The practice of infant massage is not a new phenomenon. It is a part of nature—at birth, mammals massage their newborns by licking and grooming them to encourage their body systems to normalise (Ishikawa and Shiga, 2012). In humans, midwives ‘massage’ newborns through drying to stimulate a response to take their first breath.
Early records of massage practice and research are diverse. It has been documented as early as 2760BC in China (Mitzel-Wilkinson, 2000). In Asia, infant massage is a long-established mothering tradition, passed down from generation to generation (Porter, 1996). Influenced by Florence Nightingale, massage training was provided by nurses and physicians for many health-related conditions during the 1880s and into the 1900s (Ruffin, 2011).
In a series of experiments with monkeys, Harlow (1958) found tactile stimulation in mother–infant interaction improved confidence and secure emotional behaviour. In the 1930s, research studies suggested that massage therapy could increase blood circulation and reduce muscle atrophy (Field et al, 2007).
More recent research has further suggested that the practice of infant massage provides benefits for both mother and baby. The popularity and demand for infant massage resulted in the establishment of the International Association of Infant Massage (IAIM) in 1986, whose membership spans more than 40 countries (McClure, 2001).
Infant skin
Human skin is the largest organ of the body and protects its internal components (bones, muscles, ligaments, blood vessels and internal organs) from injury (Lewis-Jones, 2012). The skin has three layers: the epidermis; the dermis; and the hypodermis. The epidermis is the first point of contact for the application of topical products. The stratum corneum is the outermost layer of the epidermis (the visible part of the skin). Its main function is to act as a barrier to penetration by external irritants and to protect against excessive water loss. Skin barrier function can be affected by genetic and environmental factors. Examples of the latter include water quality, pollution, detergents and the application of skin care products.
At term birth, the skin is sufficiently mature to withstand extrauterine life; however, infant skin does not become comparable to that of an adult until approximately 12 months of age (Stamatas et al, 2010). During this time, infant skin is more vulnerable than adult skin because it is different in several ways. For example, the stratum corneum is 30% thinner in neonates and the epidermis is 20% thinner (Stamatas et al, 2010). This puts infants at a greater risk of permeability and dryness than adults. In addition, the neonatal body surface to body weight ratio is greater than in adults, and infant skin has a greater absorption rate than that of adults. The consequence of this difference is an increased vulnerability to the effects of topical treatments (Nikolovski et al, 2008).
Dry skin is common in the first few months of a baby's life (Saijo and Tagami, 1991). The recommendation to new parents to use topical oils for the prevention or treatment of neonatal dry skin has become traditional practice (Walker et al, 2005; Cooke et al, 2011). While there is a dearth of evidence to support the practice of recommending topical application of natural oils, there is a readiness to believe that what is ‘natural’ is also ‘safe’ (Lavender et al, 2009; Bedwell and Lavender, 2012).
Maternal and neonatal benefits
Maternal benefits Some research studies have considered infant massage as an intervention to improve the mother–infant relationship and maternal mental health. Infant massage positively affected the mood state of mothers in one randomised study (n=39) (Fujita et al, 2006).
Attendance at an infant massage class was found to provide a means of postnatal peer support, reducing isolation in a mixed methods study (n=156) (Clarke et al, 2002). In Clarke's study, the quantitative data were not significant for all outcomes but the qualitative data supported this conclusion.
Infant massage has been shown to improve mother–infant interaction for mothers with postnatal depression in a randomised study (n=34) (Onozawa et al, 2001). However, the authors acknowledged that, due to the small sample size, it could not be determined which aspect of the massage class contributed to the improvement.
Infant massage was shown to be an effective method of improving attachment in a quasiexperimental non-randomised study (n=117) (Gürol and Polat, 2012). In this study, the attending doctor made the decision about the allocation of mothers to the intervention or control group; therefore, allocation bias may exist, which would affect the credibility of the findings.
Underdown et al (2013) conducted a realist evaluation to assess which parent–infant dyads would benefit most from an infant massage programme. The study found that only dyads at moderate risk (one to two risk factors above normal) would benefit, rather than those at low or high risk. The authors suggested that these parents should be targeted and that further research in the form of randomised controlled trials (RCTs) was required to assess the intervention with a targeted sample in a robust way.
