Use of Comforting Touch and Massage to Reduce Stress in Preterm Infants in the Neonatal Intensive Care Unit Part Two
Research on the need for comforting touch for preemies in the NICU.
By Lynda Law Harrison, RN, PhD, FAAN
Researchers who study prematurely born human infants have reported that these infants receive very little comforting or contingent touch during their early days of life in the neonatal intensive care unit (NICU). Most of the touch that preterm infants experience in the NICU is associated with medical or nursing procedures (procedural touch), and is not contingent in the infant's cues.1-4 In recent years there has been increased interest in implementing principles of developmental care in NICUs. These principles include individualizing the care and handling that are provided to high risk infants and decreasing handling and stimulation that have adverse effects on the infants.5,6 Peters 7 summarized the literature related to handling in the NICU and suggested that there has been little change in the "handling frequency, patterns, and trends established in the mid 1970s, and that the principles of developmental care (eg, "decreasing infant disruptions and handling by caregivers, and modulating or attenuating infant responses to the care they receive," p 86) are still not being followed."
Preterm infants often show adverse responses to procedural touch; they include hypoxia, bradycardia, sleep disruptions, or increased intracranial pressure.2-4,8-19 The results of numerous animal studies suggest that providing appropriate tactile stimulation during the early days of life is critical for promoting optimal growth and development.13,18,19 This report summarizes the findings of studies that have described the types of nonprocedural touch received by hospitalized preterm infants and includes a review of studies that have evaluated infants' responses to gentle touch and massage interventions. Because the practice of skin to skin holding (Kangaroo Care [KC]) also includes a form of tactile stimulation (in addition to vestibular stimulation and oral stimulation related to breast feeding), a brief summary of studies of KC is included, although a comprehensive review of these studies is beyond the scope of this report. Newborn and Infant Nursing Reviews, Vol 1, No 4 (December), 2001: pp 235-241.
Nonprocedural Touch Received by Hospitalized Preterm Infants
Studies that describe nonprocedural touch in the NICU have focused on examining the types of handling provided by both NICU staff and by parents. Blackburn and Barnard 2 examined the caregiving that was provided to 102 preterm infants by reviewing time lapsed video recordings and reported that social stroking was the least frequently observed type of contact. A total of 25 of the infants received no social stroking during the six 24 hour observation periods. Wernern and Conway 4 reported similar findings based on two 55 minute observations of 11 preterm infants who were 23.5 to 28.5 weeks of gestational age (GA). Only 4.4% of the sensory stimuli provided by caregivers were described as comfort measures, including soothing touch, and only 3 of the 11 infants received any comforting touch. Such touch was provided by only 1 of the nurses who were observed.4
Harrison and Woods 20 examined the types of touch provided by parents and grandparents during 3 visits to their infants in the NICU and found that the types and amounts of touch varied considerably. The most frequent types of touch provided by the parents included holding, stroking, simple contact, or rubbing. Miller and Holditch-Davis 3 observed preterm infants over a 4 hour period in the NICU and compared the types of contact provided to infants by nurses and parents. Nurses spent more time providing care such as feeding, changing, bathing, taking vital signs, or providing high level care such as respiratory care or needle sticks. Parents provided more contact that was described as touching and holding.
Symon and Cunningham 21 used time lapsed video recordings to study the handling that was received by 12 infants in a NICU in Scotland on the first and/or third days after birth. The GAs of the infants ranged from 26 to 36 weeks (mean 30.75 weeks). The mean total length of time handled was 3.45 hours on day 1 and 3.56 hours on day 2. The number of handling periods ranged from 28 to 71, with a mean of 37 periods for nonventilated infants and 45.3 periods for ventilated infants. There was no correlation between infant GA and the duration of handling, although NICU staff perceived that there was such a correlation.
Zahr and Balian 22 observed a total of 55 preterm infants in 3 NICUs (2 in California and 1 in Lebanon ) and reported that infants received an average of 8.5 episodes of nursing procedures that involved handling during the 4 hour observation period. Infants in open warmers received more nursing interventions and handling than did infants in incubators, and handling during nursing interventions was associated with a significant decrease in oxygen saturation levels.
Kitchin and Hutchinson 23 conducted a qualitative, ethnologic study to examine the types of touch received by preterm infants during resuscitation. These researchers analyzed the videotapes of 10 resuscitations and coded the types of human and mechanical touch that were provided. They noted that most of the human touch that was provided was associated with tasks designed to ensure the infant’s survival, such as controlling the infant’s movement or evaluating the infant. They noted that there were few instances of protective human touch that was defined as touch with a “smooth, gentle, and caring quality” (p 48).
The results of these studies suggest that hospitalized preterm infants continue to be handled frequently throughout the day in the NICU, but that the infants receive minimal amounts of nonprocedural touch that might be expected to promote comfort. There is a need for further research to identify types of touch that could be provided to hospitalized preterm infants by parents and by nurses to promote comfort and minimize stressors associated with the NICU environment.24–26 Further research is also needed to determine whether implementing a developmental approach to care is associated with increased provision of nonprocedural, contingent, and comforting touch in the NICU.
