- Email: [email protected]
Delay to surgery does not affect survival following osteoporotic femoral fractures
G Model JINJ 6799 No. of Pages 6
Injury, Int. J. Care Injured xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Injury journal homepage: www.elsevier.com/locate/injury
Delay to surgery does not affect survival following osteoporotic femoral fractures Kit Brogana,* , Harold Akehurstb , Elizabeth Bondc , Chris Geea , William Poolea , Nirav N. Shahb , Steve McChesneyc , Stephen Nicola a b c
Department of Orthopaedics, Royal Sussex County Hospital, Eastern Rd, Brighton BN2 5BE, UK Department of Orthopaedics, Worthing hospital, Lyndhurst Rd, Worthing BN11 2DH, UK Department of Orthopaedics, Waikato Hospital, Selwyn Street and Pembroke Street, Hamilton 3204, New Zealand
A R T I C L E I N F O
Keywords: Osteoporotic Femur Non-NOF Fragility Distal femoral Periprosthetic Mortality
A B S T R A C T
Aims: Fragility femoral fractures occur in a similar group of patients to hip fractures but they are not routinely managed along standardised guidelines. This study looked specifically at whether delay to surgery has an impact on mortality and morbidity. Patients and methods: An international, multi-centre retrospective review was carried including all patients over 60 years with fragility femoral fractures, including most periprosthetic fractures, between December 2008–2014. Results: 243 patients met the inclusion criteria with mean follow-up 25 months. 197 (81%) were female with mean age 81 years. Median time to surgery was 2 days; 39% were operated on <24 h, 23% 24–48 h, and 37% at >48 h. 3- and 12-month mortality were 14% (95% CI: 9–18%) and 26% (20–31%) respectively. On Kaplan-Meier plotting, relationships were apparent between survival and sex, and ASA grade, but not delay to surgery or fracture type. Conclusion: Fragility femoral fractures have equivalent mortality to hip fractures but we found no link between delay to surgery and mortality. We believe it is safe to delay surgery, within reason, whilst their acute and chronic medical problems are optimised. We believe this information will help develop guidelines similar to hip fracture pathways. Crown Copyright ã 2016 Published by Elsevier Ltd. All rights reserved.
Introduction Fragility fractures are a major health problem with global numbers of hip fractures being reported as 1.3 million in 1990 and projected to reach 21 million by 2050 [1]. In general, femoral fractures occur in a bimodal distribution—young patients with high-energy fractures or elderly patients with low-energy injuries [2–4] with 85% of these fractures occurring in patients over 50 years of age. Fragility femoral fractures are approximately ten times less common than hip fractures [3] but their incidence remains likely to increase at a similar rate to NOF fractures due to the ageing population. This will present a significant burden on health care provision in the future. Given that hip fractures are the commonest cause of injury related death, it is unsurprising that so many guidelines for the management of these fractures exist [5–7]. These guidelines, and the Best Practice Tariff, have created a standardised approach to the
management of these fractures which has reduced 30-day mortality to 8.2% [7]. One of the key recommendations is that hip fracture patients should be operated on within 36 h from diagnosis to improve the clinical outcome and minimise mortality [5,8]. Non-NOF femoral fractures femoral fractures in the elderly are also recognized as a difficult problem [2,9,10] that have a high twelve-month mortality [9,11,12]. Like hip-fracture patients they often have significant medical comorbidities that do influence the ultimate success and functional outcome achieved [13] but they are currently not managed in this standardised way according to national guidelines. There has been a call for further research into these fractures in order to create guidelines with the aim of improving outcomes [14]. The aim of this study is to determine the mortality and morbidity of non-NOF femoral fractures in patients over 60 years and to assess whether delay to surgery has an impact on this. This study was classified as a clinical audit as the aim was to look at outcomes of routine care and therefore was not subject to research
* Corresponding author. E-mail address: [email protected] (K. Brogan). http://dx.doi.org/10.1016/j.injury.2016.07.003 0020-1383/Crown Copyright ã 2016 Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: K. Brogan, et al., Delay to surgery does not affect survival following osteoporotic femoral fractures, Injury (2016), http://dx.doi.org/10.1016/j.injury.2016.07.003
G Model JINJ 6799 No. of Pages 6
2
K. Brogan et al. / Injury, Int. J. Care Injured xxx (2016) xxx–xxx
ethics committee approval. However, local audit committee approval was provided by all trusts.
