INTRODUCTION AND PURPOSE
Chronic respiratory insufficiency in infants and young children has been attributed to conditions that affect the lungs (eg, bronchopulmonary dysplasia), the airway (eg, tracheomalacia), central regulation of breathing (eg, hypoxic ischemic encephalopathy), and the chest wall and thorax (eg, spinal muscular atrophy).1,2 Any of these conditions may require the use of mechanical ventilation and/or supplemental oxygen as well as the need for inpatient pulmonary rehabilitation.3
Most commonly, admission to an inpatient pediatric pulmonary rehabilitation program is for infants born prematurely who have been hospitalized in a neonatal intensive care unit (NICU). Similar to adult programs in step-down units, long-term acute care, or free-standing post–acute rehabilitation hospitals, pediatric inpatient pulmonary rehabilitation programs are designed to wean infants and young children from respiratory support as well as to promote overall health and function.3
In addition to respiratory care needs, infants and young children admitted to pulmonary rehabilitation programs often have delays in gross motor development and receive physical therapy (PT).4 Literature is limited, however, on the rehabilitation intervention for young children with chronic respiratory insufficiency who receive supplemental oxygen and/or are dependent on mechanical ventilation. Although the goal of hospitalization is to wean from respiratory support, PT intervention aimed at increasing the quantity and quality of movement and fostering motor skill development may result in an increased oxygen demand and, thus, an increase in heart rate (HR) and respiratory rate (RR). As with any physical therapist intervention, cardiopulmonary response to physical demands during movement activities should be monitored to determine the intensity of therapy that best matches and challenges the child.5
Few previous studies have examined cardiorespiratory responses to developmental activity and positioning for infants. In a study of 15 infants with chronic lung disease ready for NICU discharge, it was noted that there was no difference in oxygen saturation (SaO2) with prone or supine positioning for sleep.6 In another study evaluating the effect of positioning on SaO2 of infants born preterm with respiratory distress syndrome and mechanical ventilation in a NICU, SaO2 was found to be highest in a prone position rather than in the supine or side-lying positions.7 Finally, for infants in a special care nursery that were born premature, there was minimal difference between prone and supported sitting positions on HR, RR, and SaO2 values.8
Only 1 study has been reported that evaluated the effects of developmental PT activities on SaO2 and HR for infants born preterm and hospitalized in a NICU. No significant change in SaO2 was found during passive and active movements in the side-lying position, supported sitting, and rest for those infants; however, HR was significantly increased during intervention when compared to initial baseline rest.9 No studies have evaluated cardiorespiratory responses during PT intervention in older infants and young children with chronic respiratory insufficiency during weaning from mechanical ventilation or supplemental oxygen.
The purpose of this pilot study was to document intervention activities and cardiorespiratory responses to exercise and movement activities during PT intervention for infants and young children while hospitalized with chronic respiratory insufficiency requiring supplemental oxygen and/or mechanical ventilation in a pulmonary rehabilitation program.
Participants and Setting
Twelve infants and young children born between 24 and 34 weeks gestation and now ranging from 6 to 30 months chronological age admitted to an inpatient pulmonary rehabilitation program from local NICUs were identified by the pulmonary rehabilitation unit PT staff and enrolled over an 18-month period. Children admitted to the program using mechanical ventilation and/or supplemental oxygen and less than 3 years of age were eligible for inclusion. Infants and young children with unstable cardiac defects and/or severe or progressive neurological impairment were excluded. Consent from the child's guardian to participate was obtained prior to the child's enrollment in the study.
Eight of the included children were admitted to the pulmonary rehabilitation program requiring mechanical ventilation via tracheostomy, while 4 were admitted requiring supplemental oxygen via nasal cannula. During the observed PT sessions, all children were on oxygen mist via nasal cannula or tracheostomy tube (oxygen flow rate ranged from 0.7 L/min to 8.0 L/min, FiO2 = 21%-100%). Those children with a tracheostomy tube (n = 9) and oxygen were being weaned from mechanical ventilation and were using mechanical ventilation only at night.
