Massage with kinesthetic stimulation improves weight gain in preterm infants



The aim of this study was to evaluate the effects of massage with or without kinesthetic stimulation on weight gain and length of hospital stay in the preterm infant.

Study Design:

A prospective randomized clinical trial was conducted evaluating the effects of massage with or without kinesthetic stimulation (KS) on weight gain and length of stay (LOS) in medically stable premature (<1500 g and/or 32 weeks gestational age) neonates. Infants were randomized either to receive no intervention (control), massage therapy alone (massage), or massage therapy with KS (M/KS). Linear regression analysis was performed to evaluate differences in the primary outcomes between the groups after controlling for covariates. Post hoc analysis with stratification by birthweight (BW> and <1000 g) was also performed.


A total of 60 premature infants were recruited for this study; 20 infants in each group. Average daily weight gain and LOS were similar between the groups after controlling for covariates. For infants with BW>1000 g, average daily weight gain was increased in the intervention groups compared to control. This effect was mainly attributable to the M/KS group.


Massage with KS is a relatively simple and inexpensive intervention that can improve weight gain in selected preterm infants. Length of hospital stay is not impacted by massage with or without KS. Further studies are needed to evaluate the effect of massage in the extremely low BW(<1000 g) infant.


The management and outcome of premature infants have changed in the postsurfactant era. With more premature infants surviving past the peripartum period, more focus has been devoted to optimizing the growth and development of this population. In addition to the physiologic consequences of preterm birth, the stressful environment and lack of tactile stimulation associated with care in the neonatal intensive care unit (NICU) may further compromise these vulnerable neonates. In the past two decades a number of studies have been conducted to examine the impact that tactile and kinesthetic stimulation (KS) has on the growth and development of premature babies.

Preliminary studies have suggested that massage therapy with KS may have positive effects on preterm infants including greater weight gain,1, 2, 3, 4, 5, 6, 7 improved bone mineralization,8 earlier hospital discharge,1, 2, 3 and more optimal behavioral and motor responses2, 3, 4, 9, 10 compared to controls. However, there is inconsistency of these findings across studies and methodological concerns with previous trials have led some authors to caution widespread and routine use of preterm infant massage.11 Previous studies have also varied in the type of intervention used, ranging from gentle ‘still touch’ to programs including physical activity. It has not been distinguished whether potential benefits are associated with massage alone or the combination of massage and KS or exercise.

We conducted a randomized controlled clinical trial to test the hypothesis that infant massage with or without KS (or exercise) can improve weight gain and decrease length of hospital stay in preterm infants.

Patients and methods


A prospective randomized controlled clinical trial evaluating the effects of massage with or without KS on preterm infants was performed at the George Washington University NICU between August 2003 and March 2007. This study was approved by the George Washington University Committee on Human Research Institutional Review Board and registered at

Study population

Subjects consisted of healthy preterm infants born at the George Washington University Hospital. Inclusion criteria for study participation were (1) birthweight (BW) <1500 g and/or gestational age32 weeks, (2) postnatal age>7 days and current weight>1000 g and (3) relative medical stability (that is, ‘feeders and growers’ that were no longer deemed at a critical stage in their care, including patients on nasal continuous positive airways pressure or naso/orogastric enteral feeds). Infants were excluded if they had a major congenital anomaly (including chromosomal abnormalities, neuromuscular disorders, congenital heart disease, neural tube defects and gastrointestinal malformations), or were restricted in their movement or ability to undergo the intervention (including those infants with pathological fractures, bony deformities, contractures). Informed consent was obtained from the parents of each patient enrolled in the study.

