SciELO - Scientific Electronic Library Online

vol.14 issue3Effect of high-voltage pulsed current plus conventional treatment on acute ankle sprain author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Brazilian Journal of Physical Therapy

Print version ISSN 1413-3555

Rev. bras. fisioter. vol.14 no.3 São Carlos May/June 2010 



Effect of aquatic physical therapy on pain and state of sleep and wakefulness among stable preterm newborns in neonatal intensive care units



Carine VignochiI; Patrícia P. TeixeiraII; Silvana S. NaderIII

IDepartment of Physical Therapy Universidade Luterana do Brasil (ULBRA), Canoas (RS), Brazil
IIDepartment of Neonatology, Hospital Universitário, ULBRA
IIIDepartment of Pediatrics, ULBRA

Correspondence to




OBJECTIVE: To evaluate the effects of aquatic physical therapy on pain and on the cycle of sleep and wakefulness among stable hospitalized premature infants.
METHOD: This study was characterized as an uncontrolled clinical trial on a time series and included 12 clinically stable newborns of gestational age less than 36 weeks who were hospitalized in a neonatal intensive care unit (NICU). After selection, the newborns were placed in a liquid medium for aquatic physical therapy lasting 10 minutes. Movements to stimulate flexor posture and postural organization were performed. The sleep-wakefulness cycle was assessed using the adapted Brazelton (1973)* scale and pain was assessed by the occurrence of signs of pain according to the Neonatal Facial Coding System (NFCS) scale; and physiological parameters.
RESULTS: In relation to states of sleep and wakefulness, before the physical therapy, the newborns' behavior varied from fully awake with vigorous body movements to crying. After the physical therapy, the states of sleep ranged from light sleep with closed eyes to some body movement. These values presented statistically significant differences (p<0.001). The score on the pain assessment scale also decreased from 5.38±0.91 to 0.25±0.46, with p < 0.001 after the intervention. The vital signs remained stable.
CONCLUSION: It is suggested that aquatic physical therapy can be a simple and effective method for reducing pain and improving sleep quality among preterm infants in NICUs. Controlled studies with larger numbers of subjects are needed in order to generalize the results.
Article registered of the Clinical Trials under the NCT00785837.

Key words: aquatic physical therapy, preterm, neonatology.




According to Gaspardo, Linhares and Martinez1, there is evidence that newborns, even preterm ones, have the neurological ability of feeling pain. Pain reduction among newborns avoids potential risks and waste of energy in an attempt to compensate the difficulty of adaptation to the extra-uterine environment. By studying pain among newborn infants, Anand2 concluded that it is one important harmful factor of the extra-uterine environment. In addition, when pain is not treated, it can cause several deleterious effects, such as metabolic alterations, increase in the level of circulating hormones and in the susceptibility to infections, alterations of cerebral blood flow, hypoxia and modifications in the patterns of sleep and wakefulness, other than behavioral changes.

Procedures used to relieve pain can improve homeostasis and stability among newborns and are essential for the immature neonates' care and support1. Among the various pain relief procedures, the improvement in sleep cycles is essential for the subject's neurodevelopment, learning skills, memory and preservation of brain plasticity2. Studies indicate that bathing techniques can reduce crying and anguish and improve behavior and sleep quality1. The tactile-kinesthetic stimulation has shown itself to be effective in pain reduction3.

Due to the complexity of neonatal intensive care units (NICUs), pain relief procedures and multidisciplinary therapy programs focusing on the newborns' comfort are necessary.

This study intends to assess the effects of aquatic physical therapy on the improvement of sleep quality and pain reduction among stable premature infants hospitalized in NICUs.



The study was an uncontrolled clinical trial on a time series, in which the participants acted as their own controls before undergoing the physical therapy intervention. The study sample was recruited by convenience through a consecutive selection of newborns that fulfilled the inclusion criteria.

