Association between timing of surgery and postoperative outcomes in older adults with distal femur fractures: a systematic review and meta-analysis Authors: Shota Hoshika, Norio Yamamoto, Natsumi Saka, Takahiro Tsuge, Tomoo Sato, Yuichi Isaji, Shuri Nakao Published: 2026 ABSTRACT IntroductionThe association between surgical timing and clinical outcomes in older adults with distal femur fractures remains unclear. We conducted a systematic review and meta-analysis to evaluate whether earlier surgery is associated with improved outcomes in this population.Materials and methodsIn accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, two independent reviewers searched MEDLINE, Embase, the Cochrane Central Register of Controlled Trials, ClinicalTrials.gov, and the World Health Organization International Clinical Trials Registry Platform from inception to October 8, 2024. The primary outcome was 30-day mortality, and secondary outcomes included longer-term mortality and postoperative complications. Risk of bias was assessed using the Quality in Prognosis Studies tool, and the certainty of evidence was evaluated using the Grading of Recommendations Assessment, Development and Evaluation approach.ResultsOf 6,101 records screened, 16 retrospective cohort studies comprising 31,213 patients met the inclusion criteria. Early surgery was not associated with reduced 30-day mortality (adjusted odds ratio [OR]: 0.77, 95% confidence interval [CI]: 0.56–1.05; low-certainty evidence) or with longer-term mortality at 90 days (crude OR: 0.91, 95% CI: 0.51–1.60), 180 days (crude OR: 0.46, 95% CI: 0.12–1.77), or 1 year (crude OR: 0.61, 95% CI: 0.34–1.09). Similar findings were observed for postoperative complications: cardiac complications (OR: 0.71, 95% CI: 0.48–1.07), pulmonary complications (OR: 1.07, 95% CI: 0.77–1.47), and pulmonary embolism (OR: 1.14, 95% CI: 0.62–2.16). Heterogeneity across outcomes was low to moderate, and definitions of early surgery varied among studies. Overall, this meta-analysis did not demonstrate a clinically meaningful association between surgical timing and short-term clinical outcomes in older adults with distal femur fractures. Sensitivity analyses consistently suggested lower mortality with earlier surgery, likely reflecting stricter definitions of early surgery and improved control of time-related bias.ConclusionsThis meta-analysis did not demonstrate a clinically meaningful association between surgical timing and short-term clinical outcomes in older adults with distal femur fractures. Given the heterogeneity in definitions and study designs, high-quality prospective studies with standardized timing are warranted.Supplementary InformationThe online version contains supplementary material available at 10.1007/s00402-026-06296-2. Introduction Distal femoral fractures represent a significant clinical challenge in aging societies, where optimizing management and improving outcomes are pressing priorities. Although they account for 3–7% of all femoral fractures and are relatively uncommon overall [1], in older adults they represent the second most frequent fragility fracture after hip fractures [2]. These injuries are challenging to manage due to their high mortality and complication rates, which are comparable to or even exceed those associated with hip fractures [1, 3–5]. Early surgical intervention is recommended for hip fractures. A systematic review and meta-analysis (SRMA) demonstrated that timely surgery reduces both mortality and postoperative complications in this population [6, 7]. In contrast, evidence regarding the impact of surgical timing on outcomes after distal femoral fractures is limited and inconclusive [2, 3, 5, 8–13]. Some studies suggest that early surgery may increase morbidity and mortality, whereas others report no significant association [2, 3, 5, 8–13]. Consequently, the optimal timing of surgery for distal femoral fractures has not been clearly established [1]. Notably, no SRMA has yet clarified the relationship between surgical timing and postoperative outcomes in this patient population. We aimed to examine the association between the timing of surgery and postoperative outcomes, including mortality and complications, in older adults with distal femoral fractures. Materials and methods Study design This review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [14]. The study protocol was prospectively registered in a public repository (registration ID: mu5q8). The PRISMA flow diagram is shown in Fig. 1, and the completed checklist is provided as Supplementary Table 1. Search strategy We searched MEDLINE (PubMed), the Cochrane Central Register of Controlled Trials (CENTRAL), and EMBASE (Dialog) from database inception to