2.1 Study design

This study was a single-centre, retrospective observational cohort study involving patients ≥65 years with distal femur fractures that underwent surgery at our institution.

2.2 Participants

Our cohort comprised of 111 patients who were ≥65 years, admitted with distal femoral fractures and managed operatively with a single lateral plate between January 2013 and December 2023. Patients were then classified into 3 main groups; group 1 (65–74 years), group 2 (75–84 years), and group 3 (85–98 years) respectively.^29,30 Biological sex was coded in the variable Gender, with 1 = female and 2 = male.

2.3 Variables and definitions

Using the electronic medical records available within our NHS trust, we obtained relevant information for each patient, including patient demographics, fracture classifications, treatment methods, and mortality outcomes. The AO/OTA classification system was applied for classification of our NDFFs, while the Lewis and Rorabeck classification was utilized for PDFFs ^31–34.

Additional variables such as the number of proximal and distal cortices engaged, as well as the number of proximal and distal wires used as adjuncts. We also collected the time to theatre, LOS, time to FWB status (<48 h and <30 days),^9,10,35,36 number of complications 30-day theatre returns and patient mortality rates.

2.4 Statistical analysis

Data analysis was performed using SPSS software (version 27.0; IBM Corp., Armonk, NY, USA). Descriptive statistics were calculated for the entire cohort and separately for each of the age-based subgroups. Continuous variables are reported as means and standard deviations or medians and interquartile ranges, as appropriate. The subgroups were compared using Kruskal–Wallis tests, with Pearson's χ² used for categorical variables. Associations between fixation characteristics (cortical purchase and wire use), LOS, and unplanned return to theatre were explored using Spearman rank correlation coefficients (ρ). Survival analysis was conducted using the Kaplan–Meier method to estimate postoperative survival times, with groups based on age (65–74, 75–84, and 85–98 years). Median survival times with 95% confidence intervals were reported, and between-group differences were assessed with the log-rank test. Statistical significance was set at p < 0.05 for all comparisons.

3.1 Study sample and baseline characteristics

A total of 111 patients were analysed. The mean age was 82.05 years (±8.44). Prespecified age groups comprised: Group 1 (65–74 years), 21/111 (18.9%); Group 2 (75–84 years), 46/111 (41.4%); Group 3 (85–98 years), 44/111 (39.6%). The cohort was predominantly female (Gender coded 1 = female, 2 = male): 91/111 (82.0%) female and 20/111 (18.0%) male. Laterality of injuries were equally distributed with 58/111 (52.3%) and 53/111 (47.7%) for right and left respectively. Out of our total 13 (11.7%) fractures were deemed unclassifiable as they did not fulfil our requirements for classification under AO/OTA classification or the Lewis and Rorabeck classification. Baseline characteristics are summarised in Table 1 (Table 1).

3.2 Operative fixation and construct details

All cases were treated with a SLP fixation construct. Mean proximal cortical purchase was 9.09 (±3.47; range 2–33) and distal cortical purchase 7.58 (±2.58; range 0–15). Mean number of proximal and distal wires used were 0.30 (±0.61; range 0–3) and 0.23 (±0.63; range 0–3) respectively. By age group, mean proximal cortices were 10.00 (±6.12; n = 21), 8.35 (±2.51; n = 46) and 9.44 (±2.34; n = 43); distal cortices were 8.25 (±2.61; n = 20), 7.00 (±2.77; n = 46) and 7.88 (±2.28; n = 43). Proximal wire use averaged 0.24 (±0.54), 0.39 (±0.75) and 0.23 (±0.48); distal wire use averaged 0.19 (±0.51), 0.39 (±0.83) and 0.07 (±0.33) for Groups 1–3, respectively. Kruskal–Wallis tests across age groups did not detect significant between-group differences for proximal cortices (p = 0.056), distal cortices (p = 0.089), proximal wire (p = 0.651) or distal wire (p = 0.054) (Table 2).

3.3 Peri-operative timings and length of stay

Mean time to theatre was 4.05 days (±11.57; range 0–118). Same-day surgery occurred in 11/110 (10.0%); 33/110 (30.0%) were operated within 1 day and within 2 days overall 66/110 (60.0%). Time to theatre did not differ significantly across age groups (p = 0.213).

