Early and 1-year mortality of native geriatric distal femur fractures: A systematic review and time-to-event meta-analysis

Abstract

Purpose

Distal femur fractures (DFF) account for 6% of all femoral fractures and predominate in females. The current 1-year mortality of DFF is currently reported to be between 10 and 38%, a wide margin, and confounded by multiple factors including age, high energy mechanisms, pathological and periprosthetic fractures. The purpose of this study was to assess and determine all-cause mortality following geriatric native distal femur fractures at 30 days, six months and one year.

Methods

– The databases Cochrane CENTRAL, MEDLINE, EMBASE and NHS NICE Healthcare Databases Advanced Search Interface were searched in accordance with PRISMA guidelines. Original research articles relevant to mortality outcomes in native geriatric distal femur fractures following low energy trauma were included. A time-to-event data meta-analysis model was used to estimate pooled 30-day, six month and one-year mortality. A random effects meta-regression model was performed to assess potential sources of heterogeneity when studies reported on factors affecting the mortality observed in patients with geriatric distal femur fractures.

Results

– Thirteen studies were included in the meta-analysis with a mean age of 79.6 years. Eight studies reported the 30-day mortality of distal femur fractures in patients as a pooled estimate of 8.14%. Pooled estimate for 6-month mortality reported was 19.5% and the one-year mortality reported by ten studies was 26.10%. Time-to-event modelling showed that risk of mortality at one year in elderly patients with distal femur fractures was significantly higher HR = 4.31 (p < 0.001). When evaluating prognostic predictors, age and Type C fracture were predictive of highest mortality rates.

Conclusions

– This study is the first meta-analysis to evaluate the early and long-term mortality observed in elderly patients presenting with native distal femoral fractures. Through our results we have shown the quantifiable impact patient age and fracture configuration has on one-year mortality in this patient cohort.

1. Introduction

Distal femur fractures (DFF) have a reported incidence of approximately 8 in 100 000 per year¹ and account for up to 6% of all femoral fractures. DFFs predominate in females¹ and exhibit a bimodal age distribution.² Young patients usually undergo high energy mechanisms of injury while 50–70% of cases occur in the geriatric population as a result of low energy falls.³ DFF incidence is only predicted to increase with the ageing global population and will certainly present a significant burden to health care services.⁴

DFF operative management addresses poor bone quality, extensive comminution and often, intra-articular communication.^{5, 6, 7} Furthermore, these technical challenges occur in the context of frail patients with complex comorbidities who require a multidisciplinary approach, early operative intervention, and rehabilitation to mitigate the well-recognised complications of prolonged bed rest. Unsurprisingly, patients with DFFs exhibit overlapping medical comorbidities to those with hip fractures and comparable high rates of disability, morbidity and mortality.^{8, 9, 10, 11} The Department of Health (UK) introduced Best Practice Tariffs for the management of hip fractures in 2010 and succeeded in reducing length of stay; improving patient independence and 30-day mortality.^12,13 Only recently have these Tariffs been extended to DFF patients.¹⁴

DFF 1-year mortality is currently reported to be between a wide margin of 10–38%.^{5,10,15, 16, 17} These study cohorts are often confounded by patients of all ages, high energy mechanisms, polytrauma, open, pathological or periprosthetic fractures. In order to establish accurate mortality data this meta-analysis aims to specifically evaluate early (30-day, 6-month) and 1-year mortality in native low-energy geriatric DFFs. Secondary aims were to identify risk factors to predict increased mortality.

2. Methods

This review was designed and performed in accordance with the recommendations of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines for reporting meta-analyses.

2.1. Inclusion criteria

• ●Original research articles relevant to mortality outcomes in native geriatric distal femur fractures following low energy trauma.
• ●Live human study participants.

2.2. Exclusion criteria

• ●Conference abstracts, letters, case reports, studies with fewer than 10 patients
• ●Laboratory, cadaveric, synthetic subjects.
• ●Pathological fractures.
• ●Periprosthetic fractures.
• ●High energy and open injuries.
• ●Studies lacking mortality data.