Study design

Enrolled patients were randomly assigned to receive no intervention (control), massage therapy alone (massage) or massage with exercise (M/KS) by a computer-generated random number table sequence. Treatment allocation was concealed in opaque, sequentially numbered, sealed envelopes until study entry. Informed consent was obtained from parents of eligible infants by the investigators or on-service neonatologists. Once enrolled, participants were assigned to their group by a research assistant. Clinical and demographic information were recorded including daily weight (measured by NICU nurses) and caloric and volumetric intake. Infants were fed with fortified breast milk or premature formula. Enteral feeding protocol consists of advancement by 20 ml kg−1 per day after initial stabilization. A period of trophic feeding is used in the extremely low birthweight (ELBW<1000 g) infant with target of achieving full feeds at approximately 14 days of life.

The massage group underwent a protocol that consisted of application of six strokes, each lasting 10 s, to the following areas of the baby in prone position: (1) head from crown to neck, (2) shoulders from middle of back to arms, (3) back from neck to waist, (4) legs from top of thighs to ankles and (5) arms from shoulder to wrist. The M/KS group received the massage protocol as described with the addition of KS, which consisted of transitioning the baby to a supine position and six movements of each arm at the elbow and leg at the knee. Infants were monitored continuously for heart rate, respiratory rate and percutaneous oxygen saturation throughout the intervention.

Massage and KS were performed by bedside registered nurses in the NICU who were trained by the same licensed massage therapist (BJ). This therapist trained all incoming staff nurses during the study period. Training included instruction on the study design as well as intervention techniques and procedures for each study group. All efforts were made to assign study infants to trained staff nurses. In the event that a study infant was assigned to an agency or traveling nurse, the charge nurse would perform the intervention during that shift. The LMT continued to visit the NICU 1 to 2 times per week during day and evening shifts to supervise the technique of massage performed by trained nurses. Additionally, an instructional video was available and encouraged to refresh training for nurses who had not performed the intervention regularly. The intervention was done two times per day for 15 min at a time from the time of study entry until discharge. The control group infants were managed via the nursery standard of care. Primary outcomes of average daily weight gain during the study period and length of stay (LOS) were assessed at discharge. Secondary outcomes of change in head circumference (HC) and length were also noted. Besides the bedside nurses performing the actual intervention, all other NICU personnel, including managing physicians, were not aware of the randomization code or the group orientation of their infants.

Sample-size calculation and statistical analysis

Our historical data showed that average daily weight gain in our NICU was 25±5 g per day and average LOS in our NICU was 45±6 days for <1500 g preterm infants. We proposed that the infants exposed to massage therapy±KS would gain an average of 20% more per day and have a shortened LOS by 15% when compared to controls. To detect this difference, we calculated that it would be adequate (power=0.8) to test both primary outcomes with a sample size of 60 infants: 20 infants in the control group and 20 infants in each intervention group.

Demographic and clinical information for the patients are described as means±standard error of the mean (s.e.m.) for continuous data and rates for categorical variables. Differences between the control and intervention groups were evaluated by the χ2-test for categorical variables and analysis of variance (ANOVA) for continuous variables. Posterior testing was performed by Tukey's test to evaluate mean differences between individual groups. Nonparametric analysis was also performed with the Kruskal–Wallis test (KW). Multiple regression analysis was performed to evaluate differences between the groups after controlling for effects of covariates. Due to the observed overrepresentation of more immature infants randomly assigned to the control group, post hoc analysis was performed evaluating effects of massage on infants stratified by BW (< or >1000 g). Data were analyzed using SPSS 12.0 for windows (Chicago, IL, USA).


There were 147 very low BW infants admitted to our NICU during the study period. Of these infants, 2 were excluded because of congenital anomalies, 20 expired and 21 were transferred to another institution for surgical reasons. Of the remaining eligible infants, 44 families declined consent and 60 infants were enrolled in the study, 20 in each group. Infants participating in the study were slightly smaller than those not enrolled (mean BW 1054±290 vs 1179±268 g, respectively, P=0.027, gestational age 28±2 vs 29±3 weeks, P=0.009), but gender distribution was similar. Two infants in the massage group were excluded after enrollment (one diagnosed with congenital cytomegalovirus requiring contact isolation and removal from the massage protocol, one with congenital hydrocephalus transferred to another hospital for neurosurgical intervention). The intervention was well tolerated and there were otherwise no withdrawals from the study. Demographic and clinical characteristics of the study patients are presented in Table 1. Postmenstrual age and weight at study entry were similar between the groups. The majority of infants (80%) had regained BW at time of study entry. No infants were more than 10% below BW at the start of study. There were no infants with necrotizing enterocolitis or severe (grades III or IV) intraventricular hemorrhage so these variables were not included in the regression models. Bronchopulmonary dysplasia (BPD) was defined per NICHD criteria as infants requiring oxygen (n=3) or nasal continuous positive airway pressure (n=3) at 36 weeks postmenstrual age.12

