1 Introduction

The incidence of periprosthetic fractures (PPFs) after THA represents a substantial portion of revision procedures, as they are the third most common reason for revision surgery after aseptic loosening and infection.^1–4 PPFs impose significant limitations on patient quality of life, resulting in high morbidity and frequency of subsequent complications.⁵ Katz et al. found that in THA patients who sustained PPF, there was a three-fold higher risk of hospitalization in the subsequent year of the event.⁶

Elderly patients, particularly those with osteoporosis, represent a vulnerable population at significantly increased risk for periprosthetic fractures.^7,8 However, younger, more active patients with increased physical demand also put themselves at greater risk for PPF.⁹ Depending on fracture morphology, patients may require open reduction and internal fixation (ORIF) or a complete femoral component revision bypassing the fracture.¹⁰ Despite the high success rate of these revision surgeries, an 8–21 % mortality rate has been reported following rTHA for these PPFs.^11–13

One systematic review by Khan et al. describes re-fracture rates of Vancouver B2 fractures and B3 fractures to be 2.6 % and 1.2 %, respectively, although they do not provide a comparative rate at which other non-fracture revision indications subsequently fracture.¹⁴ Therefore, this study aimed to determine the rate of femoral re-fracture following rTHA for an initial PPF. We hypothesize that older patients and those with multiple medical comorbidities are at a higher risk for re-fracture following rTHA and are likely to experience poorer outcomes.

2 Materials and methods

3 Results

3.3

3.3 Survivability analysis

After performing Kaplan-Meier analysis on the matched cohort with re-fracture as the endpoint, the 1-year survivability of patients revised for initial PPF was 95.9 % (95 % CI, 92.4–99.5 %; Fig. 1). For patients revised for all other indications, 1-year survivability was significantly higher at 99.6 % (95 % CI, 98.9–100 %; P = 0.008).

Close

Fig. 1 Kaplan-Meier survival curves based on the indication of initial revision, with time to periprosthetic fracture as the endpoint. PPFX, periprosthetic fracture.

Export to PPT

4 Discussion

As the demand for THA continues to increase yearly, so too has the number of revision procedures. A recent projection by Shichman et al. reported an estimated increase of 9.15 % in rTHAs for each 5-year period after 2020.¹⁸ As such, it is important to evaluate complications like PPF after rTHA due to its substantial cost implications for both the patient and the healthcare system.^19,20 Our study sought to determine if there is an increased risk of re-fracture of the femur in patients who underwent rTHA for an initial PPF.

The propensity score-matched cohort analysis demonstrated that patients revised for an initial PPF had a near ten-fold risk of re-fracture compared to those revised for other indications (3.68 versus 0.37 %, P = 0.017). These findings align with our hypothesis that older patients and those with higher comorbidity indices are more at risk for re-fractures. However, the strength of our main findings is difficult to interpret, as there was a low incidence of PPFs after revision in both cohorts. Furthermore, matched cohort analysis on 1-year survivorship between patients revised for the initial PPF and those revised for other indications were similar despite significant differences at 95.9 % and 99.6 %, respectively. Longer-term data could reveal higher failure rates than those currently observed. Although the re-fracture incidence may be significantly higher in those with an initial PPF, there may be limited clinical significance given the relatively similar survivorship and readmission rates. This could suggest that despite an initial PPF during rTHA, patients tend to perform well without a significant detriment to postoperative outcomes.

Previous studies allude to the higher PPF rates following rTHA in the setting of an initial PPF. From an analysis of the Mayo Clinic total joint registry, Abdel et al. reported a nearly fivefold increase in postoperative PPF rate after rTHA compared to the primary procedure (i.e., 1.8 versus 0.4 % at one year).^3,4 Although the current study's fracture rate for non-PPF revision indications was lower compared to Abdel et al., patients revised for an initial PPF had a cumulative 1-year probability that was twice as high as their cohort.³ Another study by Lindahl et al. reported 245 of 1049 patients needing reoperation after rTHA, with the most common reasons being nonunion (5.6 %) and re-fracture (5.5 %) after initial PPF.⁵ This could be explained by increasing complexity in femoral exposure due to prior scar tissue formation, loss of bone stock in revision patients, as well as increased operative time. Surgeons also have been shown to experience increased stress and strain during revision procedures, which could predispose to increased surgical errors.²¹

The secondary aim of the study was to assess risk factors that predispose patients to a greater likelihood of sustaining a subsequent PPF. A revision procedure itself is an inherent risk factor, as Singh et al. reported a two-fold likelihood of PPF in revision TKA if their previous surgery necessitated component exchange and/or removal.²² Our initial unmatched analysis reported that patients who were women (P < 0.001), with multiple comorbidities including older age and BMI (P < 0.001), and those who underwent a posterior approach (P < 0.001) were significantly more likely to be revised for an initial PPF. Previous studies demonstrate similar findings, citing older age, female sex, and BMI as risk factors for PPF in primary THA.^23–25 As a result, some of the re-fracture risk factors are the same as the initial PPF. An elevated CCI could also serve as an indirect marker for health conditions affecting the proximal femoral bone stock.²⁶ However, the posterior approach as a risk factor contrasts a previous study by Siljander et al. that reported lower PPF rates with the posterior approach compared to other approaches.²⁷