Table 1 Demographics & descriptive statistics. n (%)/mean (SD)
Methodology This was an international multi-centre retrospective cohort study conducted between December 2008 and December 2014. Waikato Hospital, New Zealand and The Royal Sussex County Hospital (RSCH), UK are busy level 1 trauma centres, and Worthing Hospital, UK is a district general hospital. Inclusion criteria were: age over 60-years and admission with a low-energy femoral fracture, excluding NOF and subtrochanteric fractures. Exclusion criteria were pathological fracture, highenergy mechanism and periprosthetic fracture with loose components requiring revision arthroplasty. The rationale for this is that these patients require subspecialty surgeons to perform their operations and in most centres the personnel is not available on a daily basis. However, periprosthetic fractures with stable implants, suitable for open reduction and internal fixation, are often managed by general orthopaedic surgeons and therefore were included in this group. There is a growing burden of periprosthetic injuries and we felt it would be useful to include this group in the analysis [15]. Plain anteroposterior and lateral radiographs were assessed by the lead author and classified according to either the AO-OTA system [16] or the Vancouver classification, for periprosthetic fractures [17]. Chart review was also conducted and information collected on a standardised spreadsheet. Medical and surgical complications were also recorded. Statistical analysis was performed in R version 3.2.2 (R Core Team, 2015) (R Core Team, 2015). Descriptive analyses were performed using quantiles, arithmetic means and medians for continuous data, and absolute and proportional frequencies for categorical data. Mortality at 3 and 12 months was calculated with 95% confidence intervals. Potential contributing factors to mortality were evaluated firstly with Kaplan-Meier curves and the log-rank test for categorical independent variables. Multivariate analysis using Cox Proportional Hazards regression was performed to identify significant contributions simultaneously, with model optimisation using a backwards selection on AIC procedure. Complication rates were described using absolute frequencies and incidence. Relationships between rates of specific complications and potentially-influencing factors (such as age, ASA, fracture type, delay to surgery and surgical technique) were analysed using Pearson’s Chi-squared tests (with continuity correction where required), student’s t-tests or Wilcoxan rank-sum tests according to determination of normality using the Shapiro-Wilk test. A 0.05 level of significance was used throughout.
Hospital
Age (years) Female sex ASA
Time to surgery (hours)
Worthing Waikato Royal Sussex County
90 (37%) 52 (21%) 101 (42%)
1 2 3 4 5
82.2 (9.1) 196 (81%) 3 (1.2%) 56 (23%) 130 (53.5%) 32 (13.2%) 1 (0.4%)
<24 24–48 >48
Length of stay (days)
95 (39%) 57 (23%) 91 (37%) 25.1 (25)
Mortality & survival analysis Mean follow-up time was 25 months during which 102 deaths were recorded. Overall 3- and 12-month mortality were 14% (95% CI: 9–18%) and 26% (20–31%) respectively [Fig. 1]. On Kaplan-Meier plotting, relationships were apparent between survival and sex, hospital and ASA grade, but not delay to surgery or fracture site [Fig. 2]. The relationship between survival and hospital and ASA grade achieved statistical significance on log-rank testing (p = 0.017 and p < 0.001 respectively). The optimised Cox Proportional Hazards regression model retained sex, age, ASA grade and length of stay, the latter of which did not achieve statistical significance. Odds ratios for the retained variables are shown in Table 2. Complications Surgical complications data were retrieved from all patients, while complete medical complications data were retrieved from 142 patients attending the three hospitals. Medical and surgical complications occurred in 39% (95% CI: 31–47%) and 39.9% (31.9– 48.0%) of patients respectively [Table 3]/[Fig. 1]. Although Kaplan-Meier plotting was suggestive of relationships between survival and LRTI, UTI and delirium [Fig. 4], only LRTI
100%
Results
75%
Data from 243 patients admitted between December 2008 and October 2014 were included in the study. 13 patients had incomplete radiographs and were excluded from analysis, except for survival. 197 (81%) were female with a mean age of 81 years (range 60–103 years). Median length of stay was 17 days (range 2– 155 days); median time to surgery was 2 days; 39% of patients were operated on within 24 h, 23% between 24 and 48 h, and 37% at >48 h [Table 1]. 40 (17%) patients had diaphyseal (AO 32-A/B/C) fractures; 80 (35%) had distal femoral (AO 33-A/B/C) fractures; 38 (16.5%) had periprosthetic THR fractures; and 72 (31.5%) had periprosthetic TKR fractures [Table 6].