October 8, 2024, and updated the search on December 14, 2025, without restrictions on publication language or year. We also searched for ongoing or unpublished studies using the World Health Organization International Clinical Trials Platform (ICTRP) and ClinicalTrials.gov. (Supplementary Tables 2–6). Additionally, we reviewed references of included articles and articles citing the included studies using CitationChaser (https://estech.shinyapps.io/citationchaser/). Eligibility criteria Study design: We included studies that reported stratified outcomes for patients who underwent surgery for distal femur fractures, including cohort studies and case series. All types of publications were eligible, including published and unpublished articles, conference abstracts, and letters. We applied no restrictions on language, country, observation period, or publication year. Case reports and biomechanical studies were excluded. Participants/population: We included studies that recruited participants aged 60 years or older from any setting, such as the general population, research cohorts, or clinical populations, as most distal femur fractures in this age group result from osteoporotic fractures [3, 11–13]. We included interventions such as open reduction and internal fixation, total knee arthroplasty (TKA), AO screws and K-wires, cannulated screws, and external fixation for distal femur fractures. Exposure: The main exposure was the timing of surgery after hospital admission. Eligible patients were classified into early or delayed surgery groups. Early surgery was defined as occurring within 2 days of admission, consistent with previous studies of distal femur and hip fractures [3, 7, 15]. We also included studies that defined early surgery as within 2 days of injury or did not specify whether timing was measured from admission or injury. Comparison: The control group consisted of patients who underwent surgery three or more days after admission for distal femur fractures. Outcome of interest The primary and secondary outcomes were selected and prioritized based on their clinical importance and in accordance with the consensus core outcome set for patients with hip fractures [16, 17]. The primary outcome was 30-day mortality, defined as death occurring during the hospital stay or within 30 days after surgery, consistent with previous studies [6, 18–20]. The secondary outcomes were: (1) 90-day, 180-day, and 1-year mortality; (2) pulmonary complications, as defined by the original authors [17]; (3) cardiac complications; (4) deep vein thrombosis (DVT); (5) pulmonary embolism (PE); and (6) additional surgery, defined as surgery or readmission related to the initial treatment [17]. Data screening and extraction Two reviewers (SH and one of TT, TS, YI, or SN) screened titles and abstracts. Full texts of potentially eligible articles were then reviewed. Other relevant studies were identified by manually searching the reference lists of included studies. Data were extracted by independent reviewers using a prepiloted data extraction form. We attempted to contact study authors to clarify information or request additional data as necessary. Disagreements were resolved by discussion, and if necessary, a third reviewer (NY, NS) acted as an arbiter. For effect measures, we extracted both univariable and multivariable estimates. For univariable analyses, we extracted the number of events and participants in each group. For multivariable analyses, odds ratios (ORs) were prioritized. When both hazard ratios (HRs) and ORs were reported, the measure most frequently used across studies was selected for the primary meta-analysis, and the alternative measure was used for sensitivity analyses. Quality assessment Two reviewers (SH and one of TT, TS, YI, or SN) assessed the quality of each included article using the Quality in Prognosis Studies (QUIPS) tool [10]. When necessary, study authors were contacted. Disagreements between reviewers were resolved by discussion, and if consensus could not be reached, a third reviewer (NY or NS) arbitrated. For the QUIPS assessment, we defined the following confounders: age, sex, comorbidity, fracture type, and the type of surgery [3, 11, 12, 21]. Studies adjusting for all these confounders were rated as having a low risk of bias in the confounding domain. Overall risk of bias was rated low if all QUIPS domains were low, moderate if one domain was moderate or high, and high if two or more domains were high. Data synthesis Binary outcomes, including mortality and postoperative complications, were summarized as ORs with 95% confidence intervals (CIs). Pooled ORs were calculated using the Mantel–Haenszel method with a random-effects model, and pooled HRs were calculated