Article Text

other Versions

Systematic review
Cochrane review: non-nutritive sucking, kangaroo care and swaddling/facilitated tucking are observed to reduce procedural pain in infants and young children
  1. Judith Meek,
  2. Angela Huertas
  1. Department of Women's Health, Neonatal Service, University College London Hospital NHS Trust, London, UK
  1. Correspondence to Judith Meek
    Department of Women's Health, Neonatal Service, University College London Hospital NHS Trust, 2nd floor North, 250 Euston Road, London NW1 2PG, UK; judith.meek{at}

Statistics from

Commentary on:

Implications for practice and research

  • Kangaroo care, swaddling/facilitated tucking and non-nutritive suckling are effective techniques for diminishing responses to procedural pain in preterm and newborn term infants.

  • The comparative effectiveness of these techniques against other commonly used methods needs further analysis.

  • There is insufficient research in the 1-month to 3-year-old age group to make any recommendations. Different techniques may be appropriate across this group according to age and disability.

  • Further research using evoked cortical responses is essential.


Painful procedures are often performed on infants and young children, and they can have life long consequences, particularly when repeated frequently and if pain is not managed effectively. The preverbal population is especially vulnerable, and we have a responsibility to protect the fragile developing nervous system. However, it has proved difficult to extrapolate analgesic techniques from adults to infants. There have been several reviews of pharmacological therapies for procedural pain and this review of non-pharmacological techniques is thorough and timely.


Interventions were studied in three age groups. Skin breaking and nursing procedures were studied. Circumcision and ophthalmic examination were excluded as they have already been reviewed. Contextual, cognitive and behavioural interventions were included, except for breast feeding and sucrose which are the focus of other reviews.

Responses were defined as reactivity (immediate) or immediate regulation (after 30 s). The outcome measure was a significant change in pain score, although eight studies used crying time. This was a comprehensive search of electronic databases and grey literature. Careful attempts were made to minimise the risk of bias.


A total of 3396 participants were analysed from 30 randomised controlled trials (RCTs) and 30 crossover trials with 81 treatment arms. Kangaroo care, non-nutritive sucking and swaddling were effective for preterm infants for reactivity and immediate regulation. For term infants only non-nutritive sucking was effective at both times, and rocking for immediate regulation. There was substantial heterogeneity between the studies which needs considering when interpreting the results. There was insufficient data to analyse the effect of any interventions in the older children.


The stimuli considered as noxious included handling for nappy changes and weighing. Although handling is clearly stressful for preterm babies, it is not strictly noxious. The fact that handling elicits behavioural and physiological responses recognised by pain scoring reflects the difficulty in differentiating between stress and pain. This is perhaps a fundamental limitation of using pain scoring scales in preterm infants. However, there is new evidence that noxious stimulation of preterm infants can elicit generalised intense delta brush activity rather than localised cortical responses.1 Therefore the most premature babies may have a generalised response to all sensory stimulation, albeit varying in intensity, leading to strong recommendations for supportive developmental care. The findings of this study reinforce the evidence for using these techniques.

To date quantitative studies have all used pain scores, which have further limitations. Although cortical haemodymanic responses to heel lance correlate with the facial component of the PIPP score,2 on a worrying number of occasions babies in this study mounted a cortical response without any facial change. This reflects the observation that stressed infants will tune out and become unresponsive. There is also a long latency to the facial response in extremely preterm babies,3 which could reduce the total score, and blur the distinction between the immediate and the delayed responses measured in the review.

A recent RCT comparing sucrose with water before heel lances4 demonstrated the expected effect on the PIPP score but none on the cortical response. Sucrose, and other behavioural modifiers, may have an effect on physiological stability which is not due to analgesia. The challenge now is to look at the long-term effects of blunting behavioural or physiological responses to pain without necessarily providing analgesia. Will these techniques reduce the long-term developmental damage caused by repeated noxious stimulation?

There was insufficient data to draw any conclusions for the 1-month to 36-month group. This is a wide ranging group in terms of cognitive development and so perhaps it is not surprising that no consistent patterns emerge. Further studies are required on the different age bands related to developmental stage and the presence of developmental delay or sensory impairment.


View Abstract


  • Competing interests None.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.