The study was conducted at the NICU of the Hospital Luterano in Porto Alegre (RS) - Brazil, from July 2005 to June 2007. A sample of 12 subjects was estimated based on a previously performed pilot study, which considered a 0,99 effect size and standard deviation of 0.313 to provide an 80% power for detecting a difference in neonatal facial coding system (NFCS) scores after the therapeutic procedure (alpha of 0.05).

The inclusion criteria were: clinically stable premature newborns with gestational age of less than 36 weeks, who presented behavioral abnormalities and were given permission to bathe by the pediatrician. Examples of behavioral abnormalities included intolerance to touch, excessive crying, pain signs according to the NFCS scale and difficulty to change from a state of crying and agitation to deep sleep after 60 minutes or less. The presence of these abnormalities was assessed by pediatricians and physical therapists at the NICU. The exclusion criteria were: clinically unstable premature infants who presented temperature alterations, contraindications to immersion baths, neurological problems, or who needed ventilatory support, as well as those with signs of infectious process or facial congenital malformations that would derail the use of the NFCS assessment scale.

Some physiological parameters were assessed in this study, such as mean blood pressure (MBP), body temperature (TAX), cardiac frequency (CF), saturation of oxygen (O2 SAT) and breathing frequency (BF). The NFCS, developed by Grunau & Craig, was used to assess pain3. Scores on the NFCS are determined by the presence or absence of eight facial movements: frowning, eyelid compression, deepening of the nasolabial furrow, open lips, vertical or horizontal mouth stretch, tense tongue and trembling of the chin (i.e. one point is assigned for each observed facial movement). The presence of pain is considered when the score exceeds three points (Appendix 1).

The neonatal behavioral assessment scale (NBAS), adapted from Brazelton4, was used for the analysis of the newborns' behavior during sleep and wakefulness states. This scale provides a score for each state of sleep or wakefulness of the infant (Appendix 2).

To perform the aquatic physical therapy, a standard plastic cot was used. The cot was sterilized with a solution of chlorhexidine and 70% alcohol and placed next to the incubator. The water temperature was kept at 37 ºC by use of a water thermometer model zls-1270.

For the verification of the physiological variables (MBP, TAX, CF, O2 SAT and BF), a Dixtal monitor with a pediatric adaptor was used. The researchers responsible for this procedure were specialists in neonatology who had received a 30-day training to follow the protocol of aquatic physical therapy and to use the assessment scales.

The infants who fulfilled the inclusion criteria were selected to receive the study intervention. Participants were wrapped in a towel with their bodies semi-inflected and then were gently placed in the liquid environment, where the physical therapy was applied for 10 minutes. The aquatic physical therapy included smooth and slow movements of the trunk and pelvis. The participant performed sliding movements in the liquid environment, which also promoted tactile-kinesthetic stimulation and facilitated a flexed posture induced by buoyancy. At the end of the procedure, the participants were removed from the cot and, while maintaining their posture, they were wrapped in a towel and taken to the incubator. The variables related to pain, sleep quality and physiological parameters were registered at the following time-points: 15 minutes before treatment; during the aquatic physical therapy; immediately, 30 and 60 minutes after treatment.

In order to compare the mean pain assessed by the NFCS at the different time-points, a repeated measures analysis of variance (ANOVA) was performed. Using the same statistical test and time-points, the mean values of physiological variables were also evaluated. For the results of the NBAS, the Friedman test with correction for Kendall's coefficient of concordance was applied, due to its equivalence to the repeated measures ANOVA for categorical variables. For data analysis, the Statistical Package for the Social Sciences (SPSS) software was used.

The study was approved by the Research Ethics Committee of the Universidade Luterana do Brasil (ULBRA), Canoas, (RS), Brazil, in December 2006 (CEP-ULBRA 2006-397H). Newborns whose guardians read and signed the informed consent form were included in the study.



The sample consisted of 12 premature newborns (six female and six male) with mean gestational age of 31.25 weeks (Table 1).



Physiological variables

Mean blood pressure

Before the aquatic physical therapy procedure, the MBP was 56.63±8.03. After aquatic physical therapy, a small decrease in the MBP was observed (55.57±4.5), but this difference was not significant (p=0.228) (Table 2).