using the inverse variance method with a random-effects model. When both crude and adjusted ORs were reported, adjusted ORs were used for the primary synthesis. Meta-analyses were conducted separately for crude and adjusted estimates. Heterogeneity was assessed using the I² statistic, with values ≥ 75% considered high. Forest plots were constructed to illustrate study outcomes, and sensitivity analyses were conducted for primary outcomes to assess robustness. Publication bias was not evaluated using funnel plots or Egger’s test because fewer than ten studies were available for each outcome. Statistical analyses were performed using RevMan version 5.4 (The Nordic Cochrane Centre, The Cochrane Collaboration, Copenhagen, Denmark). A two-sided P-value < 0.05 was considered statistically significant. Subgroup analysis To explore potential effect modifiers on the primary outcomes, subgroup analyses were prespecified according to: (1) sex (men versus women), (2) age (< 75 years versus ≥ 75 years), (3) history of comorbidities (present versus absent), (4) type of surgery (fracture fixation versus arthroplasty), and (5) presence of other implants such as TKA or total hip arthroplasty (present versus absent). Sensitivity analysis Sensitivity analyses were performed for the primary outcomes to evaluate the robustness against methodological decisions. For 30-day mortality, early surgery was redefined as within 24 h rather than 48 h. For 180-day mortality, time from admission was used instead of time from injury, and cases with uncertain timing were included. For 1-year mortality, both adjusted HRs based on time from injury and admission and crude ORs based on time from injury, admission, and uncertain cases were analyzed. Furthermore, to explore the impact of clinical heterogeneity related to surgical procedures, we conducted a sensitivity analysis excluding studies involving arthroplasty for the crude ORs of 90-day and 1-year mortality. We also planned sensitivity analyses for operative time and blood loss, if sufficient data were available. Across all analyses, we examined whether the direction and consistency of the associations were maintained compared with the main analysis. Summary of findings table We created a Summary of Findings table for 30-day mortality, pulmonary complications, cardiac complications, and pulmonary embolism. The certainty of evidence was assessed using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach [22]. This approach considered eight domains influencing certainty, including factors that may reduce certainty (risk of bias, inconsistency, indirectness, imprecision, and publication bias) and factors that may increase certainty (large effect, dose–response, and plausible confounding). Certainty was classified as high, moderate, low, or very low. One reviewer (SH) performed the assessment, which was verified by another (NY). Discrepancies were resolved through discussion or, if needed, adjudication by a third reviewer (NS). Reasons for downgrading were reported in the Summary of Findings table. The median event rate in the comparison group across included studies was used as the baseline risk. Difference between protocol and review Exposure was initially defined only as time from admission; however, in the review, it was broadened to include time from injury and cases with unclear timing. Although only 1-year mortality was prespecified as a secondary outcome, 90-day and 180-day mortality were additionally analyzed. The meta-analysis was initially planned to combine crude and adjusted estimates; however, these were analyzed separately in the final review. Results Study selection and characteristics After screening 6,280 records, we included 16 studies (31,213 patients) in the final analysis (Fig. 1). Of 30 reports sought for retrieval, 14 were excluded after full-text review because of wrong design (n = 5), wrong population (n = 7), or wrong intervention (n = 2); details of these studies and reasons for exclusion are provided in Supplementary Table 7. Fig. 1Preferred reporting items for systematic reviews and meta-analysis (PRISMA) Preferred reporting items for systematic reviews and meta-analysis (PRISMA) 2020 flow diagram illustrating the study selection process. By contacting study authors, we obtained unpublished data from four studies. The included studies were conducted in multiple regions, with the majority originating from the United States and the majority being multicenter cohorts (Table 1; Supplementary Table 8). Table 1Characteristics of included studiesStudySettingCountryFrom when to surgeryCut-off pointPatients < 48 hPatients > 48 hSurgeryFollow-up, year n Age, yearFemale (%) n Age, yearFemale (%)Chaoyong (2009)Single centerChinaInjury7 