Reliability of 3 Radiologic Classifications for the Severity of the Developmental Dysplasia of the Hip in Children Older Than 4 Years

BACKGROUND Tonnis, International Hip Dysplasia Institute (IHDI), and lateral metaphyseal height (LMH) are commonly used classifications for grading the severity of the developmental dysplasia of the hip. The reliability of these classifications is not widely studied in older children. The aim of the study was to evaluate the reliability of these 3 radiologic classifications in children older than 4 years and compared with children younger than 4 years and evaluate the cases with varied inter-rater reliability. METHODS A purposeful sample of 40 children with untreated developmental dysplasia of the hip with ages between 6 months to 8 years was studied for the assessment of the severity grading according to all 3 classifications. Six pediatric orthopaedic surgeons classified all hips for all 3 categorical classifications as per the original description. Inter-rater and intrarater reliability was calculated according to the intraclass correlation coefficient. The cases with different ratings were assessed in detail to evaluate the reasons for the varied rating. RESULTS The interobserver and intraobserver reliability of all 3 classifications were excellent [intraclass correlation coefficient (ICC): 0.935, 0.820, and 0.935 for IHDI, Tonnis, and LMH classification, respectively]. The excellent reliability was also observed in younger and older children. Interobserver reliability of only dysplastic hips (52 hips) was good for Tonnis (ICC: 0.741) and excellent for IHDI (ICC: 0.911) and LMH classification (ICC-0.9). The main reason for the varied rating was because of the varied perception of the superolateral margin of the acetabulum in few hips. CONCLUSION The inter-rater and intrarater reliability of all 3 classifications (IHDI, Tonnis, and LMH) is excellent. All classifications can be used till the age of 8 years. The difficulty in selecting the superolateral margin of the acetabulum is a major cause of inter-rater variability. LEVEL OF STUDY Level III.