Internal fixation versus revision arthroplasty for Vancouver B2–B3 fractures: mortality and functional outcomes in frail patients. Insights from the PIPPAS study of 485 patients Authors: Jesús Moreta, Héctor J Aguado, Pablo Castillón-Bernal, Josep M Muñoz-Vives, Pilar Camacho, Montsant Jornet-Gibert, Jordi Teixidor, Adela Pereda-Manso, Yaiza García-Sánchez, Cristina Ojeda-Thies, Pablo García-Portabella, Elvira Mateos Álvarez, David Noriega-González, María-Fe Muñoz-Moreno, Irene Arroyo-Hernantes, Begoña Aránzazu Álvarez-Ramos, Belén García-Medrano, Carmen Martínez-Sellés, Sergio Marín-Jiménez, Virginia García-Virto, Sergio País-Ortega, Adriana Acha, Jordi Tomás-Hernández, Jordi Selga-Marsà, José Vicente Andrés-Peiró, Carlos Piedra-Calle, Ferrán Blasco-Casado, Ernesto Guerra-Farfán, Jordi Querolt-Coll, Guillermo Triana-López de Santamaría, José M Hernández, Marina Renau-Cerrillo, Carles Gil-Aliberas, Anna Carreras-Castañer, Marian Vives-Barquiel, Eliam Ajuria Fernández, Eugenia Fernández Manzano, Unai García De Cortázar, Mirentxu Arrieta, Daniel Escobar Published: 2025 ABSTRACT AbstractBackgroundPeriprosthetic femoral fractures following hip arthroplasty (FH-PPF) represent a severe complication, especially in elderly patients with compromised health. Traditionally, revision arthroplasty is recommended for B2–B3 FH-PPF, yet internal fixation has emerged as a debated alternative in select patients. The hypothesis was that fixation, in selected patients with B2–B3 FH-PPF, decreases mortality and surgical complication rates with the same functional outcomes as revision arthroplasty.Materials and methodsPIPPAS is a multicenter prospective observational study. This cohort substudy includes 485 patients across 57 hospitals with B2–B3 FH-PPF between January 2021 and May 2023. Management strategy, revision or fixation, was at the attending surgeon’s discretion. Propensity score matching, controlled for age, age-adjusted Charlson Comorbidity Index (a-CCI), prefracture mobility, Pfeiffer scale, and ASA score, was done. Mortality risk factors were assessed using univariate and multivariate analysis.ResultsOut of 485 patients, 164 received fixation, and 321 underwent revision. Fixation patients were older (88 versus 82 years, p < 0.001) and frailer. Fixation was associated with shorter hospital stay (13 versus 15 days, p = 0.003) but higher 1-year mortality (25% versus 14.3%, p = 0.04). There were no differences in medical or surgical complications (p = 0.83 and p = 0.36) at any time, but dislocation rate was higher in the revision group (p = 0.001). The 1-year mortality rate in patients with no weight-bearing restrictions was higher for the revision group (p = 0.01). The propensity score matching showed higher 1-year mortality rate in the fixation group but no differences in functional outcomes, complications, or up to 6-months mortality. In the multivariate analysis a-CCI, cognitive impairment, B3 fractures, and prefracture independent walking impairment were independent mortality risk factors.ConclusionsRevision arthroplasty showed less 1-year mortality rate and weight-bearing restrictions than fixation. However, frail patients with B2–B3 FH-PPF managed with fixation allowing full weight-bearing showed a lower 1-year mortality rate. Fixation in B2–B3 FH-PPF is a treatment option in frail patients, while aiming for stable constructions allowing full weight-bearing.Level of Evidence II: prospective cohort study.Trial registration: ClinicalTrials.gov (NCT04663893)Graphical Abstract Supplementary InformationThe online version contains supplementary material available at 10.1186/s10195-025-00886-3.Keywords: Periprosthetic fracture, Orthogeriatrics, Vancouver B2 fracture, Vancouver B3, Mortality, Revision, Fixation, Frailty, Weight-bearing, Total hip arthroplasty Introduction Periprosthetic femoral fractures after hip arthroplasty (FH-PPF) are a severe complication. Their incidence is rising worldwide [1] owing to the increase in the global number of hip arthroplasty procedures and the high prevalence of elderly patients carrying a hip arthroplasty. According to the Unified Classification System (UCS), B type fractures sit around or just below the stem and are subdivided into B1 when the stem is fixed, B2 if the stem is loose with adequate bone stock, and B3 when the stem is loose and the bone quality is poor [2]. The management strategy recommended for B2–B3-type FH-PPF is revision arthroplasty of the stem [2–5]. As these fractures occur mainly in elder and frail patients with poor functional status, there is recent controversy about whether this treatment is ideal for all patients. Therefore, there is an ongoing debate whether management with internal fixation can be a valid treatment option in selected cases [6–9]. Recent studies comparing revision versus fixation alone present similar reoperation rates [8–14]. Our hypothesis was that fixation, in selected patients with B2–B3 FH-PPF, decreases mortality and surgical complication rates with the same functional outcomes as revision arthroplasty. Materials and methods The PIPPAS study is a collaborative multicenter prospective observational case series study (level IV evidence) evaluating PPF and peri-implant fractures in 56 Spanish hospitals and one hospital in Buenos Aires, Argentina [15]. B2–B3 FH-PPF management was the standard of care at each participating site, as determined by the attending surgeon. This cohort substudy, fixation versus revision arthroplasty, included patients aged 18 years or older who presented with a B2- or B3-type FH-PPF between January 2021 and May 2023, with available 1-year follow-up clinical data. We excluded patients managed nonoperatively, patients with intraoperative or pathologic fractures, patients with fractures between a hip stem and any other distal implant, and pregnant patients. Written consent for participation in the study was obtained from all participants or their legal representatives. The primary outcome was 1-year mortality rate; secondary outcomes were postoperative weight-bearing restrictions, 1-year surgical complication rate (including dislocation, refracture, infection, nonunion, and revision surgery), medical complication rate, worsening of mobility, cognitive impairment or clinical frailty score (CFS), moving to a nursing home, health-related quality of life, and fracture healing. Prospective data collection included patient demographics, management, and outcomes based on the Fragility Fracture Network’s Minimum Common