Table 1 Patient demographic and clinical information

Average daily weight gain over the study period was 28.9±1, 27.1±1.4 and 30±1.2 g for the control, massage and M/KS groups, respectively. These values were not significantly different by univariate analysis with ANOVA and KW or multiple linear regression controlling for the effects of gestational age, gender, caloric intake, BPD and sepsis. After stratification by BW, average daily weight gain was significantly higher in infants with BW >1000 g (ANOVA P=0.008, KW P=0.012; Figure 1). This difference was attributable to the M/KS group according to Tukey's probability table (mean difference 6.7 g kg−1 per day compared to control). This remained significant after controlling for the covariates in a regression model (Table 2). Average weight gain was not significantly different in the infants with BW <1000 g after controlling for covariates.

Figure 1

Average daily weight gain by group.

Table 2 Regression analysis for average daily weight gain in infants with BW>1000 g

Median LOS was not significantly different between the groups (Figure 2). The two outliers in the massage group both required transfer to subacute facility for oromotor dysfunction, all other infants were discharged to home. After excluding these outliers, LOS was significantly shorter in the intervention groups (ANOVA P=0.021, KW P=0.033). However this difference lost significance after controlling for gestational age, gender, sepsis and BPD in a regression model (Table 3). Likewise, LOS was not different between the groups after stratification by BW.

Figure 2

Length of stay by group.

Table 3 Regression analysis for length of stay

Change in HC and linear growth over the study period was similar between the groups (HC: 1±0.1, 0.8±0.1, and 0.9±0.1 cm per week; Length: 1.1±0.1, 0.9±0.4, 1.3±0.1 cm per week for the control, massage, M/KS groups, respectively, P>0.05).


In selected preterm infants, there appears to be a demonstrable effect on weight gain and this effect seems to be most notable in the group exposed to massage with KS. This is the first randomized clinical trial to demonstrate a difference in outcome for preterm infants that are exposed to massage with KS compared to massage alone. Although previous studies have demonstrated that massage in combination with KS is associated with improved weight gain compared to controls,1, 2, 3, 4, 5, 6, 7 our results suggest that this difference is mostly attributable to the effects of KS. This is consistent with results of other trials that have evaluated gentle still touch13 or massage alone and found no difference in weight gain between intervention groups and controls.10, 14 However, when massage has been combined with KS or physical activity, benefits have been more consistently demonstrated.

The importance of KS is clear when considering the proposed mechanisms by which massage improves growth and weight gain. Evidence suggests that improvements in weight gain are related to improved metabolic efficiency leading to acquisition of body mass. Infants receiving M/KS do not consume or retain more calories than controls, as ours and previous studies have shown similar caloric intake patterns between the groups.2, 3, 6, 7 Similarly, M/KS infants do not appear to conserve more calories by spending more time in the sleep state, because studies have actually shown increased arousal in massaged infants compared to controls.2, 3 Thus, it remains that the increase in weight gain is most likely related to hormonal alterations or changes in basal metabolic function. Increased activity levels have been associated with improved weight gain in both animal and human models. Increased growth hormones following physical activity has been shown in rat pups15, 16 and humans.17 Increased vagal activity inducing gastric motility has been demonstrated in infants exposed to M/KS.18 Animal models of repeated exercise have revealed that there is an overall anabolic effect on body protein metabolism leading to somatic growth.19 Improved bone mineralization and skeletal growth have been associated with programs of daily physical activity in preterm infants.8, 20, 21, 22, 23 Thus physical activity may improve weight gain through a variety of mechanisms affecting body mass of all types—adipose, muscle and bone.