daysNRNRNRNRNRNRORIF with plating, intramedullary nail, Fixation with K-wire or screws1.9Streubel (2011)Single centerUSAAdmission48 hNRNRNRNRNRNRORIF with plating2.8Bettin (2016)MulticenterUSAUnclear72 h676.3 ± 5NR1277.4 ± 7.8NRArthroplasty2.2Brogan (2016)MulticenterUK and NZUnclear24 h, 24–48 h, 48 h152NRNR91NRNRORIF with plating, Intramedullary nail, others2.1Moloney (2016)MulticenterUSAAdmission48 h136NRNR40NRNRPlating2.8Atrey (2017)Single centerCanadaUnclear72 h583.4 ± 3.8NR574,6 ± 12.5NREndoprosthetic replacement3.5Myers (2017)MulticenterUSAInjury48 h23576.6 ± 9.7204 (86.8)4873.6 ± 1034 (70.8)ORIF with plating, intramedullary nail1Nyholm (2017)MulticenterDanishAdmissionEvery 12 h264NR213 (81)128NR97 (76)ORIF with plating, intramedullary nail, others0.25Shangzeng (2017)Single centerChinaInjury7 daysNRNRNRNRNRNRArthroplasty with plating/nailing2.6Jennison (2018)MulticenterUKAdmission36 h46NRNR34NRNRORIF with plating, intramedullary nail5.2Merino-Rueda (2021)Single centerSpainAdmission48 h1484.8 ± 5.4NR4584.8 ± 7.7NRORIF with plating, intramedullary nail6.9Yamamoto (2021)MulticenterJapanAdmission48 h138281.3 ± 9.51225 (89)138281.4 ± 9.51214 (88)ORIF with plating, intramedullary nail, TKA, Others0.13Yamada (2022)MulticenterJapanAdmission48 h58NRNR267NRNRORIF with plating, intramedullary nail, ORIF with screw2Gutbrod (2024)MulticenterUSAAdmission24 h, 24–48 h, 48 h15,517NRNR6528NRNRFixation with K-wire or screws4.9Meca (2024)Single centerSpainAdmission48 h4067,2 ± 19.512 (30)7568.9 ± 17.717 (23)ORIF with plating, intramedullary nail, TKA1Onizuka (2025)MulticenterUSAAdmission24 h, 24–48 h, 48 h3895NR3182 (82)605NR491 (81)ORIF with plating, intramedullary nail0.25The continuous values are presented as mean ± standard deviation (range)TKA total knee arthroplasty; ORIF open reduction and internal fixation; K-wire Kirschner wire; NR not reported Table 1Characteristics of included studiesStudySettingCountryFrom when to surgeryCut-off pointPatients < 48 hPatients > 48 hSurgeryFollow-up, year n Age, yearFemale (%) n Age, yearFemale (%)Chaoyong (2009)Single centerChinaInjury7 daysNRNRNRNRNRNRORIF with plating, intramedullary nail, Fixation with K-wire or screws1.9Streubel (2011)Single centerUSAAdmission48 hNRNRNRNRNRNRORIF with plating2.8Bettin (2016)MulticenterUSAUnclear72 h676.3 ± 5NR1277.4 ± 7.8NRArthroplasty2.2Brogan (2016)MulticenterUK and NZUnclear24 h, 24–48 h, 48 h152NRNR91NRNRORIF with plating, Intramedullary nail, others2.1Moloney (2016)MulticenterUSAAdmission48 h136NRNR40NRNRPlating2.8Atrey (2017)Single centerCanadaUnclear72 h583.4 ± 3.8NR574,6 ± 12.5NREndoprosthetic replacement3.5Myers (2017)MulticenterUSAInjury48 h23576.6 ± 9.7204 (86.8)4873.6 ± 1034 (70.8)ORIF with plating, intramedullary nail1Nyholm (2017)MulticenterDanishAdmissionEvery 12 h264NR213 (81)128NR97 (76)ORIF with plating, intramedullary nail, others0.25Shangzeng (2017)Single centerChinaInjury7 daysNRNRNRNRNRNRArthroplasty with plating/nailing2.6Jennison (2018)MulticenterUKAdmission36 h46NRNR34NRNRORIF with plating, intramedullary nail5.2Merino-Rueda (2021)Single centerSpainAdmission48 h1484.8 ± 5.4NR4584.8 ± 7.7NRORIF with plating, intramedullary nail6.9Yamamoto (2021)MulticenterJapanAdmission48 h138281.3 ± 9.51225 (89)138281.4 ± 9.51214 (88)ORIF with plating, intramedullary nail, TKA, Others0.13Yamada (2022)MulticenterJapanAdmission48 h58NRNR267NRNRORIF with plating, intramedullary nail, ORIF with screw2Gutbrod (2024)MulticenterUSAAdmission24 h, 24–48 h, 48 h15,517NRNR6528NRNRFixation with K-wire or screws4.9Meca (2024)Single centerSpainAdmission48 h4067,2 ± 19.512 (30)7568.9 ± 17.717 (23)ORIF with plating, intramedullary nail, TKA1Onizuka (2025)MulticenterUSAAdmission24 h, 24–48 h, 48 h3895NR3182 (82)605NR491 (81)ORIF with plating, intramedullary nail0.25The continuous values are presented as mean ± standard deviation (range)TKA total knee arthroplasty; ORIF open reduction and internal fixation; K-wire Kirschner wire; NR not reported Characteristics of included studies The continuous values are presented as mean ± standard deviation (range) TKA total knee arthroplasty; ORIF open reduction and internal fixation; K-wire Kirschner wire; NR not reported All included studies were retrospective cohort studies, and most reported crude ORs. The study populations comprised older adults with a mean age of 77.0 years (interquartile range [IQR]: 74.4–81.9 years); females represented the majority (median: 80.7%, IQR: 78.3–83.2%) in studies reporting sex distribution. The prevalence of early surgery (median, IQR) was 50.0% (33.3–67.3%). Follow-up periods varied across studies, with outcomes assessed at 30 days, 90 days, 180 days, and 1 year; the median follow-up was 2.2 years (IQR: 1.0–3.0 years). Definitions