Physiotherapy in high-risk infants - a motor learning facilitator or not?

Follow-up of infants admitted to a neonatal intensive care unit is challenging. After having fought for their lives these infants have to learn to adapt to the challenges of daily life with their existing developmental resources, which in many may be diminished. Identification of signs suggestive of abnormal development usually leads to therapeutic interventions in the hope of improved outcome. The clinician has to decide when and what kind of intervention to start and for how long. General movements are spontaneous, endogenously generated movements that are replaced by goal-directed movements around the age of 4 months.1 Definitely abnormal general movements are characterized by serious reduction in movement variation and are considered to have a good predictive value for abnormal outcome, especially when observed in high-risk infants. Hielkema et al.2 have conducted a randomized controlled trial (RCT) study in which they identified high risk-infants showing definitely abnormal general movements at 10 weeks corrected age. They compared the effects of two different physiotherapeutic interventions – COPing with and CAring for Infants with Special Needs (COPCA) and traditional infant physiotherapy – on motor performance at 18 months using the Infant Motor Profile as the outcome measure. Secondly they analysed the content of physiotherapy using video recorded sessions and compared the physiotherapeutic actions with the Infant Motor Profile scores. In many RCT studies on physiotherapeutic interventions the new intervention is clearly described (and the therapists well trained in it prior to the study) but the current practice is not.3 Hielkema et al.2 state that in the Netherlands traditional infant physiotherapy is mostly based on the principles of neurodevelopmental treatment. However, the basic principles of neurodevelopmental treatment could not be identified from the analysis of video recorded sessions. The basic principles of COPCA were identifiable, which is primarily reflective of the training that therapists have received in this particular therapy form. Thus it is more relevant to analyse the study as a comparison of COPCA against an eclectic therapy, which may not be similar to the early physiotherapeutic intervention given in a particular country. The main result of the Hielkema et al. study was that there was no difference in motor outcome at the age of 18 months between the two intervention groups. This raises several questions that cannot be answered fully without further study. The main questions concern timing, the content of intervention, and the study design. In this study, the therapy was initiated when definite abnormal general movements were observed at the corrected age of 10 weeks. From a clinical and ethical point of view this is well understood, but was the age of 10 weeks too late to benefit optimally from the brain’s intrinsic plasticity mechanism in this high-risk group consisting mainly of preterm infants? Should we start individually tailored therapy earlier for these children and if so what should it consist of? Skilled therapists can open channels but positive change in developmental trajectories requires that these channels remain open. Only when the parents become experts on their infant’s behaviour and the interplay between their actions and the infant’s reactions (and vice versa), can the infant learn how to face potentially challenging situations. The main active part of early intervention is thus not coaching as such but confirmation that the parents have digested what has been proposed and have changed their daily practice accordingly. We are clearly in need of studies that focus on identifying how the transfer of therapeutic strategies into daily life can be affirmed. For this to happen it may even be that interactive and pedagogic skills together with the personality of the therapist are at least as important as their people handling skills. Rosenbaum4 has recently raised the important question whether RCTs are the best way to study the effect of rehabilitation interventions, a question that was identified also when analysing existing studies on physiotherapeutic interventions.5 RCTs can be used for short-term interventions with a clear focus (e.g. intensive upper extremity training) but are less suited to comprehensive treatments in which developmental change requires change in conducting daily activities and can be hindered by adverse events in daily life. In the study of Hielkema et al.2 no difference could be observed between the two intervention methods. The authors also state that it was unethical to have no-intervention as a comparison group. Can we now state that physiotherapy has no effect in high-risk infants, or that the same effect can be reached using different theoretical approaches? Or is it more truthful to say that we do not yet know the optimal timing and content of comprehensive physiotherapeutic interventions for high-risk infants? The study did not tell us how well the parents and infants were able to adopt the new strategies in their daily activities, so we can also question whether the optimal potential of the infants has yet been reached. The Hielkema et al. study was conducted according to the criteria of a RCT and its main message is important: COPCA and traditional infant physiotherapy had similar effect on motor outcome at 18 months measured with the Infant Motor Profile. But did it give any answers to what kind of intervention is appropriate and for what length of time? To our frustration it did not – a result so familiar from many RCTs on therapeutic interventions.5 Rehabilitation research urgently requires new strategies to continuously evaluate and improve clinical practice. RCTs cannot be the only option to study interventions that are not ‘active when swallowed’ but are effective only if individually tailored and implemented into daily activities.