Outcome of extensive varus and valgus stem alignment in short-stem THA: clinical and radiological analysis using EBRA-FCA

This study analyzed the outcome of extensive varus- and valgus stem alignment of a calcar-guided short stem regarding implant stability and radiological alterations in a 2-year follow-up. While pronounced femoral varus alignment does not cause increased instability in terms of varus/valgus tilt and axial subsidence, extensive valgus positioning is followed by significantly increased initial subsidence.

The philosophy of calcar-guided short stems emphasizes on the alignment of the stem alongside the medial calcar in different varus- and valgus position, according to the patient’s anatomy, using individual levels of neck resection [14, 15]. In a comparison of a modular short stem and a conventional straight stem, Schmidutz et al. found a significantly wider range of stem alignment for the short stem (6.2° varus to 8.8° valgus) than for the conventional stem (2.6° varus to 3.3° valgus) [17]. In 24%, stems were implanted in pronounced varus position; in 18%, they were placed in pronounced valgus position [17]. The present collective confirms a broad range of postoperative CCD angles using the investigated implant (Table 1). To position the stem individually, the level of neck resection has to be adapted accordingly. A high resection leads the stem alongside the calcar in a varus position; a valgus position is achieved by deep resection [14, 15]. However, the effect of extensive varus- and valgus alignment on implant migration following short calcar-guided stems has not been investigated. Aseptic loosening is considered to be the most common reason for implant revision [24]. In this context, primary stability, as well as the design specific potential to maintain proximal femoral bone stock is of great importance to predict implant failure [6]. For fixation of metaphyseal anchoring short stems, a sufficient cortical ring is recommended [5]. In varus anatomy, given a high neck resection, ring-fixation is accentuated. However, in valgus anatomy, given the need of a low resection level to place the stem in the intended valgus position, the ring-fixation might be compromised [5] with subsequent varus- or valgus tilt and axial subsidence after full weight bearing. Therefore, in addition, sufficient contact of the tip of the stem to the lateral cortex in regard to achieving a classical three-point anchoring is of great importance (Fig. 4). Again, in pronounced varus stem alignment, contact of the tip of the stem to the lateral cortex is commonly achieved. In valgus position, however, a missing cortical contact of the tip can be frequently observed, particularly in cases of “undersizing” (Fig. 4). Hence, the operating surgeon should be aware, that “undersizing” accompanied with a lack of contact to the lateral cortex, especially in valgus hips, might support initial instability with subsequent implant micromovement. Due to the different implant design of the investigated stem, in contrast to the Metha stem, in those cases, a resection to the level of the fossa piriformis should be pursued, to implant the adequate size of the stem and avoid “undersizing”. Earlier published results on the migration pattern of the investigated short stem showed a significant influence of body weight on the rate of early axial subsidence [19]. The regression analysis relating axial subsidence to CCD categories confirms the impact of heavy weight accordingly in the present investigation. However, the model accounted for only 6.4% of the variation.

Furthermore, it remains unclear how different resection heights and stem alignments affect load transfer and stress-shielding patterns, which are, besides micromovement of the implant, also generally thought to have a considerable effect on aseptic loosening [5]. Lerch et al. found no correlation between stem alignment, stem size, or resection level with changes in bone mineral density (BMD) in a recent investigation of the Metha short stem (Aesculap, Tuttlingen, Germany) using dual-energy X-ray absorptiometry (DEXA) [25]. However, the small size of the study population might not allow valid conclusions. Floerkemeier et al., in a biomechanical study using strain gauges on synthetic composite femora, measuring strain patterns of three different resection levels after implanting also the Metha short stem (Aesculap, Tuttlingen, Germany), found a decrease in strain in the region of the greater trochanter by 30, 32, and 52% for the high-, medium-, and low-neck resection level, respectively [5]. This corresponds to the results of this study, with stems aligned in extensive valgus position showing slightly enhanced stress-shielding in Gruen-zone 1, compared to implants in varus position, although not being statistically significant. In Gruen-zone 7, the medial calcar cortex, the biomechanical measurements resulted in slightly reduced strains for the low- and medium-resection level, while there was a strain increase of 43% for the high-resection level [5]. Again, this study confirms these findings, showing a slightly pronounced bone resorption on the medial part of the neck in valgus hips with low resection level, compared to varus hips with high-resection level, however, without any statistical significance.

Aligning the stem in an extensive varus position results in an increase in offset and subsequent enhanced lever-arm compared to the extensive valgus alignment, in turn, leading to an increased strain in the medial part of the proximal femur and the region around the distal tip of the stem [5]. Given increased forces, especially on the medial cortical bone, with some stem designs, a higher risk of intraoperative and postoperative periprosthetic fractures has been reported [9]. However, after 2 years in the present investigation, no periprosthetic fractures were observed. Theoretically, abnormal load distribution might lead to cortical hypertrophy in Gruen-zones 3 and 5 with possible subsequent thigh pain [10, 26]. In a recent study, analyzing the Fitmore stem (Zimmer, Warsaw, IN, USA), cortical hypertrophy was observed in 63% of hips accompanied with a low rate of thigh pain, implying distal loading and proximal stress bypass [27]. In contrast, the optimys stem is designed with a polished tip to retain strain peaks in the distal part and, therefore, avoid cortical reactions. The overall incidence after a follow-up of 2 years has been shown to be below 5% without any appearance of thigh pain or other clinical consequences [28]. A slight emphasis regarding cortical hypertrophy could be found for valgus hips, however, not being statistically significant. The small number of cortical hypertrophy in the present study does not allow valid conclusions regarding the impact of different extensive stem alignment.

This study has several limitations. First to be mentioned is the short follow-up of 2 years. Although only long-term results should be considered valid, the initial evaluation of bony alterations is necessary to identify undesirable results [29]. Early migration analysis may allow a prediction of implant survival [7] and the reaction of environmental bone in the early stage may help to predict the long-term outcome. Second, RSA provides higher accuracy in comparison to the EBRA-FCA method used in this study. The computer-assisted EBRA-FCA system was evaluated to be able to detect stem subsidence of ±1 mm and varus/valgus tilting of ±0.4° given a specificity of 100% and sensitivity of 78% [21]. However, the migration pattern after 2 years has been established in several studies using EBRA-FCA providing a reference to long-term survival [7, 30]. Furthermore, measuring the CCD angles in 2D radiographs lacks some accuracy compared to 3D imaging. However, radiographs were done using a standardized protocol to reduce a possible bias.