2.3. Literature search and study selection

A systematic search of Cochrane CENTRAL database, MEDLINE, EMBASE, PubMed and NHS NICE Healthcare Databases Advanced Search interface was executed independently by two authors (A.S. and F.J.R). The last search was conducted on the July 8, 2023. Exploded medical subject headings and corresponding appropriate terms were utilised (Appendix A). The two authors removed duplicate records and assessed relevant titles and abstracts which were tagged as included, excluded, or needing further review. Full texts were reviewed for all included studies and those needing further review. References of relevant articles were manually searched to identify additional eligible studies. Any disagreements in study selection were resolved by discussion between reviewers and any remaining discrepancies were resolved by a third reviewer (D.M.R.).

2.4. Data collection

The following data were extracted from the included studies: authors, publication year, country of origin, study design, study size, number of participants, participant demographics (including mean age, gender, body mass index, smoker, comorbidities, Charlson Comorbidity Index, ASA), pre-injury status (including residence, mobility, independence, cognitive impairment), injury characteristics (AO classification), management (time to surgery, surgical management) and outcomes (length of inpatient hospital stay, 30-day). Extracted data were entered into a pre-generated standard Microsoft® Excel (Microsoft Corporation, Redmond, Washington, USA) file. Data extraction was performed independently by two authors (A.S. and F.J.R.), and disagreements were resolved by discussion and consensus. If no agreement could be reached, a third author was consulted to act as an arbitrator (D.M.R.).

2.5. Primary outcome

The primary outcome was to determine aggregate all-cause mortality following geriatric native distal femur fractures at 30-days, six months and one year as well through a time-to-event meta-analysis to model the risk of one year mortality observed in these patients.

2.6. Secondary outcome

Secondary outcomes were to assess the impact of patient demographics i.e., age, gender, comorbidity indices when reported as well as injury demographics (e.g., fracture configuration and fixation type) on early and one year mortality of patients. Moreover, we aimed to compare morbidity and mortality of geriatric distal femur fractures against other known fragility fractures (e.g., neck of femur fractures) when reported.

2.7. Assessment for risk of bias

Three authors (A.S., D.M.R. and Y·C.) critically appraised the methodological quality of the included studies. Newcastle Ottawa scale was utilised for non-randomised studies (range 0–9). Discrepancies were resolved by reviewer discussion, and if they remained unresolved, a third investigator was consulted (F.J.R.).

2.8. Assessment for reporting bias

We have included a total of 13 studies; seven were included in the funnel plot analysis, which is not a sufficient number for reporting bias using a funnel plot.¹⁸

2.9. Data synthesis and statistical analyses

A descriptive synthesis summarised study characteristics, patient demographics and reported outcomes. Where substantial heterogeneity in study design and population demographics occurs, a narrative review was used to analyse this data. A mixed effects meta-analysis model was only performed when no evidence of substantial study heterogeneity was found. Meta-analysis was performed for patient and injury characteristics (age, gender, Charlson comorbidity index (CCI), AO classification); definitive management (operation versus no operation, fixation type) and length of stay. For early and late outcomes (30-day and one year mortality), a time-to-event data meta-analysis model was used to estimate pooled 30-day, six-month and one-year mortality through reported aggregate data. Individual study hazard ratios (HR) and standard error (SE) for outcome measures were calculated from a mixture of direct and indirect inverse variance modelling. When mean values were not available for continuous outcomes, data on median and interquartile range (IQR) were extracted and subsequently converted to mean and standard deviation (SD) using the well-practised equation described by Hozo et al..¹⁹

A random effects meta-regression model was performed to assess potential sources of heterogeneity when studies reported on factors affecting the mortality observed in patients with geriatric distal femur fractures. Known patient risk factors such as age, high comorbidity indices, fracture configuration and fixation type were incorporated into the meta-regression model. Sensitivity analysis was performed to evaluate the robustness of the observed outcomes and compare studies rated as low or moderate risk of bias and assess against potential confounders in all studies reporting adjusted and unadjusted results. A Newcastle–Ottawa Score (NOS) of 5 or more has been shown to be a moderate or good quality rating of paper, hence this cut-off was used for sensitivity analysis. Additionally, primary meta-analysis was repeated for studies specifically reporting patient mortality from large registry data to address publication bias. All analyses were performed on STATA 16 (Stata-corp, College station, Texas, USA).