Mean hospital LOS was 22.04 days (±19.73; range 2–110), with median LOS of 17 days (IQR 12-27) and then 12 days (65-74) in Group 1, 18 days (75-84) in Group 2, and 17 days (85-98) in Group 3. Between age-group differences were not significant (p = 0.112). Additionally, Spearman's correlation demonstrated only a weak positive association between LOS and age (ρ = 0.201, p = 0.040), indicating that older age was associated with slightly longer hospital stay but with a small effect size.

3.4 Post-operative weight-bearing

The time to FWB was only available for 81/111 (73.0%). Overall, the median FWB was 180 days (IQR 14–180), with a mean of 111.7 days (±104.5). By age group, medians were 180, 180, and 14 days for groups 1-3 respectively (Fig. 1).

Close

Fig. 1 Distribution of time to full weight-bearing by age group in patients ≥65 years treated with single lateral plate fixation. Median time to full weight-bearing was 180 days in the 65–74 and 75–84-year groups, and 14 days in the 85–98-year group.

Export to PPT

3.5 Complications and returns to theatre

We documented a complication rate of 7.3%, with unplanned returns to theatre occurring in 5.4%.

We identified that return to theatre was significantly correlated with proximal cortical count (ρ = 0.365, p < 0.001), distal wire use (=0.360, p < 0.001), and LOS (ρ = 0.250, p = 0.010). Proximal wire use correlated with longer LOS (ρ = 0.253, p = 0.009; n = 105). Also understandably, we identified that complications correlated with return to theatre (ρ = 0.241, p = 0.011) (Fig. 2).

Close

Fig. 2 Proximal and distal cortical purchase by unplanned return to theatre within 30 days. (A) Boxplot of the number of proximal cortices engaged according to unplanned return to theatre (<30 days; 0 = no, 1 = yes). (B) Boxplot of the number of distal cortices engaged according to unplanned return to theatre (<30 days; No vs Yes). For each plot, the central line represents the median, boxes show the interquartile range, whiskers indicate 1.5 × IQR, and circles denote outliers.

Export to PPT

3.6 Mortality and survival time

Mortality. Mortality status was available for all patients; 64/111 (57.7%) died during follow-up. Rates increased with age: 6/21 (28.6%) in 65–74 years, 24/46 (52.2%) in 75–84 years, and 34/44 (77.3%) in 85–98 years. Among decedents (n = 64), time to death ranged 2–2550 days with a median of 579 days; subgroup medians were 1359 (65–74), 418 (75–84), and 545 days (85–98). Kaplan–Meier analysis showed no statistically significant difference in survival between the age groups (log-rank χ² = 1.05, p = 0.306) (Fig. 3)

Close

Fig. 3 Kaplan–Meier survival curves by age group after distal femur fracture fixation. Kaplan–Meier estimates of postoperative survival (y-axis, probability of survival) over time from index surgery (x-axis, days) in patients ≥65 years with distal femur fractures treated using a single lateral plate. Curves are stratified by age group (65–74, 75–84, and 85–98 years). Tick marks indicate censored observations. Shaded bands (if displayed) represent 95% confidence intervals.

Export to PPT

4.1 Principal findings

In this study, we observed wide variability in time to FWB and prolonged hospital stay, as well as an expected relationship between mortality and age grouping. Constructs were consistently relatively dense, with no statistically significant age-related differences in cortical purchase or wire use, despite clear differences in chronological age across groups. The oldest group achieved earlier FWB than the younger groups, presenting a counterintuitive approach, which likely reflects a surgical impetus to mobilise frailer patients, rather than a biological advantage with age. Conversely, mortality percentages increased with age, in line with prior reports for geriatric distal femur fracture (DFF) populations.^5,6

