RSA of the Symax hip stem

We read with a great interest the recent article by Kruijntjens et al. (2020). The investigators performed a 2-year model-based radiostereometric analysis (RSA) of the uncemented Symax femoral stem. The article reported no previous RSA studies on the Symax stem but we have executed a randomized double-blind, placebo-controlled trial (RCT) on the primary stability of the Symax stem in 49 postmenopausal women (Aro et al. 2018). The trial included an extended RSA follow-up for 3–5 years and the follow-up of implant survival for 8–10 years. We want to highlight some methodological differences between the 2 studies which make comparisons of interest.

The results of the 2 studies complement each other. The stem design lead to early stabilization (within 4–12 weeks) in both studies. In our RCT, the stem migration did not respond to antiresorptive therapy. All stems, independent of the amount of initial migration, osseointegrated radiographically. No revision arthroplasty was performed. Due to the low rate of clinical failure (< 2% at 9 years), no meaningful analysis of an association between early stem migration and implant survival could be carried out. In this respect, the Symax stem resembled the outcome of 7 uncemented femoral stems recently analyzed in a meta-analysis (van der Voort et al. 2015).

Based on the literature, Kruijntjens et al. concluded that there is a substantial variability in the amount of initial subsidence between stem designs. However, the comparison of different studies is challenging because any variation of stem migration may reflect more the heterogeneity of the skeletal status of study populations than the characteristics of tested femoral stem designs. The study of Kruijntjens et al. included both sexes with a mean age of 59 (30–70) years. Osteoporosis was an exclusion criterion but the measurement of local and systemic BMDs was not reported. The mean subsidence of the stem was minimal (y-translation –1.0 mm, 95% confidence interval [CI] –3.4 to 1.4). The mean stem rotation (retroversion) was 2.4° (CI –2.2 to 7.0). In our RCT, only subjects with normal BMD had minimal stem subsidence (0.7 mm, CI 0.2–1.2) and rotation into retroversion (0.8°, CI 0.3–1.4). On the contrary, osteopenic and osteoporotic subjects exhibited more stem subsidence and rotation during the first 12 weeks after surgery. The primary stability of uncemented femoral stems is sensitive to adequate bone stock (Nazari-Farsani et al. 2020). It is reasonable that all RSA arthroplasty studies have a preoperative evaluation of local and systemic BMD, if a study protocol accepts recruitment of subjects (like postmenopausal women) at a known risk of low systemic BMD.

Our RCT was performed in collaboration with the implant manufacturer, facilitating the standard marker-based RSA. Kruijntjens et al. applied model-based RSA (Kaptein et al. 2006), with experts of this method as co-investigators. Indeed, model-based RSA is highly tempting for clinical trials. The results of marker-based and model-based RSA show high agreement (Nazari-Farsani et al. 2016). Looking at the model-based RSA data of Kruijntjens et al. there was a considerable variation (CI –1.2° to 1.8°) in double examinations of y-axis rotation. The stem rotation to retroversion also had variation (CI –2.2 to 7.0). It would be great to get a comment of the investigators. Was the variation due to a actual inter-individual difference of stem rotation or only due to inherent challenges of model-based RSA in measurement of stem rotation?

Finally, Kruijntjens et al. performed the baseline RSA prior to loading of the operated hip, during the first day after surgery. They suggested a similar approach for all RSA studies. The suggestion was made without performing a comparison of different imaging and rehabilitation protocols. The current recommendation (ISO 16087:2013) is to schedule baseline RSA measurements within 5 days postoperatively, preferably before weight-bearing. 2 published studies have performed the baseline RSA imaging when the patients still were anesthetized. Interestingly, these studies showed no migration of uncemented femoral stems (Ström et al. 2007) and acetabular cups (Wolf et al. 2010) during the first week after surgery. The RCT of Ström et al. even compared the effect of different weight-bearing regimen on stem migration. The degree of early weight-bearing (unrestricted versus partial weight-bearing) did not change the migration pattern. The initial stem migration does not seem to start with the first steps of postoperative weight-bearing but progressively only after 1 week. Thus, the current recommendation for timing the baseline RSA may be still appropriate.