Body temperature

The mean TX (measured in degrees Celsius) was 36.23 ± 0.49 before the aquatic physical therapy procedure, and 36.57±0.19 after the procedure. This difference was not significant (p=0.76) (Table 2).

Cardiac frequency

Before the aquatic physical therapy procedure, the mean CF (measured in heart beats per minute) was 172.13±12.12. The mean CF was decreased to 157.25±16.21 five minutes after the procedure (p=0.013), and to 147.63±1.10 30 minutes after the procedure (p=0.001) (Table 2).

Saturation of oxygen

The mean O2 SAT was significantly increased from pre- to post-intervention until 60 minutes after the aquatic physical therapy procedure (p<0.001). The mean O2 SAT was 91%±3.66% before the procedure and 95.75%±3.24% five minutes after the procedure (p=0.002) (Table 2).

Breathing frequency

A statistically significant reduction was observed in BF after the end of the aquatic physical therapy procedure (p<0.001). The mean BF (measured in respiratory movements per minute) was 54±3.54 before the procedure, 48.88±5.24 five minutes after the procedure (p=0.006), 45.88±4.99 30 minutes after the procedure (p=0.001) and 44.38±4.77 60 minutes after the procedure (p<0.001) (Table 2).

Neonatal behavioral assessment scale adapted from Brazelton4 (NBAS)

Figure 1 describes the changes observed in median NBAS scores at the different time-points. Before the physical therapy procedure, the newborns' behavior varied from fully awake with vigorous body movements to crying (scores ranging from 5 to 6 points). During the procedure, the median NBAS score were reduced, with the most prevalent behaviors being a wakefulness state with eyes open and minimum body movements, followed by a state of sleep with eyes opening and closing (scores of 3 and 4). At the end of the procedure, the sleep states varied between light sleep with eyes closed and some body movement and deep sleep with no movement and regular breathing (scores of 1 and 2). These differences were statistically significant (p<0.001).



Neonatal facial coding system (NFCS)

The mean score on the NFCS was 5.38±0.916 before the physical therapy procedure. During the first minutes of physical therapy, the mean NFCS score decreased to 4±1.85. At the end of the procedure, the mean score was reduced to 1.38±1.18 (p<0.001) when compared to the mean pre-intervention score. Further reductions were observed 30 minutes (0.38±0.74; p<0.001) and 60 minutes (0.25±0.463; p<0.001) after the procedure (Table 3 and Figure 2).






During the physical therapy procedure, the newborns' behavior varied from a crying state to a wakefulness state with open eyes and organized body movements. This result is consistent with that found in Sweeney's study5, which observed the maintenance of the alert and quiet state during hydrotherapy, as opposed to the sleep state demonstrated during immersion baths without body movements. After several procedures performed in newborns presenting changes in tonus, the same author observed an improvement in spasticity when the subjects underwent aquatic physical therapy. The newborns obtained, in a shorter period of time and with less effort, semi-flexed postures and improvements in the performance of spontaneous movements in a liquid environment5. Other studies investigating bathing techniques in newborns found reductions in crying and anguish and improvements in behavior6. According to Béziers and Hunsinger7, the water environment facilitates the movement and also allows infants to recover the feeling of their fundamental movements and continue to develop their "motor coordination".

When it comes to sleep and wakefulness cycles, the REM and non-REM cycles are considered essential to the subject's neurodevelopment, learning skills, memory and preservation of brain plasticity8. Nedive9 states that the facilitation and protection of sleep and sleep cycles are essential, in the long term, to learning skills and to cerebral development by the preservation of brain plasticity. In the present study, newborns' sleep and wakefulness cycles changed to a light sleep state with eyes closed and some body movement immediately after the aquatic physical therapy, and after some time, the newborns acquired a deep sleep state with regular breathing.