Dataset for hip fracture audits but adapted to the specific nature of FH-PPFs [16] (appendix 1). Cognitive status was assessed with the Pfeiffer Short Portable Mental Status Questionnaire (SPMSQ) [17]. Distinction between B1 and B2–B3 types was at the discretion of the treating team based on clinical, radiological and surgical findings. Previous clinical history of thigh pain was suggestive of loosening. X-rays showing subsidence, progressive lucencies in multiple Gruen zones, pedestal formation, and calcar remodeling were indicative of type B2 fractures or type B3 fractures if there were also sings of osteolysis and poor supportive bone stock [18]. The intraoperative assessment included direct testing with traction, rotation and visual assessment of cement–bone or stem–bone interface. We classified surgeons into two groups of expertise: those performing more than 20 revision arthroplasty procedures and/or more than 20 minimally invasive fixations per year and those performing fewer than 20 procedures in either category. The weight-bearing protocol at hospital discharge was determined by the surgical team at each site according to local practice. However, the prescription of weight-bearing restrictions did not necessarily ensure patient adherence. Several patients may have performed weight-bearing as tolerated after discharge, which was often only identified at the 30-day follow-up visit, at which time the weight-bearing recommendation was usually revised. Health-related quality of life was assessed using the EQ-5D-5L instrument at 6 and 12 months [19]. Fracture healing was defined as the presence of at least three cortical callus bridges on radiographic examination and pain-free full weight-bearing. A comprehensive list of variables is available in the Supplementary Material. Data were collected and managed using REDCap electronic data capture tools hosted at the Instituto de Estudio de Ciencias de la Salud de Castilla y León, Spain [20]. The manuscript was adapted to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement [21]. The study was conducted in accordance with the ethical standards laid down in the 1964 Helsinki Declaration and received approval from the institutional review boards of the coordinating center and each participating hospital. This study is registered at ClinicalTrials.gov (NCT04663893). Statistical analysis Descriptive statistics summarized group epidemiological data and outcomes between “fixation” and “revision arthroplasty” groups. Continuous variables were summarized as mean and standard deviation (SD) or median and interquartile range (IQR) as appropriate (p < 0.05 Shapiro–Wilk test). Categorical variables were summarized by absolute frequencies and percentages. Pearson’s chi-squared test analyzed the association between qualitative variables, the Kruskal–Wallis test for quantitative variables, or the analysis of variance (ANOVA) test for quantitative variables with normal distribution. Fisher’s exact test or the likelihood ratio test was used for variables with more than two categories. Risk factors for 1-year mortality and clinical outcomes were assessed with univariate logistic regression analysis. Variables with a p-value < 0.05 were selected for a multivariate logistic regression analysis. A predictive equation was obtained to predict mortality. A receiver-operating characteristic (ROC) curve represented the discriminative capacity of the model. Propensity score matching techniques were used to minimize risks of bias between both groups in age, prefracture mobility, prefracture place of residency, age-adjusted Charlson Comorbidity Index (a-CCI), Pfeiffer, and presence of a dysfunctional prosthesis with a 1:1 ratio. Uni- and multi-variate regression analysis were performed afterwards. Kaplan–Meier estimators were used to estimate survival functions, and comparisons were made using the log-rank test. p-values < 0.05 were considered statistically significant. Statistical analysis was conducted using RStudio (v.4.3.2; R Foundation for Statistical Computing, Vienna, Austria). Results The study involved 485 patients meeting the inclusion criteria, with two groups: 164 patients managed with internal fixation and 321 patients with revision arthroplasty. Patients managed with fixation were older (88 versus 82 years, p < 0.001), frailer according to the CFS, the American Society of Anesthesiologists physical status classification system (ASA), and the a-CCI, presented higher cognitive impairment (Pfeiffer 3 versus 2, p < 0.001), living at nursing healthcare facilities (21.3% versus 10.9%, p < 0.001), and presented worse mobility (p = 0.009). The fixation group had shorter surgical delay (128 versus 168 h, p = 0.007), more patients operated under spinal anesthesia (80.98% versus 56.39%, p < 0.001) and received a higher proportion of comanagement with geriatricians (46.34% versus 35.51%, p = 0.025) (Table 1). Table 1Demographic, baseline data for patients presenting a type B2 or B3 femoral hip periprosthetic fracture (FHPPF), management and hospital outcomes: Qualitative variables are summarized using counts and percentages (%)TotalFixationRevisionp-Valuen485164321Baseline: Age (years)*84 [14]88 (8)82 (15)< 0.001Sex, n (%): Female306 (63.09%)112 (68.29%)194 (60.44%)0.110 Male179 (36.91%)52 (31.71%)127 (39.56%)Place of residence, n (%): Community415 (85.57%)129 (78.66%)286 (89.1%)0.003 Health institution70 (14.43%)35 (21.34%)35 (10.9%)Prefracture mobility*, n (%): Independent outdoors354 (73.14%)106 (64.63%)248 (77.5%)0.009 Independent indoors86 (17.77%)37 (22.56%)49 (15.31%)Dependent44 (9.09%)21 (12.8%)23 (7.19%) Pfeiffer’s SPMSQ*2 (5)3 (4)2 (4)< 0.001 CFS*4 (3)5 (2)4 (3)0.001 ASA*3 (1)3 (1)3 (1)0.037 CCI age-adjusted*5 (2)5 (2)5 (2)0.001Osteoprotective treatment, n (%): Yes136 (28.04%)53 (32.32%)83 (25.86%)0.164 No349 (71.96%)111 (67.68%)238 (74.14%)Antiaggregant or anticoagulant medication, n (%): Yes190 (39.18%)75 (45.73%)115 (35.83%)0.038 No295 (60.82%)89 (54.27%)206 (64.17%) Hb at admission (gr/dL)^12.08 (1.87)11.87 (1.98)12.19 (1.81)0.119Fracture features: UCS type: B2389 (80.21%)137 (83.54%)252 (78.5%)0.541 B396 (19.79%)27 (16.46%)69 (21.5%)Cemented or uncemented stem Cemented142 (29.28%)55 (33.54%)87 (27.1%)0.171 Uncemented343 (70.72%)109 (66.46%)234 (72.9%) Time since stem implantation (months)*120 (131.75)130 (126.5)115 (141)0.023Dysfunctional prosthesis Yes225 (46.49%)66 (40.24%)159 (49.69%)0.061 No259 (53.51%)98 (59.76%)161 (50.31%)Management Surgical delay (h)*154.71 (593.26)128.69 (105.29)168 (795.39)0.007Anesthesia General171 (35.33%)31 (19.02%)140 (43.61%)< 0.001 Neuroaxial313 (64.67%)132 (80.98%)181 (56.39%)Team expertise: < 20 procedures per year157 (32.37%)57 (34.76%)100 (31.15%)0.484 > 20 procedures per year328 (67.63%)107 (65.24%)221 (68.85%)Surgical approach: Less invasive34 (7.01%)34 (20.73%)0< 0.001 Open451 (92.99%)130 (79.27%)321 (100%)Was the stability of the stem checked? Yes444 (91.55%)123 (75%)321 (100%)< 0.001 No41 (8.45%)41 (25%)0Cerclage for reduction Yes430 (88.66%)138 (84.15%)292 (90.97%)0.037 No55 (11.34%)26 (15.85%)29 (9.03%)Type of fixation: Plate242 (49.90%)152 (92.68%)90 (28.04%)< 0.001 Cerclage wiring144 (29.69%)12 (7.32%)132 (41.12%)Medical staff involved⋄ Geriatrics190 (39.18%)76 (46.34%)114 (35.51%)0.025 Internal Medicine or others196 (40.41%)64 (39.02%)132 (41.12%) No99 (20.41%)24 (14.63%)75 (23.36%)Hospital outcomes: Length of hospital stay* (days)14 (11)13 (10.25)15 (11)0.003 Need of transfusion Yes338 (69.98%)105 (64.42%)233 (72.81%)0.0721 No145 (30.02%)58 (35.58%)87 (27.19%)Initial postoperative mobilization out of bed < 48 h330 (68.46%)118 (71.95%)212 (66.67%)0.280 > 48 h152 (31.54%)46 (28.05%)106 (33.33%)Weight-bearing permission Full weight-bearing186 (38.51%)34 (20.86%)152 (47.5%)< 0.001 Not allowed or with restrictions297 (61.49%)129 (79.14%)168 (52.5%)Change of residence at hospital discharge Same316 (69.15%)95 (61.69%)221 (72.94%)0.010 Community to health institution137 (29.98%)59 (38.31%)78 (25.74%) Health institution to community4 (0.88%)04 (1.32%)Continuous variables were summarized as mean and standard deviation (SD)^, or median and interquartile range (IQR)*. Categorical variables were summarized by absolute frequencies and percentages. Prefracture mobility scale: registered as 1: complete independent gait, 2: outdoors independent gait with one technical aid, 3: outdoors independent gait with two technical aids, 4: only indoors independent gait with or without aids, 5: no mobility at all or with the help of two other people, and grouped as independent outdoors (nos. 1, 2, and 3), independent indoors (no. 4), and dependent (no. 5). ^Osteoprotective treatment: percentages of each individual treatment are referred to the total number of patients receiving osteoprotective treatment⋄ Medical staff involved in the patient’s care (other than trauma and anesthesia)Surgeon experience: more or less than 20 less invasive fixation procedures or revisions for PPF per yearIQR interquartile range, Pfeiffer’s SPMSQ Pfeiffer’s Short Portable Mental Status Questionnaire, CFS clinical frailty scale, ASA American Society of Anesthesiologists (ASA) physical status classification system, Hb Hemoglobin Demographic, baseline data for patients presenting a type B2 or B3 femoral hip periprosthetic fracture (FHPPF), management and hospital outcomes: Qualitative variables are summarized using counts and percentages (%) Continuous variables were summarized as mean and standard deviation (SD)^, or median and interquartile range (IQR)*. Categorical variables were summarized by absolute frequencies and percentages. Prefracture mobility scale: registered as 1: complete independent gait, 2: outdoors independent gait with one technical aid, 3: outdoors independent gait with two technical aids, 4: only indoors independent gait with or without aids, 5: no mobility at all or with the help of two other people, and grouped as independent outdoors (nos. 1, 2, and 3), independent indoors (no. 4), and dependent (no. 5). ^Osteoprotective treatment: percentages of each individual treatment are referred to the total number of patients receiving osteoprotective treatment ⋄ Medical staff involved in the patient’s care (other than trauma and anesthesia) Surgeon experience: more or less than 20 less invasive fixation procedures or revisions for PPF per year IQR interquartile range, Pfeiffer’s SPMSQ Pfeiffer’s Short Portable Mental Status Questionnaire, CFS clinical frailty scale, ASA American Society of Anesthesiologists (ASA) physical status classification system, Hb Hemoglobin Outcomes are presented in Table 2. Patients managed with fixation showed higher rates of 1-year mortality (25% versus 14.33%, p = 0.04), with no differences in mortality up to the 6-month follow-up. Patients with fixation had more weight-bearing restrictions (20.86% versus 47.5%, p < 0.001 after surgery, and 31.43% versus 56.84%, p < 0.001 at 30-day follow-up) and worse EQ5D at 6- and 12-month follow-up (0.62 versus 0.66, p = 0.003 and 0.61 versus 0.66, p = 0.014) but shorter hospital stay (13 versus 15 days, p = 0.003). The Kaplan–Meier analysis demonstrated a significantly higher 1-year survival rate for the revision arthroplasty group compared with the fixation group (p = 0.032) (Fig. 1). There were no differences in medical or surgical complication rates at any time point, although the dislocation rate was higher for the revision group (p = 0.001) (Table 3). Patients with no weight-bearing restrictions at hospital discharge showed lower 1-year mortality rates in the fixation group than those in the revision group (p = 0.014). Patients with no weight-bearing restrictions at 30-day follow-up showed no differences in 1-year mortality rates (p = 0.14) between both groups. Table 230-day, 6-month, and 1-year follow-up outcomes and mortality for patients presenting a type B2 or B3 femoral hip periprosthetic fracture (FHPPF): Qualitative variables are summarized using counts and percentages (%)TotalFixationRevisionp-Valuen48516432130-day follow-upWeight-bearing permission at 30 days: Full weight-bearing206 (48.47%)44 (31.43%)162 (56.84%)< 0.001 Not allowed or with restrictions219 (51.53%)96 (68.57%)123 (43.16%)Change of mobility at 30 days: Same154 (36.32%)37 (26.43%)117 (41.2%)0.002 Worse14 (3.3%)2 (1.43%)12 (4.23%) Better256 (60.38%)101 (72.14%)155 (54.58%)Place of residence at 30 days: Community243 (57.18%)62 (44.29%)181 (63.51%)< 0.001 Health institution182 (42.82%)78 (55.71%)104 (36.49%)6-month follow-up EQ-5D*0.65 (0.32)0.62 (0.37)0.66 (0.28)0.003Is the fracture healed? Yes229 (69.18%)95 (90.48%)134 (59.29%) No43 (13.0%)10 (9.52%)33 (14.6%)< 0.001 Does not apply59 (17.82%)059 (26.11%)1-year follow-up EQ-5D*0.64 (0.41)0.61 (0.43)0.66 (0.31)0.014Is the fracture healed? Yes202 (75.38%)73 (92.41%)129 (68.25%) No20 (7.47%)6 (7.59%)14 (7.41%)0.895 Does not apply46 (17.25%)46 (24.34%)Change of EQ-5D from 6-month follow-up Same51 (19.54%)15 (19.48%)36 (19.57%)0.766 Better103 (39.46%)28 (36.36%)75 (40.76%) Worse107 (41%)34 (44.16%)73 (39.67%)Pfeiffer’s SPMSQ2 (4.75)3 (4.5)2 (4.75)0.002Change of Pfeiffer’s SPMSQ from baseline Same83 (34.73%)25 (31.65%)63 (37.28%)0.423 Better57 (23.85%)11 (13.92%)38 (22.49%) Worse99 (41.42%)43 (54.43%)68 (40.24%) CFS*5 (2)6 (2.25)5 (3)0.008Change of CFS from baseline Same72 (27.59%)20 (28.57%)47 (25.82%)0.348 Better49 (18.77%)19 (27.14%)38 (20.88%) Worse140 (53.64%)31 (44.29%)97 (53.3%)Change of residence from baseline Same229 (84.81%)65 (81.25%)164 (86.32%)0.202 Community to health institution38 (14.07%)15 (18.75%)23 (12.11%) Health institution to community3 (1.11%)03 (1.58%)Change of mobility from baseline Same181 (67.29%)47 (58.75%)134 (70.9%)0.090 Better12 (4.46%)6 (7.5%)6 (3.17%) Worse76 (28.25%)27 (33.75%)49 (25.93%)Cumulative mortality In-hospital28 (5.77%)10 (6.1%)18 (5.61%)0.990 30-day follow-up36 (7.42%)15 (9.14%)21 (6.54%)0.395 6-month follow-up69 (14.23%)31 (18.90%)38 (11.84%)0.056 1-year follow-up87 (17.94%)41 (25.0%)46 (14.33%)0.004Medical complications In-hospital225 (46.39%)82 (50%)143 (44.55%)0.297 30-day follow-up102 (21.03%)37 (22.56%)65 (20.25%)0.636 6-month follow-up189 (38.97%)69 (42.07%)120 (37.38%)0.366 1-year follow-up248 (51.13%)92 (56.1%)156 (48.6%)0.142 At any moment412 (84.95%)138 (84.15%)274 (85.36%)0.827Surgical complications 30-day follow-up116 (23.92%)32 (18.29%)84 (26.79%)0.050 6-month follow-up167 (34.43%)55 (33.54%)112 (34.89%)0.845 1-year follow-up222 (45.77%)79 (48.17%)143 (44.55%)0.509 At any moment266 (54.85%)87 (53.05%)179 (55.76%)0.364Continuous variables were summarized as median and interquartile range (IQR)*. Categorical variables were summarized by absolute frequencies and percentages. IQR interquartile range, Pfeiffer’s SPMSQ Pfeiffer’s Short Portable Mental Status Questionnaire, CFS clinical frailty scale 30-day, 6-month, and 1-year follow-up outcomes and mortality for patients presenting a type B2 or B3 femoral hip periprosthetic fracture (FHPPF): Qualitative variables are summarized using counts and percentages (%) Continuous variables were summarized as median and interquartile range (IQR)*. Categorical variables were summarized by absolute frequencies and percentages. IQR interquartile range, Pfeiffer’s SPMSQ Pfeiffer’s Short Portable Mental Status Questionnaire, CFS clinical frailty scale Kaplan–Meier analysis for 1-year survival rate between fixation and revision arthroplasty groups Table 3Surgical complications at 30-days, at 12-month follow-up, and cumulative. * p < 0.0530-day follow-up12-month follow-upCumulative (1 year)FixationRevisionpFixationRevisionpFixationRevisionpFracture09 (2.81%)0.0702 (1.23%)5 (1.56%)13 (1.83%)15 (4.67%)0.1690Dislocation1 (0.61%)24 (7.5%)< 0.001*2 (1.23%)18 (5.62%)0.040*3 (1.83%)45 (14.02%)0.0013*Infection7 (4.29%)22 (6.88%)0.3546 (3.68%)15 (4.69%)0.78217 (10.37%)50 (15.58%)0.2284Fixation failure01 (0.31%)1001 (0.61%)1 (0.31%)1Loosening3 (1.84%)2 (0.62%)0.4404 (2.45%)2 (0.62%)0.27 (4.27%)6 (1.87%)0.4192Nonunion01 (0.31%)104 (1.25%)0.4636 Surgical complications at 30-days, at 12-month follow-up, and cumulative. * p < 0.05 After propensity score matching, 99 patients were included in each group, showing no differences in patient and fracture features between both groups. The revision group showed longer surgical delay (6 versus 4 days, p < 0.001) and received general anesthesia more frequently (54.55% versus 17.35%, p < 0.001) (Table 4). Mortality was higher for the fixation group only at 1-year follow-up, with no differences before that time point. The fixation group presented higher rates of weight-bearing restrictions after surgery and at 30-day follow-up (17.35% versus 42.42%, p < 0.001) (Table 5). Table 4Demographic, baseline data, management, and hospital outcomes for patients presenting a type B2 or B3 femoral hip periprosthetic fracture (FHPPF), included in the propensity matching score. Qualitative variables are summarized using counts and percentages (%)^TotalFixationRevisionp-Valuen1989999Baseline:Age (years)*87 (9)87 (8.5)86 (8.5)0.456Sex, n (%): Female130 (65.66%)64 (64.65%)66 (66.67%)0.881 Male68 (34.34%)35 (35.35%)33 (33.33%)Place of residence, n (%): Community165 (83.33%)81 (81.82%)84 (84.85%)0.703 Health institution33 (16.67%)18 (18.18%)15 (15.15%)Prefracture mobility*, n (%): Independent outdoors139 (70.2%)69 (69.7%)70 (70.71%)0.188 Independent indoors39 (19.7%)20 (20.2%)19 (19.19%) Dependent20 (10.1%)10 (10.1%)10 (10.1%)Pfeiffer’s SPMSQ*3 (4)3 (4)3 (4)0.781CFS*5 (2)5 (2)5 (2)0.963ASA*3 (1)3 (1)3 (1)0.519CCI age-adjusted*6 (1.75)6 (1.5)6 (2)0.992Osteoprotective treatment, n (%): Yes104 (52.53%)55 (55.56%)49 (49.49%)0.477 No94 (47.47%)44 (44.44%)50 (50.51%)Antiaggregant or anticoagulant medication, n (%): Yes85 (42.93%)43 (43.43%)42 (42.42%)0.600 No113 (57.07%)56 (56.57%)57 (57.58%)Hb at admission (gr/dL)^12 (1.79)12.13 (1.46)11.88 (2.09)0.695Fracture features:UCS type: B2160 (80.81%)82 (82.83%)78 (78.79%)0.588 B338 (19.19%)17 (17.17%)21 (21.21%)Cemented or uncemented stem Cemented64 (32.32%)36 (36.36%)28 (28.28%)0.288 Uncemented134 (67.68%)63 (63.64%)71 (71.72%) Time since stem implantation (months)123 (135)123 (123)121 (151.25)0.676Dysfunctional prosthesis Yes93 (46.97%)48 (48.48%)45 (45.45%)0.776 No105 (53.03%)51 (51.52%)54 (54.55%)Management: Surgical delay (h)*120 (119)96 (108)143.5 (120)< 0.001Anesthesia General71 (36.04%)17 (17.35%)54 (54.55%)< 0.001 Neuroaxial126 (63.96%)81 (82.65%)45 (45.45%)Team expertise: < 20 procedures per year63 (31.82%)34 (34.34%)29 (29.29%)0.542 > 20 procedures per year135 (68.18%)65 (65.66%)70 (70.71%)Surgical approach: Less invasive21 (10.61%)21 (21.21%)0< 0.001 Open177 (89.39%)78 (78.79%)99 (100%)Was the stability of the stem checked? Yes174 (87.88%)75 (75.76%)99 (100%)< 0.001 No24 (12.12%)24 (24.24%)0Cerclage for reduction Yes170 (85.86%)81 (81.82%)89 (89.9%)0.153 No28 (14.14%)18 (18.18%)10 (10.1%)Type of fixation: Plate118 (59.6%)94 (94.95%)24 (24.24%) Cerclage wiring48 (24.24%)4 (5.05%)44 (44.44%)Medical staff involved⋄ Geriatrics85 (42.93%)43 (43.43%)42 (42.42%)0.955 Internal Medicine or others78 (39.39%)38 (38.38%)40 (40.4%) No35 (17.68%)18 (18.18%)17 (17.17%)Hospital outcomes: Length of hospital stay (days)*14.5 (13)13 (12.75)17 (13.75)0.015Initial postoperative mobilization out of bed < 48 h132 (67.01%)75 (75.76%)57 (58.16%)0.013 > 48 h65 (32.99%)24 (24.24%)41 (41.84%)Weight-bearing permission Full weight-bearing59 (29.95%)17 (17.35%)42 (42.42%)< 0.001 Not allowed or with restrictions138 (70.05%)81 (82.65%)57 (57.58%)Change of residence at hospital discharge Same111 (60.66%)50 (54.95%)61 (66.3%)0.076 Community to health institution70 (38.25%)41 (45.05%)29 (31.52%) Health institution to community2 (1.09%)2 (2.17%)Continuous variables are expressed as median and interquartile range (IQR)*. Prefracture mobility scale: registered as 1: complete independent gait, 2: outdoors independent gait with one technical aid, 3: outdoors independent gait with two technical aids, 4: only indoors independent gait with or without aids, 5: no mobility at all or with the help of two other people, and grouped as independent outdoors (nos. 1, 2, and 3), independent indoors (no. 4), and dependent (no. 5). ^Osteoprotective treatment: percentages of each individual treatment are referred to the total number of patients receiving osteoprotective treatment.