3. Results

1716 citations were identified. Following rigorous screening, thirteen full text articles were included in the review (Fig. 1).

Fig. 1.

Shows the PRISMA diagram for search strategy and search selection.

3.1. Patients’ characteristics in meta-analyses

The 13 papers included in the meta-analysis reported a mean age of 79.6 (95% CI 73.1–85.4) years. The proportion of females was 80% (95% CI 78–82%). A pooled estimate for CCI was 4.1 (95% CI 2.15–6.13). The most common distal femur fracture configurations were 33-A (69% (95% CI 66–72%)), followed by 33-C (25% (95% CI 22–28%)) and 33-B (11% (95% CI 9–14%)). Moreover, 98% (95% CI 97–99%) of these patients were managed surgically, of which 95% (95% CI 95–96%) underwent intramedullary nail fixation in five studies and 92% (95% CI 90–94%) had plate fixation in three studies. The mean estimate of length of stay was 13.9 days (95% CI 12.2–15.6). Three studies^{20, 21, 22} stratified patient comorbidities using the Charlson Comorbidity index (CCI) which pooled values of 4.12 (95% CI 2.15–6.13). Four studies^{21,23, 24, 25} assessed length of hospital stay where mean length of stay was 13.90 days (95% CI 12.18–5.61). Patient characteristics have been summarised in Table 1.

Table 1.

Study characteristics and patient demographics.

Author, Year	Country	Data collection (years)	Patient Demographics					Fracture Characteristics and management				Outcome			Newcastle Ottowa Score
			Number	Age		Gender/F	CCI	Fracture definition	Nail	Plate/Screws	Knee Replacement/prostheses	30-day Mortality	6 Month Mortality	1 Year Mortality
Appleton et al., 200623	UK	1987–2004	52 (54 fractures)	>55	82 (55–98)	49	NR	33A3 - 9 33C - 45	NIL	NIL	54	7/51 (15%)	NR	22 (42%)	5
Brogan et al., 201626	UK	2008–2014	80 (72–8 with TKR)	NR	81 (60–103)	NR	NR	33A–56 33B–6 33C - 11	9	62	NR	NR	NR	22%	6
Dunlop et al., 199924	UK	1994–1997	28	>55	83 (Std 10.2)	26	NR	33A - 33C -	28	0	0	2/28 (7.1%)	5/28 (17%)	9/28 (32.1%)	5
El-Kawy et al., 200725	UK	1998–2002	23	NR	75 (65–97)	16	NR	33A–23	23	NR	NR	NR	NR	8%	5
Jennison et al., 201915	UK	2011–2016	88	>65	82.4 (65–103)	80	NR	33A - 67 33B - 5 33C - 75	5	75	NR	8 (9.1%)	18 (20.5%)	30 (34.1%)	7
Larsen et al., 202010	Denmark	2005–2010	293	≥60	62.6	66.2%	3.4 (SD 2.6)	Non-periprosthetic distal femur fractures	NR	NR	NR	8%	26%	35%	6
Lundin et al., 202127	Sweden	2001–2016	14 918	≥50	71	73%	NR	Distal femur fracture	NR	NR	NR	NR	NR	20%	6
Mubark et al., 202020	UK	2010–2014	189	>60	81.1 (60–99, ± 8.1)	65%	5.1 ± SD 2.1	Lower 1/3 femur or distal end of femur with or without intra-articular extension	NR	NR	NR	18.3 (9.68%)	38.4 (20.32%)	34.31%	5
Muller et al., 202028	Germany	–	45	>60	–	–	–	–	–	–	–	-/NR	–	20%	5
Myers et al., 20189	USA	2002-2012	149	>60	NR	NR	NR	AO 33 A/B/C	NR	NR	NR	NR	NR	25/149 (16.8%)	5
Nyholm et al., 201729	Denmark	2012–2016	392	NR	76 (50–101)	310	NR	33A - 255 33B - 60 33C - 77	16	354 + 10	8 Ex-fix 4 other	28 (7.1%)	NR	NR	5
Pean et al., 201521	USA	NR	555	NR	PF - 71.22 IM nail - 74.5	133	PF - 3.18 IM nail - 3.60	Distal femur fractures	44	511	NR	3.72%	NR	NR	5
Streubel et al., 201122	USA	1999- 2009	44	≥60	78.3 (60–100, SD 10)	35	2.16 ± SD 2.1	33A–22 33C - 22	NR	44	NR	1 (2%)	5 (13%)	8 (23%)	5

