Ultrasound of the Hip Joints in DDH in Infants: AIUM 2024
Why the AIUM 2024 Protocol is Needed
AIUM Practice Parameter 2024, prepared with ACR, SPR, and SRU, describes how to perform ultrasound to detect and assess developmental dysplasia of the hip (DDH). The practical meaning of the revision is not to oppose approaches but to combine Graf's static morphometry and Harcke's dynamic stability assessment. In one study, the physician must answer two questions: is the bony-cartilaginous roof of the acetabulum formed, and is the femoral head held under load?
Indications and Timing of the Study
Ultrasound is used when there is clinical suspicion of instability, abduction asymmetry, click/Ortolani or Barlow signs, as well as risk factors: breech presentation, family history of DDH, combined orthopedic anomalies. The method is especially useful in infants when a significant part of the femoral head is still cartilaginous and radiography is less informative.
The study can be performed when clinically necessary in the early neonatal period, but for screening, the AIUM parameter considers the practice of delaying until 4–6 weeks: some immature joints normalize spontaneously in the first weeks. As the femoral head ossifies, the acoustic window deteriorates; in older children, the choice of imaging method is aligned with the clinical task.
Equipment and Positioning
A linear high-frequency transducer with sufficient resolution is used to visualize the cartilaginous head, bony edge of the acetabulum, and labrum. The child is positioned so that the pelvis is stable, without rotation; usually, both joints are examined sequentially. Short contact time, a warm gel medium, and stabilization of the pelvis with a hand or positioner are important: pelvic tilt artifact directly distorts angle α.
Each joint is marked by side. Depth and focus settings are adjusted so that the iliac bone, acetabulum, femoral head, femoral neck, and soft tissue landmarks are in the frame. Angles should not be measured on a random 'beautiful' slice: a standardized coronal plane is first obtained, then lines are drawn.
Static Graf Method
The basic slice is the coronal plane through the hip joint. Signs of a suitable image: a straight line of the lateral wing of the iliac bone, visible bony roof of the acetabulum, cartilaginous roof with labrum, femoral head, and lower edge of the iliac bone. If the iliac bone is curved or the slice is oblique, α will be underestimated or overestimated.
For Graf, a baseline is drawn along the lateral contour of the iliac bone, a bony roof line, and a cartilaginous roof line. Angle α reflects the development of the bony roof; angle β characterizes the position of the cartilaginous roof and labrum. Measurement is performed on a saved static image, and if in doubt, the slice is repeated until a reproducible result is achieved.
Graf Threshold Values
| Graf Class | Angle α | Angle β / Age | Clinical Interpretation |
|---|---|---|---|
| Type I | ≥60° | β does not define dysplasia | Mature, centered hip joint |
| Type IIa | 50–59° | Age under 3 months | Physiologically immature joint with preserved centering |
| Type IIb | 50–59° | Age over 3 months | Dysplasia with persistent bony roof deficiency |
| Type IIc | 43–49° | β <77° | Dysplastic but centered joint |
| Type D | 43–49° | β >77° | Decentration/instability against dysplasia |
| Type III–IV | <43° | Evaluate labrum and head position | Severe dysplasia, subluxation or dislocation |
The table does not replace the description of the femoral head position. With the same α, the clinical significance changes if the head is not centered or dislocates under load.
Dynamic Harcke Method
Harcke complements Graf by assessing stability. The joint is examined at rest and under gentle stress, usually in the coronal plane with a flexed thigh and in the transverse plane. The task is to see if the femoral head remains centered in the acetabulum, shifts laterally/posterolaterally, or reduces with position change.
The dynamic test should be gentle and controlled. The ultrasound physician stabilizes the pelvis, avoids excessive pressure with the transducer, and does not turn the examination into an orthopedic manipulation. If dislocation is evident on static frames, repeated forceful provocation attempts are unnecessary; it is sufficient to document the position and reducibility if it is safely assessed.
Minimum Image Set
- Right and left joints in standardized coronal plane for Graf.
- Frames with drawn lines and measured angles α; if necessary, β.
- Documentation of femoral head centering at rest.
- Dynamic images or cine loops under stress by Harcke.
- Transverse dynamic slices showing head displacement or stability.
- Mandatory side, position, and mode labeling: rest or stress.
What to Write in the Conclusion
The conclusion should be structured. For each joint, indicate: slice quality, angle α, if necessary β, Graf type, head centering, dynamic test data, and symmetry. The phrase 'signs of DDH' without angles and stability description is of little use to the orthopedist and does not allow comparison of studies over time.
Example of brief logic: 'Right joint: α 62°, type I, head centered, stable under stress. Left joint: α 48°, β 72°, type IIc, head centered, no dislocation/with subluxation under stress.' Numbers and dynamic part should correspond to saved images.
Typical Errors
- Measuring α on an oblique slice where the iliac bone is not a straight line.
- Assessing only angles without dynamic instability check.
- Comparing studies performed in different planes and without saved lines.
- Excessive pressure with the transducer, which artificially centers an unstable head.
- Ignoring age in type II: α 50–59° in an infant under and over 3 months is interpreted differently.
Practical Algorithm
- Check indications, age, clinical data, and risk factors.
- Stabilize the pelvis and obtain a coronal slice of each joint.
- Confirm anatomical landmarks, then measure α and β by Graf.
- Classify the joint by Graf thresholds.
- Perform dynamic Harcke assessment at rest and under gentle load.
- Save static frames, measurements, and dynamic fragments.
- Formulate a conclusion separately for the right and left joints.
Frequently asked questions
Is it enough to perform only Graf angles?
No. In the AIUM 2024 parameter, Graf's static morphology is considered together with Harcke's dynamic assessment because a joint with a roof deficit can be stable or unstable.
What is the main normal threshold for angle α?
For a mature joint by Graf, α ≥60° is used. Values 50–59° relate to type II and are interpreted considering the infant's age.
Why is screening often not done in the first days of life?
Early ultrasound is possible with clinical suspicion, but for screening, delaying until 4–6 weeks reduces overdiagnosis of transient physiological immaturity.