⋄ Medical staff involved in the patient’s care (other than trauma and anesthesia)Surgeon experience: more or less than 20 less invasive fixation procedures or revisions for PPF per yearIQR interquartile range, Pfeiffer’s SPMSQ Pfeiffer’s Short Portable Mental Status Questionnaire, CFS clinical frailty scale, ASA American Society of Anesthesiologists (ASA) physical status classification system, Hb hemoglobin Demographic, baseline data, management, and hospital outcomes for patients presenting a type B2 or B3 femoral hip periprosthetic fracture (FHPPF), included in the propensity matching score. Qualitative variables are summarized using counts and percentages (%)^ Continuous variables are expressed as median and interquartile range (IQR)*. Prefracture mobility scale: registered as 1: complete independent gait, 2: outdoors independent gait with one technical aid, 3: outdoors independent gait with two technical aids, 4: only indoors independent gait with or without aids, 5: no mobility at all or with the help of two other people, and grouped as independent outdoors (nos. 1, 2, and 3), independent indoors (no. 4), and dependent (no. 5). ^Osteoprotective treatment: percentages of each individual treatment are referred to the total number of patients receiving osteoprotective treatment. ⋄ Medical staff involved in the patient’s care (other than trauma and anesthesia) Surgeon experience: more or less than 20 less invasive fixation procedures or revisions for PPF per year IQR interquartile range, Pfeiffer’s SPMSQ Pfeiffer’s Short Portable Mental Status Questionnaire, CFS clinical frailty scale, ASA American Society of Anesthesiologists (ASA) physical status classification system, Hb hemoglobin Table 530-days, 6-month, and 1-year follow-up outcomes and mortality for patients presenting a type B2 or B3 femoral hip periprosthetic fracture (FHPPF) included in the propensity matching score. Qualitative variables are summarized using counts and percentages (%).TotalFixationRevisionp-Valuen198999930-day follow-upWeight-bearing permission at 30 days: Full weight-bearing69 (38.98%)24 (27.91%)45 (49.45%)0.005 Not allowed or with restrictions108 (61.02%)62 (72.09%)46 (50.55%)Change of mobility at 30 days: Same37 (20.9%)13 (15.12%)24 (26.37%)0.184 Worse27 (15.25%)14 (16.28%)13 (14.29%) Better113 (63.84%)59 (68.6%)54 (59.34%)Place of residence at 30 days: Community91 (51.41%)39 (45.35%)52 (57.14%)0.156 Health institution86 (48.59%)47 (54.65%)39 (42.86%)6-month follow-up EQ-5D*0.62 (0.34)0.62 (0.34)0.64 (0.29)< 0.001Is the fracture healed? Yes106 (71.14%)62 (88.57%)44 (55.7%)< 0.001 No22 (14.77%)8 (11.43%)14 (17.72%) Does not apply21 (14.09%)021 (26.58%)1-year follow-up EQ-5D*0.62 (0.42)0.62 (0.38)0.6 (0.45)0.416Is the fracture healed? Yes113 (79.58%)57 (90.48%)54 (70.13%)0.004 No12 (8.45%)6 (9.52%)6 (7.79%) Does not apply17 (11.97%)17 (22.08%)Change of EQ-5D from 6-month follow-up Same25 (18.12%)12 (19.35%)13 (17.11%)0.943 Better54 (39.13%)24 (38.71%)30 (39.47%) Worse59 (42.75%)26 (41.94%)33 (43.42%)Pfeiffer’s SPMSQ*3 (5)3 (5)3 (5)0.929Change of Pfeiffer’s SPMSQ from baseline Same40 (29.63%)18 (29.03%)22 (30.14%)0.4294 Better34 (25.19%)13 (20.97%)21 (28.77%) Worse61 (45.19%)31 (50%)30 (41.1%)CFS*6 (3)6 (2.25)5 (3)0.794Change of CFS from baseline Same44 (31.21%)19 (29.69%)25 (32.47%)0.230 Better25 (17.73%)8 (12.5%)17 (22.08%) Worse72 (51.06%)37 (57.81%)35 (45.45%)Change of residence from baseline Same116 (81.69%)53 (82.81%)63 (80.77%)0.654 Community to health institution25 (17.61%)11 (17.19%)14 (17.95%) Health institution to community1 (0.7%)01 (1.28%)Change of mobility from baseline Same87 (61.27%)37 (57.81%)50 (64.1%)0.745 Better6 (4.23%)3 (4.69%)3 (3.85%) Worse49 (34.51%)24 (37.5%)25 (32.05%)Cumulative mortality In-hospital15 (7.58%)8 (8.08%)7 (7.07%)1 30-day follow-up20 (10.1%)12 (12.12%)8 (8.08%)0.479 6-month follow-up43 (21.72%)26 (26.26%)17 (17.17%)0.168 1-year follow-up56 (28.28%)35 (35.35%)21 (21.21%)0.040Medical complications In-hospital101 (51.01%)48 (48.48%)53 (53.54%)0.570 30-day follow-up36 (18.18%)17 (17.17%)19 (19.19%)0.854 6-month follow-up58 (29.29%)32 (32.32%)26 (26.26%)0.435 1-year follow-up67 (33.84%)38 (38.38%)29 (29.29%)0.230 At any moment154 (77.78%)77 (77.78%)77 (77.78%)1Surgical complications 30-day follow-up39 (19.7%)15 (15.15%)24 (24.24%)0.153 6-month follow-up44 (22.22%)21 (21.21%)23 (23.23%)0.864 1-year follow-up54 (27.27%)29 (29.29%)25 (25.25%)0.632 At any moment72 (36.36%)33 (33.33%)39 (39.39%)0.460Continuous variables are expressed as median and interquartile range (IQR)*. IQR interquartile range, Pfeiffer’s SPMSQ Pfeiffer’s Short Portable Mental Status Questionnaire, CFS clinical frailty scale 30-days, 6-month, and 1-year follow-up outcomes and mortality for patients presenting a type B2 or B3 femoral hip periprosthetic fracture (FHPPF) included in the propensity matching score. Qualitative variables are summarized using counts and percentages (%). Continuous variables are expressed as median and interquartile range (IQR)*. IQR interquartile range, Pfeiffer’s SPMSQ Pfeiffer’s Short Portable Mental Status Questionnaire, CFS clinical frailty scale Table 6 shows risk factors for mortality in the univariate and multivariate analysis. The predictive model for mortality presented a ROC curve with an area under the curve of 0.7784 (Fig. 2). The equation predicting mortality obtained from this model is presented in Table 6. Mortality risk predictive cutoff values in the multivariate regression analysis could be established for Pfeiffer > 2 (odds ratio, OR: 2.52; 95% CI (1.438,4.492), p = 0.001) and a-CCI > 5 (OR: 2.35; 95% CI (1.373,4.064), p = 0.002). Table 7 shows risk factors for mortality after