Teens and young adults with severe cerebral palsy (CP), Gross Motor Functional Classification Scale (GMFCS) levels IV-V, have a 15% to 80% chance of developing neurological scoliosis.1–3 In this population, 30% develop scoliosis with a Cobb angle of more than 40°.2 Neurological scoliosis affects individuals across all International Classification of Function (ICF)4 domains reducing cardiopulmonary function, which impedes chest excursions and abdominal movements,5–8 and contributing to morbidity and mortality.9 The asymmetrical pull of spastic musculature can cause the torso to collapse and rotate, causing increased rib prominence and pelvic obliquity,10 which increases the individual's risk for skin ulcerations.2,3 Functionally, this trunk collapse and rotation can limit motor function,11 reduce tolerance for sitting upright,12 restrict upper extremity reaching,13 and may cause pain from pressure.2 Combined, these body structure and functional limitations can reduce the individual's participation,14 use of augmentative communication devices,1,2,5,10,11 and may negatively affect academic performance.15 School-based physical therapists (PTs) can use different seating systems to prevent and/or decrease body structure changes, with the aims of improving function, school participation, educational success, and overall quality of life.
Effect of Positioning on Impairment Outcomes
Limited studies on positioning of young adults with severe CP and scoliosis or standardized measures to determine effectiveness of adaptive seating are available.14,15 Impairments shown to improve with customized seating include vital capacity, forced expiratory volume in 1 second, and expiratory time in children with spastic CP aged 5 to 12 years.16 Oxygen saturation (SaO2) levels were significantly higher in the sitting than in the supine position, and respiration rates (RRs) decreased, approaching more normal levels, in 3 of 5 of young adults with CP and severe scoliosis.1 Cardiorespiratory fitness typically decreases as individuals age1,9; this decline may be faster in individuals functioning at GMFCS level V because they have little opportunity for fitness activities.17 Thus, the influence of customized positioning devices on cardiorespiratory measures may be more significant in those with the greatest mobility limitations.8
In clinical and school-based settings, cardiorespiratory levels are measured using a pulse oximeter and direct measures. Normal SaO2 levels range from 97% to 99% and an SaO2 of 95% is “clinically acceptable”18(p.77) in individuals with normal hemoglobin levels. Low SaO2 levels indicate poor body oxygenation, and levels below 87% indicate impending respiratory failure.18,19
Body temperature (BT) may influence SaO2 readings, alertness, and visual attention.20 Because BT can vary with age, sex, or the site of measurement, establishing a BT baseline before implementing positional changes is reasonable. Typical BT for those older than 18 years is 97.6° to 99°F.21
Custom seating systems maximize the body surface support area and reduce pressure points by distributing contact over broader surfaces.2,22 A customized seat might also increase heat production or retention as the seating system surface sits closer to the person's skin, potentially trapping body heat. Traditional seating has less contact with the body, allowing heat dissipation, but it also creates focal pressure points and puts the individual with significant skeletal asymmetries at risk for skin ulcerations.2,14,22 Customized functional seating is highly individualized for people with severe CP and scoliosis11 because standard seating solutions do not accommodate significant body structure asymmetries.5
Effect of Positioning on Activity and Participation Outcomes
A limited number of studies examined activity and/or participation changes using adaptive seating.15 Although authors suggest that customized seating improves sitting posture and head control while participating in daily routines, such as family meals12 and fine motor tasks in diverse settings,11,12,15,23,24 conflicting evidence exists. A 1995 systematic review found no differences in upper extremity control with adaptive seating versus nonadaptive seating,25 but a more recent case report of a child with quadriplegic CP accessing an augmentative communication device demonstrates improved upper extremity movement and participation.11 Studies of children with multiple disabilities who are nonambulatory indicate that sitting time,12 attention and engagement,12,13 number of classroom activities,12,15 and participation time13,26 increase when seating is customized. These studies investigate younger (9 months to 5 years), more flexible, and still developing children, but offer insights into how custom seating may affect older, less flexible young adults with CP. Given the small body of evidence and the cost of customized seating, objective measures that demonstrate positive changes can justify their use.
This single subject study compares the effects of 2 seating systems used in a school setting and their effects on measures in 3 ICF domains: cardiopulmonary functioning (body function and structure), time to activate communication switches (activity), and accuracy of answering questions (participation) for a young adult with quadriplegic CP and neurological scoliosis. It was hypothesized that SaO2 and processing time would improve, but that BT would rise causing increased sweating or skin irritation.
At the time of the study, “John” was a 19-year-old man diagnosed with spastic quadriplegic CP, GMFCS level V, who had received school-based PT for 16 years. He was in a local, public high school class with another 7 adolescents with disabilities. He had a severe right thoracolumbar scoliosis, with a Cobb angle of more than 90° and significant pelvic obliquity. At 53 lb, he was not a surgical candidate, had not used a spinal orthosis, and did not use medication to manage the spasticity. John's right hip was dislocated and he was most comfortable in a “windswept” position, with knees to the left side. He wore bilateral ankle-foot orthoses when sitting in a mobile base, positioning chair (James Leckey Design, Lisburn, Ireland, UK), which he had used for the last 7 years. The original chair configuration had a planar cushioned back and seat with standard order lateral trunk supports, medial thigh support, chest straps, and a tray; and is referred to as the original chair position (OCP). The tray was used for positioning 2 BIGmack communication switches (AbleNet, Inc, Roseville, Minnesota), with the “yes” switch on John's left and the “no” switch on his right (Figure 1).
John had reliable but limited communication skills. He smiled to indicate happiness, excitement, and agreement. He looked down to the left when upset or to refuse participation, cried when in pain, and used distinct vocalizations to gain attention. To disagree, John made eye contact, then looked away, and did not smile. His typical head position in the OCP was forward flexion, tilted to the right, and looking down to the left (Figure 2). John had reliable “yes/no” responses when using his tray switches, with an average response time of 41 seconds per answer.
At the time of this study, John received monthly, consultative PT sessions to assist teachers with positioning to achieve 2 of his individual educational plan (IEP) goals: to answer 5 “yes/no” questions accurately within 30 seconds each; and to complete computer schoolwork within 1 hour while in his positioning chair.
Teachers were concerned that his scoliosis made the OCP uncomfortable, impeded his cardiorespiratory system, limited sitting to about 30 minutes, and hindered his speed to reach communication switches. His BT tended to run high, which increased sweating and concerns of skin irritation. School staff members were interested in providing a custom-molded back (CMB) to better support his trunk and improve the use of communication switches.
This prospective single-subject alternating treatment design had 2 conditions: a baseline phase with the OCP (A1), an intervention phase with the CMB and original seat (B), and a return to baseline (A2). The independent variable was the seating system type (planar vs custom molded), and the dependent variables included SaO2, heart rate (HR), RR, BT, processing time to activate a switch, and response accuracy. The study and use of photographs had approval from the student, his family, and the school in the absence of a formal Institutional Review Board process.
Body structures and function measures of SaO2 and HR were measured with a digital fingertip pulse oximeter (Beijing Choice Electronic Tech. Co., Beijing, China); RR was measured as the number of chest excursions in 1 minute by palpation of the upper chest wall; and BT was measured with a factory-calibrated noncontact infrared thermometer (Santa Medical, Model: RY220, Santa Medical, Tustin, California). Each session, John responded to 5 “yes/no” questions, with at least 1 that required a correct “no” answer, by reaching for the appropriate switch. Activity was defined as the time it took to reach a switch from a standardized starting position after a question was asked, measured in seconds with a manual stopwatch. Participation was defined as the number of correct answers to the 5 questions (Table 1); accurate responses to questions aligned with his IEP goals.
Data were collected across 22 sessions in 18 weeks, at the same time of day. Phase A1 had 10 sessions over 10 weeks in the OCP (Figure 1) and was the longest phase as it included Thanksgiving and Winter breaks. Phase B had 10 sessions over 6 weeks using the CMB (Figure 3). Phase A2 had 2 sessions over 2 weeks in the OCP for the testing protocol; however, the CMB was available for classroom use for up to 3 hours each day. Phase A2, scheduled for 10 testing sessions, was terminated after 2 sessions because of John's refusal to use the OCP after accommodating to the CMB.
Measures of SaO2, HR, RR, and BT were taken twice: 3 minutes after John was positioned in the seating system, when his cardiorespiratory measures typically stabilized from the excitement of the transfer, and at the end of each session to determine seating system effects. Processing time to activate a switch in response to a question, and the number of correct answers to 5 questions were recorded once per session. Formal reliability measures were not calculated; however, the same tester (KL) performed and recorded all measurements using the same procedures and equipment.
Two seating systems were compared. John's OCP had a Contoured Advance Seat (www.leckey.com/products/contoured-advance-seat, James Leckey Design, Lisburn, Ireland, UK) with standard order seat and back supports, lateral trunk supports, and a medial thigh support, upholstered with standard factory-issued fabric, and a tray for his communication devices. The CMB used the same Leckey mobile base and seat with the new CMB. A durable medical equipment vendor and an Ottobock manufacturing representative used the Ottobock Shape System's (OBSS, Ottobock Group, Berlin, Germany) special fitting base and bags to accurately model the curves of John's back (www.ottobockus.com/mobility/mobility-101/seating-and-positioning/). A 3-D digital image file of the shape was used by the Ottobock Fabrication Center to replicate the shape in a contoured cushion. A trial cushion was tested and revisions were made before John received the final upholstered CMB. Darlexx, a spandex-like fabric with 4-way stretch and increased breathability, was used to upholder the CMB to minimize sweating from total contact of the trunk and to accommodate the extreme contours needed to match John's curves (Figure 4). During each study phase, John spent a combined total of 2 to 3 hours per day in his OCP or CMB, as tolerated for classroom activities, inclusive of testing sessions.
Dependent t tests were used to determine whether physiological measures changed within treatment sessions. Visual and split middle analyses were used to illustrate trend changes and daily variability of the averaged physiological measures and the activity and participation measures for all phases, though results only address Phases A1 and B because of insufficient data points in Phase A2.
Body Structure and Function
Dependent t-tests comparing the initial and final cardiopulmonary measures were not significantly different except for BT in the CMB (t(9) = 3.973; P = .003). The increase of approximately 1° was not considered clinically significant given John's variability, so the average of the 2 SaO2, HR, RR, and BT measures were graphed for trend analysis. In all graphs, Phase A2 trends were not analyzed, as only 2 measures were available.
Within-session changes in SaO2 were not significant in Phase A1 (t(9) = −0.258; P = .802) or Phase B (t(9) = −0.429; P = .678). John's average SaO2 was 93.5% ± 3.09% (range = 83%-98%) in Phase A1, 96.5% ± 0.71% (range = 95%-99%) in Phase B, and 94.8% ± 1.77% (range = 93%-96%) in Phase A2 (Table 2). Functionally, he moved from a typical state of oxygen deprivation on 6 of the 10 days in Phase A1 to clinically acceptable and borderline normal levels18 for all of Phase B. Split-middle analysis illustrates a decelerating trend in Phase A1, but in Phase B, he was consistently oxygenated with reduced variability (Figure 5).
Normal adolescent HR ranges from 60 to 100 beats per minute (bpm).27 Within-session HR changes were not significant in Phase A1 (t(9) = −0.297; P = .773) or Phase B (t(9) = 0.069; P = .946). The average HR was 93.4 ± 13.19 bpm (range = 67-136) in Phase A1, 98.7 ± 6.33 bpm (range = 52-140) in Phase B, and 96.3 ± 32.17 bpm (range = 52-126) in Phase A2 (Table 2). Split-middle analysis suggests an accelerating trend in Phase A1, reduced variability in Phase B, but no change in level (Figure 6).
