Children's Hospital Colorado

Twelve-Week Standard of Care Pavlik Harness Treatment for Infants with IIc Hip Dysplasia Longer than Median Time to Normalization

For families

  • Developmental dysplasia of the hip (DDH) is a common orthopedic condition in infants.
  • Treatment varies with severity, but the Pavlik harness is frequently used.
  • The standard treatment for type IIc DDH is 12 weeks of Pavlik harnessing.
  • Approximately half of the infants studied did not need 12 weeks for hips to normalize.

For health professionals

  • Infants with Graf type IIc DDH receive 12 weeks of Pavlik harnessing at Children’s Colorado.
  • Study authors sought to determine if the standard of care treatment duration is necessary.
  • They found 7 weeks was the median time to normalization.
  • Femoral hip coverage and instability were the only significant variables associated with time to normalization.

Background: Pavlik harnessing for 12 weeks is typical initial developmental hip dysplasia treatment

Developmental dysplasia of the hip (DDH) is a common orthopedic condition that ranges from stable, dysplastic hips to frank dislocation. Untreated DDH can result in gait abnormalities, degenerative arthritis and chronic pain.

Since DDH may not show up in an examination, ultrasound screening is recommended for infants with risk factors.

The Graf method ultrasound classification system for DDH describes the alpha angle to determine the type of DDH. Graf type can be used in conjunction with femoral head coverage (FHC) on an ultrasound to determine DDH severity.

Upon diagnosis, reported DDH treatment protocols vary greatly, and there is no consensus on best practices. More intense interventions and treatment are typical for later detection, and there is a higher likelihood of failure or other complications.

A frequently used initial approach, the Pavlik harness is considered the gold standard of treatment. Specialists in the Orthopedics Institute at Children’s Hospital Colorado treat most dysplastic hips with the Pavlik harness full time for 12 weeks, regardless of Graf type or severity.

Guidelines are general for harness utilization with decisions often left up to the individual clinician, and many patients achieve “normalization” earlier than 12 weeks.

This retrospective analysis was led by Children’s Colorado’s Gaia Georgopoulos, MD, Nancy Hadley-Miller, MD, and Margaret Siobhan Murphy-Zane, MD, and involved many other experts from the Orthopedics Institute and with the University of Colorado Anschutz School of Medicine Orthopedics Department. The study sought to decrease the currently recommended length of harness treatment by depicting treatment norms for patients with Graf type IIc hips (43-49° bony angle, still FHC) and identifying average time to normalization and factors indicative of faster or slower normalization.


Study authors collected data from patients diagnosed at Children’s Colorado between December 2009 and November 2018. They included 271 patients under 3 months old with a Graf type IIc hip on initial or ultrasound visit. After exclusion criteria, the final study population included 106 patients and 132 hips. All patients were initially treated with the full-time Pavlik harness.  

Normalization, or a normal hip, was defined as alpha angle greater than or ≥60° and FHC ≥50%. The Kaplan–Meier and Cox proportional hazards regression analyses were used to model the time to normalization and identify factors associated with earlier normalization. 

Results: IIc hips normalized at a median of seven weeks 

Only two diagnostic variables were significantly associated with time to normalization in the final model: 

  • Percent FHC (hazard range (HR per) 1% increase: 1.03; 95% CI: 1.01–1.06; P = 0.0024)  
  • Hip instability at the initiation of treatment (HR unstable vs. stable: 0.64; 95% CI: 0.44–0.93; P = 0.0068)

Discussion and conclusion: Femoral head coverage and instability may be better indictors for bracing duration

There are no universally agreed upon, evidence-based best practices or protocols for DDH treatment, regardless of severity or Graf type. 

Causes for failure to normalize after 12 weeks in 14.2% of hips were varied. They included:

  • 8 hips normalized just past 12-week mark 
  • 6 hips normalized after week 12 with secondary bracing or surgical intervention 
  • Risk factors and age of treatment initiation varied 4 had instability 
    • 3 had breech presentation 
    • 2 had family history 
    • 5 were female  
  • 27% was average FHC 
    • Less than 25th percentile of entire population 
    • Possible causes unclear 
    • Instability, decreased FHC are concerning findings

All patients who achieved normalization had normal radiographs at follow-up. This led study authors to question the necessity of the 12-week treatment duration for hips that respond well and normalize before 12 weeks.

There was a significant difference in percent FHC between stable and unstable hips:  

  • Greater coverage of the femoral head by the acetabulum suggests shorter time to normalization, shorter treatment lengths 
  • Percent FHC on ultrasound could be a diagnostic tool, particularly at ends of its range 
  • FHC may be stronger measurement of hip morphology than its current interpretation (50% FHC a guideline of normalcy)

The widespread variability in the study cohort likely reflects the general patient population, thus any defined treatment should not be considered one-size-fits all.

Further research is needed to determine if:

  • Length of treatment to achieve normalization correlates with severity across Graf types  
  • Other diagnostic variables are indicative of longer or shorter time to normalization

Study authors from Children’s Hospital Colorado and University of Colorado School of Medicine

This study involved a multidisciplinary team within the Children’s Colorado Orthopedics Institute and the University of Colorado School of Medicine Department of Orthopedics. Study authors included: Gaia Georgopoulos, MD; Patrick Carry, MS; Reba Salton; Tyler Freeman, MD; Kaley Holmes; Nancy Hadley-Miller, MD; Brian Kohuth, PA-C; Debbie Burke, PA-C; Matthew Belton, MD; Margaret Siobhan Murphy-Zane, MD.