHg]). PP varied by up to 42% during the wheelchair locomotion cycle. The PTI was similar between static (30.1 + 9.3kPa [225.75 + 69mmHg]) and dynamic conditions (36.2 + 18.1kPa [271 + 135.7mmHg]). Conclusions: The results from this study are consistent with some of the previous work on the nondisabled and a single case study, but with greater external validity because of the nature of the sample chosen and the methodology employed. PPs were greater during dynamic wheelchair locomotion compared with static seating interface pressures, with the peak varying up to 42% during the wheelchair locomotion cycle. The PTI indicates that the cumulative effect of the loading was comparable between conditions. The question that remains is whether this
From the Physical Therapy Department, University of Wisconsin-LaCrosse, LaCrosse, WI (Dr. Kernozek), and Division of Kinesiology, University of Minnesota, Minneapolis, MN (Mr. Lewirl). Submitted for publication April 22, 1997. Accepted in revised form August 18, 1997. No commercial party having a direct financial interest in the results of the research supporting this article as or will confer a benefit upon the authors or upon any organization with which the authors are associated. Reprints are not available. © 1998 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation 0003-9993/98/7903-447753.00/0
dynamic loading, resulting in a change in PP throughout the cycle, has a significant effect on tissue health.
© 1998 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation IEAT INTERFACE PRESSURE assessment is of interest to ~both researchers and clinicians because 25% to 85% of all persons with spinal cord injury (SCI) develop pressure ulcers that result in more than 2.3 million Medicare hospital days (1987) and approximately $I.1 billion in direct insurance costs. ~-3 Complications associated with pressure ulcers are attributed to between 4% and 8% of deaths. 4,5 While both intrinsic and extrinsic risk factors are associated with the etiology of pressure ulcers, pressure is thought to be one of the key extrinsic factors in the development of sores. 6 Some of the other risk factors include shear forces, skin temperature, the nutritional status of the patient, existence of moisture, body build and anatomic structure, and local perfusion. Skin breakdown usually occurs in the areas of the ischial tuberosities because of the concentration of load due to bony structure and the amount of time loaded. 7,8 Currently, seat interface pressure distributions are meas u r e d - i f they are measured at all--statically in a clinical or research environment with the patient in a fixed position on various cushions or in various body positions. 912 Measuring static seating pressures provides a baseline, but it is likely that the seat interface is loaded differently throughout the day with different activities of daily living (ADL) skills. One of the most common ADL skills used by individuals with paraplegia to move from one place to another is wheelchair locomotion. A dynamic assessment of the seat interface pressures allows the quantification of load throughout the wheelchair locomotion cycle. The literature on dynamic assessment of seat interface pressures is sparse. Few studies have examined reaching tasks, 13,14 the average pressures over time, 15,I6 and the average pressure with various cushions. 17 The last three reports on average pressures collapsed the discrete pressure values over a fixed time period. This caused the time-dependent fluctuations in the values to be lost. Currently, the only published research that assessed dynamic pressures examined two nondisabled participants on a specially designed wheelchair seat during wheelchair locomotion. 18 This was the first study to demonstrate the difference between static and dynamic seat interface pressures, but it lacked external validity because of the participant selection method and because no seat cushion was used as an interface. Kalpen and colleagues 19 also examined a single elite wheelchair athlete on a treadmill at high speed (5 and 7mph) on two cushions. Based on the results of the Eckrich and Patterson 18 study, Eckrich and Patterson claimed that "the effect of these dynamic pressure changes may act in a manner similar to a 'wheelchair pushup' supporting the vascular and lymphatic pumping mechanisms." This claim warrants a further investigation of the dynamic assessment of seat interface
S
Arch Phys Med Rehabil Vo| 79, March 1998
314
SEAT INTERFACE PRESSURES, Kernozek
pressures of individuals with paraplegia during wheelchair locomotion. The purpose of our study was to investigate the peak pressure (PP) and pressure time integral (PTI) during static seating in comparison with the ADL skill of wheelchair locomotion in a group of SCI participants. The hypothesis was that PPs and the peak PTI would be greater during the wheelchair locomotion cycle compared with static measures. METHODS A convenience sample of 15 patients was selected, each of whom had been using a manual wheelchair for at least 5 hours a week for 6 months and was functioning with an SCI/disability of T1 or below. The average time between a subject's injury and testing was 17.5 -+ 10.25 years; the mean body weight of this sample was 77.5 + 22.4kg (170.5 + 49.31bs). Thirteen men and 2 women were involved in the study. The Novel Pliance SystemTM, a which consists of a flexible 32 × 32 capacitive sensor mat (each sensor, 1.5cm 2) interfaced with a PC sampling at 10Hz, was used to measure seat interface pressures. The system is hardwired, requiring that cabling be tethered between the external analog-to-digital converter hardware and PC. This configuration seemed to have little effect on patients' performances. The sensor mat was calibrated by homogenous air pressure throughout the measurement range before data collection. Within a pressure chamber, verification of pressures from the sensor mat was performed between data collection sessions at 15 +- 1.0kPa (112.5 + 7.5mmHg). If measured pressures deviated by more than this criterion, the sensor mat was recalibrated. This sensor technology was chosen over other sensor technologies because of its excellent reliability as shown in the analysis of gait. a°,21 This mat can be stretched 5% of its length and this was thought to minimize hammocking at the seat interface. 