A prospective cohort study on the short collum femoris-preserving (CFP) stem using RSA and DXA. Primary stability but no prevention of proximal bone loss in 27 patients followed for 2 years

Abstract

Background and purpose: Short femoral stems have been introduced in total hip arthroplasty in order to save proximal bone stock. We hypothesized that a short stem preserves periprosthetic bone mineral density (BMD) and provides good primary stability.

Methods: We carried out a prospective cohort study of 30 patients receiving the collum femoris-preserving (CFP) stem. Preoperative total hip BMD and postoperative periprosthetic BMD in Gruen zones 1-7 were investigated by dual-energy x-ray absorptiometry (DXA), stem migration was analyzed by radiostereometric analysis (RSA), and the Harris hip score (HHS) was determined.

Results: 2 patients were excluded intraoperatively and 1 patient was revised due to a deep infection, leaving 27 patients for analysis. The mean HHS increased from 49 (24-79) preoperatively to 99 (92-100) after 2 years. DXA after 1 year showed substantial loss of BMD in Gruen zone 7 (-31%), zone 6 (-19%), and zone 2 (-13%, p < 0.001) compared to baseline BMD determined immediately postoperatively. The bone loss in these regions did not recover after 2 years, whereas the more moderate bone loss in Gruen zones 1, 3, and 5 partially recovered. There was a correlation between low preoperative total hip BMD and a higher amount of bone loss in Gruen zones 2, 6 and 7. RSA showed minor micromotion of the stem: mean subsidence was 0.13 (95% CI: -0.28 to 0.01) mm and mean rotation around the longitudinal axis was 0.01º (95% CI: -0.1 to 0.39) after 2 years.

Interpretation: We conclude that substantial loss in proximal periprosthetic BMD cannot be prevented by the use of a novel type of short, curved stem, and forces appear to be transmitted distally. However, the stems showed very small migration-a characteristic of stable uncemented implants.

Figure 1.

The collum femoris-preserving (CFP) stem is available with two different curvatures in order to accommodate variations in femoral anatomy.

Figure 2.

Changes in periprosthetic BMD in all Gruen zones. The p-values are derived from a general linear model with a repeated-measures design (* p < 0.05; n.s.: not statistically significant compared to baseline or between the groups). Error bars show 95% CI.

Figure 3.

Migration of all individual components along the y-axis, representing stem subsidence measured in mm. Precision error = 0.09 mm. 1 stem continued to subside up to 2 years after surgery, but absolute subsidence was 1.67 mm and the patient was clinically asymptomatic.

Figure 4.

Migration of all individual components around the y-axis, representing stem rotation measured in degrees. Precision error = 0.83°. The same stem that was found to subside more than average in Figure 3 showed an abnormal rotation pattern, but maximum rotation was 1.4°.