Successful Pavlik treatment in late-diagnosed developmental dysplasia of the hip

PURPOSE: The objective of this study was to evaluate success and complication rates of the Pavlik harness in late-diagnosed hip dislocation (developmental dysplasia of the hip, DDH). We also set out to assess the additional value of an abduction brace for three to six months, when successful reduction using the Pavlik harness was achieved. METHODS: We included 31 patients (31 hips, 28 female, right/left = 4/27) with late-diagnosed hip dislocation treated with a Pavlik harness between 1995 and 2008. The average age at the start treatment was 27 weeks (21-57). None were lost to follow-up; the mean follow-up was 4.2 years (two-10). Of these patients, 61 % were classified as Tönnis type 2, 32 % as type 3 and 7 % as type 4. RESULTS: Of 31 hips, 20 (65 %) were successfully reduced after the use of the Pavlik harness. The average duration of Pavlik treatment was seven weeks (three-12). The mean time to stable reduction was six weeks (three-11), after which the patients were weaned off the Pavlik harness. The mean age at final treatment was 19 months (12-28). Five patients (15.0 %) developed radiological signs of osteonecrosis (Kalamchi and MacEwen classification; two group I, one group II and two group IV). When we compared the different Tönnis types a significant difference in successful reduction was found. Seventeen (81 %) of Tönnis type 2 dislocated hips were successfully reduced, while only two (25 %) of Tönnis type 3-4 hips were (odds ratio 25, p = 0.001). Clinical examination (e.g. limited abduction, positive Ortolani, Barlow or Galeazzi) at the time of diagnosis was not significantly related to the success rate of the Pavlik treatment. The mean acetabular index (AI) significantly improved from 36.5 to 30.5° after initial Pavlik treatment to 22.3° at final follow-up (4.1 years). CONCLUSIONS: Prolonged use of the Pavlik harness in late-diagnosed hip dislocation (DDH) Tönnis type <3 is a safe and successful treatment option in the older infant. Although the AI was significantly reduced after the abduction brace, its additional use remains debatable, as no control group was evaluated.