Long-term results of revision total hip arthroplasties using cementless stems and allografts for paprosky type IIIB and IV femoral defects short title: Revision hip arthroplasty in type IIIB and IV femoral defects.

Purpose: We determined long-term (1) clinical results of revision total hip arthroplasty (THA) in the presence of Paprosky type IIIB and IV femoral defects using validated scoring instrument; (2) osseointegration and bone remodeling; (3) the fate of cortical strut onlay allograft; (4) rates of revision and osteolysis; and (5) survivorship of the cementless stem.

Methods: We reviewed the results of 240 revision THAs in 220 patients (mean age, 59 years, range, 36-67 years) performed with an extensively porous-coated femoral stem (Solution stem; DePuy, Warsaw, Indiana) combined with a cortical strut onlay allografts for Paprosky Type IIIB and IV femoral diaphyseal bone defects from February 1994 to June 2003. Demographic data, Harris hip score, WOMAC score, UCLA activity score, and radiographic data were recorded. We determined the fate of strut cortical strut allograft and component survival rates at a mean of 26.5 years using revision and aseptic loosening as end points. Minimum follow-up was 21 years (range, 21-30 years).

Results: The clinical results improved significantly for the Harris hip score, WOMAC, and UCLA activity scores (p < 0.001). At the final follow-up, mean Harris hip, WOMAC, and UCLA activity scores were 83 ± 15 (34-100), 20 ± 15 (11-52) and 6.7 ± 1.3 (5-8) points, respectively. Of the 240 femoral stems, 218 (91%) had bone ingrowth, and 22 (9%) were unstable. Allografts were well incorporated in the host femur in all hips. The resorption of allografts was graded as mild in 192 hips (80%) and moderate in 48 hips (20%). A Kaplan-Meier survivorship analysis at a mean of 26.5 years follow-up showed that the survival rate of the femoral component was 91% (95% CI, 0.88-0.96) with re-revision for any reason as the endpoint for failure. The survival rate at a mean of 26.5 years for worst case scenario was 76% (95% CI, 0.71-0.88).

Conclusion: We found good results at a mean of 26.5 years after the revision surgery in terms of longevity and functional outcome using an extensively porous-coated stem combined with cortical strut allografts in the Paprosky Type IIIB and IV femoral diaphyseal defects. We agree that the initial quality of an uncomplicated revision of THA using supportive cortical strut allografts maintains relatively well beyond minimum 21 years of follow-up. Future studies might compare this approach with allograft-prosthesis composites, proximal femoral replacement, or modular fluted tapered stems.