Age was adjusted for 11 of the 12 participating children who were less than 24 months' chronological age (adjusted age range = 3-18 months) and 10 children had a gastrostomy tube (g-tube). All children scored between less than the 1st and the 61st percentile on the Peabody Developmental Motor Scales, second edition (PDMS-2) Gross Motor Subtests.10 Additional demographic characteristics of the individual children are detailed in Table 1.
Approximately 1 week prior to the first study measurement session, the child's primary physical therapist completed the 3 Gross Motor subtests of the PDMS-2,10 to provide baseline descriptive information on the gross motor skills of each participant. The PDMS-2 is a standardized norm-referenced assessment used to measure gross motor (reflexes, stationary, and locomotion) as well as fine motor skills for children from birth to 6 years of age. Reliability and validity for the PDMS-2 have been well-documented.10–14
Subsequently, within a 4-week period, 3 measurement sessions were carried out during the child's regularly scheduled therapy times. The child's primary physical therapist provided the intervention. During the 30- to 45-minute intervention session, HR, SaO2, and RR were recorded every 30 seconds from the screen's digital display of the individual electronic physiological monitoring devices that are standard practice within the pulmonary rehabilitation program (Phillips Intellivue MP30, Andover, Massachusetts). As needed, new sensors were applied just before the monitoring session with the oxygen sensor applied to the same extremity for all 3 measurement sessions. If a lead came off during the intervention or no data were evident on the screen, no reading was recorded for that parameter until the lead was reapplied or replaced and data were once again visually displayed on the monitor.
To ensure the reliability of HR, SaO2, and RR recordings, interrater reliability was tested over 8 sessions in which 2 physical therapists designated as measurement recorders for the study were present and documented cardiorespiratory responses at the same time. The 2 measurement recorders were physical therapists, each with more than 25 years of clinical experience and experience of reading the electronic monitors. Intraclass correlation coefficients (ICCs) for interrater reliability of measurement recording were as follows: 0.997 (95% confidence interval [CI] = 0.996-0.997) for HR (n = 320 measures), 0.980 (CI = 0.995-0.984) for SaO2 (n = 325 measures), and 0.923 (CI = 0.905-0.938) for RR (n = 344 measures).
In this pulmonary rehabilitation program, it is typical for children to receive PT twice per week. Physical therapy intervention is based on the strategies described in NICU practice,15–17 early intervention family-centered care,18,19 and motor development principles and practices.20–24 Intervention is directed at preventing or reducing impairments in flexibility, posture, strength, and sensation, while promoting appropriate positioning, motor development, and functional mobility; and participation in family life. Intervention also includes the application of orthotics, prescription of adaptive equipment, and caregiver instruction.5 Procedural (direct) interventions observed during this study consisted of therapeutic exercise (developmental activities training, motor training, and movement pattern training); balance and task-specific performance training; and strength and endurance training.
Initially, 1 of 2 standardized intervention plans developed by the study authors was used. The primary physical therapist determined which intervention plan was most similar to the child's current PT plan of care and gross motor abilities. Plan A was designed for younger participants and included activities in the supine, side-lying, and supported sitting positions. Plan B was designed for older participants and included activities in the supine, prone, sitting and kneeling and standing positions. Both plans included initial rest and rest periods throughout. Two children received intervention as outlined in Plan A and 2 children as outlined in Plan B for up to 2 of the 3 scheduled measurement sessions. As it was determined that the children did not need such frequent rest periods and children stayed more engaged if allowed to initiate and direct some of the intervention activities, the prescribed plans were discontinued and intervention was directed by the primary physical therapist for the remainder of the study.
Each PT session began and ended with a 2- to 5-minute initial rest period, primarily in a supine or sitting position. Physical therapy intervention occurred in each child's hospital crib or on a mat on the floor next to the crib. Standard PT equipment (eg, bolster, bench) and age-appropriate toys (eg, rattles, books) were used as indicated by the PT plan of care. In addition to HR, SaO2, and RR, intervention activities (categorized by the study authors as initial rest, rest, supine, prone, sitting, and kneeling/standing activities) were recorded (Table 2). Rest periods were given as the physical therapist providing the intervention deemed necessary (based on the child's responses to intervention), during which time cardiorespiratory measurements continued to be recorded every 30 seconds. No changes were made to the rate or flow of oxygen before, during, or after the PT sessions. The intervention did not occur on the same day as an invasive medical procedure or on a day when a child had an acute illness.
