The use of short stems in primary hip arthroplasty has grown considerably in recent years, with a large variety of designs and stabilization methods available. Few studies have directly compared how these designs and fixation modes influence the bone mineral density (BMD) that is said to be better preserved with shorter implants. This led us to carry out a medium-term retrospective comparative study to (1) specify how the design of two different types of short stems influences the periprosthetic BMD, (2) compare the radiological and clinical outcomes of these stems.

The periprosthetic BMD of a short stem varies based on the implant's design.

Ninety-three patients underwent total hip arthroplasty (THA) with a short stem between July 2012 and July 2014. Two groups of patients were formed: Vitae™ short stem (Adler Ortho, Milan, France), Optimys™ short stem (Mathys, Bettlach, Switzerland). The Vitae group consisted of 46 patients, while the Optimys group consisted of 47 patients. The mean age was 66 years, and the mean body mass index (BMI) was 26 kg/m². The groups had comparable age, BMI, sex, Harris Hip score and preoperative radiological features. Periprosthetic BMD of the two types of implants was determined between the 1^st and 4th year postoperative using dual-energy x-ray absorptiometry (DEXA) in the Gruen zones. The radiological and clinical outcomes were also determined.

There was a significant decrease in BMD in zone 7(−5.8%±0.1 CI_95%[−0.09; −0.02] (p=0.003)), zone 2 (−4.8% ±0.1 CI_95%[−0.093; −0.003] (p=0.038)) and zone 6 (−11% ±0.2 CI_95%[−0.19; −0.03] (p=0.009)) in the patients with Vitae stems and in zone 2 (−13.7%±0.3 CI_95%[−0.25; −0.03] (p=0.018)) in the patients with Optimys stems. A significant difference in BMD in zone 7 at 1 year (p=0.014) and 4 years (p=0.001) postoperative and in zone 6 (p=0.011) at 4 years postoperative existed between groups, with the Optimys group having higher BMD. There was a significant increase in femoral offset in the Optimys group but not the Vitae group: Δoffset=–0.50 mm±5.2 CI_95%[−2.05; 1.05] (p=0.522) and Δoffset=2.79 mm±4.2 CI_95%[1.21; 4.37] (p=0.001). The Harris Hip score was not significantly different between the two groups at 4 years postoperative (95.3±2.5 [88; 99] vs. 95.2±2.6 [88; 99] (p=0.991)).

The design of short stems appears to influence the distribution of loads and the periprosthetic BMD. These stems are associated with very good radiological and clinical results in the medium term. Bone remodeling appears to continue beyond 1 year after the THA procedure.

III; Retrospective case-control study.