It should be noted that the majority of re-fractures were B2 fractures that occurred in patients who had initial B2 fractures (Table 2). Vancouver B2 fractures are defined as loose stems in the setting of adequate bone stock.¹⁵ These were managed with modular diaphyseal-engaging stems that bypassed the fracture site, consistent with procedure techniques described by previous literature that maximize rotational and axial stability.^28–30 As a result, future studies should evaluate whether an initial fracture with a loose stem predisposes to a future risk of fracture with a loose stem despite addressing the initial fracture with diaphyseal fixation.

Of the 183 patients who were revised for an initial fracture, 24 (13 %) of them received cemented femoral components. We found that cemented femoral stems were more commonly used in older patients and patients with higher comorbidity indices (Table 4). Cemented femoral stems were more commonly utilized in Vancouver B3 or C patterns compared to uncemented stems (P < 0.001). This finding parallels previous studies that recommend uncemented components to be used in younger patients with adequate bone stock to avoid undue femoral cortical strain.^26,31 However, the risk for postoperative periprosthetic re-fracture did not significantly differ between cemented and uncemented subgroups, even with all five fractures occurring in the cemented subgroup. These findings coincide with Abdel et al., where there was no increased risk of fracture between the cemented and uncemented femoral component groups in rTHA.³ Interestingly, the literature would suggest that uncemented stems have a higher risk given their press-fit implantation method and slow time to full osseointegration, acting as a potential stress riser in the early postoperative period.^2,32–35 One study reported that in patients aged 65 years and older with uncemented fixation of primary THA, there was a 7.7 times increased risk for PPF,³⁵ which was a risk factor shared among the five re-fracture patients (Table 2). Additionally, postoperative outcomes and reoperations also reflected the propensity for the uncemented revisions to re-fracture, as they experienced a shorter time to failure and re-revision than cemented revisions (46.2 versus 121 days, P = 0.047).

The results from our study should be interpreted within the context of the design. The retrospective nature can limit the interpretation of results, as selection and misclassification biases can be present when analyzing existing data. Our findings are from a single-center, private academic institution in the Northeast region, which can limit their generalizability when compared to orthopaedic practices with different systems or dissimilar demographics. Although we performed a matched cohort analysis to strengthen the study's findings and eliminate confounding factors for re-fracture rate, there were a limited number of outcome events. Since PPFs are rare, prospective studies with a larger sample size are needed to further evaluate this complication's incidence. Likewise, the subgroup analysis of cemented versus uncemented stems required greater patient numbers to power the conclusions from the study sufficiently. Longer-term follow-up would also provide additional information on any heightened PPF risk in the postoperative revision setting.

In the present study, patients who underwent rTHA for an initial PPF had approximately a 3.7 % chance of sustaining a subsequent PPF. Despite a tenfold increased risk for re-fracturing compared to patients revised for other indications in rTHA, matched cohorts ultimately demonstrated comparable postoperative complication rates, readmission rates, and 1-year survivorship. Nonetheless, the high cost and resource utilization for the treatment of PPF emphasize the importance of close management to mitigate the elevated re-fracture rates. Further investigation with a greater number of outcome events is necessary to strengthen the findings of our study.

CRediT authorship contribution statement

Alan D. Lam: Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Writing – original draft, Writing – review & editing, Visualization, Supervision. Nihir Parikh: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Writing – original draft, Writing – review & editing, Supervision, Project administration. Adam S. Kohring: Validation, Formal analysis, Writing – original draft, Writing – review & editing. R. Craig Juniewicz: Methodology, Validation, Data curation, Visualization. Elizabeth Abe: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Visualization. Samuel P. Alfonsi: Methodology, Validation, Data curation, Visualization. Ryan M. Sutton: Validation, Writing – review & editing, Supervision. Chad A. Krueger: Conceptualization, Methodology, Validation, Investigation, Resources, Writing – review & editing, Supervision, Project administration.

Guardian/patient's consent

This was a retrospective study. No informed consent was obtained. Waiver of approval obtained from the Thomas Jefferson University IRB.

Ethical approval and patient consent

This study received ethical approval from the Thomas Jefferson University IRB (approval #iRISID-2023-2579). This is an IRB-approved retrospective study that received an official waiver of approval; all patient information was de-identified, and patient consent was not required. Patient data will not be shared with third parties.

Ethical statement

This study received ethical approval from the Thomas Jefferson University IRB (approval #iRISID-2023-2579). This is an IRB-approved retrospective study that received an official waiver of approval; all patient information was de-identified, and patient consent was not required. Patient data will not be shared with third parties.

Funding statement

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.