According to the NFCS scores, there was a significant decrease in pain after the aquatic physical therapy. The physiological effects of water cause pain relief and general muscle relaxation by the reduction of sensibility in nerve endings and the increase of peripheral circulation, which provides more blood to the muscles10. According to Sobrinho, Kreulish e Habib11, as pain receptors are located in the skin, physical approaches to reduce pain are basically related to the spinal dorsal horn. The stimulation of larger afferents that deliver less harmful messages can reduce the impact of smaller afferents that deliver pain messages. Therefore, painless stimulus such as touch can be used to ease pain.

Aquatic physical therapy has shown to be safe when it comes to vital signs. In 1983, Sweeney5 compared exercises in water with immersion only and found normal vital signs values during both procedures. However, the authors observed a decrease in mean blood pressure during the immersion procedure. The temperature was kept at normal values, without variations. This finding might be related to the water temperature, which was closer to the infants' body temperature. According to Masi12, if the water is at the same temperature as the subject's body, the convection and conduction mechanisms are not operating in heat loss. These mechanisms, however, are very important for heat dissipation during aquatic exercises due to the great body surface in contact with water and the conduction capacity of the environment. According to Cloherty and Stark13, loosely wrapping the infant in a warm wet cloth, as performed in this study, assists the child's relaxation and temperature stability.

Before the procedure, the infants demonstrated tachycardia, with CF of 172.13±12.11, probably due to crying and/or presence of pain. This finding matches the findings of Clohery and Stark's13 study, where newborns' CF was increased during crying. Studies suggest that pain and stress aggravate the clinical instability of newborns in intensive care14. In the present study, after the physical therapy intervention, the subjects' CF decreased, especially 30 and 60 minutes after the intervention. This finding is in line with those of Sweeney et al.5, who detected a decrease in heartbeats after hydrotherapy. Caromano et al.10 observed that exercises in warm water could modify physiological responses such as CF, however this did not occur in the present study. It is suggested that the decrease in physiological responses is associated to changes in behavior (i.e. crying to deep sleep).

The BF was significantly reduced after the aquatic physical therapy, especially 60 minutes after the procedure. As previous studies demonstrate that tachypnoea can be a consequence of anxiety or pain15, this finding might be related to pain reduction and the change in neonatal behavior to a deep sleep state.

Before the aquatic physical therapy, the infants were agitated and tearful, demonstrating a mean O2 SAT of 91%±3.66%. After the aquatic physical therapy, there was a statistically significant increase in O2 SAT. It is believed that this result is related to the decrease in the neonates' crying and pain. Authors report that the O2 SAT decreases due to the stressful environment of the NICU2. According to Campion16, the therapeutic effects of either exercising in warm water or simple immersion improve blood circulation and respiratory muscle functioning. Beziers7, Getz, Hutzler and Vermer17 also argue that the water environment increases vital capacity and reduces body weight, allowing the newborns to recover the movements performed during their fetal life.

The present study demonstrated that the aquatic physical therapy was safe for preterm newborns. However, the indication for this technique should be precise and follow well-established criteria in addition to the caution with water temperature, cot sterilization and observation of vital signs.

The present study has several limitations that need to be acknowledged. First, it is possible that the scores on the pain assessment scale may have been biased due to the difficulty in identifying and scoring the newborns' facial movements (i.e. observer bias). Second, the limited sample size makes it difficult to generalize the results (i.e. low external validity). A total of 14 and 24 patients would be required to provide 80% power and 90% power to detect important differences, respectively; whereas the actual sample size (n=12) provided only 77.3% power with an alpha value of 0.05. The main limitation of this study was the absence of a control group. As such, the observed effects could have been due to factors other than the intervention provided (e.g. change in the environment conditions, effect of the time, oscillations of the circadian cycle, etc.). Based on the discussions that were raised and the limitations of this study, future controlled studies with larger sample sizes are needed to examine the presence of possible confounding factors that could interfere with the results.



The aquatic physical therapy was shown to be safe and effective in reducing pain and improving sleep quality among premature infants hospitalized in NICUs. This intervention, when well performed, can be used as a non-pharmacological method for pain relief and improvement in deep sleep quality and duration. As such, this procedure can contribute to the multidisciplinary principles of humanization in NICUs.