19 The participants were measured in their own wheelchair with a new Jay ActiveTM seat cushion that was prescribed for each participant. The order of measurements was randomized (static versus dynamic) with the pressure-sensitive mat placed between the patient and the cushion. To identify a wheeling cycle from the data, a deformable hand switch interfaced with a light-emitting diode (LED) on the chair was placed in the participant's right hand. When the participant made contact with the push rim, the switch closed, illuminating the LED. From the Pliance SystemTM a synch pulse was discharged from the analyzer, illuminating a second LED on the chair. Both LEDs were captured on videotape with a camcorder (30Hz) to determine the onset (right hand on push rim) and termination (right hand on push rim for the second time) of a wheeling cycle and onset of pressure mat sampling. The dependent variables obtained from the sensor mat were PP and PTI. PP was the peak value that occurred from any single sensor during the static and dynamic trials. This PP value occurred in the region of one of the ischial tuberosities in all trials. PTI was calculated by integrating the PP time curve for each trial, enabling a cumulative effect of the PP to be calculated over a specific time period. Three consecutive wheeling cycles from four wheeling trials at a speed of 1.3m/sec (+10%) were averaged and compared with the mean from two static measurements. A photoelectric timing system was used to monitor wheeling speed over a 6-meter interval in the middle of the 15.2m (50ft). Trials that were not within the criterion speed were discarded. Measurement times for the static trials were truncated to the length of time required to complete an average dynamic wheeling cycle for that participant, and both the PP and PTI were calculated and compared. To determine differences between static and dynamic loading variables, a single factor repeated measures multivariate analysis of variance (MANOVA) was used. Follow-up dependent t tests using the Arch Phys Med Rehabil Vol 79, March 1998
Bonferoni alpha were used in the event that a significant multivariate effect was found. RESULTS The results of the repeated measures MANOVA showed a significant difference in the PP and PTI between the static and dynamic wheeling measurements (Wilk's = .00, p < .05). Follow-up dependent t tests yielded a difference in PP between the static and dynamic trials (t = 5.40, p < .025) and no difference in the PTI between static and dynamic trials (t = 1.45,p > .025). Table 1 shows the mean and standard deviation for static and dynamic trials for each of the dependent variables. The PPs during static seating conditions were less than during dynamic seat interface pressures during wheeling. Figure 1 shows the mean and 95% confidence interval from a single wheelchair locomotion trial of three consecutive locomotion cycles. The three trials were interpolated and ensembleaveraged to obtain the resulting curve. Note that PP varies by about 40% from the peak to the minimum PP throughout the wheelchair locomotion cycle. Minimum PP decreased below the PP during static loading in all trials for each subject. However, PP varied throughout the wheeling cycle by 35% to 42% below the static loading conditions. DISCUSSION There is no agreement on a pressure threshold for tissue damage. Some researchers I have indicated that tissue damage can occur with a PP of 6kPa (45mmHg) with friction, and with PPs ranging from 1 to 9kPa (7.5 to 67.5mmHg). a2,23 Both the static and dynamic PPs in our study were greater than the tissue damage thresholds reported in the literature. The differences between static and dynamic seat interface pressures were similar to the results found by Eckrich and Patterson. is As is often the case with pressure measurement research, the pressure values cannot be directly compared with those in the present study because of the different instruments used, with differing sensor technologies and sensor sizes. However, the PPs reported by Eckrich and Patterson is varied by a mean of 23% compared with a mean of 36% found in our study. Their protocol may account for these differences, since they had their two participants start and stop in a distance of 25 feet in 5.6 (_+.35) seconds, and these two participants were both nondisabled. It is likely that nondisabled persons possess different "body build" characteristics than individuals with paraplegia; this would produce different interface pressures. 9 Kalpen I9 reported PPs of 23.2 -- 1.0kPa (174 _+ .75mmHg) at 3.1mph (83.3m/sec) on a latex foam cushion and 20.0kPa (150 _+ .75mmHg) at 7.5mph (200m/sec). Reduced PP was found with an air-filled cushion compared with foam. Static PP was 18kPa (135mmHg) on the air cushion and 25kPa (187.5mmHg) on foam. The maximum PP was approximately 40% greater for the foam cushion than for the air. While Kalpen utilized the same instrumentation as in our study, differences in the PP were much greater comparing static and dynamic trials because of the greater wheelchair locomotion speeds employed. Since PP changes throughout the wheelchair locomotion Table 1: Means and Standard Deviations for Peak Pressure and Pressure lime Integral for Static and Dynamic (Wheeling) Trials (n = 15) Variables
Static
Dynamic
Peak pressure (kPa) Pressure time integral (kPa/sec)
16,2 _+ 5.0 30.1 ± 9.3
20.3 ~ 6.6 ~ 36.2 _+ 18.1
* p < .025.
