What Is the Cumulative Incidence of Femoral Stem Revision and Stem Complication in Cemented and Uncemented Hip Arthroplasty for Proximal Femoral Metastatic Bone Disease?

Abstract

BACKGROUND Hip arthroplasty is often indicated in metastatic bone lesions of the proximal femur, with or without pathologic fracture. Conventional knowledge is that cemented fixation is best, although uncemented fixation has potential advantages of shorter operative time, avoidance of the physiologic stress of cement, and the chance for osseointegration. However, both techniques are options that are employed, and there is no clear evidence to guide this choice. QUESTIONS/PURPOSES In patients with proximal femoral metastatic bone lesions who were carefully selected either to receive cemented or uncemented fixation based on patient age, bone quality, tumor histology type, and the anatomic location of the lesion, we asked: (1) What is the cumulative incidence of femoral stem revision and stem complication in patients treated with cemented and uncemented hip arthroplasty for proximal femoral metastatic bone disease? (2) Are perioperative radiation and uncemented fixation independently associated with stem complication? METHODS Between January 2011 and December 2022, six centers performed 337 primary hip arthroplasties (THA or hemiarthroplasty) for proximal femoral metastatic bone disease. While these relative indications for fixation technique varied by center and surgeon, cemented fixation was used in some centers exclusively; where used selectively, it was generally used more frequently in older patients (> 65 years), any patient with poorer radiographic proximal femoral bone quality, or in the setting of pathologic fractures and/or lesions requiring intralesional resection rather than complete resection. Uncemented fixation was often selectively used in younger patients (< 65 years) with adequate radiographic proximal femoral bone quality and often for lesions where all macroscopically visible disease was removed with sufficient remaining bone to accept uncemented fixation. A total of 287 cemented reconstructions (of which 19% [55 of 287] were THAs and 81% [232 of 287] were hemiarthroplasties) and 50 uncemented reconstructions (of which 50% [25 of 50] were THAs and 50% [25 of 50] were hemiarthroplasties) were performed. A total of 66% (190 of 287) and 36% (18 of 50) of patients, respectively, had died before 2 years, and 21% (61 of 287) and 42% (21 of 50), respectively, were lost to follow-up before 2 years but were not known to have died. As expected, the groups were substantially different at baseline, with the uncemented group being younger, less likely to have had a pathologic fracture, more likely to have received attempted wide resection rather than intralesional resection, more likely to have received this fixation technique at certain centers, and more likely to have received a THA, indicating a generally better preoperative functional status. Because of those substantial baseline differences between the fixation groups, we did not compare them but rather will report each separately in terms of survivorship with respect to stem revision and stem complication and factors associated with stem complication in this retrospective study. Those lost before 2 years were included if they reached a study endpoint before being lost. Patients who underwent a resection of the proximal femur and proximal femoral replacement were not included. Femoral stem revision was defined as any femoral reoperation including femoral stem revision, femoral stem explant with or without spacer, fixation around the stem, and head-liner exchange for infection or dislocation. A stem complication was defined as aseptic loosening, periprosthetic fracture around the stem, stem breakage or fracture of the implant, or tumor recurrence around the stem. A patient with a stem complication did not have to undergo a reoperation to be included. Competing risk analysis was performed to estimate cumulative incidence (95% confidence interval [95% CI]) of femoral stem revision and stem complication, with death as a competing risk. Logistic regression assessed whether radiation or uncemented fixation were independently associated with stem complication when controlling for each other. RESULTS In all patients, the cumulative incidence (considering death as a competing risk) of femoral stem revision at 2 years in the uncemented group was 4.4% (95% CI 0.8% to 13.6%) and 1.5% (95% CI 0.5% to 3.5%) in the cemented group. The cumulative incidence of stem complication at 2 years in the uncemented group was 2.0% (95% CI 0.2% to 9.4%) and 5.2% (95% CI 3.0% to 8.4%) in the cemented group. In patients who received radiation, the cumulative incidence (considering death as a competing risk) of femoral stem revision at 2 years in the uncemented group was 0% and 3.3% (95% CI 1.1% to 7.8%) in the cemented group. The cumulative incidence of stem complication at 2 years in the uncemented group was 0% and 7.8% (95% CI 3.8% to 13.6%) in the cemented group. We did not compare the groups statistically because they were so dissimilar at baseline. The percentage of patients who underwent femoral stem revision for periprosthetic fracture in the uncemented group was 2% (1 of 50) and 2% (6 of 287) in the cemented group. The percentage of patients who developed an inpatient venous thromboembolism in the uncemented group was 0% and 2.8% (8 of 287) in the cemented group; there was one patient with bone cement implantation syndrome in the cemented group. When controlling for each other, radiation (OR 1.6 [95% CI 0.7 to 3.9]; p = 0.30) and uncemented fixation (OR 0.2 [95% CI 0.01 to 1.2]; p = 0.17) were not independently associated with stem complication. CONCLUSION Because of substantial baseline differences between our study groups (which reflect careful patient selection), we cannot say whether uncemented stems are equivalent to or superior to cemented stems. Fixation choice remains multifactorial based on patient age, bone quality, tumor histology, and the anatomic location of the lesion. These data suggest that cemented fixation remains a reliable option for all patients. However, this study found that for well-selected patients-generally those who were younger (< 65 years) with adequate radiographic proximal femoral bone quality and with lesions where all macroscopically visible disease was removed with sufficient remaining bone to accept uncemented fixation-uncemented stems can be a reasonable choice regardless of radiation status. Future comparative studies should focus on that subgroup of patients to see whether there are any specific advantages to uncemented reconstruction, such as shorter operative time, less physiologic stress of cement, and the chance for osseointegration, and if there are, whether those advantages come with any important tradeoffs. LEVEL OF EVIDENCE Level III, therapeutic study.