Pain is a sensory experience and can be managed by pharmacological or non-pharmacological procedures, or by the simultaneous use of both. Non-pharmacological interventions, such as tactile approaches, massages, non-nutritive sucking and immersion baths aim to prevent the aggravation of a painful process, disorganization of the neonates, stress and agitation; in other words, to minimize the pain repercussions. In the present study, the improvement in physiological parameters, pain and sleep quality after the intervention suggest that aquatic physical therapy can be beneficial to stable preterm newborns. This technique can contribute to the reduction of harmful effects among the infants, without depriving them of the tactile-kinesthetic stimulation necessary to their neurodevelopment. More controlled studies are necessary to prove these benefits.



Lidiane Zanelatto, Rafael Luiz Souza Do Amaral and Paulo Nader



1. Gaspardo CM, Linhares MBM, Martinez FE. A eficácia da sacarose no alívio de dor em neonatos: revisão sistemática da literatura. J Pediatr (Rio J). 2005;81:435-42.         [ Links ]

2. Anand KJ. Effects of perinatal pain and stress. Prog Brain Res. 2000;122:117-29.         [ Links ]

3. Grunau RE, Weinberg J, Whitfield MF. Neonatal procedural pain and preterm infant cortisol response to novelty at 8 months. Pediatrics. 2004;114(1):e77-84.         [ Links ]

4. Brazelton TB. Neonatal Behavioral Assessment Scale. London: Clinics in Developmental Medicine; 1973.         [ Links ]

5. Sweeney JK. Neonatal hydrotherapy: an adjunct to developmental intervention in an intensive care nursery setting. Phys Occup Ther Pediatr. 1983;3(1):39-52.         [ Links ]

6. Santos PF, Cunha MCB, Franco RC. Estimulação e aquisição do controle cervical e do rolar por meio do método Halliwick associado ao conceito Bobath, em crianças com atraso motor. Fisioter Bras. 2006;7(3):224-8.         [ Links ]

7. Béziers M, Hunsinger Y. O bebê e a coordenação motora: os gestos apropriados para lidar com a criança. São Paulo: Summus; 1994.         [ Links ]

8. Graven S. Sleep and brain development. Clin Perinatol. 2006;33:693-706.         [ Links ]

9. Nedive E. Molecular analysis of developmental plasticity in neocortex. J Neurobiol. 1999;41(1):135-47.         [ Links ]

10. Caromano FA, Themudo-Filho MRF, Candeloro JM. Efeitos fisiológicos da imersão e do exercício na água. Fisioter Bras. 2003;4(1):1-5.         [ Links ]

11. Sobrinho GM, Kreulich LS, Habib AL. O efeito do toque na dor do bebê. Fisioter Bras. 2004;65:26-35.         [ Links ]

12. Di Masi F. Hidro: propriedades físicas e aspectos fisiológicos. Rio de Janeiro: Sprint; 2000.         [ Links ]

13. Cloherty J, Stark AR. Manual de neonatologia. Minas Gerais: MEDSI; 2000.         [ Links ]

14. Calasans MTA, Kraychette DC. Dolor de recién-nacido. Revista Recrearte. 2005;4:01-11.         [ Links ]

15. Scalan GD, Stoller J, Wilkins R. Fundamentos da terapia respiratória de Egan. São Paulo: Manole; 2000.         [ Links ]

16. Campion MR. Hidroterapia: princípios e prática. São Paulo: Manole; 2000.         [ Links ]

17. Getz M, Hutzler Y, Vermeer A. Effects of aquatic interventions in children with neuromotor impairments: a systematic review of the literature. Clin Rehabil. 2006;20(11):927-36.         [ Links ]



Correspondence to:
Carine Moraes Vignochi
Rua Dona Gabriela, 264/504
CEP 90850-010, Porto Alegre (RS), Brazil

Received: 24/10/2008
Revised: 31/07/2009
Accepted: 18/08/1009



* Brazelton TB. Neonatal Behavioral Assessment Scale. London: Clinics in Developmental Medicine; 1973.





Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License