Key: NR – not reported, TKR – total knee replacement. PF - plate fixation, IM nail - intramedullary nail.

3.2. Study characteristics and methodological assessment (Newcastle-ottawa scale)

There are various study designs and characteristics in the 13 studies included in the meta-analysis. Of these, 10 studies^{9,10,15,20,22,25, 26, 27, 28, 29} were carried out retrospectively, and the remaining three studies^21,23,24 were carried out prospectively. Five out of the 13 studies were large multi-site retrospective studies utilising national databases,^{10,21,26,27,29} with the remaining studies gathering patient data from single sites. None of the studies included in the meta-analyses were blinded. Studies often compared outcomes of distal femur fractures to patients with hip fractures,^15,22,27 femoral shaft fractures,²⁷ proximal femur fractures²⁰ and periprosthetic fractures.^9,10 Follow-up in the assessment of mortality rates are essential and of these 13 studies, a few had short-term follow-up (30–90 days),^21,29 the vast majority had medium-term follow-up (12–24 months)^{9,15,20,22,24, 25, 26} and others covering between 5 and 16 years.^10,23,27,28 The methodological assessment of these studies has been summarised in Appendix B.

3.3. Primary outcomes

Eight studies^{10,15,20, 21, 22, 23, 24,29} reported the 30-day mortality of distal femur fracture patients, showing a pooled estimate of 8.14% (95% CI 5.36–10.91). The pooled estimate for 6-month mortality, reported by five studies^{10,15,20,22,24} was 19.54% (95% CI 18.67–20.42%), whilst one-year mortality reported by 11 studies^{9,10,15,20,22, 23, 24, 25, 26, 27, 28} had a pooled proportion of 26.1% (95% CI 25.5–26.6%) as shown in Fig. 2 (A-C). We performed a time-to-event meta-analysis using an inverse variance random effects model to obtain a pooled hazard ratio HR of 4.31 (95% CI 2.74–5.89) [Z = 0.00, p < 0.001] as shown in Fig. 2D. Variation in the pooled HR attributable to heterogeneity was high with I² = 98.63%. Three studies^22,23,28 mean survival of patients during their follow-up period with the pooled values being 37.71 months (95% CI 7.86–67.57).

Fig. 2.

Shows the forest plots for reported mortality at (A) 30-days, (B) Six-months and (C) One-year and the associated risk of mortality (D) predicted at one year.

3.4. Secondary outcomes

The results of secondary outcomes have been summarised in Appendix C. When assessing the reported predictors of mortality from the included studies, the following variables were included in our meta-regression model.

3.4.1. Patient age

Six out of the 13 studies further analysed the impact of patient age on early and one-year mortality. Two studies showed that age was a significant risk factor for increased mortality at 30-days (p = 0.015)¹⁵ and overall survival with an odds ratio of 1.052 (p = 0.000).²⁶ Three studies found a significant increase in 30-day and one-year mortality with increasing age,^10,21,29 however one paper showed using regression analysis that age did affect survival.²² In our meta-regression analysis, the risk of mortality increased proportionally irrespective of time to follow-up with increasing age (β = 16.16 p = 0.04) as shown in Fig. 3-A and 3-B.

Fig. 3.

Shows metaregression plots assessing impact of age on (A) 30-day mortality and (B) one year mortality in contrast to 33-A fractures on 30-day mortality (C) and 33-C fractures on 30-day mortality (D).