Additional mechanisms have been proposed that provide a basis for the benefits of massage alone in the promotion of growth. Non-nutritive sucking leading to stimulation of proprioreceptors in the oral mucosa have been shown to increase gastrin, insulin and cholecystokinin release and similar hormonal changes have been shown in animals exposed to tactile stimulation in other parts of the skin.24 Stress behaviors may increase caloric expenditures affecting weight gain. Massage and other modalities of somatosensory input (for example, Kangaroo care) have been associated with attenuated adverse reactions to stress.3, 25 Decreasing cortisol levels have been demonstrated in preterm infants following massage.26 It is not clear whether these benefits would be observed in infants only exposed to KS.

Our findings suggest that the effects of massage alone are not sufficient to significantly impact weight gain, and that potential metabolic changes require addition of KS. This is consistent with other studies that evaluated daily physical activity programs alone, without the additional periods of massage. Moyer-Mileur et al.,20, 21 and others utilizing a similar protocol,22, 23 demonstrated that infants receiving 5 min of daily range-of-motion exercise had improved weight gain compared to control infants who received a 5 min daily interactive period of holding and stroking. One might argue that this ‘holding and stroking’ may not have the same effects as a systematic protocol of moderate pressure massage. Thus, our study included such a protocol for massage alone and a control group receiving no intervention to further elucidate the level of intervention required to promote weight gain.

These findings are important because recent meta-analyses11, 27 raised important methodological concerns with previous studies (for example, blinding of treatment allocation and control of performance bias via blinding of managing neonatologists to group assignment) that cast uncertainty on the benefits of massage and physical activity in preterm infants. The authors concluded that evidence was weak to support widespread routine use of these interventions and questioned whether providing massage was a cost-effective use of nursing and NICU staff time. Our findings provide further evidence that M/KS improves weight gain in selected preterm infants and demonstrate the importance of inclusion of KS or range-of-motion exercise in any protocol used. This may aid in defining the role of M/KS in NICU care. In consideration of nursing time, perhaps focus on abbreviated programs of physical activity without massage may be more practical and just as beneficial. Ferber et al.6 demonstrated that training of parents or caregivers is associated with similar effect size on weight gain in preterm infants when compared to infants massaged by trained professionals. Involvement of the parent may also lead to other benefits such as decreased parental stress and improved caregiver satisfaction, because there is evidence that similar interventions lead to improved mother–infant interaction.28 If parents and caregivers are to be trained in massage, focus on KS should be included.

We did not find an effect of massage/KS on weight gain in the overall study population. In post hoc analysis, this appeared to be due to the wide variability of the primary outcomes in the lower BW, more immature infants. Obviously, the most immature preterm infants have a range of comorbidities that could not be adequately controlled for in this study. At the time of study conception, the entry criteria were designed to capture infants of similar postmenstrual age at enrollment, allowing for evaluation of weight gain over a comparable time period in all infants. However, because weight gain and LOS are so highly variable in the smallest of infants, stratification by BW at study entry, with adequate sample size to account for this variability, would have lead to more interpretable results in this population. Future studies are needed targeting the ELBW population before conclusions can be made regarding the effects of massage in this population.

We did not find a significant difference in LOS between our intervention and control groups, in contrast with other studies.1, 2, 3 Similar to weight gain in the ELBW infant, this may be due to the higher variance than expected in this outcome. LOS is a difficult outcome to assess given that it depends not only on gestational age, but also on the medical and social condition of the infant and is, to some extent, parent and individual physician driven. These factors are difficult to control for in any trial design, and may be accountable for the variability in findings for this outcome. Similarly, we did not detect a difference in linear growth or HC between the groups. However, it should be noted that this study was not powered to detect differences in this secondary outcome. Although HC and linear growth may be a useful reflection of postnatal growth, these are more prone to measurement error and variability, making detection of meaningful differences more problematic. For this reason, weight gain was chosen as the primary outcome of interest.

