Commentary on "Clinical Feasibility of 2-Dimensional Video Analysis of Active Cervical Motion in Congenital Muscular Torticollis".

Abstract

“How could I apply this information?” For infants with congenital muscular torticollis, measurements of active cervical range of motion in the transverse and coronal planes are important to assess severity, prognosis, and response to intervention. Despite insufficient validity of visual estimation (VE) to examine cervical rotation and lateral flexion in these infants, there is no widely practiced alternative (but see the Section on Pediatrics, Clinical Practice Guideline1 for evidence-based recommendations for these measurements). In the current study differences between techniques were observed and presumed to indicate greater accuracy of video analysis (VA). However, VA may be impractical in clinical settings because of the time required. Video analysis has possible utility in settings with support staff to assist with setup, measurement, video processing, and with access to the necessary equipment. It also has potential usefulness in research studies if it were to be validated against a reference standard measure. The reported data confirm concerns with VE and underscore the remaining need for the use of evidence-based objective measures of active cervical range of motion for infants with congenital muscular torticollis. “What should I be mindful about when applying this information?” The authors' primary objective of determining the feasibility of the VA method in clinical practice was met by evaluating the time and effort required. However, the comparison of the VA technique to a flawed method of measurement (VE) rather than a reference standard, such as 3-dimensional motion analysis, did not allow the validity of the video technique to be established, and the authors acknowledge this fundamental limitation. Using a reference standard comparison would have also allowed the contribution of the other procedural limitations, such as marker placement, equipment setup, and small sample size to be properly investigated. Finally, with the small number of observations for each therapist (9 and 3, respectively) and the variability in results, the magnitude of the actual differences between the techniques shown in Table 1 is likely more meaningful than whether the differences were statistically significant or not. Kristen Kouvel, PT, DPT, PCS Laura Prosser, PT, PhD Center for Rehabilitation The Children's Hospital of Philadelphia Philadelphia, Pennsylvania