Mortality
Patient demographics
50%
25%
0% 3 months
12 months
Fig. 1. Mortality.
Please cite this article in press as: K. Brogan, et al., Delay to surgery does not affect survival following osteoporotic femoral fractures, Injury (2016), http://dx.doi.org/10.1016/j.injury.2016.07.003
G Model JINJ 6799 No. of Pages 6
K. Brogan et al. / Injury, Int. J. Care Injured xxx (2016) xxx–xxx
100%
Survival
Survival
100% 75% 50% 25%
75%
Sex
50%
F M
25%
0%
0% 0
1
2
3
4
5
0
1
Time (years)
3
4
5
100%
75%
Hospital Waikato RSCH Worthing
50% 25%
Survival
Survival
2
Time (years)
100%
ASA grade
75%
1 2 3 4 5
50% 25%
0%
0% 0
1
2
3
4
5
0
1
Time (years)
2
3
4
5
Time (years) 100%
75%
Delay to surgery (days)
50%
<24h 24−48h >48h
25%
Survival
100%
Survival
3
75%
Fracture site AO 32 AO 33 Periprosthetic hip Periprosthetic knee
50% 25%
0%
0% 0
1
2
3
4
5
0
Time (years)
1
2
3
4
5
Time (years) Fig. 2. Kaplan-Meier plots.
Medical Complications 100 75 50 25 0 None
AF
AKI
C. diff
CHF
Delirium
LRTI
MI
Pressure sore
UTI
VTE
Surgical Complications 100 75 50 25 0 None
Dislocation
Delayed union
Further fracture
Deep infection
Superficial Malunion infection
Non− union
Symp metal
Fig. 3. Complication counts.
achieved statistical significance (p < 0.001) on log-rank testing. No medical complication was significantly associated with time to surgery. Non-union, delayed union and malunion occurred in 11.9%, 1.6% and 1.2% of patients respectively. Union complications were not significantly associated with medical complications, age or ASA grade (Fig. 3). Although no individual surgical complication was significantly associated with time to surgery, a greater time to surgery was
significantly associated with absence of any surgical complication (p < 0.001) [Table 4]. Kaplan-Meier plotting was suggestive of worsened survival in periprosthetic hip fractures and, to a lesser degree, AO 32 fractures [Fig. 5], but this trend was not significant on log-rank testing (p = 0.067). There was a non-significant trend towards more frequent union complications in periprosthetic TKR fractures than AO 32, 33 or periprosthetic THR fractures [Table 5]. Within these categories
Please cite this article in press as: K. Brogan, et al., Delay to surgery does not affect survival following osteoporotic femoral fractures, Injury (2016), http://dx.doi.org/10.1016/j.injury.2016.07.003
G Model JINJ 6799 No. of Pages 6
4
K. Brogan et al. / Injury, Int. J. Care Injured xxx (2016) xxx–xxx
Table 2 Cox Proportional Hazards models.
Mortality
Variable
Odds ratio
p-value
Initial model
Sex: male Age Delay Hospital: RSCH Hospital: UK ASA Length of stay
1.793 1.054 0.990 0.716 0.869 3.043 0.994
0.020 0.000 0.670 0.189 0.731 0.000 0.395
Optimised model
Sex: male Age ASA Length of stay
1.825 1.052 3.247 0.992
0.017 0.000 0.000 0.126
Table 3 Complication rates.
