Total hip arthroplasty (THA) often requires complex reconstruction for acetabular bone defect in patients with developmental dysplasia. We performed autogenous bulk structural bone grafting to deal with the lack of acetabular bone stock. The purpose of the present study was to assess the clinical and radiographic results after mean follow-up of 12 years.

Between April 1992 and December 1997 the single senior author performed 75 consecutive primary THA for patients with degenerative osteoarthritis. Acetabular bone grafting was performed for 27 joints. Of these, six patients (six hips) were lost to follow up. Left 20 patients (21 hips) included in the study. There were two male and 18 female with a mean age at the time of the operation of 54.5 years (40–66 years). The mean duration of follow-up was 12 years (8–15 years). The diagnosis for all hips at the time of operation was secondary osteoarthritis due to developmental dysplasia. The degree of subluxation as categorized according to the classification of Crowe et al was group I in 11 hips, group II in 6 hips, group III in 4 hips. All operations were performed through a posterolateral approach using the femoral head for the graft. The grafts were screwed to the superolateral aspect of the acetabular roof. We used the Bioceram implant (Kyocera, Kyoto, Japan) with a 26 mm alumina-ceramic head. Both acetabular and femoral components were fixed with cement using the double-thumb technique in all procedures. Harris hip score was used for clinical evaluation. Standard anteroposterior radiographs were used for radiographic evaluation. The presence of a radiolucent line at the cement-bone interface in the three zones of DeLee and Charnley was recorded. Loosening of the acetabular component was classified according to the criteria of Hodgkinson et al. The remodeling process of the grafted bone was analyzed according to the method described by Knight et al. The initial postoperative anteroposterior radiographs were measured to define the proportion of the socket covered by bone graft according to the method described by Inao and Matsuno.

The mean Harris hip score improved from 45.0 (24–60) before operation to 90.4 (77–100) at the final follow-up. At the final follow-up, 13 sockets showed the presence of a radiolucent line at the cement-bone interface and three sockets showed radiological evidence of loosening. According to the criteria of Hodgkinson et al, two sockets were type 3 and one were type 4. Bridging tarbeculation across the graft-host interface was seen in all cases. One case had the graft collapsed with migrated socket. The mean proportion of the socket covered by bone graft was 23.1% (9.8–42.3%). Three patients with loosed sockets had candidate for revision surgery, but no revision surgery was done because they had mild pain and did not demand the operation.

Autogenous bulk structural bone grafting for reconstruction of the acetablum presented favorable results during mean follow-up of 12 years in the condition that the proportion of coverage of graft was less than 50%.

Though over ten-year follow-up results of impaction bone grafting for acetabular reconstruction from European countries are available in literatures, clinical reports from Asian countries are rare. The purpose of the present study is to assess mid-term clinical and radiographic follow-up results at least three years after acetabular reconstruction with impaction bone grafting technique by single surgeon in Japanese cohort.

The senior author performed 24 acetabular revisions with impaction bone grafting technique in 24 patients from February 2001 to June 2005. The average age of the patients at the revision was 67.5 years (36–82 years). The average follow-up period was 5 years and 5 months (3–7.3 years). The reasons for the operation were aseptic loosening of sockets in 17 hips and migration of bipolar heads in seven. The acetabular bone defects were classified as cavitary in 3 hips and as combined segmental-cavitary in 21 hips according to AAOS classification. For clinical assessment, Merle d’Aubigné and Postel hip score was assessed. Peri-operative complications were recorded. For radiological assessment, antero-posterior hip radiograph was analyzed pre-operatively, and post-operatively at one month, 6 months and every 6 months thereafter. Clear lines more than 2 mm around the sockets using DeLee and Charnley zone classification, and migration of the sockets were assessed. Hodgikinson’s type 3 (complete demarcation line) and type 4 (migration more than 5 mm or change of the angle more than 5 degrees) were classified as “loosening”. Kaplan-Meier survival analysis was performed with radiographic loosening and any re-operation (including recommendation for the re-operation) for the sockets as the end point, respectively.

The mean Merle d’Aubigné and Postel hip score improved from 11.5 points before operation to 15.7 points at the final follow-up. Though, intra-operative blow-out fracture of the acetabular floor was detected in 3 hips, re-containment had been achieved by adding metal mesh or bone graft. Clear lines at cement-bone interface were detected at zone 3 in 2 hips. Migration more than 5 mm was detected in 2 hips of type III defect at 2 years and 6 months. Re-revision was recommended for one migrated hip at 3 years and 6 months after the operation, and the other hip was stable with no clinical symptom without progressive migration at the final follow-up of 5 years. The Kaplan-Meier survival analysis, with loosening and re-operation as the end point, predicted a rate of survival of the socket of 91.7% and 95.2% at 5 years, respectively.

In conclusion, acetabular reconstruction with impaction bone grafting is attractive, but technical demanding procedure. The survival rate of the present series was compatible with the results of previous literatures. However, careful follow-up is essential, especially for the cases with massive bone defect.