Lisa C. Howard MD, Bas A. Masri MD, Don S. Garbuz MD, and Clive P. Duncan MD Department of Orthopaedics, University of British Columbia, Vancouver, BC, Canada The question of risk factors in periprosthetic fractures is debated extensively in the literature and as such, a review of the current evidence is helpful to identifying these risks as they pertain to this common complication. Periprosthetic femur fractures after THA can be a difficult complication to manage. Recognition of features associated with periprosthetic fractures could allow for prophylactic measures to prevent a fracture in the at‐risk patient. One level II prospective registry study was identified,1 while the remainder were generally level III,2–9 including several large database and registry studies.10–14 A summary of cohort and case‐control studies classified as level III evidence noted a tendency of periprosthetic fracture patients to be female gender.2–5,10–12 A meta‐analysis concluded an odds ratio (OR) of 1.534 (95% confidence interval [CI]: 1.345–1.749, p <0.001), while a Swedish registry study showed a relative risk of 1.6 (95% CI: 1.1–2.3).6,13 A prospective registry study as well as a retrospective cohort did not conclude gender to be a risk factor.1,7 Abdel et al. in their retrospective review of 32 644 primary THAs found female gender to be a risk factor of intraoperative (CI: 1.2–1.7; p <0.001) but not postoperative fractures.14 Several retrospective studies have shown age to be a risk factor for fracture.7,8,10,12,13 One meta‐analysis concluded that age greater than 80 is a risk factor (OR = 4.203; 95% CI: 2.859–6.181; p <0.001)6 while another prospective cohort study concluded that age is a risk factor with a relative risk of 1.4 for each decade of increasing age (95% CI: 1.2–1.6; p <0.001). In contrast, Lindahl et al. found a significantly higher fracture rate in younger patients (p <0.001),1 while a large‐scale retrospective cohort study did not find age to be a risk factor.14 In the prospective Swedish National Registry study, only 6% overall were patients with rheumatoid arthritis. They had an increased prevalence in the periprosthetic fracture group, with an incidence of 11 and 10% in the primary and revision populations, respectively (p <0.001).1 Zhu et al. in their meta‐analysis showed that rheumatoid arthritis is a statistically significant risk factor for periprosthetic fracture (p <0.001).6 A large‐scale prospective cohort study reported osteoporosis to increase fracture risk by 2.8 times (95% CI: 1.6–4.8; p <0.001).12 Wu et al. found that preoperative osteoporosis was a significant predictor for fracture.8 In a small case control study, Sarvilinna reported a risk ratio of 4.4 (95% CI: 1.4–14) for periprosthetic fracture if the primary diagnosis for arthroplasty was fracture.9 A large‐scale 40‐year follow‐up retrospective cohort documented the incidence of fracture to be 0.4% at 1 year, 0.8% at 5 years, 1.6% at 10 years, and 3.5% at 20 years.14 Lindahl et al. reported an average of 7.4 years post primary THA (688 fractures), and 3.9 years post revision hip arthroplasty (361 fractures).1 The timing of periprosthetic fractures from the index procedure, however, seems to relate in part to the type of prosthesis used. A large prospective cohort study reported that 77% of all prostheses used were uncemented and the incidence of fracture was 2.4% for uncemented and 0.9% for cemented (p <0.001) hips.12 The Nordic registry reported that fracture risk for uncemented stems was highest in the first six months, while fractures with cemented stems tended to occur later.13 Abdel et al. found that, overall, uncemented stems had a higher prevalence of fracture at all time points up to 20 years (p <0.001).14 The reason for early postoperative fractures in uncemented components appears to be related to technical intraoperative errors that lead to fractures in the period of latency before osteointegration.13,15 Late postoperative fractures have been associated with loosening and osteolysis.15,16 For instance, in the Swedish Hip Arthroplasty Register 70% of implants were noted to be loose prior to periprosthetic fracture.1 The Vancouver Classification is the most widely used periprosthetic fracture classification and the evidence behind its use is important to review in the context of this complication. The variability of pathology seen in periprosthetic fractures of the femur necessitates an effective classification system to aid in the communication of diagnoses among surgical colleagues and to develop a management plan. The ability to classify these fractures properly will assist in their management. The two studies available to address this question are validation studies.17,18 A useful classification system incorporates clinical and radiographic information to guide management, allowing for appropriate treatment and comparison of similar fractures. In 1995, Duncan and Masri published the Vancouver Classification System (VCS), which emphasized the quality of the prosthetic–bone interface (stability) as well the host bone stock in the therapeutic decision‐making process.19,20
30 Periprosthetic Hip Fractures
Clinical scenario
Top three questions
Question 1: In patients who sustain a periprosthetic femur facture, are there factors that may be predictive of this complication after primary THA?
Rationale
Clinical comment
Available literature and quality of the evidence
Findings
Gender
Patient age
Rheumatoid arthritis
Osteoporosis and prior fragility fracture
Timing to fracture and implantation method
Resolution of clinical scenario
Question 2: In patients with periprosthetic fractures of the femur, is there a validated classification system that has satisfactory intraobserver and interobserver reliability and validity that aids in therapeutic planning?
Rationale
Clinical comment
Available literature and quality of the evidence
Findings
Stay updated, free articles. Join our Telegram channel