Nil medical MI LRTI VTE Delirium UTI AF CHF AKI Pressure sore C. diff Nil surgical Superficial infection Deep infection Malunion Delayed union Non-union Symptomatic metal Further fracture Dislocation
Absolute
Incidence
95% CI
87 3 27 3 6 16 3 3 3 5 1 97 8 2 3 4 29 7 4 1
35.8% 1.2% 11.1% 1.2% 2.5% 6.6% 1.2% 1.2% 1.2% 2.1% 0.4% 39.9% 3.3% 0.8% 1.2% 1.6% 11.9% 2.9% 1.6% 0.4%
27.9% 0.6% 5.9% 0.6% 0.1% 2.5% 0.6% 0.6% 0.6% 0.3% 0.6% 31.9% 0.4% 0.7% 0.6% 0.4% 6.6% 0.1% 0.4% 0.6%
43.7% 3.1% 16.3% 3.1% 5.0% 10.7% 3.1% 3.1% 3.1% 4.4% 1.5% 48.0% 6.2% 2.3% 3.1% 3.7% 17.3% 5.6% 3.7% 1.5%
there were no fracture subcategories with significantly raised frequency of union complications. There was no statistically significant relationship between operation type and frequency of union complications (IM nail versus distal femoral locking plates p = 0.49). Discussion Elderly patients sustaining osteoporotic femoral fractures are a challenging group of patients and we can expect an increase in the incidence of these fractures in the future. The aim of this study was to look at the morbidity and mortality of patients sustaining fragility femoral fractures and to look at the variables that influence them. In particular, we wanted to see whether there was a link between delay to surgery and mortality. The result of this can then be used to develop guidelines for improved future management. Our study is unique in looking at patients from 3 different centres internationally and is also the largest series to date of these injuries. Waikato hospital in New Zealand receives no financial incentives for treating hip fractures within a certain time frame but still adheres to guidelines where possible [6]. As there is no financial incentive in preferentially treating hip fracture patients all femoral fracture patients are treated in a similar fashion. In contrast the two UK hospitals adhere to NICE guidelines and consequently have the Best Practice Tariff applied, which creates financial incentives to prioritise hip fracture patients.
Our study shows that the vast majority of patients with these injuries are successfully discharged from hospital and go on to union. However, this injury represents a life changing event for these patients with similar 3 and 12 month mortality rates to hip fractures [18]. Currently we mostly extrapolate data from hip fracture evidence to clinically guide the timing of surgery in these patients. However, despite extensive investigation in this area there is still much debate surrounding timing of surgery in hip fractures. There are multiple studies that have found delay beyond 48 h in hip fractures increases mortality [19,20] but they have also noted there is a significant effect from pre-existing disease severity. Parker et al. [21] found that after adjusting for these confounding factors patients operated on within 12 h had a lower mortality but they could not support the view that an operation between 12 and 48 h reduced mortality. There have been earlier studies that have also adjusted for these confounding factors and conversely found that surgical delay greater than 48 h does in fact lead to increase in 30day and 1-year mortality [8,22]. Despite this conflicting evidence, national guidelines recommend that NOF fracture fixation takes place in the first 24–36 h [5,23]. In order to achieve this, hip fracture patients have their medical comorbidities optimised rapidly so as not to delay surgery. This is often achieved with the use of a multidisciplinary team (MDT) model and a cohesive approach by all involved, which has been further shown to improve outcomes in hip fractures [24]. Despite many similarities to hip fractures, osteoporotic femoral fractures are often not currently optimised or prioritised for early surgery in this formalised MDT approach. Specifically, in non-NOF femoral fractures there is very limited evidence to guide the timing of surgery. Streubel et al. [25] did find a link between delay to surgery and mortality, although this was again linked to pre-existing morbidity was in a much smaller patient cohort, 92 patients. These medical comorbidities are a common finding in this population with 32% of patients having pre-existing cardiovascular disease, 21% with diabetes and 11% with osteoporosis [26]. In our study a delay to surgery was not associated with increased mortality, an important negative finding. In the light of this finding, the perhaps intuitive association between increased ASA grade and mortality may offer more opportunity to improve outcomes by resolving acute medical problems and optimising chronic conditions before surgery. The most significant factors influencing survival were age, ASA grade and sex. The first of these is unmodifiable but it’s confirmation, especially considered together with the high mortality demonstrated in our study, should inform discussions with the patient and their family. The influence of male sex may require further study, as it is likely to be at least partly attributable to the unequal sex distribution of osteoporosis and, especially, chronic medical conditions; this was beyond the scope of our study. We used an automated selection method to optimise our model based on the quality of information provided by putative independent variables. This process retained length of stay with a non-significant odds ratio of 0.99. While this is not straightforward to interpret, it suggests a trend towards worse survival in patients discharged earlier and warrants further study. Initial Kaplan-Meier plotting was suggestive of a relationship between survival and hospital, which was supported by statistical significance on log-rank testing. However, this effect was small and not statistically significant in the multivariate analysis, suggesting that any difference was accounted for by confounding factors (such as the age, sex and ASA grades of their respective cohorts). A
Please cite this article in press as: K. Brogan, et al., Delay to surgery does not affect survival following osteoporotic femoral fractures, Injury (2016), http://dx.doi.org/10.1016/j.injury.2016.07.003
G Model JINJ 6799 No. of Pages 6
K. Brogan et al. / Injury, Int. J. Care Injured xxx (2016) xxx–xxx
5
Survival
100% 75%
LRTI
50%
No Yes
25% 0% 0
1
2
3
4
5
Time (years)
Survival
100% 75%
UTI
50%
No Yes
25% 0% 0
1
2
3
4
5
Time (years)
Fig. 5. Kaplan-Meier plot for fracture sites. Table 5 Union complications by fracture site.