of outcomes differed among studies, and reporting of early postoperative and preoperative complications was frequently incomplete or based on inconsistent time points. Most included studies had a moderately high risk of bias, primarily due to insufficient adjustment for confounders or small sample sizes (Table 2). Table 2Risk of bias assessment using the quality in prognosis studies (QUIPS) toolStudyStudy participationStudy attritionPrognostic factor measurementOutcome measurementStudy confoundingStatistical analysis and reportingOverallChaoyong (2009)LowHighModerateLowHighLowHighStreubel (2011)LowLowModerateLowHighModerateModerateBettin (2016)LowHighLowLowHighHighHighBrogan (2016)LowHighLowLowHighModerateHighMoloney (2016)LowModerateLowLowHighModerateModerateAtrey (2017)LowHighLowLowHighHighHighMyers (2017)LowLowLowLowModerateModerateModerateNyholm (2017)LowHighLowLowHighHighHighShangzeng, 2017LowLowLowLowHighHighHighJennison (2018)LowLowLowLowHighModerateModerateMerino-Rueda (2021)LowLowLowLowHighLowModerateYamamoto (2021)LowHighLowLowLowLowModerateYamada (2022)LowModerateLowLowHighModerateModerateGutbrod (2024)LowLowLowLowHighLowModerateMeca (2024)LowHighLowHighModerateModerateHighOnizuka (2025)LowHighLowLowHighLowHigh Table 2Risk of bias assessment using the quality in prognosis studies (QUIPS) toolStudyStudy participationStudy attritionPrognostic factor measurementOutcome measurementStudy confoundingStatistical analysis and reportingOverallChaoyong (2009)LowHighModerateLowHighLowHighStreubel (2011)LowLowModerateLowHighModerateModerateBettin (2016)LowHighLowLowHighHighHighBrogan (2016)LowHighLowLowHighModerateHighMoloney (2016)LowModerateLowLowHighModerateModerateAtrey (2017)LowHighLowLowHighHighHighMyers (2017)LowLowLowLowModerateModerateModerateNyholm (2017)LowHighLowLowHighHighHighShangzeng, 2017LowLowLowLowHighHighHighJennison (2018)LowLowLowLowHighModerateModerateMerino-Rueda (2021)LowLowLowLowHighLowModerateYamamoto (2021)LowHighLowLowLowLowModerateYamada (2022)LowModerateLowLowHighModerateModerateGutbrod (2024)LowLowLowLowHighLowModerateMeca (2024)LowHighLowHighModerateModerateHighOnizuka (2025)LowHighLowLowHighLowHigh Risk of bias assessment using the quality in prognosis studies (QUIPS) tool In the QUIPS appraisal, frequent concerns were noted in the Outcome Measurement domain and, to a lesser extent, in the Study Attrition domain, reflecting the variable definitions and incomplete reporting of complication outcomes. Primary outcomes Low-certainty evidence suggested that early surgery may result in little to no reduction in 30-day mortality (3 studies, 29,309 patients: adjusted OR: 0.77, 95% CI: 0.56–1.05; I² = 28%, Fig. 2A; Table 3). Thirty-day readmission was a prespecified outcome but could not be included in the SRMA due to insufficient and nonuniform reporting. Fig. 2Forest plot of the primary outcome. A Pooled adjusted odds ratio for 30-day mortality. B Crude odds ratio for 90-day mortality. C Crude odds ratio for 180-day mortality. D Crude odds ratio for 1-year morality. SE standard error; IV inverse variance; CI confidence interval Forest plot of the primary outcome. A Pooled adjusted odds ratio for 30-day mortality. B Crude odds ratio for 90-day mortality. C Crude odds ratio for 180-day mortality. D Crude odds ratio for 1-year morality. SE standard error; IV inverse variance; CI confidence interval Table 3Summary of findings from 16 studies evaluating the timing of surgery for distal femur fracturesOverview of study designPatients or study population: Patients over sixty with distal femur fracturesExposure: Patients who underwent early surgery within 2 days for distal femur fracturesComparison: Patients who underwent delayed surgery ≥ 3 daysOutcomeIllustrative comparative risks* (95% CI)Relative effect (95% CI)No. of participants (studies)Certainty of the evidence (GRADE)CommentsRisk with delayed surgeryRisk with early surgery30-day mortalityStudy populationOR 0.77(0.56 to 1.05)29,309(3 studies)⊕⊕⊝⊝Lowa, bEarly surgery may result in little to no difference in reducing mortality18 per 100014 per 1000(10 to 19)Pulmonary complicationsStudy populationOR 1.07(0.77 to 1.47)7641(4 studies)⊕⊝⊝⊝Very Low a, bThe evidence is very uncertain about the effect of early surgery in reducing pulmonary complications58 per 100062 per 1000(45 to 83)Cardiac complicationsStudy populationOR 0.71(0.48 to 1.07)7408(2 studies)⊕⊕⊝⊝Lowa, bEarly surgery may result in little to no difference in reducing cardiac complications25 per 100018 per 1000(12 to 26)Pulmonary embolismStudy populationOR 1.14(0.62 to 2.16)7335(2 studies)⊕⊝⊝⊝Very Lowa, bThe evidence is very uncertain about the effect of early surgery in reducing pulmonary embolisms21 per 100024 per 1000(13 to 44)GRADE