SEAT INTERFACE PRESSURES, Kernozek
...............................................
i ...............
i ...............
11 o
Peak
315
be determined to have similar effects on tissue health. Similar analyses have surfaced in the literature relative to the diabetic foot, where the repetitive mechanical trauma is thought to be of importance in the etiology of decubiti.26 Much greater pressures, however, are experienced during gait due to the greater ground reaction forces and the much smaller bony prominences associated with the foot.
Prs$sur8
................................ .;l ................ -.............................. {
09.0
(kPa)
; ................ ~ . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . .
,
...............
, ................
O.O%
20.0 %
~-. . . . . . . . . . . . . . .
40.0 % Percentage
, . . . . . . . . . . . . . . . .
~. . . . . . . . . . . . . . .
~.0 %
.
. . . . . . . . . . . . . .
, ...............
Em3.O%
06.0 03.0
iOO.O %
of Cycle
Fig 1, Mean and 95% confidence interval for peak pressure of three consecutive cycles from a single wheeling trial.
CONCLUSIONS Our results are consistent with the results of some previous work on the nondisabled and in a single case, but our study has greater external validity because of the size and nature of the sample chosen and the protocol used. PPs were greater during dynamic wheelchair locomotion compared with static seating, with the peak varying up to 42% during the wheelchair locomotion cycle. The PTI indicated that the cumulative effect of the loading was comparable between the two conditions. The question remains as to how this dynamic loading affects tissue health in comparison with static loading. References
cycle, the question remains whether these interface pressures are as damaging to tissue health as the static loading. To answer, we must first recognize and heed Brand's 24 warning that we do not know enough about the etiology of pressure sores to speculate on the direct causes of tissue breakdown. Much more research is needed on the etiology of sore development and on the development of a threshold for injury. Even during wheelchair locomotion, seat interface pressures do not decrease below the pressure thresholds discussed in the literature. Conversely, does the fluctuation in PP facilitate a sort of "pumping mechanism" as suggested by Eckrich and Patterson? 18 Since there are other therapeutic techniques that use external manipulation of the skin to stimulate blood and lymphatic activity, it seems logical that this fluctuation in PP throughout the wheelchair locomotion cycle could promote this activity. Thus, the cycling loading may be beneficial to the patient rather than a risk factor for tissue damage. Further research needs to examine these factors. A notable fact is that the sensor mat was sampled at 10Hz. The ability to capture the peak may be diminished because of this sampling rate. As a result, the PPs that were measured may have been less than what was actually occurring. Three cycles and four trials were used to obtain these PP measurements. Standard deviations were similar between trials of dynamic wheelchair locomotion measures, providing some merit to the measurement frequency used. Another limitation is that the protocol restricted wheeling to a small speed range. However, research on wheeling kinematics by Veeger and coworkers 25 showed that the relation between trunk flexion and upper arm flexion was related to wheelchair locomotion speed. This speed of 1.39m/sec elicited the kinematic performance found at higher speeds without placing excessive physiologic demands on the pal~icipant. Thus, the values obtained are relative to this speed and the resultant data must be viewed in this context. The amount of impulsive load from the PP was calculated by the PTI. This variable was thought to describe the cumulative effect of the PP at the seat interface. Between the static and dynamic wheelchair locomotion conditions, the impulsive loading at the seat interface was similar, whereas the rate of change and range of PP were quite different. If the rate and rate of change in PP are thought not to be of importance irrespective of time, then both modes of loading (static and dynamic) could
