Differences between proximal bone remodeling in femoral revisions for aseptic loosening and periprosthetic fractures using the Wagner SL stem

Abstract

Background: Monoblock taper fluted stems have been reliably used to treat proximal femoral periprosthetic fractures (PFF) and femoral aseptic loosening (AL). Although proximal femoral remodeling has been observed around the Wagner Self-Locking (SL) stem, the exact characteristics of this process are yet to be established. Our aim was to compare the remodeling that takes place after femoral revisions for PFF and AL.

Methods: Consecutive patients between January 2015 and December 2017 undergoing femoral revision using the Wagner SL stem for PFF or AL without an extended trochanteric osteotomy (ETO) or bone grafting were selected from our database. Radiological follow-up was performed using plain antero-posterior hip radiographs taken postoperatively and at 3, 6, 12 months and at 24 months. The Global Radiological Score (GRxS) was utilized by four blinded observers. Intra and interobserver variability was calculated. Secondary outcome measures included the Oxford Hip Score and the Visual Analog Scale for pain.

Results: We identified 20 patients from our database, 10 PFF and 10 AL cases. The severity of AL was Paprosky 2 in 2 cases, Paprosky 3A in 2 cases and Paprosky 3B in 6. PFF were classified as Vancouver B2 in 7 cases and Vancouver B3 in 3 cases. Patients undergoing femoral revision for PFF regained 89% (GRxS: 17.7/20) of their bone stock by 6 months, whilst patients with AL, required almost 2 years to achieve similar reconstitution of proximal femoral bony architecture 86% (GRxS: 17.1/20). Inter-observer reproducibility for numerical GRxS values showed a "good" correlation with 0.68, whilst the intra-observer agreement was "very good" with 0.89. Except immediate after the revision, we found a significant difference between the GRxS results of the two groups at each timepoint with pair-wise comparisons. Functional results were similar in the two groups. We were not able to show a correlation between GRxS and functional results.

Conclusions: Proximal femoral bone stock reconstitutes much quicker around PFF, than in the cases of AL, where revision is performed without an ETO. The accuracy of GRxS measurements on plain radiographs showed good reproducibility, making it suitable for everyday use in a revision arthroplasty practice.

Keywords: Bone remodeling; Bone-prosthesis Interface; Hip prosthesis; Periprosthetic fractures; Prosthesis loosening; Revision arthroplasty.

Fig. 1

Example for aseptic loosening of the stem. a: Preoperatively massive, Paprosky 3B type resorption of proximal femoral bone stock around a cemented stem. b: Wagner SL stem fixed in the healthy diaphysis below the lytic region without an ETO (Patient 11)

Fig. 2

Example for periprosthetic femur fracture. a: Vancouver B2 type periprosthetic fracture with component loosening around a cementless stem; b: Anatomical reduction with two cables around a Wagner SL stem (Patient 10)

Fig. 3

Example for evaluation of Gruen zones on AP radiograph (Patient 18)

Fig. 4

Example for SBS (white) and O-SS (yellow) measurements in AL group. a: IBS -2, O-SS 14 immediate after the revision, GRxS: 4; b: SBS 18, O-SS 20 at latest follow-up, GRxS: 18 (Patient 18)

Fig. 5

Example for SBS (white) and O-SS (yellow) measurements in PFF group. a: IBS 10, O-SS 14 at first immediate after the operation, GRxS: 13; b: SBS 20, O-SS 20 at latest follow-up, GRxS: 20 (Patient 07)

Fig. 6

Example of a PFF case. a: Immediate postop radiograph after revision. The poor quality of the proximal bone stock is unequivocal. b: obvious new bone formation seen at 6 month follow-up. There is some subsidence. c: Radiograph 2 years after the operation. Complete reconstitution of bone stock. Ai-Bi-Ci: Shows the same pictures in inverted view (Patient 07)

Fig. 7

Example for AL case. a: Postoperative radiograph after revision. Obvious proximal femoral lysis. b: After 12 months there is mild subsidence, and continuous bony remodeling. c: Almost complete reconstitution of bone stock at 2 years. Ai-Bi-Ci: Shows the same pictures in inverted view (Patient 17)

Fig. 8

The GRxS means of AL and PFF groups at the follow-up timepoints. There is a significant difference between the results of the groups at each timepoint (except immediate after the operation) with paired Wilcoxon signed-rank test (0 = immediate after the operation (p-value = 0.08198), after 3 months (p-value = 0.03412), 6 months (p-value = 0.008492), 12 months (p-value = 0.0213), and 24 months (p-value = 0.01788))

Fig. 9

Change of GRxS group classification between immediate postoperative and last follow-up measurements