the propensity score matching. Table 6Univariate and multivariate regression analysis for mortality.UnivariateMultivariateMultivariate significant values (p-value < 0.1)ORCIp-ValueORCIp-ValueCoefficientsORCIp-ValueCoefficientsPatient related at admission−5.48716−5.79221* Age1.070(1.038, 1.106)< 0.001 Sex: female1.391(0.843, 2.285)0.194 Place of residency: healthcare facility1.881(0.984, 3.507)0.0501.029(0.451, 2.256)0.9440.02259Prefracture mobility: Independent indoors3.312(1.820, 5.999)< 0.0011.345(1.175, 1.551)< 0.0010.743031.913(0.958, 3.768)0.0620.64876 Dependent2.850(1.062, 7.518)0.0341.924(0.959, 3.81)0.0620.988632.729(1.121, 6.617)0.0261.00404 Pfeiffer’s SPMSQ1.223(1.130, 1.326)< 0.0012.636(1.047, 6.628)0.0380.146511.155(1.052, 1.268)0.0030.14371 CFS1.562(1.334, 1.850)< 0.001 ASA2.214(1.585, 3.181)< 0.001 age-adjusted CCI13.71(1.213, 1.562)< 0.0011.345(1.175, 1.551)< 0.0010.288591.324(1.158, 1.524)< 0.0010.28064 Prefracture osteoprotective treatment0.823(0.467, 1.411)0.489 Antiaggregant or anticoagulant medication1.818(1.111, 2.979)0.017Hb at admission0.817(0.713, 0.933)0.003Fracture related at admissionB3 type1.864(1.065, 3.220)0.0271.17(1.065, 1.293)0.0020.501521.155(1.052, 1.268)0.0030.14371Cemented stem1.502(0.895, 2.496)0.119Time since stem implantation0.999(0.991, 1.009)0.960Dysfunctional prosthesis0.859(0.527, 1.399)0.542Treatment relatedSurgical delay0.999(0.991, 1.009)0.960Revision0.477(0.289, 0.787)0.004Open approach (only fixation)0.634(0.282, 1.525)0.284Spinal anesthesia0.991(0.601, 1.650)0.970Expert team0.792(0.477, 1.329)0.370Medical staff involved:Geriatrics2.530(1.149, 6.188)0.0290.644(0.354, 1.150)0.143−0.38847Internal Medicine or others3.540(1.641, 8.546)0.002Weight-bearing restrictions18.563(1.084, 3.283)0.0282.306(1.967, 3.002)< 0.0011.126383.004(1.245, 8.015)0.0191.10008The equation is: 1/(1 + e−y) where y= −5.79221 + (0.64876 × "Pre-Fracture mobility independent indoors") + (1.00404 × "Pre-Fracture mobility dependent") + (0.14371 × Pfeiffer SPMSQ) + (0.28064 × age-adjusted CCI) + (0.14371 × "B3 type") + (1.10008 × "Weight-bearing restriction"). For qualitative variables, values are 0 or 1, depending on the absence or presence of the variable, and for quantitative variables, the value is the numeric value of the variableCI Confidence Interval, OR Odds Ratio* Intercept for the equation predicting mortality obtained from this model. Univariate and multivariate regression analysis for mortality. The equation is: 1/(1 + e−y) where y= −5.79221 + (0.64876 × "Pre-Fracture mobility independent indoors") + (1.00404 × "Pre-Fracture mobility dependent") + (0.14371 × Pfeiffer SPMSQ) + (0.28064 × age-adjusted CCI) + (0.14371 × "B3 type") + (1.10008 × "Weight-bearing restriction"). For qualitative variables, values are 0 or 1, depending on the absence or presence of the variable, and for quantitative variables, the value is the numeric value of the variable CI Confidence Interval, OR Odds Ratio * Intercept for the equation predicting mortality obtained from this model. ROC curve corresponding to the predictive model for mortality Table 7Univariate and multivariate regression analysis for mortality for the propensity score matching groups UnivariateMultivariateMultivariate significant variables (p < 0.05)ORCIp-ValueORCIp-ValueCoefficientsORCIp-ValueCoefficientsPatient related at admission−10.77446−11.57057* Age1.061(1.020, 1.110)0.0061.063(1.013, 1.121)0.0170.061151.076(1.033, 1.128)0.0010.07359 Sex: male2.294(1.212, 4.356)0.0113.178(1.463, 7.083)0.0041.156363.336(1.592, 7.197)0.0021.20482 Place of residency: healthcare facility1.333(0.581, 2.923)0.481Prefracture mobility: Independent indoors2.424(1.132, 5.140)0.0213.307(1.423, 7.672)0.0051.196023.911(1.769, 8.661)< 0.0011.3639 Dependent2.850(1.062, 7.518)0.0343.764(1.234, 11.308)0.0181.325435.421(1.911, 15.098)0.0011.69027 Pfeiffer’s SPMSQ1.187(1.071, 1.320)0.0011.094(0.971, 1.231)0.1340.09024 CFS1.447(1.175, 1.812)< 0.001 ASA2.257(1.476, 3.641)< 0.001 Age-adjusted CCI1.401(1.187, 1.682)< 0.0011.061(0.864, 1.292)0.5590.05953 Prefracture osteoprotective treatment0.740(0.364, 1.448)0.389 Antiaggregant or anticoagulant medication1.650(0.885, 3.088)0.115 Hb at admission0.801(0.481, 1.271)0.357Fracture related at admission B3 type1.639(0.763, 3.432)0.195 Cemented stem1.237(0.638, 2.363)0.522 Time since stem implantation0.691(0.370, 1.285)0.244 Dysfunctional prosthesis1.013(0.964, 1.061)0.581Treatment related Surgical delay1.013(0.964, 1.061)0.581 Revision0.492(0.258, 0.921)0.0290.697(0.343, 1.427)0.318-0.36108 Open approach (only fixation)0.526(0.213, 1.350)0.168 Spinal anesthesia0.880(0.464, 1.691)0.697 Expert team0.782(0.409, 1.518)0.460Medical staff involved: Geriatrics4.444(1.418, 19.672)0.022 Internal Medicine or others5.973(1.912, 26.421)0.0061.507(0.746, 3.050)0.2510.41021 Weight-bearing restrictions 30 days1.548(0.773, 3.256)0.2312.394(1.158, 5.171)0.0210.873152.706(1.347, 5.701)0.0061.2731The equation is: 1/(1 + e−y) where y= −5.79221 + (0.64876 × "Pre-Fracture mobility independent indoors") + (1.00404 × "Pre-Fracture mobility dependent") + (0.14371 × Pfeiffer SPMSQ) + (0.28064 × age-adjusted CCI) + (0.14371 × "B3 type") + (1.10008 × "Weight-bearing restriction"). For qualitative variables, values are 0 or 1, depending on the absence or presence of the variable, and for quantitative variables, the value is the numeric value of the variableCI Confidence Interval, OR Odds Ratio*Intercept for the equation predicting mortality obtained from this model Univariate and multivariate regression analysis for mortality for the propensity score matching groups The equation is: 1/(1 + e−y) where y= −5.79221 + (0.64876 × "Pre-Fracture mobility independent indoors") + (1.00404 × "Pre-Fracture mobility dependent") + (0.14371 × Pfeiffer SPMSQ) + (0.28064 × age-adjusted CCI) + (0.14371 × "B3 type") + (1.10008 × "Weight-bearing restriction"). For qualitative variables, values are 0 or 1, depending on the absence or presence of the variable, and for quantitative variables, the value is the numeric value of the variable CI Confidence Interval, OR Odds Ratio *Intercept for the equation predicting mortality obtained from this model Discussion According to previous research recommendations [2–5], B2–B3 FH-PPF should be managed with stem revision. Recent literature, however, has questioned whether all patients should adhere to these recommendations, particularly frail and elderly individuals with low functional demands and certain fracture patterns [22, 23]. A