3.4.2. Gender

Four out of the 13 studies also further analysed the impact of gender on early and one-year mortality. One study identified gender to be a significant risk factor for overall survival with an odds ratio of 1.825 (p = 0.017)²⁶ with men less likely to survive than females. Two further studies concluded similar results such that the male gender significantly increased both 30-day (p = 0.010) and one-year mortality (p = 0.04).^10,29 Similarly, one study did not show gender to have an independent effect on mortality.²²

3.4.3. Fracture configuration

Two studies investigated the fracture configuration and classification in relation to mortality. There was overall no statistical difference in fracture configuration and classification in the overall one-year mortality.^9,22 Meta-regression analysis showed that an increasing proportion of extra-articular (33A) distal femur fractures predicted reduced 30-day mortality in comparison to an increasing proportion of intra-articular (33C) fractures (β = 4.21 p = 0.05) as shown in Fig. 3-C and 3-D.

3.4.4. Timing of surgery

Seven studies^{9,10,15,20,22,26,29} further assessed whether timing of surgery affected 30-day, 6-month and one-year mortality in patients with distal femur fractures. Two studies have shown no overall association with delay to surgery and mortality.^26,29 Two studies identified no significant association with time to surgery and 30-day mortality,^15,22 whereas one study exhibited significant patient mortality at 30-days when surgery occurred more than 2 days from injury (p = 0.036).⁹ Furthermore, 6-month mortality was significantly associated with increased time to surgery in two studies.^9,22 The same study also found a significant association with the effect of timing surgery on one-year mortality (p = 0.005).²²

3.4.5. Place of residence

One study showed that residence status (home versus institution) was not a risk factor for mortality at 30 days (p = 0.42).¹⁵

3.4.6. Other fragility fractures

Seven studies^{9,10,15,20,22,26,27} included the mortality rate of other fragility fractures and compared this with DFF mortality rates. Two studies compared mortality of DFF to hip fractures, which showed no significance at 30 days (p = 0.63),¹⁵ 6-months and one-year mortality.²² There is, however, an increased odds ratio of mortality in DFF compared with hip fractures (Supplementary Fig. 1). One study showed DFF had higher mortality rates compared with proximal femur fractures at 30-days (9.68% vs 6.99%), 6-months (20% vs 14.5%) and one-year (34.4% vs 21.5%), though no comment on its statistical significance.²⁰ On the other hand, one study showed no difference in one-year mortality between pelvic fractures, femoral shaft and DFF.²⁷ DFF mortality was also compared with periprosthetic fractures, which showed no statistical significance in mortality rates between the two,^9,22 though there is reported increased cumulative risk in mortality.¹⁰

4. Discussion

Native geriatric distal femur fractures (DFF) are a growing burden on health services, particularly given the expanding elderly global population. Despite posing overlapping management challenges with fragility hip fractures, mortality data is less robust for DFFs. This meta-analysis quantifies early and one year mortality following DFFs in studies that pre-date the extended scope of the NHS England Best Practice Tariff (BPT) for fragility fractures and the COVID-19 pandemic. Thirty-day, 6-month and 1-year mortality were 8.1%, 19.5% and 26.1%, respectively. We report data which align with pre-pandemic hip fracture mortality at 30-days as reported by the National Hip Fracture Database (6.1%)¹³ and DFF 1-year mortality reported in literature.^{5,10,15, 16, 17}

The introduction of the BPT has resulted in year-on-year reduction of 30-day hip fracture mortality from 8.4% (2012)³⁰ to 6.1% (2018). The latest report which included data on DFFs acknowledges lagging performance including timely orthogeriatric reviews (19%), early operation (47% within 36 h) and post-operative mobilisation (74%).¹³ DFFs are yet to receive equivalent priority as hip fractures with the concurrent overburdening COVID-19 pandemic undoubtedly contributing to suboptimal patient care. Unsurprisingly, COVID-19 infection in the frail hip fracture patient cohort confers increased 30-day mortality,^13,31 which could suggest a similar prognostic risk as DFF patients are of similar frailty. Incentivising DFF management is necessary and will result in improved patient outcomes.

