Table of Contents:
Background: Since magnetic resonance imaging (MRI) has been introduced as diagnostic
parameter for congenital dislocation of the hip (CDH) in the 1980ies, it is considered to be the preferred
method of imaging to control hip repositioning.
The aim of this study was to investigate the advantage of using MRI in comparison to X-rays after reposition
of CDH with special attention to the incidence of redislocation in the course of treatment. With no special
classifications to evaluate hips on MR images existing, we developed one for each plane.
Methods: To develop the new classifications, 117 MR images were evaluated. Three independent
investigators applied the classifications twice. Intra- and inter-observer reliability and alpha reliability
coefficient were calculated for both classifications. Within the observed group different imaging (radiography
or MRI) was correlated with items of therapy and follow-up, with special regard to redislocation. Pearsons coefficient was calculated and considered significant when α > 0.7 and p < 0.05.
Results: The evaluation of the newly developed classifications showed an intra-observer reliability up to
82.4% accordance for coronal MR images and up to 87.5% for transversal images. Inter-observer reliabilitiy
revealed for coronal images α = 0.7878 and for transversal images α = 0.7307.
The evaluation of correlations between imaging and aspects of therapy and follow-up did not show any
significance. With regard to the diagnosis of redislocation pelvigraphy and coronal MRI showed no
advantage, but transversal imaging, using the new classification, could reveal by then undiagnosed
Conclusions: We propose an algorithm for the use of MRI within CDH repositioning control: Balancing the risks of sedation and ionizing radiation, the sedation used for reposition should be used for subsequent MR
imaging. If no pathological findings are found in either plane subsequent imaging controls should be
performed by radiography. If pathologies are identified, reposition has to be repeated as well as MRI. In
case of no pathological findings, imaging should be performed by radiography.