1. Dinsdale SM. Decubitus ulcers in swine: light and electron microscopy study of pathogenesis. Arch Phys Med Rehabil 1974;54:147-52. 2. Staas WE, Cioschi HM. Pressure sores--a multifaceted approach to prevention and treatment. West J Med 1991;154:539-44. 3. Fuhrer MJ, Garber SL, Rintala DH, Clearman R, Hart KA. Pressure ulcers in community-resident persons with spinal cord injury: prevalence and risk factors, Arch Phys Med Rehabil 1993;74:1172-7. 4. Cooney TG, Reuler JB. Pressure sores. In: Topics in primary care medicine. West J Med 1984;154:539-44. 5. Shields RK, Cook TE. Effect of seat angle and lumbar support on seated buttock pressure. Phys Ther 1988;68:1682-6. 6. Mawson AR, Biundo JJ, Neville R Linares HA, Winchester Y, Lopez A. Risk factors for early occurring pressure ulcers following spinal cord injury. Am J Phys Med Rehabil 1988;67:123-7. 7. Kosiak M. Etiology and pathology of ischemJc ulcers. Arch Phys Med Rehabil 1959;40:62-9. 8. Reswick JB, Rogers J. Experience at Rancho Los Amigos Hospital with devices and techniques to prevent pressure sores. In: Kenedi RM, Cowden JM, Scales JT, editors. Bedsore biomechanics. London: Macmillan; 1976. p. 301-10. 9. Garber SL, Krouskop TA. Body build and its relationship to pressure distribution in the seated wheelchair patient. Arch Phys Med Rehabil 1982;63:17-20. 10. Garber SL Wheelchair cushions: a historical review. Am J Occup Ther 1985;39:453-9. 11. Garber SL, Dyerly LR. Wheelchair cushions for persons with spinal cord injury: an update. Am J Occup Ther 1991;45:550-4. 12. Henderson JH, Price SH, Brandstater ME, Mandac BR. Efficacy of three measures to relieve pressure in seated persons with spinal cord injury. Arch Phys Med Rehabil 1994;75:535-9. 13. Park CA. Activity positioning and ischial tuberosity pressure: a pilot study. Am J Occup Ther, 1992;46:904-9. 14. Swarts AE, Krouskop TA, Smith DR. Tissue pressure management in a vocational setting. Arch Phys Med Rehabil 1988;69:97-100. 15. Fisher SV, Patterson RP. Long term pressure recordings under the ischial tuberosities of tetraplegics. Paraplegia 1983;21:99-106. 16. Patterson RP, Fisher SV. Sitting pressure-time patterns in patients with quadriplegia. Arch Phys Med Rehabil 1986;67:812-4. 17. Bar CA. Evaluation of cushions using dynamic pressure measurement [review]. Prosthet Orthot Int 1991;15:232-40. 18. Eckrich KM, Patterson KE. Dynamic interface pressure between seated users and their wheelchairs. Int J Indust Ergon 1991;8:1l 523. Arch Phys Med Rehabil Vol 79, March 1998
316
SEAT INTERFACE PRESSURES, Kernozek
19. Kalpen A, Bochdansky T, Seitz E The dynamic body pressure distribution in a moving wheelchair. Workshop MW04: RESNA Conference, June 7-12, 1996. 20. McPoil T, Cornwall MW, Yamada W. A comparison of two in-shoe measurement systems. Lower Extremity 1995;2:1-9. 21. Kernozek TW, LaMott EE, Dancisak MJ. Reliability of an in-shoe measurement system during treadmill walking. Foot Ankle Int 1996; 17:204-9. 22. Kosiak M. A mechanical resting surface: its effect on pressure distribution. Arch Phys Med Rehabil 1976;57:481-4. 23. Eriksson E. Etiology: microcirculatory effects of pressure. In: Zacharkow D, editor. Posture: sitting, standing, chair design and exercise. Springfield (IL): Charles Thomas; 1988. p. 115 -30.
Arch Phys Med Rehabil Vol 79, March 1998
115-30. 24. Brand PW. Comments on the article "Development of test methods for the evaluation of wheelchair cushions" [editorial]. Bull Prosth Res 1980;10:3-4. 25. Veeger HEJ, Van der Woude LHV, Rozendal RH. Wheelchair propulsion technique at different speeds. Scand J Rehabil Med 1989;21:197-203. 26. Cavanaugh PR, Ulbrecht JS. Biomechanics of the foot in diabetes. In: Levin MD, O'Neal LW, Bowker JH, editors. The diabetic foot. 5th ed. St. Louis (MO): Mosby; 1993. p. 199-232.
Supplier a. Novel Electronics, Inc., 964 Grand Avenue, St. Paul, MN 55105.