A selection of studies investigating the maternal benefits of infant massage have been presented as examples but, overall, there is very little research to support the suggested benefits. The majority of research is methodologically weak. Further research is required to establish the genuine benefits for parents who perform infant massage.
Term infant benefits
A Cochrane systematic review of infant massage incorporated 34 studies totalling 3984 healthy term babies aged up to 6 months (Bennett et al, 2013). More than half of the studies (n=20) were rated as being at a high risk of bias with regard to their design and conduct. The studies included addressed outcomes of physical health and mental health and development. Physical health outcomes included weight, growth, sleep duration, crying/distress times, blood bilirubin levels and illness episode frequency. Mental health and development outcomes included motor skills, personal and social behaviour and psychomotor development. The findings of the review and associated meta-analyses do not support the use of infant massage in the low-risk population of parent and term infant dyads. The authors acknowledged that the reason for this conclusion may have been the poor methodological quality of 20 of the studies, and the lack of attention to the biological plausibility of the outcomes being measured. The review considered only papers looking at the benefits for the term baby; studies on benefits for mothers and preterm babies were excluded. The review authors concluded that future research should concentrate on higher-risk population samples, such as those in the study by Underdown et al (2013).
Preterm infant benefits
Weight gain is the most consistent outcome associated with preterm infant massage (Scafidi et al, 1990; Mathai et al, 2001; Sankaranarayanan et al, 2005; Field et al, 2010; Kulkarni et al, 2010). One explanation put forward for this weight gain is the separate finding that significantly lower levels of energy and stress behaviour were expended among preterm babies in the intervention groups (Lahat et al, 2007), which meant that they were able to ‘sleep and grow’ (Lampl and Johnson, 2011). Another explanation is that infant skin has a high rate of absorption (Nikolovski et al, 2008), so topical oil may have been absorbed systemically to provide a nutritional function (Fernandez et al, 1987; Solanki et al, 2005). The most important finding from preterm infant massage studies is the significant reduction in mortality and infection (Darmstadt et al, 2005; Darmstadt et al, 2008; Mendes and Procianoy, 2008). Darmstadt's research demonstrated a 26% reduction in mortality (Darmstadt et al, 2008), and that preterm babies were 41% less likely to develop a nosocomial infection when massaged with sunflower seed oil compared to no treatment (Darmstadt et al, 2005). Reduction in infection in a preterm population is related to reduction in mortality.
How to massage an infant
There are no standardised guidelines describing a routine method for performing infant massage. Field (2002) suggests massage in 15 minute sessions, three times a day. Each block of 15 minutes consists of 5 minutes of tactile stimulation, followed by 5 minutes of kinaesthetic stimulation, followed by 5 minutes of tactile stimulation. Parents and health professionals can provide the massage equally effectively (Ferber et al, 2002). Massage should not be carried out within an hour of feeding, to minimise the risk of vomiting. The whole body should be included in the massage technique and a moderate pressure is recommended for optimal effect (Field, 2002).
Mathai et al (2001) suggests a slightly different technique where sessions start with two phases of tactile stimulation, before a final phase of kinaesthetic stimulation. The first phase includes placing the baby in a prone position and providing 12 strokes of 5 seconds, each administered from the head through the neck and shoulder to the buttock. The second phase includes placing the baby in a supine position and providing 12 strokes of 5 seconds each administered from the face, through the cheeks, chest, abdomen, upper limb, lower limb, palms then down to the soles of the feet. These two phases are followed by a third phase of kinaesthetic stimulation providing flexion and extension of the major joints.
Using a lubricant during infant massage is recommended to avoid friction (Kulkarni et al, 2010). Many natural oils have been documented as topical applications routinely used in infant massage including mustard oil (Darmstadt and Saha, 2002; Mullany et al, 2005), sunflower oil (Ahmed et al, 2007), coconut oil (Sankaranarayanan et al, 2005), olive oil (Cooke et al, 2011) and sesame oil (Agarwal et al, 2000). However, topical oils may induce an allergic response (Solanki et al, 2005; Kulkarni et al, 2010) in the recipient and/or the provider. There have been documented case studies of allergic contact dermatitis on the hands of workers using olive oil (Malmkvist et al, 1990; Kränke et al, 1997; Isaksson and Bruze, 1999; Wong and King, 2004). No studies have looked at the long-term effects of using oils on baby skin, but the question has arisen as to whether early use of topical oils with babies has a connection with the development of atopic eczema (Danby et al, 2013).