Survival
100% 75%
Delirium
50%
No Yes
25% 0% 0
1
2
3
4
Fracture
Count
%
p-valuea
AO 32 AO 33 Periprosthetic hip Periprosthetic knee
9 10 7 10
22.0 13.7 18.4 13.7
0.6103
5
Time (years) Fig. 4. Kaplan-Meier plots for medical complications.
Table 4 Surgical complications & time to surgery.
None Superficial infection Deep infection Malunion Delayed union Non-union Symptomatic metal Further fracture Dislocation a
Count
Mean delay (days)
p-valuea
97 8 2 3 4 29 7 4 1
4.06 2.13 3.00 1.67 2.00 3.24 2.43 3.75 1.00
0.00 0.88 0.86 0.59 0.27 0.62 0.98 0.41 0.34
Wilcoxan rank-sum test.
comprehensive comparison of outcomes between hospitals is beyond the scope of this study.
more union complications in patients with periprosthetic TKR fractures. This may be due to biological factors such as scarring and an effect on the metaphyseal blood supply following knee replacement surgery. It may also relate to biomechanical factors with the knee replacement changing the biomechanical load on the bone. This was not a primary research question and more basic science and clinical research would be required to answer this finding. However, it is useful information for the patient and surgeon to be aware of the higher rate of non-union and to be meticulous about preserving the blood supply and producing a biomechanically favourable construct for optimising union. Another finding in studies is that Vancouver B1 fractures have a higher rate of non-union and late failure when treated with locking plates [29], whereas we found no such association. In order to avoid these issues all centres are very careful to ensure that they accurately differentiate between B1 and B2 fractures. If there is a concern, then intra-operative screening is performed and the decision to convert to revision THR is made once the diagnosis of a B2 fracture is made. It is therefore important when managing B1 fractures that a surgeon capable of performing revision hip arthroplasty is present.
Morbidity The goals of managing these patients are to prevent further falls, manage their bone loss and to restore function and mobility. Non-operative management is very challenging, with poor results compared to operative management [27,28]. We believe it should be reserved for only the most obtunded patients. Operative fixation is challenging and is often complicated by the high frequency of periprosthetic fractures around hip or knee replacements—45% in this series. We are unable to explain the finding of fewer surgical complications in patients with longer time to surgery and the clinical relevance of this is unclear. We are reluctant to suggest that this constitutes a rationale for deliberately delaying surgery without a clear need for medical optimisation. We also found a trend towards
Table 6 Fixation method by fracture type. Fixation Classification
DHS IM Nail Distal femoral locking plate Conservative
32-A/B/C 33-A 33-B 33-C Vancouver B Vancouver C Periprosthetic TKR
2 1 0 0 0 0 0
32 9 0 0 1 0 3
6 44 6 11 7 30 69
0 1 0 0 0 0 0
Please cite this article in press as: K. Brogan, et al., Delay to surgery does not affect survival following osteoporotic femoral fractures, Injury (2016), http://dx.doi.org/10.1016/j.injury.2016.07.003
G Model JINJ 6799 No. of Pages 6
6
K. Brogan et al. / Injury, Int. J. Care Injured xxx (2016) xxx–xxx
Weight bearing
References
One of the goals of surgery is to restore function and there is no consensus on the level of post operative weight bearing in these patients. Accepted practice has historically been to protect weight bearing based on small cohort series [30]. Ehlinger et al. [31] reviewed this topic and stated that it is the quality of the surgical technique that is the most important factor in terms of outcome and therefore the optimal weight bearing strategy is still to be decided. This particular patient group often struggles to mobilise non or partial weight bearing and therefore the only real options are to allow patients to weight bear as tolerated or to non-weight bear them for a period of time. The majority of RSCH patients (84/101) were allowed to weight bear as tolerated from day one whereas patients from Worthing and Waikato were managed with protected weight bearing until signs of clinical or radiological union were seen. It is interesting to note therefore that there were no increased failures of metalwork, non-unions or further fractures in the RSCH subgroup of patients. In fact, the RSCH series has the lowest incidence of non-union. It is beyond the scope of this paper to draw absolute conclusions from this result as it was not something the study was designed to investigate. This would be better done with a prospective randomised study. However, we feel that it is probably important that all patients should have a construct that allows for immediate full weight bearing as tolerated and that there should be no detrimental effects associated with this.