Geriatric periprosthetic fractures of the distal femur pose equivalent or greater challenges to that of native femoral fractures and are reported to have worse prognosis.^{7,22,32, 33, 34},^{7,28,31, 32, 33} The possibility of systematic differences in patients with the presence of stress risers predisposing to fractures; delays due to patient transfer to specialist centres; surgical planning; surgeon availability and longer, complex operations³¹ influenced the decision to exclude studies pertaining to periprosthetic fractures and where the authors were unable to delineate native distal femur fractures from periprosthetic fractures.

There are several limitations to this meta-analysis. Although our pooled population exhibited a mean age of 79.6 years, and the majority of patients were female. There is likely significant between study heterogeneity, which is likely due to varying study sizes, implant types and fixation methods (as shown in Supplementary Fig. 2). We analysed patient groups from small case series to large registry data with a wide variety of age cut-offs ranging from 50 to 60 years. The studies included American and European patient cohorts which will likely compromise the generalisability of our patient sample. Furthermore, a significant number of studies, particularly registry data, did not declare patient characteristics, fracture classification or definitive operative management thereby limiting sub-group analysis. At the stage of full-text review, the authors endeavoured to extract data pertaining to native geriatric DFFs however, certain studies were contaminated with open injuries, periprosthetic fractures and high energy mechanisms. This significant heterogeneity in reviewed papers with missing data limited subsequent conclusions. Despite these limitations, this study is the first meta-analysis investigating mortality in native geriatric distal femur fractures. Prospective research is required following implementation of the BTP to determine the benefit of the holistic, multidisciplinary approach of native geriatric DFFs.

5. Conclusion

Our study is the first meta-analysis to evaluate the early and long-term mortality observed in elderly patients presenting with native distal femoral fractures. We demonstrate the early 30-day mortality of 8.14% and one-year mortality of 26.1%. Our analysis also highlights the importance of age and fracture configuration on the mortality of patients within these cohort groups. Given their high mortality and associated risk factors, we believe patients with DFFs should be counselled and treated similar to other fragility fractures i.e. neck of femur fractures which also carry a significant morbidity and mortality.

Funding

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

Credit author Statement

Yanjinlkham Chuluunbaatar: Methodology, Data Curation, Writing – Original Draft, Writing – Review and Editing Nawal Banacher: Writing – Original Draft Harnoor Khroud-Dhillon: Conceptualization, Writing – Original Draft Ananth Srinivasan: Methodology, Writing – Original Draft Djamila Rojoa: Methodology, Writing- Original Draft Firas J Raheman: Conceptualization, Supervision, Methodology, Formal analysis, Writing – Review & Editing.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A

MeSH search team:

((Distal femur fractures).ti, ab OR (DFF).ti, ab OR (Femur fractures).ti,ab) AND ((mortality).ti, ab OR (Morbidity*).ti, ab OR (outcome*).ti, ab OR (morbidity and mortality).ti, ab AND (fixation).ti, ab OR (ORIF).ti, ab OR (intramedullary fixation).ti,ab)

Appendix B. Newcastle Ottawa Score – selection, comparability and outcomes of each study included in the meta-analysis

Study	Selection				Comparability		Outcome			Total Score
	1	2	3	4	5	6	7	8	9
Appleton et al., 200623	★	–	★	–	–	–	★	★	★	5/9
Brogan et al., 201626	★	–	★	–	★	–	★	★	★	6/9
Dunlop et al., 199924	★	–	★	–	–	–	★	★	★	5/9
El-Kawy et al., 200725	★	–	★	–	–	–	★	★	★	5/9
Jennison et al., 201915	★	–	★	–	★	★	★	★	★	7/9
Larsen et al., 202010	★	–	★	–	★	–	★	★	★	6/9
Lundin et al., 202127	★	–	★	–	★	–	★	★	★	6/9
Mubark et al., 202020	★	–	★	–	–	–	★	★	★	5/9
Muller et al., 202128	★	–	★	–	–	–	★	★	★	5/9
Myers et al., 20189	★	–	★	–	–	–	★	★	★	5/9
Nyholm et al., 201729	★	–	★	–	★	–	★	–	★	5/9
Pean et al., 201521	★	–	★	–	★	–	★	–	★	5/9
Streubel et al., 201122	★	–	★	–	–	–	★	★	★	5/9