Role of the health professional
There is no national or international guidance on what constitutes best practice with regard to infant massage. There are no recognised standards nor official regulation of infant massage instructors. The IAIM advises not to use any oil topically with a high oleic acid content, such as olive oil (Bond, 2015). The IAIM statement is followed by the abstract of one small randomised study of adult volunteers (n=19) (Danby et al, 2013), which reflects on the association between the use of topical oil and the development of atopic eczema caused by the effect on skin barrier function. The study relates the findings to the potential implications for infant skin care.
Health professionals have a duty to protect the public from harm (Nursing and Midwifery Council, 2015). They should therefore recommend only treatments and practices that are beneficial. The evidence-base for infant massage is methodologically limited, and the recommendation arising from a large systematic review (Bennett et al, 2013) is that further research is required to investigate exactly what the benefits are.
Massage lubricants
Many instructors will advise parents to use a topical oil on their baby's skin during massage (Cooke et al, 2011; Bond, 2015). Acceptability of massage with oil was considered in a randomised study in Bangladesh (Ahmed et al, 2007). The study found that the majority of participants started infant massage within 1 hour of birth (61%) with mustard oil (88%), which was applied all over the body (89%). Babies who were admitted to hospital were randomised to sunflower oil or Aquaphor ointment application. Parents perceived that these were superior to mustard oil and suggested they would use these products in preference to mustard oil in their massage practice in future. This demonstrates that health professionals may hold a certain level of influence over parental practices. The reasons given by parents for using topical oil included keeping the baby warm (22%), preventing infection (18%), improving the skin condition (6%) and improving the overall health of the baby (8%).
A previous study showed that 96% of caregivers had practised infant oil massage (Darmstadt and Saha, 2002). Although a popular practice, particularly in Eastern countries, there is a dearth of research considering the effect of topical oils on infant skin. Despite the lack of evidence, midwives in the UK commonly recommend topical oils for the prevention or treatment of infant dry skin or for infant massage, namely olive oil (80%) and sunflower oil (20%) (Cooke et al, 2011). Several functional mechanistic studies have been carried out in mice and adult volunteers, which suggest that oils with a high oleic acid content may be harmful to skin barrier function and be connected to the development of atopic eczema, while oils with a high linoleic acid content may have a repairing effect on skin barrier function (Darmstadt et al, 2002; Danby et al, 2013).
The OBSeRvE (Oil in Baby SkincaRE) pilot study (Cooke et al, 2013), which is currently in progress, is intended to establish evidence of the effects of topical olive oil and sunflower oil on term infant skin barrier function and to assess the feasibility of conducting a definitive RCT; results are expected soon.
Results will be of interest, particularly in view of the recent small study of preterm babies (n=22) (Kanti et al, 2014), which found that sunflower oil may impede skin barrier development, a finding that is contrary to the current evidence-base. Caution is required so that previous research (Darmstadt et al, 2008), which found that using topical sunflower oil on the skin of preterm babies resulted in a significant reduction in infection and mortality, is not ignored. Kanti's study did not include mortality or infection as an outcome. Sunflower oil possibly has an antimicrobial effect on preterm skin, which produced the significant results. The effect of topical sunflower oil on the different outcome measures of skin barrier function and mortality/infection have produced diverse findings; this warrants further investigation.
The combination of the lack of clinical trial evidence and the negative mechanistic data from the mouse and adult studies means that recommendation of topical oils should be made with caution. In view of our desire as midwives not to cause harm to babies, a recommendation to avoid the use of topical oils should be considered until evidence is available to support this practice.
Conclusions
Newborn infant massage is becoming more popular. Midwives and other maternity health professionals require a sound evidence-base to provide the best advice to new parents about their newborn baby's skin care. For infant massage, the majority of the available evidence is methodologically limited and further robust research is required.
There are no clinical guidelines and there is no regulation of practice. This is of concern, particularly in view of research that suggests some topical oils may have an adverse effect on skin barrier function, possibly contributing to the development of atopic eczema.
Further robust research, preferably in the form of RCTs, will provide midwives and other maternity health professionals with a sound knowledge of the benefits of infant massage to both mothers and infants, and the optimal choice of massage lubricant to ensure that recommendations and practices benefit infant skin.