Working Group grades of evidenceHigh certainty: We are very confident that the true effect lies close to that of the estimate of the effect.Moderate certainty: We are moderately confident in the effect estimate; the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different.Low certainty: Our confidence in the effect estimate is limited; the true effect may be substantially different from the estimate of the effect.Very low certainty: We have very little confidence in the effect estimate; the true effect is likely to be substantially different from the estimate of effect*The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect (and its 95% CI) estimated for the intervention group. Assumed risk was estimated from the meta-analysis of control risks Table 3Summary of findings from 16 studies evaluating the timing of surgery for distal femur fracturesOverview of study designPatients or study population: Patients over sixty with distal femur fracturesExposure: Patients who underwent early surgery within 2 days for distal femur fracturesComparison: Patients who underwent delayed surgery ≥ 3 daysOutcomeIllustrative comparative risks* (95% CI)Relative effect (95% CI)No. of participants (studies)Certainty of the evidence (GRADE)CommentsRisk with delayed surgeryRisk with early surgery30-day mortalityStudy populationOR 0.77(0.56 to 1.05)29,309(3 studies)⊕⊕⊝⊝Lowa, bEarly surgery may result in little to no difference in reducing mortality18 per 100014 per 1000(10 to 19)Pulmonary complicationsStudy populationOR 1.07(0.77 to 1.47)7641(4 studies)⊕⊝⊝⊝Very Low a, bThe evidence is very uncertain about the effect of early surgery in reducing pulmonary complications58 per 100062 per 1000(45 to 83)Cardiac complicationsStudy populationOR 0.71(0.48 to 1.07)7408(2 studies)⊕⊕⊝⊝Lowa, bEarly surgery may result in little to no difference in reducing cardiac complications25 per 100018 per 1000(12 to 26)Pulmonary embolismStudy populationOR 1.14(0.62 to 2.16)7335(2 studies)⊕⊝⊝⊝Very Lowa, bThe evidence is very uncertain about the effect of early surgery in reducing pulmonary embolisms21 per 100024 per 1000(13 to 44)GRADE Working Group grades of evidenceHigh certainty: We are very confident that the true effect lies close to that of the estimate of the effect.Moderate certainty: We are moderately confident in the effect estimate; the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different.Low certainty: Our confidence in the effect estimate is limited; the true effect may be substantially different from the estimate of the effect.Very low certainty: We have very little confidence in the effect estimate; the true effect is likely to be substantially different from the estimate of effect*The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect (and its 95% CI) estimated for the intervention group. Assumed risk was estimated from the meta-analysis of control risks Summary of findings from 16 studies evaluating the timing of surgery for distal femur fractures OR 0.77 (0.56 to 1.05) 29,309 (3 studies) ⊕⊕⊝⊝ Lowa, b 14 per 1000 (10 to 19) OR 1.07 (0.77 to 1.47) 7641 (4 studies) ⊕⊝⊝⊝ Very Low a, b 62 per 1000 (45 to 83) OR 0.71 (0.48 to 1.07) 7408 (2 studies) ⊕⊕⊝⊝ Lowa, b 18 per 1000 (12 to 26) OR 1.14 (0.62 to 2.16) 7335 (2 studies) ⊕⊝⊝⊝ Very Lowa, b 24 per 1000 (13 to 44) *The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect (and its 95% CI) estimated for the intervention group. Assumed risk was estimated from the meta-analysis of control risks Secondary outcomes For 90-day, 180-day, and 1-year mortality, early surgery was not significantly associated with lower mortality (90-day: 3 studies, 536 patients, crude OR: 0.91, 95% CI: 0.51–1.60, I² = 9%; 180-day: 2 studies, 354 participants, crude OR: 0.46, 95% CI: 0.12–1.77, I² = 55%; 1-year: 3 studies, 351 patients, crude OR: 0.61, 95% CI: 0.34–1.09, I² = 0%, Fig. 2B–D). Low-certainty evidence suggested that early surgery may result in little to no difference in cardiac complications (2 studies, 7,408 patients: OR: 0.71, 95% CI: 0.48–1.07; I² = 5%, Fig. 3A). Very low-certainty evidence indicated that the effect of early surgery on pulmonary complications (4 studies, 7,641 patients: OR: 1.07, 95% CI: 0.77–1.47; I² = 0%, Fig. 3B) and pulmonary embolisms (2 studies, 7,335 patients: OR: 1.14, 95% CI: 0.62–2.16; I² = 0%, Fig. 3C) remained highly uncertain. No significant differences were observed for cardiac complications (low-certainty evidence), pulmonary complications (very low-certainty evidence), or pulmonary embolism (low-certainty evidence). Early surgery was not significantly associated with reduced rates of DVT (3 studies, 7,283 patients, crude OR: 0.75, 95% CI: 0.54–1.04, I² = 0%, Fig. 3D) and additional surgery (3 studies, 4,635 