Descriptive statistics were used to portray the participants' characteristics, amount of time spent in each intervention activity, and cardiorespiratory responses to intervention. To account for missing data, nonnormality and variability inherent in the dependent variables, median HR, SaO2, and RR rate measurements were calculated across all sessions for the total group in each activity to compare to normal physiological parameters. To evaluate intersession reliability, we used the median values for each of the 6 intervention activities (initial rest, rest, supine, prone, sit, and kneeling and standing) during each observation session (1, 2, and 3) and calculated relative reliability [ICC (2,1) with CIs] for the 3 dependent variables (HR, SaO2, RR). An ICC (2,1) greater than 0.7 was considered reliable and indicated stability in the physiologic response across sessions. For those variables with an ICC of 0.7 or greater, the grand medians for each dependent variable during each intervention activity were then calculated and analyzed using a Friedman analysis of variance by ranks nonparametric test to determine whether there was a difference in HR, SaO2, and RR between activities. The .05 level of significance was used for all statistical tests. Bland-Altman plots25 were used to graphically examine absolute reliability across the multiple activity positions and across multiple sessions. The 95% limits of agreement Bland-Altman plots examine median difference between Session 1 and Session 3 for HR, SaO2, and RR. Medcalc (ver. 126.96.36.199; www.medcalc.org, Ostend, Belgium) software was used for the Bland-Altman analyses and plots.
Measurements were recorded for 11 children over 3 sessions and 1 child had measurements recorded over 2 sessions due to an early discharge. In total, 848 minutes of cardiorespiratory measures were recorded at 30-second intervals for a total of 1696 measurements. Measures taken during initial rest accounted for 15% (n = 127 minutes) of all measurements. The supine position was the most commonly used position for initial resting as side-lying rest (8 minutes total) and prone rest (2 minutes total) accounted for less than 1% of all initial rest activity recorded. Sitting was the most frequent intervention activity recorded (n = 390 measurements; 23%) (Table 3). All 12 children participated in sitting activities, 9 in prone activities, 10 in supine activities, and 5 in kneeling and standing activities during 1 or more of the measurement sessions.
No adverse effects occurred during the PT intervention and no activities had to be discontinued. While a small percentage of measures were outside normal physiological parameters (SaO2: n = 40, 3%; RR: n = 66, 4% below normal range, n = 115, 11% above normal range), no 2 measures in succession were outside the parameters. Median HR across subjects, activities, and observation sessions was 142 beats per minute (bpm) (range = 96–180). Median SaO2 was 98 (range = 71–100) and mean RR was 29 breaths per minute (range = 14–62).
Median HR measures for all 6 activities across the 3 observation sessions had an ICC 0.7 or greater, indicating intersession reliability. For absolute reliability (see Figure 1), the median bias was 1.9 bpm with median HR ranging from 18.3 bpm greater in session 3 to 22.2 bpm greater in session 1. Two values were outside of the range with 1 child exhibiting high initial rest HR, while another child exhibited a high session 3 resting HR. Median SaO2 measures greater than or equal to 0.7 included those for initial rest, rest, prone, and sitting. Bland-Altman absolute reliability of SaO2 is depicted in Figure 2. SaO2 ranged from a 7.0% increase in session 3 to a 6.9% increase in session 1. One child was highly variable and outside of the range for session 3 during prone, kneeling, and standing activities. Three other children were variable in 3 separate activities—supine, kneeling, and standing. No RR measures, however, approached a relative reliability of 0.7 (Table 4). The Bland-Altman measure for RR (see Figure 3) was similar to the HR measures. The average bias was 0.7 breaths per minute greater in session 3 with the limits extending from 12.5 breaths per minute greater in session 3 to 11.1 breaths per minute in session 1. Results of the Friedman analysis of variance for HR (6 activities) and SaO2 (4 activities) found no significant difference in HR (n = 7, χ2 = 2.273, df = 5, P = .810) or SaO2 (n = 6, χ2 = 8.837, df = 5, P = 0.116) between intervention activities.
