Aim: To assess whether observational studies provide evidence to support general ultrasound screening for developmental dysplasia of the hip (DDH).
Method: Systematic Review. Following a predefined study protocol, observational studies on ultrasound screening in unselected newborns were identified by search through 23 electronic databases and by hand search. Two reviewers selected the studies independently of each other and extracted the data.
Results: 49 observational studies were included. The prevalence of DDH ranged from 0.5 % to 30 % depending largely on the various possible definitions of DDH. Less than 0.1 % of patients with DDH were missed by ultrasound regardless of the technique employed (Graf or Terjesen). About 90 % of newborns with Graf type IIa hips requiring ultrasound control did not develop DDH. Only six studies with 23 108 newborns reported on complications, and there was only one infant with an avascular necrosis of the femoral head. The effectiveness of a general ultrasound screening cannot be evaluated reliably for several reasons: reports were often incomplete, follow-up of newborns with normal findings at the time of screening was often not carried out, clinically meaningful data about outcome as well as control groups were missing.
Conclusion: General ultrasound screening for developmental dysplasia of the hip cannot be sufficiently assessed by the observational studies available. The lack of evidence does not mean that ultrasound screening is ineffective, but randomized controlled trials comparing the effectiveness of different screening regimens are needed.