Our findings support existing literature that earlier weight-bearing after stable fixation of DFF can be done safely within the elderly and frail populations.^9,25,26 However, surgical approaches to management remains variable, with many patients still prescribed to protective weight-bearing despite frailty-care principles favouring early mobilization.^28,35,36 The paradoxical pattern whereby oldest-old patients had earlier recorded FWB than younger-old groups is unlikely to reflect superior biological healing in patients ≥85 years. Rather, it likely represents a combination of clinician behaviour, pathway effects and documentation artefact. Clinicians may be more willing to accept mechanical risk in the frailest patients, prioritising early mobilisation to mitigate the well-recognised hazards of immobility in older adults, including delirium, chest infection, pressure injury and loss of independence,^10,35,36 whereas younger-old patients may be perceived as better able to tolerate a period of protected loading and, in some cases, may have had more complex fracture patterns prompting conservative prescriptions despite broadly similar constructs. Frailer patients are also more frequently managed on orthogeriatric and rehabilitation pathways, informed by hip-fracture evidence, where early sitting, standing and walking are actively promoted,^35,36 while geriatric DFFs are less consistently embedded in formal hip-fracture–specific frameworks and their weight-bearing protocols remain more heterogeneous^25–28. Finally, our mobilisation measure captured the first documented date of FWB/WBAT, which is sensitive to recording practices: ward-based loading in younger-old patients may not be charted until routine follow-up (for example, around 6 months), whereas oldest-old patients undergoing in-patient therapy and discharge planning may have earlier entries. Together, these behavioural, pathway-related and documentation factors probably explain much of the apparent age effect in FWB timing and underscore that our mobilisation data are hypothesis-generating rather than definitive evidence of true differences in loading patterns across age strata ^{9,10,25–28,35,36}. The overall distribution of FWB times, with nearly half of documented patients still not fully weight-bearing by 30 days and many recorded at 180 days, underscores how far routine distal femur practice still sits from hip-fracture-style early mobilisation targets.

Within our cohort the proportion of deaths overall and then within the age based subgroups is reflective of prior DFF literature.^5,6 Early mobilization is a modifiable post-operative factor linked to reduced short-term complications and mortality after geriatric hip fracture; while evidence in DFF is less mature, converging observational and randomized data support more permissive loading after stable constructs.^9,27,28 Our data add an age-stratified perspective from a DGH setting, suggesting that even within a relatively homogeneous SLP practice, mobilisation decisions are influenced by clinicians’ perceptions of frailty and competing risks, and that the oldest-old may more often be prioritised for earlier ambulation when teams judge the consequences of immobility to outweigh the perceived mechanical risks.

We did not detect strong age-group differences in time-to-theatre or LOS. The literature tying “surgery within 48 h” to outcome is well established for hip fracture but less definitive for DFF; nevertheless, pathway delays and medical optimization issues likely remain relevant. Standardized co-management and mobilization bundles tailored to distal femur injury merit evaluation. In our series, 60% of patients were operated within 48 h, which is broadly aligned with hip-fracture standards, but the prolonged LOS and delayed mobilisation suggest that downstream rehabilitation and weight-bearing protocols may be a more important lever for improving functional trajectories than further marginal gains in time to theatre alone.^10,35,36

This study has several limitations that should be kept in mind when interpreting the findings. First, it reflects practice in a single UK district general hospital and may not be representative of other centres with different case-mix, implants, or rehabilitation resources. Second, the design was retrospective and relied on routine clinical records, so we can only describe associations rather than prove cause and effect. Third, we reduced construct complexity down to simple counts of engaged cortices and adjunct wires; we did not systematically measure plate length, working length, screw position or fracture gap, and we did not formally re-check extreme values on the original images. Fourth, time to full weight-bearing was taken from the first documented FWB/WBAT entry and was missing in over a quarter of patients, so it is strongly influenced by how and when clinicians chose to write this down rather than the exact day patients started loading the limb. Fifth, the numbers of complications (n = 8) and reoperations (n = 6) were small and we did not adjust for comorbidity, fracture pattern or periprosthetic versus native status, so the observed links between denser constructs and adverse events should be viewed as exploratory rather than definitive. Finally, we did not collect patient-reported outcomes, so we cannot say how different fixation strategies translated into pain, mobility, or independence from the patient's perspective.

For older patients with DFF treated using a single lateral plate, the focus should be on how the construct works rather than simply how much metal is used. In practical terms, this means favouring longer bridge spans and selective screw placement instead of filling every hole, and keeping nail–plate or dual-plate constructs for clearly unstable or very osteoporotic patterns. When fixation is felt to be stable, teams should agree clear early mobilisation goals and record what patients actually achieve. Future studies should describe plate geometry in more detail, track both prescribed and achieved loading (including within the first 48 h and first 30 days), use consistent time frames for reoperation, and include patient-centred outcomes. Ideally, pragmatic trials would compare different construct and weight-bearing strategies by fracture pattern and bone quality to test whether a mechanics-first approach really improves outcomes.

Guardian/patient's consent

Not applicable.