Studies of Comforting Touch and Massage
Studies of comforting touch and massage have classified touch and massage as (1) still, gentle touch; (2) stroking/massage; and (3) stroking/massage combined with tactile or kinesthetic stimulation. Findings from these studies are difficult to compare because the samples have included infants of varying GAs and levels of morbidity, and because the types, amounts, and timing of the touch interventions have varied widely. Some samples have included very low–birth weight infants who might be considered physiologically fragile, whereas others have included infants of older post conceptional ages (PCA) who might be considered more physiologically stable.
Studies of Still, Gentle Touch
Six studies were identified that examined the effects of still, gentle touch without stroking or massage.27–32 Jay 30 provided gentle touch (placing 1 hand on the infant’s head and the other on its abdomen) to 13 preterm infants for 12 minutes, 4 times a day, for 10 days. The infants were mechanically ventilated, which suggests that they were physiologically fragile. Infants who received the gentle touch had higher hematocrit levels and required less oxygen than infants in a matched control group. Tribotti 32 provided gentle touch (placing 1 hand on the infant’s head and one on its lower back while the infant was prone) for 15 minutes, 3 times a day, for 3 days. During the first session infants had decreased transcutaneous oxygen (TcPO2) levels, but by the third session there were no changes in TcPO2 levels. This may suggest that the infants had habituated to the touch.
Harrison and colleagues conducted 3 separate studies to evaluate the effects of a gentle touch intervention similar to the one provided by Tribotti.27–29,32 The infants in these studies were from 26 to 33 weeks of GA at birth, and the touch was provided for 10 to 15 minute periods, once or 3 times a day, for 5 to 10 days. Findings from all 3 studies indicated that there were decreased levels of motor activity and behavioral distress during the touch, compared with baseline and post touch periods. In addition, there were no clinically significant changes in oxygen saturation or heart rate levels during the touch.
In the Harrison et al study,28 3 infants were randomly assigned to the gentle touch experimental (E) group, and 3 were assigned to a control (C) group. Infants in the E group had fewer total days nonsupplemental oxygen, greater mean daily weight gain, decreased serum cortisol levels, fewer blood transfusions, and fewer days in the hospital. However, this finding may be due to the fact that group E infants had slightly higher GAs and birth weights. There were no differences between group E and group C infants on any of these variables in the other 2 studies of gentle touch that were conducted by Harrison and colleagues. The samples in these studies included 30 and 84 infants, respectively.27,29
Modrici-McCarthy 31 provided a similar gentle touch intervention to 10 infants (27 to 32 weeks of GA) for 20 minutes daily for 10 days. Like Harrison and colleagues, Modrici-McCarthy found that infants had less motor activity and behavioral distress and more quiet sleep during touch, compared with baseline periods. However, there were no differences between the 10 group E infants and 10 infants in a randomly assigned group C on outcome variables including use of supplemental oxygen, weight gain, or number of blood transfusions.
The findings from these studies suggest that gentle touch has immediate positive and comforting effects, and that it is a safe type of touch for physiologically fragile infants. However, the findings do not indicate that interventions consisting of still, gentle touch improve longer term outcomes such as weight gain, morbidity status, or length of hospital stay.
Studies of Stroking/Massage Interventions
Studies of stroking and massage interventions have been reported for
the past 26 years and have included examination of both immediate and
longer term effects of the tactile stimulation. The findings from these
studies are contradictory. The results of some studies suggested positive
immediate effects of the interventions, such as decreased apnea or increased
oxygenation. 33,34 Other studies suggested positive longer term outcomes
including increased weight gain, 35,36 improved body tone, habituation,
alertness and consolability,37 and increased motor development.38 However,
findings from other studies suggested negative immediate and longer term
effects, including lower oxygen levels and more behavioral distress.39–41
Immediate Effects
Kattwinkel and colleagues 33 provided 5 minutes of stroking 3 times each hour for 3 hours to 6 preterm infants who were 26 to 31 weeks of GA at birth, and 2 to 35 days old at the time of the intervention. 33 Infants had decreased apnea during the intervention periods. Daga and colleagues 34 studied the effects of maternal stroking of the infant’s back during alternate gavage feedings on 7 preterm infants who were less than 32 weeks of GA.34 Oxygen saturation levels were significantly higher in infants who received the stroking at 20 and 30 minutes after the feeding, compared with infants in a control group.
In contrast, others have reported that preterm infants (27 to 32.5 weeks GA) had lower oxygen levels, more behavioral signs of distress (eg, grunting or gasping), and more movement and state transitions during periods of stroking or stroking and talking simultaneously, compared with periods of only visual and/or auditory stimulation. 39–41
Hayes and Adamson-Macedo 42 reported a case study of a 27 week GA infant who was observed under 4 different tactile stimulation conditions using a counterbalanced study design over a 3 day period that began when the infant was 5 days old. The conditions included a cephalocaudal stroking procedure, called “touching and caressing — tender in caring” (TAC TIC) therapy, with light stroking pressure; TAC TIC therapy with deep stroking pressure; a comfort condition that involved 3 minutes of still, gentle touch; and a control condition of spontaneous activity when the infant was lying alone with no intervention. There were no changes in mean heart rate values across the 4 conditions, or before, during, or after any of the conditions. However, there were more distress behaviors during the TAC TIC deep stroking procedure, compared with the control or comfort positions, and there were fewer distress behaviors during the comfort condition than during the spontaneous condition.
The differences in findings across these studies may be due to differences in the GAs or morbidity status of the infants who were studied, or in the type and amount of stroking that was provided. The differences might also be explained by differences in the environments of the NICUs in which the studies were conducted.