patients, crude OR: 0.98, 95% CI: 0.57–1.67, I² = 0%, Fig. 3E). Analysis of mobility status was not performed due to insufficient reporting of relevant data in the primary studies. Fig. 3Forest plot of the secondary outcomes. A Crude odds ratio for cardiac complications. B Crude odds ratio for pulmonary complications. C Crude odds ratio for pulmonary embolism. D Crude odds ratio for deep vein thrombosis. E Crude odds ratio for additional surgery. SE standard error; IV inverse variance; CI confidence interval Forest plot of the secondary outcomes. A Crude odds ratio for cardiac complications. B Crude odds ratio for pulmonary complications. C Crude odds ratio for pulmonary embolism. D Crude odds ratio for deep vein thrombosis. E Crude odds ratio for additional surgery. SE standard error; IV inverse variance; CI confidence interval Subgroup analyses Subgroup analyses were not performed because of insufficient reporting of relevant data in the primary studies. Sensitivity analyses Sensitivity analyses indicated that early surgery may be associated with reduced mortality. Differences were observed in 30-day (3 studies, 27,506 patients: crude OR: 0.74, 95% CI: 0.56–0.97; I² = 0%, Fig. 4A), 180-day (3 studies, 402 patients: crude OR: 0.35, 95% CI: 0.17–0.71; I² = 0%, Fig. 4B), and 1-year mortality (2 studies, 642 patients: adjusted HR = 0.56, 95% CI: 0.38–0.83; I² = 0%; 4 studies, 424 patients: crude OR: 0.61, 95% CI: 0.34–1.08; I² = 0%, Fig. 4C, D). For both 90-day and 1-year mortality, exclusion of arthroplasty increased heterogeneity and resulted in imprecise estimates (90-day: 2 studies, 529 patients, crude OR: 1.07, 95% CI: 0.29–3.96, I² = 46%; 1-year: 2 studies, 244 patients, crude OR: 0.68, 95% CI: 0.22–2.05, I² = 47%, Fig. 4E, F). Sensitivity analyses for operative time and blood loss could not be performed because too few studies reported these outcomes. Fig. 4Forest plot of sensitivity analyses. A Crude odds ratio for 30-day mortality. B Crude odds ratio for 180-day mortality. C Hazard ratio for 1-year mortality. D Crude odds ratio for 1-year mortality. E Crude odds ratio for 90-day mortality excluding arthroplasty. F Crude odds ratio for 1-year mortality excluding arthroplasty. SE standard error; IV inverse variance; CI confidence interval Forest plot of sensitivity analyses. A Crude odds ratio for 30-day mortality. B Crude odds ratio for 180-day mortality. C Hazard ratio for 1-year mortality. D Crude odds ratio for 1-year mortality. E Crude odds ratio for 90-day mortality excluding arthroplasty. F Crude odds ratio for 1-year mortality excluding arthroplasty. SE standard error; IV inverse variance; CI confidence interval Discussion Our meta-analysis of 16 studies found that the pooled evidence was insufficient to determine whether early surgery reduces mortality in older adults with distal femur fractures. The certainty of the evidence regarding surgical timing in this population was low. No clear benefit was observed for complications, with certainty ratings ranging from low to very low. Therefore, the effect of surgical timing on clinical outcomes remains uncertain. Early surgery in older adults with distal femur fractures was associated with little or no difference in 30-day mortality. Evidence from comparable hip-fracture cohorts and current guidelines supports surgery within 24–48 h [6, 7]. In our analysis of 29,309 patients, earlier surgery was not significantly associated with reduced mortality (adjusted OR: 0.77, 95% CI: 0.56–1.05), likely reflecting limited statistical power, residual confounding, or clinical heterogeneity rather than a definitive absence of effect. Sensitivity analyses also favored early surgery, likely reflecting the use of stricter timing definitions and analytic strategies intended to reduce time-related bias. Overall, because the evidence was of low certainty and the primary analysis was nonsignificant, we cannot make a strong recommendation. However, given consistent findings from sensitivity analyses, a cautious, conditional suggestion to operate within 48 h of admission appears reasonable when feasible. Our SRMA found no significant association between early surgery and an overall reduction in complications among older adults with distal femur fractures. This result aligns with several previous studies that also failed to demonstrate a clear benefit of earlier intervention in reducing postoperative complications [2, 3, 10]. Several factors may explain these findings as follows: (1) heterogeneity in fracture type and surgical technique [2, 15]; (2) variability in definitions and assessments of complications [4, 10]; (3) differences in patient comorbidities and perioperative optimization [2, 9, 12, 13]; and (4) attenuation of any potential