systematic review published in 2017 with 22 studies reported that internal fixation for B2–B3 FH-PPF was associated with a higher reoperation rate [24]. By contrast, a 2022 systematic review found no significant difference in surgical complication rates between fixation alone (24%) and revision arthroplasty (18%), with fixation being superior preventing dislocation [25]. A 2023 meta-analysis further revealed that fixation was associated with reduced blood loss, shorter operative time, decreased need for transfusions, fewer complications, lower reoperation rates, and shorter hospital stays [26]. Mortality rates and overall outcomes were comparable between the groups. Additional studies have suggested that B2 and B3 can be effectively managed with fracture fixation, without stem revision, yielding favorable and, in several cases, superior results compared with revision in select patient populations [8–14, 27]. B2–B3 FH-PPF represent a severe complication because they commonly affect frail and elderly patients. The mean age from previous reviews was around 75 years old [25, 28]. Other studies have observed that fixation is more frequently performed in older patients, with the mean age in the fixation group ranging from 78.8 to 80 years old to and from 77 to 84 years in the revision group [12, 29, 30]. In the present study, patients were notably older, potentially influencing the prognosis. However, age is not always an accurate indicator of a patient’s medical status. Several studies have reported worse medical status and higher proportion of patients with ASA ≥ 3 and a-CCI, > 5, particularly among those treated with fixation [29, 30]. González-Martín et al. [31] proposed a treatment algorithm suggesting that patients with a-CCI ≥ 5, ASA ≥ 3, and low functional demand can be treated with fixation, provided that the anatomical reduction of the fracture can be achieved. Preoperative gait ability is another critical factor influencing the choice between fixation and revision. Previous studies suggest that fracture fixation maybe suitable for patients with low functional demand [29–31]. In the current study, the percentage of patients without autonomous ambulation was higher in the fixation group (12.8% versus 7.19%). Prefracture mobility has been identified as a mortality risk factor and should be carefully considered during the decision-making process. Surgical delay was longer in the revision group (128 versus 168 h), as well as hospital stay (13 versus 15 days). Across both cohorts, 46.39% of patients experienced medical complications during hospitalization. Although an economic analysis was not conducted, the longer hospital stay likely incurs higher costs. A recent study reported that the average cost for patients undergoing fixation was €14,239.07 compared with €21,498.45 for those treated with revision [30]. Only 20.86% of patients treated with fixation were permitted full weight-bearing at hospital discharge, compared with 47.5% in the revision group. This situation may explain the higher proportion of patients managed with fixation being discharged to a nursing home (38.31% versus 25.74%, p = 0.010). Like hip fractures, FH-PPF are associated with high mortality rates. Comorbidities play a critical role in determining the optimal treatment strategy for each patient. Some studies correlated a-CCI ≥ 5 with increased mortality risk [9, 32, 33]. One study reported a 1-year mortality rate of 22.3%, associated with poorer prefracture walking ability and a-CCI ≥ 3 [32]. In our study, an a-CCI > 5 was associated with 1-year mortality. Additionally, patients in the fixation group were older and had more comorbidities compared with those in the revision group, likely influencing the choice of management strategy. To account for potential selection bias, propensity score matching was performed. This analysis identified male gender, a-CCI, and cognitive impairment as independent risk factors for 1-year mortality. Early weight-bearing is a well-established protective factor following hip fractures, decreasing morbidity and mortality rates [34]. In the PIPPAS study, weight-bearing restriction was common practice in the fixation group (42.42% versus 17.35%) and was identified as an independent risk factors for mortality. Among patients treated with fixation, full weight-bearing within the first 30 days was associated with reduced risk of mortality. A review by Haider et al. [27] reported similar rates of full weight-bearing between revision and fixation groups. While a higher rate of stem subsidence was observed in the fixation group, there was no statistically significant difference in radiological loosening during the first year of follow-up, even though pain referred to the thigh was not evaluated. This finding suggests that full weight-bearing may be safely achieved after stable fixation, highlighting the relevance of the fixation technique. Minimally invasive plating osteosynthesis has been shown to reduce blood loss and local complications [35, 36]. A recent retrospective study comparing a mini-invasive approach with an open approach demonstrated shorter operative times, reduced intraoperative bleeding, and earlier fracture consolidation by 2 months [37]. In our study, less invasive approaches were used in 20.73% of the cases of fixation group. However, the surgical approach did not influence weight-bearing restrictions or mortality rates. Surgical complications should be expected in both treatment groups. In the present study, rates of infection were high in both groups, but the dislocation rate was higher in the revision group, similar to in a recent systematic review [11]. A systematic review by Haider et al. [10] found no significant differences in overall complication rates between fixation and revision arthroplasty. Although nonunion is a concern in fixation cases, nonunion rates in our study were not relevant, and there were no differences in the rate of subsequent surgeries performed for aseptic loosening during the first year of follow-up. The primary objectives for stem revision in B2–B3 FH-PPF are to address loosening and restore function, enabling unrestricted weight-bearing [2–5, 8, 11, 24]. Based on our findings, stem revision did not result in a higher re-revision rate for aseptic loosening at 1-year follow-up. Notably, in the fixation group, full weight-bearing was not associated with fixation failure, suggesting that surgeons may be overly cautious in restricting weight-bearing in certain