IntroductionThe Extreme™ modular stem was developed for implant revision with metaphyseal-diaphyseal defect. Due to the high breakage rate, a new “reduced modularity” design has been introduced, but without reported results. We therefore conducted a retrospective assessment of (1) overall stem survival, (2) functional results, (3) osseointegration, and (4) the rate of complications, and notably of mechanical failure. HypothesisReduced modularity reduces the risk of revision surgery for mechanical failure. Material and methodsForty-five prostheses were implanted between January 2007 and December 2010 in 42 patients with severe bone defect (Paprosky≥III) or periprosthetic shaft fracture. Mean age was 69.6years (range: 44–91years). Minimum follow-up was 5years, for a mean 115.4months (range: 60–156months). The main study endpoint was femoral stem survival, counting all-cause explantation as event. Functional assessment comprised subjective rating of satisfaction, Postel Merle d’Aubigné (PMA) and Harris Hip scores, and Forgotten Joint Score (FJS). Whether the revision assembly was carried out in situ, in the patient's hip, or outside, on the operating table, was not known in 2 cases; in the other 43, assembly was in situ in 15 cases (35%) and on the operating table in 28 (65%). ResultsFive-year stem survival was 75.7% (95% CI: 61.9–89.5%), taking all causes of change together. Seventeen patients (45.9%) had complications, 13 (35.1%) requiring revision surgery, including 10 (27.0%) for stem replacement. Five patients (13.5%) had steam breakage at the junction between the metaphysis and the diaphyseal stem, 4 of which occurred within 2 years of implantation or of fixation of a periprosthetic fracture. Mean preoperative Harris score was 48.4 [IQR (25–75% interquartile range): 37–58] and PMA score 11.1 (IQR: 10–12), compared to respectively 74 (IQR: 67–89) and 13.6 (IQR: 12.5–16) at follow-up. Mean FJS at follow-up was 71.5 (IQR: 61–94.5). In the 15 in situ assemblies, there were 3 breakages (20%), compared to 2 (7.1%) in the 28 table assemblies (p=0.21). DiscussionThe stem breakage rate was high despite the reduced modularity, which concentrated all stress on a single junction but without reducing the risk of mechanical failure. Surgical technique was faulty in some cases, with in situ assembly of the metaphysis after implanting the diaphyseal stem, which does not respect the manufacturer's recommendations. Level of evidenceIV; retrospective study.