mortality benefit by temporary external stabilization, which allows pain control and upright positioning [23]. In contrast, some evidence suggests that early surgery may reduce specific complications, such as urinary tract infections or acute coronary syndromes [2, 4]. Taken together, these considerations indicate that postoperative complications are influenced by multiple interacting factors, and surgical timing alone is unlikely to account for their occurrence. The primary strength of this study is that it is the first SRMA to quantify the effect of early surgery on mortality and complications in older adults with distal femur fractures. Another strength is our adherence to transparent reporting and rigorous methodology by following PRISMA guidelines, the Cochrane Handbook, and GRADE recommendations. We also contacted original authors and obtained unpublished data, including separate results where available. As a clinical implication, we conditionally recommend targeting surgery within 48 h of admission once the patient is medically optimized. Previous studies have reported that earlier surgery is associated with shorter hospital stays and lower healthcare costs, and in some analyses with reduced mortality [3, 4, 24]. However, evidence also indicates that surgery should proceed only after comprehensive preoperative assessment to ensure patient safety [2, 25]. Although our analysis found the low-certainty evidence and no clear mortality benefit, these findings support medically optimized early surgery—ideally within 48 h—as a clinically and economically meaningful strategy. Our study has several limitations. First, the number of patients included was small, resulting in wide CIs and reduced statistical precision. Given the rarity of distal femur fractures [3], larger, multicenter prospective studies are needed to strengthen the evidence base. Second, all included studies were retrospective cohort studies, making them prone to selection bias and information bias. Third, although some studies reported adjusted estimates [4, 8], many relied on crude ORs. Several studies did not adequately control for key confounders, such as comorbidities, fracture type, and surgical techniques, highlighting the need for future investigations with rigorous adjustment for confounding. Additionally, the inclusion of different surgical procedures may have contributed to clinical heterogeneity. Because procedure-specific outcomes were rarely reported, we were only able to assess this issue by excluding studies involving arthroplasty in sensitivity analyses for crude 90-day and 1-year mortality. Excluding arthroplasty increased heterogeneity and widened the CIs, resulting in imprecise estimates. Fourth, our findings were not validated in other settings or populations. Reports were concentrated in a small number of countries—primarily the United States, with limited data from Japan, parts of Europe, and China—which may limit generalizability. Fifth, definitions of early surgery varied across studies, some measured timing from admission, whereas others used time from injury, and the time zero for follow-up was not aligned. These issues likely introduced exposure misclassification and immortal time bias, potentially distorting effect estimates. Sixth, heterogeneous outcome definitions and incomplete or inconsistently timed reporting of early postoperative and preoperative complications limited cross-study comparability and may have biased between-study comparisons. Finally, we did not assess outcomes such as healthcare costs, functional recovery, or quality of life. Although our review focused on mortality and major complications, future research should incorporate these broader clinical and economic outcomes. Addressing these limitations will be crucial to generating high-quality evidence that informs clinical practice and guideline development. In conclusion, our findings suggest that the effect of early surgery for distal femur fractures in older adults remains uncertain due to low-certainty evidence. Notably, when more stringent timing definitions and strategies to reduce time-related bias were used, sensitivity analyses indicated a potential benefit of earlier surgery; however, these findings should be interpreted cautiously due to the observational character of the data. High-quality prospective studies are needed to clarify whether surgical timing improves clinical outcomes. These findings support ongoing efforts to clarify the role of early surgery in older adults with distal femur fractures, better understand processes of care that may influence postoperative outcomes, and confirm these findings through further high-quality research. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1