The purpose of this pilot study was to document PT intervention activities and cardiorespiratory responses to exercise and movement activities during PT intervention for infants and young children being weaned from supplemental oxygen and/or mechanical ventilation. All of the children in this study were born prematurely and exhibited associated respiratory insufficiency severe enough to warrant admission to an inpatient pulmonary rehabilitation program.
The children in this study were described by their primary physical therapist as alert and interactive but all were nonverbal due to their age or tracheostomy tube. Nine of the 12 children (75%) scored at or below the 2nd percentile on the PDMS-2, indicating a significant delay in gross motor skill development. This motor skill delay may be due to neurological complications associated with prematurity26 and/or due to prolonged hospitalization with limited opportunity for motor skill development.27 It was noted by the PT staff that the children spent most of their time in supine in their crib or in a supported sitting position [eg, stroller, infant seat] throughout the day. Because of this delay in the attainment of motor skills, children are referred for PT services and therapists must develop and implement intervention programs that promote gross motor skill development as well as promote cardiopulmonary health. While not showing a delay on the PDMS-2, the other 3 children had physical impairments (eg, alterations in muscle tone), reduced endurance for functional mobility activities, and family instruction and equipment needs, thus were still receiving PT while hospitalized in the inpatient pulmonary rehabilitation program and were included in the study.
In general, a sitting position (sitting activities) was most frequently used during both rest periods and active PT intervention. This may be attributed to several factors. First, sitting allows for face-to-face interaction with handling and play activities, thus enabling the therapist to observe the child's facial expressions. This is important because all of the children in this study were nonverbal and unable to verbalize distress or discomfort. Visualizing the child's face allows the therapist to observe changes in color, tears, and distress via facial expression. A focus on sitting activities may also be due to the children's current level of gross motor development. Within “sitting activities,” we included supported sitting (which may include a focus on promoting head as well as trunk control), active independent sitting, transitioning in and out of sitting, and active reaching while sitting. Sitting activities allow the therapist to address several impairments (eg, trunk strength, balance, upper extremity active motion) and can be used with a wide age range and for children with a range of motor development abilities, such as those in this study.
Children in this study also participated in PT activities in prone, quadruped, and kneeling and standing positions. These activities were used to develop postural control, transitional movements and to develop a means of mobility such as crawling, creeping, and cruising. Although 9 of the 12 children participated in prone activities, the least amount of time during PT intervention was spent in prone activities. We have noted in this pulmonary rehabilitation program that, often, physical therapists are the first and only care providers to use the prone position. This may be due to a child's extended tracheotomy tube, short oxygen tubing, or the presence of a g-tube making prone positioning physically difficult to achieve or due to a child's presumed discomfort. In addition, prone activities may be avoided because they are medically contraindicated, or as a precaution to avoid vomiting if a g-tube feeding is in progress. In addition, while prone activities are encouraged to improve neck and trunk strength, they may prove to be more challenging than supine, side-lying, or sitting positions, specifically requiring the use of the cervical and thoracic extensor musculature, which, if not used frequently (eg, just during therapy), may have an effect on children's tolerance for the prone position.
Reference standard ranges of physiologic values for infants and children younger than 3 years are as follows: HR = 100-180 bpm, SaO2 = 87-100, and RR = 20-40 breaths per minute.28 Median values for the children in this study were within these parameters, but individual measures at the extremes of the observed ranges were not consistently within these reference standards. Therapists reported relying on clinical observations as well as the electronic monitors and adjusting activities as needed to the child's tolerance. It was noted that no 2 observed measures in succession were outside the parameters and these extreme highs and lows were most likely due to monitor error since children did not display signs of distress. We might hypothesize that the children in this study were not being taxed as activities were primarily self-directed and thus HR, SaO2, and RR did not remain outside the reference standard ranges.
Heart rate measures were reliable session to session but not significantly different between activities. This is different from the findings by Kelly et al,9 who did note differences in HR based on developmental activity. The sample in this study is older however, and less variability in HR is reasonable to expect. SaO2 measures were reliable across sessions for initial rest, rest, prone, and sitting activities but showed no difference between activities. The children were all receiving supplemental oxygen and were medically stable and, thus, we anticipated that SaO2 would stay in the reference standard range.
Because of the low intersession reliability of RR measures, we did not examine the difference in RR between intervention activities. This poor reliability could be attributed to several factors, including that children may have been receiving more or less oxygen support from session to session, depending on how the weaning was progressing and observation session 3 could have been as much as 4 weeks after session 1; the different amount of time spent in each activity from session to session; and/or because the electronic monitors were not adequately assessing respiratory rate during activity. Data related to reliability of the monitors were unavailable and we relied on the hospital maintenance schedule for safety monitoring and calibration.
We did have trouble with the monitor leads coming off or occasionally not providing a reading during intervention. When this occurred, no measures were recorded for that 30-second interval. This was sometimes attributed to the child's movement of the lower extremity (eg, kicking in supine) or during weight-bearing the oxygen sensor lead might come off for the children participating in kneeling and standing activities. Of the possible total 1,696 measures, 5% (n = 86) of HR measures, 9% (n = 147) of SaO2 measures, and 11% (n = 184) of RR measures were not documented.
Although a limited amount of physical therapy literature is available examining the effects of PT intervention9 and body position7,8 on the cardiorespiratory responses of preterm infants; this study examined intervention activities for older infants and young children who were born preterm and no longer in a NICU. Although this was a small sample, this study provides new information about PT intervention for children admitted to inpatient pulmonary rehabilitation and their cardiorespiratory responses to exercise and movement. This work may also be applicable to physical therapists who work with infants and young children discharged directly to their home from the NICU. We hope that this pilot work will allow for the development of a larger intervention study to directly evaluate the intensity of PT to improve the attainment of motor skills and cardiorespiratory endurance and potentially, earlier weaning from oxygen and mechanical ventilation.
Pediatric physical therapists commonly provide services for infants and young children with chronic respiratory insufficiency and gross motor delays admitted to inpatient pulmonary rehabilitation. Physical therapy intervention may consist of activities in supine, sitting, prone, and kneeling and standing positions. These young children who are in the process of weaning from supplemental oxygen and/or mechanical ventilation were able to tolerate current “standard of care” PT interventions with close monitoring of their cardiorespiratory responses by the physical therapist.
The authors thank the children and families who participated in this study. We also thank the pulmonary rehabilitation program staff for their cooperation.
1. Mallory GH, Stillwell PC. The ventilator-dependent child: issues in diagnosis and management. Arch Phys Med Rehabil. 1991;72:43–55.
2. Wheeler WB, Maguire EL, Kurachek SC, Lobas JG, Fugate JH, McNamara JJ. Chronic respiratory failure of infancy and childhood: clinical outcomes based on underlying etiology. Pediatr Pulmonol. 1994;17:1–5.
3. O'Brien JE, Haley SM, Dumas HM, et al. Outcomes of post-acute hospital episodes for young children requiring airway support. Dev Neurorehabil. 2007;10:241–247.
4. Dumas HM, Rosen EL, Haley SM, et al. Measuring physical function in children with airway support: a pilot study using computer adaptive testing. Dev Neurorehabil. 2010;13:95–102.
5. American Physical Therapy Association. Guide to Physical Therapist Practice. Rev 2nd ed. Alexandria, VA: American Physical Therapy Association; 2003.
6. Elder DE, Campbell AJ, Doherty DA. Prone or supine for infants with chronic lung disease at neonatal discharge? J Paediatr Child Health. 2005;41:180–185.
7. Bjornson KF, Dietz JC, Blackburn S, et al. The effect of body position on the oxygen saturation of ventilated preterm infants. Pediatr Phys Ther. 1992;4:109–115.
8. Stap LJ, Overbach A, Reinhart AP, et al. A comparison of prone and sitting on heart rates, respiration rates, and arterial oxygen saturation levels of premature infants. Pediatr Phys Ther. 1990;2:196–201.
9. Kelly MK, Palisano RJ, Wolfson MR. Effects of a developmental physical therapy program on oxygen saturation and heart rate in preterm infants. Phys Ther. 1989;69:467–474.
10. Folio M, Fewell R. Peabody Developmental Motor Scales: Examiner's Manual. 2nd ed. Austin, TX: rpo-Ed Inc; 2000.
11. Darrah J, Magill-Evans J, Volden J, et al. Scores of typically developing children on the Peabody Developmental Motor Scales: infancy to preschool. Phys Occup Ther Pediatr. 2007;27:5–19.
12. Wang HH, Liao HF, Hsieh CL. Reliability, sensitivity to change and responsiveness of the Peabody Developmental Motor Scales–Second Edition for children with cerebral palsy. Phys Ther. 2006;86:1351–1359.
13. Kolobe TH, Bulanda M, Susman L. Predicting motor outcome at preschool age for infants tested at 7, 30, 60, and 90 days after term age using the Test of Infant Motor Performance. Phys Ther. 2004;82:1144–1156.
14. Provost B, Heimerl S, Loez BR. Levels of gross and fine motor development in young children with autism spectrum disorder. Phys Occup Ther Pediatr. 2007;27:21–36.
15. Sweeney JK, Heriza CB, Blanchard Y; American Physical Therapy Association. Neonatal physical therapy. Part I: clinical competencies and neonatal intensive care unit clinical training models. Pediatr Phys Ther. 2009;21:296–307.
16. Mahoney MC, Cohen MI. Effectiveness of developmental intervention in the neonatal intensive care unit: implications for neonatal physical therapy. Pediatr Phys Ther. 2005;17:194–208.
17. Dusing SC, Murray T, Stern M. Parent preferences for motor development education in the neonatal intensive care unit. Pediatr Phys Ther. 2008;20:363–368.
18. Chiarello L, Effgen SK. Updated competencies for physical therapists working in early intervention. Pediatr Phys Ther. 2006;18:148–158.
19. Chiarello LA. Serving infants, toddlers, and their families: Early intervention services under IDEA. In: Campbell SK, Palisano RJ, Orlin MN, eds. Physical Therapy for Children. 4th ed. St Louis, MO: Elsevier; 2012:944–967.
20. McCoy SW, Dusing S. Motor control: developmental aspects of motor control in skill acquisition. In: Campbell SK, Palisano RJ, Orlin MN, eds. Physical Therapy for Children. 4th ed. St Louis, MO: Elsevier; 2012:87–150.
21. Lekskulchai R, Cole J. Effect of a developmental program on motor performance in infants born preterm. Aust J Physiother. 2001;47:169–176.
22. Arndt SW, Chandler LS, Sweeney JK, et al. Effects of a neurodevelopmental treatment-based trunk protocol for infants with posture and movement dysfunction. Pediatr Phys Ther. 2008;20:11–22.
23. Heathcock JC, Galloway JC. Exploring objects with feet advances movement in infants born preterm: a randomized controlled trial. Phys Ther. 2009;89:1027–1038.
24. Bartlett DJ, Kneale Fanning JE. Relationships of equipment use and play positions to motor development at eight months corrected age of infants born preterm. Pediatr Phys Ther. 2003;15:8–15.
25. Bland MJ, Altman DG. Agreement between methods of measurement with multiple observations per individual. J Biopharm Stat. 2007;17:571–582.
26. Kerstjens JM, de Winter AF, Bocca-Tjeertes IF, et al. Developmental delay in moderately preterm-born children at school entry. J Pediatr. 2011;159:92–98.
27. Carli G, Reiger I, Evans N. One-year neurodevelopmental outcome after moderate newborn hypoxic ischaemic encephalopathy. J Paediatr Child Health. 2004;40:217–220.
28. Dumas HM, Kelly MK. Children requiring long-term mechanical ventilation. In: Campbell SK, Palisano RJ, Orlin MN, eds. Physical Therapy for Children. 4th ed. St Louis, MO: Elsevier; 2012:756–780.
heart rate; humans; oxygen consumption; physical therapy modalities; posture; premature infant/physiology; respiratory insufficiency© 2013 Lippincott Williams & Wilkins, Inc.