The International Hip Dysplasia Institute (IHDI) classification is more informative than the Tönnis classification

A new, supposedly more reproducible radiographic classification, set to replace the Tönnis classification of hip dislocations, was proposed in 2015: the International Hip Dysplasia Institute (IHDI) classification.

To compare the IHDI classification with the Tönnis classification when evaluating the severity of hip dislocations as well as their respective inter- and intra-observer reliability.

Since January 2000, Swedish-born children with a hip dislocation were prospectively registered. From this registry, radiographs of 97 hips in 79 patients (91% girls; median age = 7 months), born in 2000-2009, were analyzed. Two observers, one consultant and one resident, classified each hip both by IHDI and Tönnis twice.

The IHDI classification had a more even distribution of grades with the majority in grade 2-3. The Tönnis classification graded the majority (77%) of the patients as grade 2 and equally among the other grades. There was moderate inter-observer agreement using both methods calculated with Kappa, 0.61 (95% confidence interval [CI] = 0.44-0.79) for Tönnis and 0.62 (95% CI = 0.49-0.74) for IHDI. The resident calculated Tönnis with weak intra-observer reliability of 0.57 (95% CI = 0.40-0.74) compared to high intra-observer reliability of 0.86 (95% CI = 0.74-0.98) for the consultant. Both observers graded IHDI with high intra-observer reliability.

IHDI is more discriminative than the Tönnis classification when evaluating the severity of a hip dislocation in infants.