Seat Interface Pressures of Individuals With Paraplegia: Influence of Dynamic Wheelchair Locomotion Compared With Static Seated Measurements Thomas W. Kernozek, PhD, Jeff E. Lewin, MS ABSTRACT. Kernozek TW, Lewin JE. Seat interface pressures of individuals with paraplegia: influence of dynamic wheelchair locomotion compared with static seated measurements. Arch Phys Med Rehabil 1998;79:313-316. Objective: To provide a comparison of the seat interface pressures between static seating and dynamic seating during wheelchair locomotion of individuals with paraplegia. Design: Repeated measures multivariate analysis of variance (MANOVA) comparing two conditions: static seat and dynamic seat interface pressures. Setting: University campus and clinic. Participants: Fifteen participants, each of whom propelled a manual wheelchair for at least 5 hours per week over the previous 6 months and functioned with a spinal cord injury/ disability level of T1 or below. Main Outcome Measures: Peak pressure (PP) and pressure time integral (PTI) as measured by the Novel Pliance System TM, which consists of a flexible 32 × 32 capacitive sensor mat (each sensor 1.5cm 2) interfaced with a PC, was sampled at 10Hz. The participants were measured in their own wheelchair with a new Jay Active seat cushion. Results: The repeated measures MANOVA showed a difference in the PP and PTI between the static and dynamic measurements (Wilk's = .00, p < .05). Follow-up dependent t tests yielded a difference in PP (t = 5.40, p < 0.025) and no difference in the PTI between static and dynamic conditions (t = 1.45, p > 0.025). The PP during static seating (mean = 16.2 -+ 5.0kPa [121 + 37.5mmHgD was less than during dynamic seat interface pressures during wheelchair locomotion (20.03 _+ 6.6kPa [152.3 + 49.5mmHg]). PP varied by up to 42% during the wheelchair locomotion cycle. The PTI was similar between static (30.1 + 9.3kPa [225.75 + 69mmHg]) and dynamic conditions (36.2 + 18.1kPa [271 + 135.7mmHg]). Conclusions: The results from this study are consistent with some of the previous work on the nondisabled and a single case study, but with greater external validity because of the nature of the sample chosen and the methodology employed. PPs were greater during dynamic wheelchair locomotion compared with static seating interface pressures, with the peak varying up to 42% during the wheelchair locomotion cycle. The PTI indicates that the cumulative effect of the loading was comparable between conditions. The question that remains is whether this
From the Physical Therapy Department, University of Wisconsin-LaCrosse, LaCrosse, WI (Dr. Kernozek), and Division of Kinesiology, University of Minnesota, Minneapolis, MN (Mr. Lewirl). Submitted for publication April 22, 1997. Accepted in revised form August 18, 1997. No commercial party having a direct financial interest in the results of the research supporting this article as or will confer a benefit upon the authors or upon any organization with which the authors are associated. Reprints are not available. © 1998 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation 0003-9993/98/7903-447753.00/0
dynamic loading, resulting in a change in PP throughout the cycle, has a significant effect on tissue health.
© 1998 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation IEAT INTERFACE PRESSURE assessment is of interest to ~both researchers and clinicians because 25% to 85% of all persons with spinal cord injury (SCI) develop pressure ulcers that result in more than 2.3 million Medicare hospital days (1987) and approximately $I.1 billion in direct insurance costs. ~-3 Complications associated with pressure ulcers are attributed to between 4% and 8% of deaths. 4,5 While both intrinsic and extrinsic risk factors are associated with the etiology of pressure ulcers, pressure is thought to be one of the key extrinsic factors in the development of sores. 6 Some of the other risk factors include shear forces, skin temperature, the nutritional status of the patient, existence of moisture, body build and anatomic structure, and local perfusion. Skin breakdown usually occurs in the areas of the ischial tuberosities because of the concentration of load due to bony structure and the amount of time loaded. 7,8 Currently, seat interface pressure distributions are meas u r e d - i f they are measured at all--statically in a clinical or research environment with the patient in a fixed position on various cushions or in various body positions. 912 Measuring static seating pressures provides a baseline, but it is likely that the seat interface is loaded differently throughout the day with different activities of daily living (ADL) skills. One of the most common ADL skills used by individuals with paraplegia to move from one place to another is wheelchair locomotion. A dynamic assessment of the seat interface pressures allows the quantification of load throughout the wheelchair locomotion cycle. The literature on dynamic assessment of seat interface pressures is sparse. Few studies have examined reaching tasks, 13,14 the average pressures over time, 15,I6 and the average pressure with various cushions. 17 The last three reports on average pressures collapsed the discrete pressure values over a fixed time period. This caused the time-dependent fluctuations in the values to be lost. Currently, the only published research that assessed dynamic pressures examined two nondisabled participants on a specially designed wheelchair seat during wheelchair locomotion. 18 This was the first study to demonstrate the difference between static and dynamic seat interface pressures, but it lacked external validity because of the participant selection method and because no seat cushion was used as an interface. Kalpen and colleagues 19 also examined a single elite wheelchair athlete on a treadmill at high speed (5 and 7mph) on two cushions. Based on the results of the Eckrich and Patterson 18 study, Eckrich and Patterson claimed that "the effect of these dynamic pressure changes may act in a manner similar to a 'wheelchair pushup' supporting the vascular and lymphatic pumping mechanisms." This claim warrants a further investigation of the dynamic assessment of seat interface
S
Arch Phys Med Rehabil Vo| 79, March 1998
314
SEAT INTERFACE PRESSURES, Kernozek
pressures of individuals with paraplegia during wheelchair locomotion. The purpose of our study was to investigate the peak pressure (PP) and pressure time integral (PTI) during static seating in comparison with the ADL skill of wheelchair locomotion in a group of SCI participants. The hypothesis was that PPs and the peak PTI would be greater during the wheelchair locomotion cycle compared with static measures. METHODS A convenience sample of 15 patients was selected, each of whom had been using a manual wheelchair for at least 5 hours a week for 6 months and was functioning with an SCI/disability of T1 or below. The average time between a subject's injury and testing was 17.5 -+ 10.25 years; the mean body weight of this sample was 77.5 + 22.4kg (170.5 + 49.31bs). Thirteen men and 2 women were involved in the study. The Novel Pliance SystemTM, a which consists of a flexible 32 × 32 capacitive sensor mat (each sensor, 1.5cm 2) interfaced with a PC sampling at 10Hz, was used to measure seat interface pressures. The system is hardwired, requiring that cabling be tethered between the external analog-to-digital converter hardware and PC. This configuration seemed to have little effect on patients' performances. The sensor mat was calibrated by homogenous air pressure throughout the measurement range before data collection. Within a pressure chamber, verification of pressures from the sensor mat was performed between data collection sessions at 15 +- 1.0kPa (112.5 + 7.5mmHg). If measured pressures deviated by more than this criterion, the sensor mat was recalibrated. This sensor technology was chosen over other sensor technologies because of its excellent reliability as shown in the analysis of gait. a°,21 This mat can be stretched 5% of its length and this was thought to minimize hammocking at the seat interface. 19 The participants were measured in their own wheelchair with a new Jay ActiveTM seat cushion that was prescribed for each participant. The order of measurements was randomized (static versus dynamic) with the pressure-sensitive mat placed between the patient and the cushion. To identify a wheeling cycle from the data, a deformable hand switch interfaced with a light-emitting diode (LED) on the chair was placed in the participant's right hand. When the participant made contact with the push rim, the switch closed, illuminating the LED. From the Pliance SystemTM a synch pulse was discharged from the analyzer, illuminating a second LED on the chair. Both LEDs were captured on videotape with a camcorder (30Hz) to determine the onset (right hand on push rim) and termination (right hand on push rim for the second time) of a wheeling cycle and onset of pressure mat sampling. The dependent variables obtained from the sensor mat were PP and PTI. PP was the peak value that occurred from any single sensor during the static and dynamic trials. This PP value occurred in the region of one of the ischial tuberosities in all trials. PTI was calculated by integrating the PP time curve for each trial, enabling a cumulative effect of the PP to be calculated over a specific time period. Three consecutive wheeling cycles from four wheeling trials at a speed of 1.3m/sec (+10%) were averaged and compared with the mean from two static measurements. A photoelectric timing system was used to monitor wheeling speed over a 6-meter interval in the middle of the 15.2m (50ft). Trials that were not within the criterion speed were discarded. Measurement times for the static trials were truncated to the length of time required to complete an average dynamic wheeling cycle for that participant, and both the PP and PTI were calculated and compared. To determine differences between static and dynamic loading variables, a single factor repeated measures multivariate analysis of variance (MANOVA) was used. Follow-up dependent t tests using the Arch Phys Med Rehabil Vol 79, March 1998
Bonferoni alpha were used in the event that a significant multivariate effect was found. RESULTS The results of the repeated measures MANOVA showed a significant difference in the PP and PTI between the static and dynamic wheeling measurements (Wilk's = .00, p < .05). Follow-up dependent t tests yielded a difference in PP between the static and dynamic trials (t = 5.40, p < .025) and no difference in the PTI between static and dynamic trials (t = 1.45,p > .025). Table 1 shows the mean and standard deviation for static and dynamic trials for each of the dependent variables. The PPs during static seating conditions were less than during dynamic seat interface pressures during wheeling. Figure 1 shows the mean and 95% confidence interval from a single wheelchair locomotion trial of three consecutive locomotion cycles. The three trials were interpolated and ensembleaveraged to obtain the resulting curve. Note that PP varies by about 40% from the peak to the minimum PP throughout the wheelchair locomotion cycle. Minimum PP decreased below the PP during static loading in all trials for each subject. However, PP varied throughout the wheeling cycle by 35% to 42% below the static loading conditions. DISCUSSION There is no agreement on a pressure threshold for tissue damage. Some researchers I have indicated that tissue damage can occur with a PP of 6kPa (45mmHg) with friction, and with PPs ranging from 1 to 9kPa (7.5 to 67.5mmHg). a2,23 Both the static and dynamic PPs in our study were greater than the tissue damage thresholds reported in the literature. The differences between static and dynamic seat interface pressures were similar to the results found by Eckrich and Patterson. is As is often the case with pressure measurement research, the pressure values cannot be directly compared with those in the present study because of the different instruments used, with differing sensor technologies and sensor sizes. However, the PPs reported by Eckrich and Patterson is varied by a mean of 23% compared with a mean of 36% found in our study. Their protocol may account for these differences, since they had their two participants start and stop in a distance of 25 feet in 5.6 (_+.35) seconds, and these two participants were both nondisabled. It is likely that nondisabled persons possess different "body build" characteristics than individuals with paraplegia; this would produce different interface pressures. 9 Kalpen I9 reported PPs of 23.2 -- 1.0kPa (174 _+ .75mmHg) at 3.1mph (83.3m/sec) on a latex foam cushion and 20.0kPa (150 _+ .75mmHg) at 7.5mph (200m/sec). Reduced PP was found with an air-filled cushion compared with foam. Static PP was 18kPa (135mmHg) on the air cushion and 25kPa (187.5mmHg) on foam. The maximum PP was approximately 40% greater for the foam cushion than for the air. While Kalpen utilized the same instrumentation as in our study, differences in the PP were much greater comparing static and dynamic trials because of the greater wheelchair locomotion speeds employed. Since PP changes throughout the wheelchair locomotion Table 1: Means and Standard Deviations for Peak Pressure and Pressure lime Integral for Static and Dynamic (Wheeling) Trials (n = 15) Variables
Static
Dynamic
Peak pressure (kPa) Pressure time integral (kPa/sec)
16,2 _+ 5.0 30.1 ± 9.3
20.3 ~ 6.6 ~ 36.2 _+ 18.1
* p < .025.
SEAT INTERFACE PRESSURES, Kernozek
...............................................
i ...............
i ...............
11 o
Peak
315
be determined to have similar effects on tissue health. Similar analyses have surfaced in the literature relative to the diabetic foot, where the repetitive mechanical trauma is thought to be of importance in the etiology of decubiti.26 Much greater pressures, however, are experienced during gait due to the greater ground reaction forces and the much smaller bony prominences associated with the foot.
Prs$sur8
................................ .;l ................ -.............................. {
09.0
(kPa)
; ................ ~ . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . .
,
...............
, ................
O.O%
20.0 %
~-. . . . . . . . . . . . . . .
40.0 % Percentage
, . . . . . . . . . . . . . . . .
~. . . . . . . . . . . . . . .
~.0 %
.
. . . . . . . . . . . . . .
, ...............
Em3.O%
06.0 03.0
iOO.O %
of Cycle
Fig 1, Mean and 95% confidence interval for peak pressure of three consecutive cycles from a single wheeling trial.
CONCLUSIONS Our results are consistent with the results of some previous work on the nondisabled and in a single case, but our study has greater external validity because of the size and nature of the sample chosen and the protocol used. PPs were greater during dynamic wheelchair locomotion compared with static seating, with the peak varying up to 42% during the wheelchair locomotion cycle. The PTI indicated that the cumulative effect of the loading was comparable between the two conditions. The question remains as to how this dynamic loading affects tissue health in comparison with static loading. References
cycle, the question remains whether these interface pressures are as damaging to tissue health as the static loading. To answer, we must first recognize and heed Brand's 24 warning that we do not know enough about the etiology of pressure sores to speculate on the direct causes of tissue breakdown. Much more research is needed on the etiology of sore development and on the development of a threshold for injury. Even during wheelchair locomotion, seat interface pressures do not decrease below the pressure thresholds discussed in the literature. Conversely, does the fluctuation in PP facilitate a sort of "pumping mechanism" as suggested by Eckrich and Patterson? 18 Since there are other therapeutic techniques that use external manipulation of the skin to stimulate blood and lymphatic activity, it seems logical that this fluctuation in PP throughout the wheelchair locomotion cycle could promote this activity. Thus, the cycling loading may be beneficial to the patient rather than a risk factor for tissue damage. Further research needs to examine these factors. A notable fact is that the sensor mat was sampled at 10Hz. The ability to capture the peak may be diminished because of this sampling rate. As a result, the PPs that were measured may have been less than what was actually occurring. Three cycles and four trials were used to obtain these PP measurements. Standard deviations were similar between trials of dynamic wheelchair locomotion measures, providing some merit to the measurement frequency used. Another limitation is that the protocol restricted wheeling to a small speed range. However, research on wheeling kinematics by Veeger and coworkers 25 showed that the relation between trunk flexion and upper arm flexion was related to wheelchair locomotion speed. This speed of 1.39m/sec elicited the kinematic performance found at higher speeds without placing excessive physiologic demands on the pal~icipant. Thus, the values obtained are relative to this speed and the resultant data must be viewed in this context. The amount of impulsive load from the PP was calculated by the PTI. This variable was thought to describe the cumulative effect of the PP at the seat interface. Between the static and dynamic wheelchair locomotion conditions, the impulsive loading at the seat interface was similar, whereas the rate of change and range of PP were quite different. If the rate and rate of change in PP are thought not to be of importance irrespective of time, then both modes of loading (static and dynamic) could
1. Dinsdale SM. Decubitus ulcers in swine: light and electron microscopy study of pathogenesis. Arch Phys Med Rehabil 1974;54:147-52. 2. Staas WE, Cioschi HM. Pressure sores--a multifaceted approach to prevention and treatment. West J Med 1991;154:539-44. 3. Fuhrer MJ, Garber SL, Rintala DH, Clearman R, Hart KA. Pressure ulcers in community-resident persons with spinal cord injury: prevalence and risk factors, Arch Phys Med Rehabil 1993;74:1172-7. 4. Cooney TG, Reuler JB. Pressure sores. In: Topics in primary care medicine. West J Med 1984;154:539-44. 5. Shields RK, Cook TE. Effect of seat angle and lumbar support on seated buttock pressure. Phys Ther 1988;68:1682-6. 6. Mawson AR, Biundo JJ, Neville R Linares HA, Winchester Y, Lopez A. Risk factors for early occurring pressure ulcers following spinal cord injury. Am J Phys Med Rehabil 1988;67:123-7. 7. Kosiak M. Etiology and pathology of ischemJc ulcers. Arch Phys Med Rehabil 1959;40:62-9. 8. Reswick JB, Rogers J. Experience at Rancho Los Amigos Hospital with devices and techniques to prevent pressure sores. In: Kenedi RM, Cowden JM, Scales JT, editors. Bedsore biomechanics. London: Macmillan; 1976. p. 301-10. 9. Garber SL, Krouskop TA. Body build and its relationship to pressure distribution in the seated wheelchair patient. Arch Phys Med Rehabil 1982;63:17-20. 10. Garber SL Wheelchair cushions: a historical review. Am J Occup Ther 1985;39:453-9. 11. Garber SL, Dyerly LR. Wheelchair cushions for persons with spinal cord injury: an update. Am J Occup Ther 1991;45:550-4. 12. Henderson JH, Price SH, Brandstater ME, Mandac BR. Efficacy of three measures to relieve pressure in seated persons with spinal cord injury. Arch Phys Med Rehabil 1994;75:535-9. 13. Park CA. Activity positioning and ischial tuberosity pressure: a pilot study. Am J Occup Ther, 1992;46:904-9. 14. Swarts AE, Krouskop TA, Smith DR. Tissue pressure management in a vocational setting. Arch Phys Med Rehabil 1988;69:97-100. 15. Fisher SV, Patterson RP. Long term pressure recordings under the ischial tuberosities of tetraplegics. Paraplegia 1983;21:99-106. 16. Patterson RP, Fisher SV. Sitting pressure-time patterns in patients with quadriplegia. Arch Phys Med Rehabil 1986;67:812-4. 17. Bar CA. Evaluation of cushions using dynamic pressure measurement [review]. Prosthet Orthot Int 1991;15:232-40. 18. Eckrich KM, Patterson KE. Dynamic interface pressure between seated users and their wheelchairs. Int J Indust Ergon 1991;8:1l 523. Arch Phys Med Rehabil Vol 79, March 1998
316
SEAT INTERFACE PRESSURES, Kernozek
19. Kalpen A, Bochdansky T, Seitz E The dynamic body pressure distribution in a moving wheelchair. Workshop MW04: RESNA Conference, June 7-12, 1996. 20. McPoil T, Cornwall MW, Yamada W. A comparison of two in-shoe measurement systems. Lower Extremity 1995;2:1-9. 21. Kernozek TW, LaMott EE, Dancisak MJ. Reliability of an in-shoe measurement system during treadmill walking. Foot Ankle Int 1996; 17:204-9. 22. Kosiak M. A mechanical resting surface: its effect on pressure distribution. Arch Phys Med Rehabil 1976;57:481-4. 23. Eriksson E. Etiology: microcirculatory effects of pressure. In: Zacharkow D, editor. Posture: sitting, standing, chair design and exercise. Springfield (IL): Charles Thomas; 1988. p. 115 -30.
Arch Phys Med Rehabil Vol 79, March 1998
115-30. 24. Brand PW. Comments on the article "Development of test methods for the evaluation of wheelchair cushions" [editorial]. Bull Prosth Res 1980;10:3-4. 25. Veeger HEJ, Van der Woude LHV, Rozendal RH. Wheelchair propulsion technique at different speeds. Scand J Rehabil Med 1989;21:197-203. 26. Cavanaugh PR, Ulbrecht JS. Biomechanics of the foot in diabetes. In: Levin MD, O'Neal LW, Bowker JH, editors. The diabetic foot. 5th ed. St. Louis (MO): Mosby; 1993. p. 199-232.
Supplier a. Novel Electronics, Inc., 964 Grand Avenue, St. Paul, MN 55105.