Our findings add to previous evidence to support that preterm infant M/KS is associated with an improvement in weight gain in a subset of premature infants. The clinical impact of this benefit, especially if it does not lead to earlier hospital discharge, is unclear. Long-term follow up of patients is needed to evaluate if this effect on weight gain is sustained and translates into accelerated catch-up growth into childhood. Future studies evaluating other outcomes such as neurobehavioral and developmental effects should be undertaken with protocols that focus on KS with or without massage. Additionally, focused trials targeting the ELBW are needed.


Massage when combined with KS is associated with an improvement in daily weight gain in selected preterm infants. Length of stay is not impacted by massage.


  1. 1

    White JL, Labarba RC . The effects of tactile and kinesthetic stimulation on neonatal development in the premature infant. Dev Psychobiol 1976; 9: 569–577.

  2. 2

    Field T, Schanberg SM, Scafidi F, Bauer CR, Vega-Lahr N, Garcia R et al. Tactile/kinesthetic stimulation effects on preterm neonates. Pediatrics 1986; 77: 654–658.

  3. 3

    Scafidi FA, Field TM . Massage stimulates growth in preterm infants: a replication. Infant Behav Dev 1990; 13: 167–188.

  4. 4

    Scafidi F, Field T, Schanberg S . Factors that predict which preterm infants benefit most from massage therapy. J Dev Behav Pediatr 1993; 14: 176–180.

  5. 5

    Mathai S, Fernandez A, Modkar J, Kanbur W . Effects of tactile-kinesthetic stimulation in preterms: a controlled trial. Indian Pediatr 2001; 38: 1091–1098.

  6. 6

    Ferber SG, Kuint J, Weller A, Feldman R, Dollberg S, Arbel E et al. Massage therapy by mothers and trained professionals enhances weight gain in preterm infants. Early Hum Dev 2002; 67: 37–45.

  7. 7

    Dieter JNI, Field T, Hernandez-Reif M, Emory EK, Redzepi M . Stable preterm infants gain more weight and sleep less after five days of massage therapy. J Pediatr Psychol 2003; 28: 403–411.

  8. 8

    Aly H, Moustafa MF, Hassanein SM, Massaro AN, Amer HA, Patel K . Physical activity combined with massage improves bone mineralization in premature infants: a randomized trial. J Perinatol 2004; 24: 305–309.

  9. 9

    Scafidi FA, Field TM, Schanberg SM, Bauer CR, Vega Lahr N, Garcia R et al. Effects of tactile/kinesthetic stimulation on the clinical course and sleep/wake behavior of preterm neonates. Infant Behav Dev 1986; 9: 91–105.

  10. 10

    Solkoff N, Matuszak D . Tactile stimulation and behavioral development among low-birthweight infants. Child Psychiatry Hum Dev 1975; 6: 33–37.

  11. 11

    Vickers A, Ohlssom A, Lacy JB, Horsley A . Massage for promoting growth and development of preterm and/or low birth-weight infants. Cochrane Database of Systematic Reviews 2004, Issue 2. Art. No.: CD000390. doi:10.1002/14651858.CD000390.pub2.

  12. 12

    Ehrenkranz RA, Walsh MC, Vohr BR, Jobe AH, Wright LL, Fanaroff AA, et al., National Institutes of Child Health and Human Development Neonatal Research Network. Validation of the National Institutes of Health consensus definition of bronchopulmonary dysplasia. Pediatrics 2005; 116: 1353–1360.

  13. 13

    Harrison LL, Williams AK, Berbaum ML, Stern JT, Leeper J . Physiologic and behavioral effects of gentle human touch on preterm infants. Res Nurs Health 2000; 23: 435–446.