Appendix C Table showing information regarding secondary outcomes – to assess the impact of age, gender, fracture configuration, timing of surgery, place of residence and other fragility fractures on early and one-year mortality of patients. KEY: NR – not reported

	Age	Gender	Fracture Configuration	Timing of Surgery	Place of Residence	Other fragility fractures
Appleton et al., 200623	NR	NR	NR	NR	NR	NR
Brogan et al., 201626	Significant factor influencing survival: Odds ratio 1.052 (p = 0.000).	Significant factor influencing survival: Odds ratio 1.825 (p = 0.017)	NR	Delay to surgery was not associated with increased mortality	NR	16.5% has periprosthetic THR fractures – suggestive of worse survival but not significant on log-rank testing (p = 0.067).
Dunlop et al., 199924	NR	NR	NR	NR	NR	NR
El-Kawy et al., 200725	NR	NR	NR	NR	NR	NR
Jennison et al., 201915	Age is a significant risk factor for mortality at 30 days (p = 0.015)	NR	NR	Time to surgery was not a significant risk factor for 30-day mortality (p = 0.376)	Residential status was not a significant risk factor in mortality at 30-days (p = 0.43).	No significant difference in 30-day mortality between the two fracture types – hip fractures and distal femur fractures (p = 0.63).
Larsen et al., 202010	10% increase (p < 0.00) in the one-year mortality for each year increase in age.	Males were 2.6x more likely (p = 0.04) to die within the first year following a DFF in comparison to women.	NR	No association between delay of surgery and death within one year (p = 0.28)	NR	No report on overall statistical significance. Patient above 60 with a DFF had an increased cumulative risk of death when compared with patients with a peri-prosthetic fracture.
Lundin et al., 202127	Mortality in patients aged 18–49 was approx. 1.5% Patients aged 50 and above mortality was 20%. No report on statistical significance.	NR	NR	NR	NR	One-year mortality for pelvic fractures = 21% Femoral shaft fracture 21% DFF 20% NR on statistical significance.
Mubark et al., 202020	NR	NR	NR	Time to surgery in DFF was significantly longer than proximal femur group (p = 0.04) but not reported on mortality.	NR	DFF had ‘higher’ mortality at all times when compared with proximal femur. No report on statistical significance.
Muller et al., 202128	NR	NR	NR	NR	NR	NR
Myers et al., 20189	NR	NR	No statistical difference in OTA/AO classification and overall mortality up to one year.	Patient mortality was significantly higher at: 30-days (p = 0.036) 6-months (p = 0.019) One year (p = 0.018) when surgery occurred more than 2 days from the injury.	NR	No statistical significant difference in overall mortality between native bone and periprosthetic factures, IM nails or ORIF.
Nyholm et al., 201729	Increasing age significantly increased both 30-day and 90-day mortality: 80–90 years p < 0.001 90+ years p < 0.001	Male gender significantly increased both 30-day and 90-day mortality (p = 0.009)	NR	No association between surgical delay and mortality following surgery.	NR	NR
Pean et al., 201521	Increasing age had a significant increase in serious adverse effects (SAE p = 0.05) and mortality (p = 0.03) at 30-days.	NR	NR	NR	NR	NR
Streubel et al., 201122	Regression analysis did not show an effect of age on survival.	Regression analysis also did not show an independent effect on mortality.	Regression analysis also did not show an independent effect on mortality for fracture classification	No statistical difference found for 30-day mortality to surgical delay (p > 0.32) Significance at: 6-months mortality (p = 0.02) One-year mortality (p = 0.005).	NR	No statistical difference in the overall mortality, 30-day, 6-month and 1-year mortality between non-periprosthetic DFF and periprosthetic DFF (p = 0.08 & hip fractures (p = 0.95).

Appendix B. Supplementary data

The following are the Supplementary data to this article: