Compromised respiratory function, often evident in adolescents and adults with severe cerebral palsy (CP), originates with underdevelopment of the upper chest wall. This underdevelopment inhibits chest expansion and can result in rapid and paradoxical breathing patterns. Gravitational forces interact with forces generated from muscle contractions to facilitate musculoskeletal development of an infant's thorax. Weak musculature of a young child with CP frequently cannot balance forces of gravity optimally, and chest development is adversely affected.1,2
The incidence of scoliosis and atypical chest wall structure and function are interrelated in CP. Scoliosis occurs in 25% of individuals with CP.3,4 It alters physiological functioning and respiratory measurements such as oxygen saturation (SaO2), heart rate (HR), respiratory rate (RR), and chest wall excursion (CWE).5–9
Severe spinal deformities such as scoliosis not only are associated with respiratory compromise but also impact quality of life as well as life expectancy, often resulting in early mortality.10–12 The natural history of spinal deformity in untreated neuromuscular scoliosis is a progressive curve with increased morbidities and mortality. Nonoperative management including custom wheelchair seating and spinal orthoses are used as options for treatment but offer limited benefit in influencing the natural history of the deformity.5 Surgical intervention such as spinal stabilization has also been used in management of the deformity but is associated with high risk of complications, especially for those most severely involved.13 The life expectancy of individuals with CP after development of scoliosis has not been reported, but the natural history of pulmonary function in severe CP and intellectual disability is one of progressive decline independent of scoliosis.14 Respiratory complications are the most common cause of death of individuals with CP.15,16 Understanding how best to optimize pulmonary function in these individuals is imperative to treatment at all ages and especially as they age.
The use of positioning to improve respiratory function has been studied in both healthy and clinical populations across different ages.17–23 Studies investigating the effects of positioning in adolescents and adults with CP and scoliosis were not found in the literature. In related studies, the supine position was reported to be the least efficient for respiratory outcomes compared with other postures such as the sitting, sidelying (SL), and prone positions.17,21,23,24
As more people with severe disability inclusive of pulmonary compromise due to secondary musculoskeletal changes live longer, cost-effective treatment strategies capable of enhancing pulmonary function need to be identified and employed. This study examines this critical gap in our treatment knowledge. Previous research supports that sitting and, to a lesser extent, SL positions may have a supportive role in the physical management of adolescents and adults with CP accompanied by scoliosis, lack of upright control, and chronic immobility.17–23 The extent of this population's respiratory compromise supports the need for evidence-based strategies for physical therapy management.
The purpose of the study was to investigate effects of positioning on SaO2, RR, HR, and CWE in adolescents and adults with severe CP and scoliosis. On the basis of the available literature and clinical experience, we hypothesized that SaO2 and CWE would increase with change of positioning from supine to sitting and SL. In addition, we hypothesized that HR and RR would decrease with change from the supine to the sitting position and from the supine to the SL position.
Five participants, aged 17 to 37 years, receiving services through the Medicaid Waiver Program in Tennessee's Division of Intellectual Disabilities Services were recruited via a letter to the parents or conservators of potential participants. Parents or conservators of the prospective participants contacted the principal investigator if interested in the study. Inclusion criteria included (1) diagnosis of CP (quadriplegia) and severe scoliosis with a fixed deformity; (2) severe functional limitations including inability to sit without support, stand, or ambulate (Gross Motor Functional Classification System level 5); (3) history of respiratory illnesses; (4) ability to sustain positioning on one side or the use of a custom sidelyer and ability to sustain sitting while using a wheelchair; (5) wheelchair and SL reassessments within 6 months prior to the study; and (6) provision of physical therapy services on a direct or consultative basis. Exclusion criteria included (1) history of spinal stabilization surgery; (2) current use of a spinal orthosis; (3) sensory defensiveness affecting tolerance for handling techniques; (4) change in seizure activity, medications, or dosage in the 3 months prior to the study; and (5) change in spasticity management including dosage change within the 3 months prior to the study. Table 1 provides participant demographics including age, sex, comorbidities, and respiratory history.
This study was approved by the Institutional Review Board at Rocky Mountain University of Health Professions, Middle Tennessee Regional Office Human Rights Committee, and East Tennessee Regional Office Human Rights Committee, and by the state of Tennessee Division of Intellectual Disabilities Services Research Project Review Board. Informed consent was obtained from parents or conservators before participation.
A single-subject research design (SSRD) strategy was used, as individual variation was an important consideration and too few participants were available to make comparisons across groups.25 Specifically, an alternating-treatment SSRD was used to assist in delineating preferred positions associated with optimal respiratory efficiency. Alternating treatment designs replicated over time allow subjects to act as their own control and help define effects of intervention such as positioning and handling on physiological stability.21
An initial baseline phase, encompassing each of the cardiorespiratory measurements (SaO2, RR, HR, and CWE measurements), was completed across 5 sessions with the participants in the supine position. The supine position is frequently used in the population studied and, therefore, the most natural baseline condition for this study. During the second phase, an alternating intervention phase, 3 positions (supine, sitting, and SL) were randomly alternated across 12 sessions to control for order effects. At each session, all of the respiratory measurements were recorded in each of the 3 positions. After the conclusion of the alternating-treatment phase of the study, a follow-up phase of 5 sessions in the supine position was completed with measurements recorded for each session.
In the supine position, participants were placed in their beds with pillows supporting their head/neck, trunk, and extremities to attain as optimal a posture as possible, given varying degrees of contractures. In the sitting position, participants were placed in upright positions in their custom wheelchair seating systems with the wheelchairs' back-to-seat angles set at an average of 110°. These seating systems involved a tilt in space component, head support, custom molded back with lateral support, custom molded seat, pelvic positioning device, and foot supports. In the SL position, the participants were placed on their side in their beds or sidelyers and on the side of their convexity primarily. To achieve consistency with the placement of each of the participants in the positions, all participants had positioning instructions written and pictures taken for the 3 positions. The instructions and pictures thus depicted optimal alignment for that particular participant in the positions being studied.
Outcome variables were obtained through noninvasive measures that did not require cooperation from the participants, given their cognitive and physical impairments. Although vital capacity is a common index for characterization of the restrictive and ventilatory dysfunction in individuals with scoliosis,26 this assessment is difficult in participants who have cognitive impairments.27 Because of these factors, CWE measurements were obtained from the participants' tidal volume (ie, a normal breath).
A research assistant, who was not a physical therapist and was masked to the study design, was trained by the first author. Intrarater and interrater reliability for the 4 respiratory measurements was assessed in a pilot study. One individual aged 24 years with CP quadriplegia participated in the pilot study for 5 sessions for intrarater agreement and 5 sessions for interrater agreement. Intrarater agreement was 100% for SaO2, 83% for RR, 80% for HR, and 100% for CWE. Interrater agreement was 100% for SaO2, 94% for RR, and 100% for HR and CWE.
During the study, intrarater and interrater reliability data were collected on 30% of all respiratory measurements distributed across 3 phases of the study. Intrarater reliability with the research assistant was obtained, while interrater reliability was compared between the first author and the research assistant using percent agreement of 80% or higher as an acceptable level of agreement. By convention, percent agreement is widely used for calculating reliability in applied or clinical settings.25,28 Recordings were considered in agreement when they did not deviate from each other by more than 3% for SaO2,21,29 5 beats per minute for HR,30 3 breaths per minute for RR,30 and 0.2 cm for CWE.31
During the study, data for intrarater and interrater agreement were collected on 35 sessions distributed over 3 phases, including 8 sessions in the baseline phase, 23 sessions in the intervention phase, and 4 sessions in the follow-up phase. Intrarater agreement was 97% for SaO2, 86% for HR, 83% for RR, and 94% for CWE. Interrater agreement was 100% for SaO2, 97% for HR, 100% for RR, and 94% for CWE.
Time of day for intervention was held constant for each session. Data were collected across 22 sessions for each participant as follows: 5 in the baseline phase, 12 in the intervention phase, and 5 in the follow-up phase. For the first 4 participants, baseline testing occurred over a mean of 12.5 ± 2.38 days (range, 11–15). Intervention testing occurred over a mean of 35.2 ± 14.57 days (range, 26–57) for these participants. Follow-up testing occurred over a mean of 10.25 ± 0.50 days (range, 10–11). Duration of each phase for participant 5 was extended because of the distance required for travel for the sessions. For this participant, number of testing days per phase was as follows: baseline, 31 days; intervention, 99 days; and follow-up, 22 days. During the intervention phase, each participant remained in each position for 20 minutes before change to another position. SaO2, HR, RR, and CWE measurements were assessed after a 10-minute interval in the position.
A Nonin 9500 Onyx Digital Fingertip Pulse Oximeter (Nonin Medical, Inc, Plymouth, Minnesota) was the instrument used to measure SaO2 and HR. The accuracy for SaO2 from 70% to 100% is reported as ±2 digits, and for HR as ±3%. Respiratory rate was defined as frequency of chest/abdominal excursions in 1 minute based on observation and palpation at chest wall along the upper abdominal region. A digital clock with a second indicator was used to determine the 1-minute time interval. Chest wall excursion measurements of the participant's normal quiet breathing pattern or tidal volume expansion were obtained with a fiberglass tape measure placed circumferentially at the xiphoid process.31
Measurements collected were recorded on 4 graphs per participant (1 graph for each dependent variable) for a total of 20 graphs. Visual analysis of graphs was used to analyze changes in data over time of the 4 respiratory measurements to determine clinical significance.32 Tests of statistical significance for SSRDs supplemented the visual analysis.33 Celeration lines using the split-middle method of trend estimation were used to detect changes in trend and slope across all phases providing information on clinical significance.32 Tests of statistical significance were analyzed using celeration lines and a probability table.33 The combined application represents a directional (1-tailed) test of significance.33 Hypothesis testing was conducted using change in trend and slope and tests of statistical significance between phases (baseline to alternating intervention phase and alternating intervention phase to follow-up phase) and within phase (alternating intervention phase).
Throughout the results, an increasing trend line for SaO2 and CWE indicates an increase in SaO2 and CWE, whereas a decreasing trend suggests a reduction in SaO2 and less CWE. For RR and HR, a decreasing trend indicates a reduction in RR and HR, whereas an increasing trend suggests an increase for RR and HR.
Comparison of baseline supine to sitting and sitting to follow-up supine. Decreasing trends in SaO2 levels occurred in the baseline phase for participants 1, 2, and 3, with slopes of −1.01, −1.07, and −1.03, respectively (Figure 1). An immediate increase in SaO2 in the sitting position was observed for these participants followed by a flat trend for participants 1 and 2 and an increasing trend, slope 1.02, for participant 3. Oxygen saturation in the sitting position was significantly higher than in the baseline phase for these 3 participants.
Decreasing trends in SaO2 occurred in the follow-up phase for all participants. Only participant 5 showed a significant decrease, with a slope of −1.04 compared with a sitting slope of 1.03.
Comparison of baseline to SL and SL to follow-up. Decreasing trends in SaO2 levels occurred in the baseline phase for participants 1 and 2, with slopes of −1.01 and −1.07, respectively. An immediate increase in SaO2 in the SL position was seen for these participants followed by an increasing trend, slope 1.01, for participant 1 and a flat trend for participant 2. Oxygen saturation in the SL position was significantly higher than in the baseline phase for these 2 participants.
Increasing trends occurred in SL for participants 4 and 5, with slopes of 1.01 and 1.06, respectively, compared with decreasing trends in the follow-up phase, with slopes of −1.04 for both participants. Although SaO2 decreased in the follow-up phase for all participants, SaO2 in the SL position was significantly higher only for participants 4 and 5.
Comparison of sitting, SL, and supine positions in alternating phase. No significant differences were found in any of the participants between the supine and sitting positions. Oxygen saturation in the SL position was significantly higher than in the supine and sitting positions for participant 4, with a slope of 1.01 in the SL position compared with slopes of −1.02 for both the supine and sitting positions. Results indicate that SaO2 was greater in the SL position than in the supine and sitting positions.
Comparison of baseline to sitting and sitting to follow-up. Increasing trends in RR occurred in the baseline phase for participants 1 and 3, with slopes of 1.10 and 1.08, respectively, compared with decreases in the sitting position, with slopes of −1.24 and −1.50 (Figure 2). RR in the sitting position was significantly lower than in the baseline phase.
Decreasing trends in RR occurred in sitting for participants 1 and 3, with slopes of −1.24 and −1.50, respectively, compared with slopes of −1.18 and 1.33 in the follow-up phase. RR in the sitting position was significantly lower than in the follow-up phase.
Comparison of baseline to SL and SL to follow-up. Increasing trends in RR occurred in the baseline phase for participants 1 and 3, with slopes of 1.10 and 1.08, respectively, compared with decreasing trends in the SL position, with slopes of −1.24 and −1.36, respectively. RR in the SL position was significantly lower than in the baseline phase for these 2 participants.
Decreasing trends in the SL position occurred for participants 1, 3, and 5, with slopes of −1.24, −1.36, and −1.45, respectively, compared with slopes of −1.18, 1.33, and −1.18, respectively, in the follow-up phase. RR in the SL position was significantly lower than in the follow-up phase.
Comparison of sitting, SL, and supine in alternating phase. No significant differences were found for any of the participants between the supine and sitting positions, the supine and SL positions, or the sitting and SL positions.
Comparison of baseline to sitting and sitting to follow-up. Increasing trends in HR occurred during baseline measures for participants 3 and 4, with slopes of 1.05 and 1.03, respectively, compared with a decreasing trend in the sitting position (slope −1.12) for participant 3 and an increasing trend in the sitting position (slope 1.07) for participant 4 (Figure 3). Heart rate in the sitting position was significantly lower than in the baseline phase. A decreasing trend in the sitting position (slope −1.12) was observed for participant 3 compared with an increasing trend in the follow-up phase (slope 1.04). Heart rate in the sitting position was significantly lower than in the follow-up phase.
Comparison of baseline to SL and SL to follow-up. An increasing trend in the baseline phase, slope 1.05, for participant 3 was compared with a decreasing trend in the SL position, slope −1.09. Heart rate in the SL position was significantly lower than in the baseline phase. Decreasing trends occurred in the SL position for participants 2 and 3 with slopes of −1.26 and −1.09, respectively, compared with a decreasing trend, slope −1.10, in the follow-up phase for participant 2 and an increasing trend, slope 1.04, for participant 3. Heart rate in the SL position was significantly lower than in the follow-up phase.
Comparison of sitting, SL, and supine positions in alternating phase. For participant 4, trends increased in sitting, supine, and SL positions with slope values of 1.07, 1.09, and 1.13, respectively. Although HR in the sitting position was lower than in the supine and SL positions, HR was increased for all positions rather than decreased as expected. Heart rate in the sitting position was significantly lower than in the supine and SL positions. No significant differences were found in any of the participants between the supine and SL positions.
Chest Wall Excursion
Comparison of baseline with SL and SL with follow-up. An increasing trend in SL was found, with a slope of 1.23, for participant 3 compared with a flat trend in baseline. CWE in the SL position was significantly higher than in the baseline phase. Increasing trends occurred in the SL position for participants 2 and 3, with slopes of 1.67 and 1.23, respectively, compared with an increasing trend with a less steep slope of 1.30 for participant 2 and a decreasing slope, −1.09, for participant 3 in the follow-up phase. CWE in the SL position was significantly higher than in the follow-up phase.
Comparison of sitting, SL, and supine positions in the alternating phase. Participant 4 demonstrated a flat trend in the SL position compared with increasing trends in the supine and sitting positions, with slopes of 1.38 and 1.67, respectively. Although the trend was flat in SL, CWE was significantly higher than in the supine and sitting positions. For participant 3, increasing trends in the sitting and SL positions were found, with slope values of 1.31 and 1.23, respectively. Although trends increased in both positions, the trend was steeper for the sitting position, indicating that CWE was significantly higher than in the SL position.
Participant 1 demonstrated increased activity and communicative interactions in the sitting and SL positions compared with the supine position. Participant 3 showed increased activity and alertness and a decreased frequency of suctioning in the sitting position compared with the supine position. Observations for participant 4 included quieter breaths, decreased coughing, and increased relaxation in the sitting and SL positions compared with the supine position. In addition, increased activity and communicative interactions were observed in the SL position. Participant 5 showed increased alertness (more eye opening and tracking and less lethargy) and increased mouth closure in the sitting position and increased relaxation and mouth closure in the SL position.
We hypothesized that SaO2 and CWE would increase and HR and RR would decrease with change of position from supine to sitting and SL. The alternating treatment design permits 2 sets of comparisons: (1) between the baseline and intervention phase and (2) among interventions.32 After the alternating treatment phase, additional measurements in the supine position were taken for follow-up. Comparisons between baseline and intervention, among interventions, and from intervention to follow-up with significant differences (clinically and statistically) demonstrated meaningful differences. Significant differences from intervention only to follow-up were not considered meaningful. Clinical significant differences for participants are indicated in Figures 1 through 4; statistical significant differences are noted in Table 2. Varied levels of support were found for the hypotheses of the study. Preference for the sitting or SL positions over the supine position differed depending on participant and outcome variables.
When examining the responses of participants by outcome variable, the varied effect of positioning and position changes can be identified. SaO2 was greater in the sitting and SL positions than in the supine position for 3 of the 5 participants. Overall, RR appeared to be the variable that was most sensitive to change in positions, especially the sitting and SL positions. Heart rate decreased in the sitting and SL positions compared to baseline and follow-up in only 1 of the 5 participants. Heart rate also decreased in the sitting position compared to baseline for 1 other participant. This participant also showed decreased HR in the sitting position compared with the supine and SL positions in the alternating intervention phase.
Chest wall excursion was greater in the sitting and SL positions than in baseline and follow-up in 1 of the 5 participants. This participant also showed greater CWE in the sitting position compared with the SL position in the alternating intervention phase. One other participant showed greater CWE in the SL position than in the supine and sitting positions in the alternating intervention phase.
The severity of respiratory compromise may affect ventilatory response to position and positional change. Participant 3 had the most severe respiratory compromise, including the need for supplemental oxygen, a tracheotomy, and nighttime ventilator. This participant demonstrated responses to position changes across all 4 variables and had the greatest number of significant differences. During sessions in which this participant was more congested, needed suctioning, and demonstrated fatigue, his improved status in the upright position was evident, especially for SaO2. Thus, individuals with the most severe respiratory compromise may respond less favorably to the supine position.
Age may also be a factor in the response to position and positional change among individuals with motor and respiratory compromise. The 3 oldest participants (participants 1, 3, and 4) had the most significant differences in positional comparisons. These 3 participants also had SaO2 drop below 95% in more sessions than the younger participants. Participants 1 and 3 also had sessions in which their RR was frequently more than 20 breaths per minute. Participant 1 also had frequent sessions in which his HR was more than 100 beats per minute. Typical cardiac and pulmonary function changes, which occur with the aging process, may be accelerated in individuals with preexisting respiratory compromise and may increase the influence of positioning changes on respiratory measurements.
Improved outcomes with RR, HR, and CWE associated with the changes between the supine and sitting positions corroborated the evidence reported by Noble-Jamieson et al17 for changes in vital capacity among children with neuromuscular diseases who were nonambulatory and had scoliosis. Differences in outcomes between participants with the most severe respiratory compromise and participants with less involvement in our study were also similar to findings of Noble-Jamieson et al.17 Improved outcomes with CWE also support the evidence reported by Harris et al24 for changes between the supine and sitting positions for 60 participants who were healthy and aged 19 to 34 years.
Improved measures of SaO2 with change between the supine and SL positions corroborated the evidence by Bjornson et al21 for infants born preterm. Improved outcomes with RR and HR for this positional change corroborated evidence by Yeaw23 for individuals aged 18 to 79 years with unilateral lung pathology.
Clinical observations in the sitting and SL positions for 4 of the 5 participants included increased alertness and activity level, increased communicative interactions, decreased suctioning, decreased effort for breathing, and an altered breathing pattern from primarily mouth breathing to somewhat of a mouth/nasal pattern of breathing. These observations may be important for treatment of individuals such as those in this sample. Physiological changes along with behaviors and interactions in different positions should be used to determine optimal positions. Caregivers should be informed of specific observations associated with their children's optimal performance, which are linked to improved physiological changes, and should be instructed in the use of measurements such as SaO2 and RR.
Despite individual differences, positioning in sitting and SL versus supine with this population may offer parents and other caregivers options to enhance respiratory status in individuals who are otherwise compromised. Although it is unlikely that supine positioning could be eliminated from a population whose positioning options are limited, reduced time in the supine position in contrast to other positions may provide benefit to overall respiratory status. Long-term prognosis for individuals with severe scoliosis and CP is associated with progressive decline due to respiratory system involvement. Physical therapists should use opportunities for consultation to determine and recommend optimal positions that could possibly influence overall quality of life, morbidity, and mortality for this population.
In general, the results of this study support tolerance for a variety of positions in this population. In addition to variance between individuals, there are within-individual variances, which indicate that there are days when tolerance for a position may not be as great as on other days. The need for individualizing to a patient's tolerance and specific status is important in that person's overall care.
Limitations and Implications
Replication using single-subject and group designs is warranted. Individuals without respiratory compromise should be studied to clarify typical responses to positioning. Consideration of possible movement effects from the change of positions should also be studied in healthy and clinical populations. In this study, 10 minutes elapsed after position change before measurements were obtained. Likely, this time should have been of longer duration to avoid the influence of movement effects involved with the transfer. Dean34 reports physiological effects from the increased arousal and mobilization associated with frequent changes in positions, which are distinct from the benefits resulting from a particular position.
Duration in positions as well as time frame for examining the participants' responses would benefit from additional investigation. Additional time in positions that reflects typical schedules would add to the validity or practical significance of this study. Optimal duration for each position was not addressed but is an important parameter to consider in further investigation. An increase in the number of sessions to compare changes over time may reveal other positional effects. An increase in the duration of the baseline phase would have been preferable to ensure more stable initial conditions in the supine position for each participant. An extended time frame for investigating these effects may have allowed study of effects such as changes associated with illnesses and hospitalizations. A longitudinal study investigating the effects of maximizing optimal positions within a daily routine over a course of time may be insightful to compare illnesses and hospitalizations prior to the study with these episodes during the study.
Other factors in addition to positioning effects that were not controlled during this study include the following: (1) participant's activity level during some of the measurements; (2) incidents in which the participants were not feeling well, demonstrating lethargy, congestion, coughing episodes, or difficulty clearing secretions; (3) irritability or crying episodes; (4) unexpected changes in feeding schedules; and (5) the severity (Cobb angles) of the participant's scoliosis or kyphoscoliosis. Any of these factors could have influenced changes in the respiratory measurements in addition to positioning effects.
Restricted mobility in chest walls may be related to the site of measurement, the xiphoid process. This site may not capture the greatest CWE possible as expansion of the chest wall at xiphoid process primarily measures intercostal activity. Because of severe scoliosis or kyphoscoliosis, participants primarily demonstrated a diaphragmatic breathing pattern. Limited intercostal activity likely contributed to small changes in CWE values. Other sites such as at the axilla may have allowed more direct measure of the change in chest volume for these participants.
Effects of positioning on 1 adolescent and 4 adults with CP and scoliosis were investigated in a single-subject, alternating-treatment research design. Results of this preliminary investigation support the use of therapeutic positioning in sitting and SL positions as a noninvasive intervention for a population with respiratory compromise. Visual analysis of data, providing information on clinical significance, was corroborated by statistical significance to demonstrate positive changes in the variables of SaO2, RR, HR, and CWE expansion. Severity of respiratory status and age are likely factors contributing to differences in the participants' sensitivity to position changes. Further research is needed to substantiate the efficacy of positioning with a larger population, who otherwise face a predicted decline in their respiratory status.
The authors gratefully thank the study participants, their families, and staff for making this study possible. We also thank the therapists who assisted with the recruitment of the subjects.
1. Massery M. Chest development as a component of normal motor development: implications of pediatric physical therapists. Pediatr Phys Ther. 1991;3(1):3–8.
2. Moerchen V. Respiration and motor development: a systems perspective. Neurol Rep. 1994;18:8–10.
3. Kotwicki T, Jozwiak M. Conservative management of neuromuscular scoliosis: personal experience and review of literature. Disabil Rehabil. 2008;30(10):792–798.
4. Banta JV, Romness MJ, Thomson JD, Zahradnik JL. Treatment of neuromuscular scoliosis. AAOA Instruct Course Lect. 1999;48:551–562.
5. Berven S, Bradford DS. Neuromuscular scoliosis: causes of deformity and principles for evaluation and management. Semin Neurol. 2002;22(2):167–177.
6. DiRocco P, Vaccaor P. Cardiopulmonary functioning in adolescent patients with mild idiopathic scoliosis. Arch Phys Med Rehabil. 1988;69:198–201.
7. Margonato V, Fronte F, Rainero G, Merati G, Veicsteinas A. Effects of short term cast wearing on respiratory and cardiac responses to submaximal and maximal exercise in adolescents with idiopathic scoliosis. Eur Medicophys. 2005;41:135–140.
8. Mori N, Kurosawa H, Matsumoto K, et al. Relationships between spinal deformities and respiratory function in patients with severe motor and intellectual disabilities syndrome. No To Hattatsu. 2006;38(1):10–14.
9. Kotani T, Minami S, Takahashi K, et al. An analysis of chest wall and diaphragm motions in patients with idiopathic scoliosis using dynamic breathing MRI. Spine. 2004;29(3):298–302.
10. Nachemson A. A long term follow-up study of non-treated scoliosis. Acta Orthop Scand. 1968;39:466–476.
11. Massery M. What's positioning got to do with it? Neurol Rep. 1994;18:11–14.
12. Seddon P, Khan Y. Respiratory problems in children with neurological impairment. Arch Dis Child. 2003;88(1):75–78.
13. Sawin PD, Menezes AH. Neuromuscular scoliosis: diagnostic and therapeutic considerations. Semin Pediatr Neurol. 1997;4(3):224–242.
14. Tsirikos A, Chang W, Dabney K, Miller F, Glutting J. Life expectancy in pediatric patients with cerebral palsy and neuromuscular scoliosis who underwent spinal fusion. Dev Med Child Neuro. 2003;45:677–682.
15. Cassidy C, Craig C, Perry A, Karlin L, Goldberg M. A reassessment of spinal stabilization in severe cerebral palsy. J Pediatr Orthop. 1994;14(6):731–739.
16. Strauss D, Cable W, Shavelle R. Causes of excess mortality in cerebral palsy. Dev Med Child Neuro. 1999;41:580–585.
17. Noble-Jamieson C, Heckmatt J, Dubowitz V, Silverman M. Effects of posture and spinal bracing on respiratory function in neuromuscular disease. Arch Dis Child. 1986;61:178–181.
18. Hardie J, Morkve O, Ellingsen I. Effect of body position on arterial oxygen tension in the elderly. Respiration. 2002;69(2):123–128.
19. Badr C, Elkins M, Ellis E. The effect of body position on maximal expiratory pressure and flow. Aust J Physiother. 2002;48(2):95–102.
20. Jones A, Dean E. Body position change and its effect on hemodynamic and metabolic status. Heart Lung. 2004;33:281–290.
21. Bjornson K, Deitz J, Blackburn S, Billingsley F, Garcia J, Hays R. The effect of body position on the oxygen saturation of ventilated preterm infants. Pediatr Phys Ther. 1992;4:109–115.
22. Remolina C, Khan A, Santiago T, Elelman N. Positional hypoxemia in unilateral lung disease. N Engl J Med. 1981;304:523–525.
23. Yeaw E. Effect of body positioning upon maximal oxygenation of patients with unilateral lung pathology. J Adv Nurs. 1996;23:55–61.
24. Harris J, Johansen J, Pederson S, LaPier T. Site of measurement and subject position affect chest excursion measurements. Cardiopulm Phys Ther. 1997;8:12–17.
25. Wolery M, Dunlap G. Reporting on studies using single-subject experimental methods. J Early Interv. 2001;24:83–89.
26. Weiss H. Effect of an exercise program on vital capacity and rib mobility in patients with idiopathic scoliosis. Spine. 1991;16:88–93.
27. Leopando M, Moussavi Z, Holbrow J, Chernick V, Pasterkamp H, Rempel G. Effect of a soft Boston orthosis on pulmonary mechanics in severe cerebral palsy. Pediatr Pulmonol. 1999;28:53–58.
28. Portney L, Watkins M. Single-subject designs. In: Foundations of Clinical Research Applications to Practice. 3rd ed. Upper Saddle River, NJ: Prentice Hall Health; 2008:235–275.
29. Tamura F, Shishikura J, Mukai Y, Kaneko Y. Arterial oxygen saturation in severely disabled people: effect of oral feeding in the sitting position. Dysphagia. 1999;14:204–211.
30. Edmonds Z, Mower W, Lovato L, Lomeli R. The reliability of vital sign measurements. Ann Emerg Med. 2002;39:233–237.
31. LaPier T, Cook A, Droege K, et al. Intertester and intratester reliability of chest excursion measurements in subjects without impairment. Cardiopulm Phys Ther. 2000;11(3):94–98.
32. Bloom M, Fischer J, Orme J. Designs for Comparing Interventions. Evaluating Practice Guidelines for the Accountable Professional. Boston, MA: Allyn and Bacon; 2006:477–538.
33. Ottenbacher K. Statistical analysis of single system data. In:Ottenbacher K, ed. Evaluating Clinical Change. Strategies for Occupational and Physical Therapists. Baltimore, MD: Williams & Wilkins; 1986:137–195.
34. Dean E. Body positioning. In:Frownfelter D, Dean E, eds. Cardiovascular and Pulmonary Physical Therapy. St Louis, MO: Mosby Elsevier; 2006:307–321.