  14. 14

    White-Traut RC, Tubeszewski K . Multimodal stimulation of the premature infant. J Pediatr Nurs 1986; 1: 90–95.

  15. 15

    Pauk J, Kuhn CM, Field TM, Schanberg SM . Positive effects of tactile versus kinesthetic or vestibular stimulation on neuroendocrine and ODC activity in maternally-deprived rat pups. Life Sci 1986; 39: 2081–2087.

  16. 16

    Schanberg SM, Field TM . Sensory deprivation stress and supplemental stimulation in the rat pup and preterm human neonate. Child Dev 1987; 58: 1431–1447.

  17. 17

    Van Wyk JJ, Underwood LE . Growth hormone, somatomedins, and growth failure. Hosp Pract 1978; 13: 57–67.

  18. 18

    Diego MA, Field T, Hernandez-Reif M . Vagal activity, gastric motility, and weight gain in massaged preterm neonates. J Pediatr 2005; 147: 50–55.

  19. 19

    Young VR, Torun B . Physical Activity: Impact on Protein and Amino Acid Metabolism and Implications for Nutritional Requirements. Nutrition in Health and Disease and International Development: Sumposia from the XII International Congress on Nutrition. Liss: New York, 1981.

  20. 20

    Moyer-Mileur L, Luetkemeier M, Boomer L, Chan GM . Effect of physical activity on bone mineralization in premature infants. J Pediatr 1995; 127: 620–625.

  21. 21

    Moyer-Mileur LJ, Brunstetter V, McNaught TP, Gill G, Chan GM . Daily physical activity program increases bone mineralization and growth in preterm very low birth weight infants. Pediatrics 2000; 106: 1088–1092.

  22. 22

    Nemet D, Dolfin T, Litmanowitz I, Shainkin-Kestenbaum R, Lis M, Eliakim A . Evidence for exercise-induced bone formation in premature infants. Int J Sports Med 2002; 23: 82.5.

  23. 23

    Litmanovitz I, Dolfin T, Friedland O, Arnon S, Regev F, Shainkin-Kestenbaum R et al. Early physical activity intervention prevents decrease of bone strength in very low birth weight infants. Pediatrics 2003; 112: 15–19.

  24. 24

    Uvnas-Moberg K, Widstrom AM, Marchini G, Winberg J . Release of GI hormones in mother and infant by sensory stimulation. Acta Paediatr Scand 1987; 76: 851–860.

  25. 25

    Charpak N, Ruiz JG, Zupan J, Cattaneo A, Figueroa Z, Tessier R et al. Kangaroo mother care: 25 years after. Acta Paediatr 2005; 94: 514–522.

  26. 26

    Acolet D, Modei N, Giannakoulopoulos X, Bond C, Clow A, Glover V . Changes in plasma cortisol and catecholamine concentrations in response to massage in preterm infants. Arch Dis Child 1993; 68: 29–31.

  27. 27

    Schulzke SM, Trachsel D, Patole SK . Physical activity programs for promoting bone mineralization and growth in preterm infants. Cochrane Database of Systematic Reviews 2007, Issue 2. Art. No.: CD005387. doi:10.1002/14651858.CD005387.pub2.

  28. 28

    White Traut RC, Nelson MN . Maternally administered tactile, auditory, visual, and vestibular stimulation: relationship to later interactions between mothers and premature infants. Res Nurs Health 1988; 11: 31–39.

Download references


We thank Adenike Oloade, Inderjeet Sandhu and Mary Rivas for their assistance with patient enrollment and compiling and maintaining the dataset.

Author information

Correspondence to A N Massaro.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Massaro, A., Hammad, T., Jazzo, B. et al. Massage with kinesthetic stimulation improves weight gain in preterm infants. J Perinatol 29, 352–357 (2009) doi:10.1038/jp.2008.230

Download citation


  • preterm infants
  • very low birthweight
  • massage
  • kinesthetic stimulation
  • exercise

Further reading