Normal adult RR is 12 to 18 rpm, with more than 20 rpm considered moderate tachypnea, and more than 24 rpm considered severe tachypnea19; however, no reference values for individuals with CP were found even though many demonstrate respiratory distress syndrome in early childhood.6 Within-session changes in RR were not significant in Phase A1 (t(9) = −1.395; P = .196) or Phase B (t(9) = −1.152; P = .279). The average RR was 38.0 ± 3.89 rpm (range = 28-44) in Phase A1, 36.0 ± 3.27 rpm (range = 30-44) in Phase B, and 42.5 ± 2.12 rpm (range of 38-48) in Phase A2 (Table 2). Split-middle analysis indicated a slight accelerating trend in Phase A1 and reduced variability in Phase B, with 5 of the 10 days measured at less than 35 rpm (Figure 7); in both phases, RR was consistent with severe tachypnea.
Typical BT for more than 18 years of age is 97.6° to 99°F.21 Within-session BT changes were not significant in Phase A1 (t(9) = −1.00; P = .343) but were significant in Phase B (t(9) = 3.973; P = .003). The difference between a mean starting BT of 100.3° versus 99.2° at the end of the session was not considered clinically significant as John's temperature tended to run warm and this represented a decrease toward normal. Thus, the average of the 2 measures was used for graphing. The average BT was 99.6° ± 1.07°F (range = 98.0°-102.0°) in Phase A1, 99.8° ± 0.82°F (range of 98.0°-103.0°) in Phase B, and 99.2° ± 0.12°F (range = 99.1°-99.3°) in Phase A2 (Table 2). Split-middle analysis suggested a slightly lower level of BT in the OCP, but both Phase A1 and Phase B lines fell within a single degree range (99°-100°), so this change in level was not considered clinically meaningful (Figure 8). In Phase B, daily variability was slightly reduced.
The average processing time for the 5 questions was 40.9 ± 29.54 seconds (range = 2-87 seconds) in Phase A1, 55.7 ± 46.54 seconds (range = 4-133 seconds) in Phase B, and 48.0 ± 29.7 seconds (range = 27-69 seconds) in Phase A2. Split-middle analysis indicated that the increasing trend in processing time in Phase A1 was reversed in Phase B, though the variability increased (Figure 9).
The accuracy of correct responses per session averaged 3.7 ± 0.82 (range = 2-5; mode = 4) in Phase A1 and 3.8 ± 1.32 (range = 1-5; mode = 5) in Phase B. Split-middle analyses indicated a decelerating trend that was reversed in Phase B; however, no distinctive change in level was seen (Figure 10). During Phase B, John was upset on day 15. He took the longest to respond to questions (301.3 seconds) and answered questions inaccurately on purpose. For example, John looked directly at a staff member when asked, “Is Ms Y here?” After 258.5 seconds, he answered “no.” Other times in the study, John demonstrated his sense of humor as he answered a question inaccurately and laughed, indicating he knew the answer was incorrect and thought it was funny. Secondary analysis of Phase B data without the 2 extreme points did not change the deceleration slope as determined by visual analysis.
Study measures were based on physiological concerns and John's IEP goals; however, additional changes in comfort, communication, and social approachability were reported when John used the CMB. Teaching staff and John's mother reported that he appeared more comfortable because he was more upright, “not breathing as hard,” and John's mother reported hearing more “belly laughing.”
In the CMB, John's head turning frequency and visual searching improved his overall communication. During Phases A1 and A2, when John was asked, “Is Ms X here?” he would pause, usually smile, and then activate his communication switch without searching to see whether people were present. In Phase B, John began to turn his head to scan for the person in question, visually greet visitors to his class, or look at a vocalizing classmate; these behaviors noticeably increased from his usual response of smiling without eye contact. Improved head mobility and control in the CMB also allowed John to experiment with head-activated switches that were not previously accessible. Consequently, he learned to scan with a 4-phrase head switch to greet others, comment on the weather, tell staff when he completed an activity, and say goodbye. He learned to use a different head switch to request page turning to complete schoolwork. John continued to use his “yes” and “no” communication switches on his tray during the study; when data collection was completed, he received both the 4-phrase and page turning head switches, which increased the frequency and appropriateness of his communication with those around him.