patients. There is an ongoing debate regarding who should manage periprosthetic fractures: arthroplasty-focused orthopedic surgeons, orthopedic trauma surgeons specialized in fracture fixation, or perhaps a combined team. Pohl et al. did not find differences in outcomes based on surgeon subspecialty training [38]. In our study, the level of expertise was not a risk factor for mortality, although it may influence treatment indication, complication rates and weight-bearing indications. The impact of expertise and subspecialty focus on the management of periprosthetic fractures deserves further research. The weaknesses of this study include the fact that the decision to pursue fixation or revision was made by the attending surgeon on the basis of the patient’s functional status, comorbidities, and the fracture pattern. Consequently, a selection bias was inherent, although we attempted to mitigate this using propensity score matching. There was variability in the treatment methods that were employed. In addition, the distinction between B1 and B2–B3 types was at the discretion of the treating surgeon based on clinical, radiological, and surgical findings. Although criteria were standardized, several radiographic or intraoperative details might have not been captured uniformly across sites. Surgeon experience might influence the choice of treatment (fixation versus revision), the weight-bearing protocol, as well as the risk of complications, including dislocation. Weight-bearing restrictions prescribed at hospital discharge could only be verified at the 30-day follow-up visit, depending on the patient’s actual behavior at home after discharge. Nonetheless, to the best of our knowledge, and excluding meta-analyses, this represents the largest series of B2–B3 FH-PPF cases reported to date. In conclusion, revision arthroplasty showed a lower 1-year mortality rate and fewer weight-bearing restrictions than fixation with similar functional outcomes; fixation showed reduced dislocation rates, shorter surgical delay, and hospital stay. Frail patients with B2–B3 FH-PPF managed with fixation and without weight-bearing restrictions presented a lower 1-year mortality rate. Age, frailty, and weight-bearing restrictions were identified as mortality risk factors following B2–B3 FH-PPF. Elderly patients treated with a strategy to achieve full weight-bearing, regardless of the treatment selected, fare better than patients who were prescribed weight-bearing restrictions. Supplementary Information Additional file 1. Acknowledgements We thank Dasha Gorbenko del Blanco (PhD, medical illustrator, dasha.blanco@gmail.com) for her work art producing the figures for the PIPPAS study. Abbreviations Periprosthetic femoral fractures following hip arthroplasty Unified classification system Clinical frailty score Pfeiffer short portable mental status questionnaire Age-adjusted Charlson Comorbidity Index American Society of Anesthesiologists physical status classification system Author contributions Conceptualization: JM, HJA, PCB, JMMV, PC, MJG, JT, APM, YGS, COT, and PGP; methodology: JM, HJA, PCB, JMMV, PC, MJG, JT, APM, YGS, COT, and PGP; validation HJA, APD, MMM, and IAH; formal analysis: JM, HJA, MMM, and IAH; investigation: JM, HJA, PCB, JMMV, PC, MJG, JT, APM, YGS, COT, PGP, EMA, DNG, MMM, IAH, BAAR, BGM, CMS, SMJ, VGV, SPO, AA, JTH, JSM, JVAP, CPC, FBC, EGF, JQC, GTLS, JMH, MRC, CGA, ACC, MVB, EAF, EFM, UGDC, MA, DE, EC, PBB, MD, MCM, SP, AL, BOL, PAE, JBM, DGM, DLD, JCBC, CMP, PAF P, PCC, MASS, LLL, AGA, JCF, JGC, JVC, JSDS, JMF, PLA, CBR, MACL, ABC, ALDP, AFJ, CSDLP, MJM, JGA, ESS, SGI, PSG, FJRR, EMGG, FCA, MIPN, PDPM, FMGNG, EGP, AGV, TBR, INP, IDVJ, TSP, CCB, VON, SPP, IFBC, FFD, AMM, VVG, AMA, EGM, CMM, AMV, NPS, CGA, JBS, PST, OPP, ACC, BGM, LAGV, JMR, RBM, ABB, EEM, LCM, MNT, AGT, ADI, JMP, JSS, ARA, CMH, MMM, JCSC, JER, GCA, HGCB, DAN, JGB, JGS, JCP, JMIB, JMBD, ARR, JRCP, FML, SMD, PIS, GSC, IA, BRB, JCCP, CHP, ARG, IDP, APB, JMPS, AMM, JDCR, JMM, MBM, OAG, ICE, MRGP, NEPS, and MDSD, resources: HJA; data curation, JM, HJA, SMG, MJG, APM, MMM, and IAH; writing—original draft preparation: JM and HJA; writing—review and editing: JM, HJA, PCB, JMMV, PC, MJG, JT, APM, YGS, COT, PGP, EMA, DNG, MMM, IAH, BAAR, BGM, CMS, SMJ, VGV, SPO, AA, JTH, JSM, JVAP, CPC, FBC, EGF, JQC, GTLS, JMH, MRC, CGA, ACC, MVB, EAF, EFM, UGDC, MA, DE, EC, PBB, MD, MCM, SP, AL, BOL, PAE, JBM, DGM, DLD, JCBC, CMP, PAF P, PCC, MASS, LLL, AGA, JCF, JGC, JVC, JSDS, JMF, PLA, CBR, MACL, ABC, ALDP, AFJ, CSDLP, MJM, JGA, ESS, SGI, PSG, FJRR, EMGG, FCA, MIPN, PDPM, FMGNG, EGP, AGV, TBR, INP, IDVJ, TSP, CCB, VON, SPP, IFBC, FFD, AMM, VVG, AMA, EGM, CMM, AMV, NPS, CGA, JBS, PST, OPP, ACC, BGM, LAGV, JMR, RBM, ABB, EEM, LCM, MNT, AGT, ADI, JMP, JSS, ARA, CMH, MMM, JCSC, JER, GCA, HGCB, DAN, JGB, JGS, JCP, JMIB, JMBD, ARR, JRCP, FML, SMD, PIS, GSC, IA, BRB, JCCP, CHP, ARG, IDP, APB, JMPS, AMM, JDCR, JMM, MBM, OAG, ICE, MRGP, NEPS, and MDSD; supervision, PCB, JTS, YGS, JMMV, PCC, MJG, COT, PGP, EMA, and APM; funding acquisition, HJA. All authors have read and agreed to the published version of the manuscript. Funding This research was funded by grants from AO Trauma Foundation (Spain) grant number PI-20–2041 and Gerencia Regional de Salud de “Salud de Castilla y León” (GRS SACYL) grant numbers GRS 2371/A/21 and GRS 2794/A1/23. The funding body did not participate in the design of the study, collection, analysis, or interpretation of data, nor in writing the manuscript. Data availability The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. Declarations Ethics approval and consent to participate The study was conducted in accordance with the ethical standards laid down in the 1964 Helsinki Declaration and received approval from the institutional review boards of the coordinating center (Comité de Ética de la Investigación con Medicamentos CEIm Área de Salud de Valladolid) code number PI-20–2041 and each participating hospital. Written consent for participation in the study was obtained from all participants or their legal representatives. This study is registered at ClinicalTrials.gov (NCT04663893). Consent for publication Not applicable. Competing interests The authors declare that they have no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Jesús Moreta and Héctor J. Aguado have contributed equally to this work. References Associated Data Supplementary Materials Additional file 1. Data Availability Statement The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.