During Phase B, school staff noted that John's increased head control allowed more eye contact with staff, classmates, and people unfamiliar to him, increasing his approachability. Other students and staff began to address John more frequently and ascribed it to his ability to hold his head up and make eye contact to attract their attention.
The decision to order a customized seating solution is often fraught with concerns over what the actual benefits will be for an individual. This study illustrates that objective physiological and functional measures can demonstrate the effect of a CMB.
For this comparison of an OCP with a CMB on the same mobile base, we hypothesized that SaO2 and processing time would improve, and less desirably, that BT would increase while in the CMB. The results suggest that using a CMB reduced respiratory distress to clinically acceptable levels, and that BT did not change significantly. The 1° BT variability may not be clinically meaningful, given John's typical fluctuations.
John's average processing time increased while in the CMB, although the trend in Phase A1 is accelerating, while the trend in Phase B is decelerating. In the OCP, it seemed easier for John to extend his left arm by lifting his head and using an asymmetrical tonic neck reflex to activate the “yes” switch. Activating the “no” switch was more challenging: he had to lift his head, move it toward midline to clear the headrest, and turn to his right to use his right asymmetrical tonic neck reflex.
During Phase B, processing time was perceived to be increasing in the CMB because John had to change his body mechanics and experiment with alternative movement patterns to activate switches. Ideally, there would be a deceleration during Phase B, but the daily data collection and the initial graphing suggested no change or a possible increase. Although greater variability was seen during Phase B, his average processing time decreased, apparent after celeration lines were examined. This underscores the usefulness of graphing clinical data with celeration lines.
Initially, neither John nor the school staff liked the CMB. By the end of Phase B, it had become John's new norm and he no longer liked the OCP, refusing data collection after the second session in Phase A2. The school staff struggled to place John's pelvis correctly in the CMB. Overall, the staff took longer to get John positioned because they had to wait for him to relax into the CMB before fastening support straps. For John, more trunk contact and support from the chair required an increase in upright posture. Despite the initial challenges, the school staff was willing to use the CMB as they valued the improvements in head control, use of communication switches, and social interactions.
Although not fully reflected in the measure of accurately answering 5 questions, John's school-based participation improved by using the CMB. This participation measure was used as it aligned with 1 of his educational goals, but in hindsight, the goal was too limited. Although no objective measures were used, teachers and staff observed that his social approachability increased, and he noticeably increased initiating eye contact with visitors when they entered his class, with noisy classmates and with staff members to get their attention. His increased head control allowed him to use head switches for the first time, increasing communication with others with reduced effort. With improved upright posture, more staff and students initiated greetings, and John could more easily respond. Given his enhanced socializing, in the future researchers may want to use a definition of school-based participation that includes eye contact, head control, and vocalizations.
A single-subject study is practical in the educational setting, though it has limited generalization. The absence of complete data in Phase A2 because of John's refusal to continue the testing sessions in the OCP is a limitation. But John's development of a clear preference for the CMB is an important patient-centered outcome.
Caution should be used when interpreting HR data, as HR can be influenced by external factors such as emotional stress,28–30 the student's age,30,31 or medications.29,32 John did not use any medications during this study, but he is very excitable, and that may have influenced selected HR readings.
The accuracy of answers to yes/no questions is only reliable if the participant answers the questions to the best of his ability each time. Neither test-retest nor intrarater reliability was formally established on the study measures.
Withdrawing the CMB during Phase A2 presented an ethical dilemma and feasibility problems. John developed greater social access and function with his CMB, and removing the CMB meant temporarily limiting his newfound skills.
About 30 minutes was needed to switch seat backs, adding about 1 hour per session when John did not have access to his chair for school participation. He spent that time on a mat, which is not supportive of SaO2.1 After John accommodated to the CMB, he complained verbally when having to reassume a less supported position in the OCP for testing. A minimum of 3 data points is required to interpret trend results; however, Phase A2 was stopped after 2 data points as John refused to continue. Despite the lack of data for Phase A2, improvements in all ICF domains were seen with the CMB. John's preference for a particular seat and what it affords is also a valuable outcome.
The chosen outcome measures reflect health concerns and IEP goals, but the activity and participation measures were not encompassing or sensitive enough to demonstrate significant change. Other individualized measures of activity and participation might have included active neck range while scanning to identify objects, control of a head pointer in reaching targets, the number of times he initiated a greeting with eye contact, or increased his directed verbalizations, as standardized participation measures are not yet available for this population.14
This study provides evidence that a CMB can positively affect physiological and functional measures in 1 young adult with CP and severe scoliosis. Measures of SaO2 levels had clinically relevant increases, approaching levels that are more normal. BT increased on average by 1° but did not achieve significance. Trends in communication processing time decreased with an increase in variability using a CMB, but this was easily influenced by subject motivation. Participation had a slight increase in accuracy of answers, though this measure was limited. Unexpected improvements included teachers, staff, and family observations of increased social approachability, improved head control that enabled use of a wider variety of assistive technologies, leading to greater participation in classwork and more frequent and precise communication. Future studies should use a broad array of participation measures to better capture the spectrum of possible changes. This single-subject study contributes to our understanding of potential changes in body structures and function, activity, and participation in 1 older adolescent with GMFCS level V CP and a fixed scoliosis when using a CMB in an educational setting.
We express our appreciation to “John,” his mother, and his teachers who willingly participated in this study. We are particularly grateful for the assistance of Michele Sutphin and Kristie Johnson and for their support of the study protocol. The authors thank Ron Graham, National Seating and Mobility, and Darrell Burnette, Ottobock, for donating their time and sharing their expertise.
1. Littleton SR, Heriza CB, Mullens PA, Moerchen VA, Bjornson K. Effects of positioning on respiratory measures in individuals with cerebral palsy
and severe scoliosis
. Pediatr Phys Ther. 2011;23(2):159–169.
2. Koop SE. Scoliosis
in cerebral palsy
. Dev Med Child Neurol. 2009;51(Suppl 4):92–98.
3. Morris C, Bowers R, Ross K, Stevens P, Phillips D. Orthotic management of cerebral palsy
: Recommendations from a consensus conference. NeuroRehabilitation. 2011;28:37–46.
4. World Health Organization W. The International Classification of Functioning, Disability and Health—ICF. WHO; 2001.
5. Hatta T, Nishimura S, Inoue K, et al. Evaluating the relationships between the postural adaptation of patients with profound cerebral palsy
and the configuration of the Seating Buggy's seating support surface. J Physiol Anthropol. 2007;26(2):217–224.
6. Seddon PC, Khan Y. Respiratory problems in children with neurological impairment. Arch Dis Child. 2003;88(1):75–78.
7. Barks L. Therapeutic positioning, wheelchair seating, and pulmonary function of children with cerebral palsy
: a research synthesis. Rehabil Nurs. 2004;29(5):146–153.
8. Barks L, Shaw P. Wheelchair positioning and breathing in children with cerebral palsy
: study methods and lessons learned. Rehabil Nurs. 2012;36(4):146–152, 174.
9. Rimmer JH. Physical fitness levels of persons with cerebral palsy
. Dev Med Child Neurol. 2001;43(3):208–212.
10. Viehweger E. Importance of hip problems in daily activities for cerebral palsy
patients. J Child Orthop. 2013;7:401–406.
11. Costigan FA, Light J. Effect of seated position on upper-extremity access to augmentative communication for children with cerebral palsy
: preliminary investigation. Am J Occup Ther. 2010;64(4):596–604.
12. Hulme JB, Gallacher K, Walsh J, Niesen S, Waldron D. Behavioral and postural changes observed with use of adaptive seating by clients with multiple handicaps. Phys Ther. 1987;67(7):1060–1067.
13. Ryan SE, Campbell KA, Rigby PJ, Fishbein-Germon B, Hubley D, Chan B. The impact of adaptive seating devices on the lives of young children with cerebral palsy
and their families. Arch Phys Med Rehabil. 2009;90(1):27–33.
14. Field D, Livingstone R. Clinical tools that measure sitting posture, seated postural control or functional abilities in children with motor impairments: a systematic review. Clin Rehabil. 2013;27(11):994–1004.
15. Ryan SE. An overview of systematic reviews of adaptive seating interventions for children with cerebral palsy
: where do we go from here? Disabil Rehabil Assist Technol. 2012;7(2):104–111.
16. Nwaobi OM, Smith PD. Effect of adaptive seating on pulmonary function of children with cerebral palsy
. Dev Med Child Neurol. 1986;28:351–354.
17. Bartlett DJ, Hanna SE, Avery L, Stevenson RD, Galuppi B. Correlates of decline in gross motor capacity in adolescents with cerebral palsy
in Gross Motor Function Classification System levels III to V: an exploratory study. Dev Med Child Neurol. 2010;52:e155–e160.
18. Schutz SL. Oxygen saturation
monitoring by pulse oximetry. In: Lynn-McHale DJ, Carlson KK, eds. AACN Procedure Manual for Critical Care. 4th ed. Philadelphia: W.B. Saunders; 2001:77–82.
19. Cretikos MA, Bellomo R, Hillman K, Chen J, Finfer S, Flabouris A. Respiratory rate: the neglected vital sign. Med J Aust. 2008;189(9):532.
20. Wright KP, Hull JT, Czeisler CA. Relationship between alertness, performance, and body temperature in humans. Am J Physiol Regul Integr Comp Physiol. 2002;283(6):R1370–R1377.
21. Sund-Levander M, Forsberg C, Wahren LK. Normal oral, rectal, tympanic and axillary body temperature in adult men and women: a systematic literature review. Scand J Caring Sci. 2002;16(2):122–128.
22. Silva F, Beretta EM, Prestes RC, Kindlein W. Design and milling manufacture of polyurethane custom contoured cushions for wheelchair users. Aust Med J. 2011;4(9):500–506.
23. Miedaner J, Finui L. Effects of adaptive positioning on psychological test scores for preschool children with cerebral palsy
. Pediatr Phys Ther. 1993;5(4):177–182.
24. Washington K, Deitz JC, White OR, Schwartz IS. The effects of a contour foam seat on postural alignment and upper-extremity function in infants with neuromotor impairments. Phys Ther. 2002;82(11):1064–1076.
25. Roxborough L. Review of the efficacy and effectiveness of adaptive seating for children with cerebral palsy
. Assist Technol. 1995;7(1):17–25.
26. Rigby PJ, Ryan SE, Campbell KA. Effect of adaptive seating devices on the activity performance of children with cerebral palsy
. Arch Phys Med Rehabil. 2009;90(8):1389–1395.
27. Fleming S, Thompson M, Stevens R, et al. Normal ranges of heart rate
and respiratory rate in children from birth to 18 years: a systematic review of observational studies. Lancet. 2011;377(9770):1011–1018.
28. Eston RG, Rowlands AV, Ingledew DK. Validity of heart rate
, pedometry, and accelerometry for predicting the energy cost of children's activities. J Appl Physiol. 1998;84(1):362–371.
29. Stauss HM. Heart rate
variability. Am J Physiol Regul Integr Comp Physiol. 2003;285:R927–R931.
30. Rowlands AV, Eston RG. The measurement and interpretation of children's physical activity. J Sports Sci Med. 2007;6:270–276.
31. Furukawa A, Nii E, Iwatsuki H, Nishiyama M, Uchida A. Factors of influence on the walking ability of children with spastic cerebral palsy
. J Phys Ther Sci. 1998;10(1):1–5.
32. Peel C, Mossberg KA. Effects of cardiovascular medications on exercise responses. Phys Ther. 1995;75(5):387–396.