[1] Gullberg B, Johnell O, Kanis JA. World-wide projections for hip fracture. Osteoporos Int 1997;7(5):407–13. [2] Ebraheim NA, Martin A, Sochacki KR, Liu J. Nonunion of distal femoral fractures: a systematic review. Orthopaed Surg 2013;5:46–50. [3] Higgins TF. Distal femoral fractures. J Knee Surg 2007;20(1):56–66. [4] Hoffman MF, Jones CB, Sietsema DL, Tornetta P, Koenig SJ. Clinical outcomes of locked plating of distal femoral fractures in a retrospective cohort. J Orthopaedic Surg Res 20138(43). [5] Committee NG, CG124 Hip fracture: the management of hip fracture in adults, 2014, 21st April 2014. [6] Zealand ON, Bone Care, 2020, A systematic approach to hip fracture care and prevention for New Zealand, 2012. [7] A. Johansen, WR, C. Boulton, F. Plant, J. Roberts, A. Williams, National Report 2013—Summary, 2013. [8] Bottle A, Aylin P. Mortality associated with delay in operation after hip fracture: observational study. BMJ 2006;332(7547):947–51. [9] Appleton PMM, Houshian S, Robinson CM. Distal femoral fractures treated by hinged total knee replacement in elderly patients. J Bone Jt Surg (Br) 2006;88B(8):1065–70. [10] Forster MC, Komarsamy B, Davison JN. Distal femoral fractures: a review of fixation methods. Injury Int J Care Injured 2006;37:97–108. [11] Butt MS, Ali MS, Ali MS. Displaced fractures of the distal femur in elderly patients. J Bone Jt Surg (Br) 1995;77-B:110–4. [12] Moore TJ, Watson T, Green SA, Garland DE, Chandler RW. Complications of surgically treated supracondylar fractures. J Trauma 1987;27(4):402–6. [13] Kammerlander C, Riedmüller P, Gosch M, Zegg M, Kammerlander-Knauer U, Schmid R, et al. Functional outcome and mortality in geriatric distal femoral fractures. Injury 2012;43(7):1096–101. [14] Smith JR, Halliday R, Aquilina AL, Morrison RJ, Yip GC, McArthur J, et al. Distal femoral fractures: the need to review the standard of care. Injury 2015;46 (6):1084–8. [15] Shields E, Behrend C, Bair J, Cram P, Kates S. Mortality and financial burden of periprosthetic fractures of the femur. Geriatr Orthop Surg Rehabil 2014;5 (4):147–53. [16] Marsh JL, Slongo TF, Agel J, Broderick JS, Creevey W, DeCoster TA, et al. Fracture and dislocation classification compendium—2007: Orthopaedic Trauma Association classification, database and outcomes committee. J Orthopaedic Trauma 2007;21(10):S1–133. [17] Masri BA, Meek RM, Duncan CP. Periprosthetic fractures evaluation and treatment. Clin Orthop Relat Res 2004;420:80–95. [18] Goldacre MJ, Roberts SE, Yeates D. Mortality after admission to hospital with fractured neck of femur: database study. BMJ 2002;325(7369):868–9. [19] Gdalevich M, Cohen D, Yosef D, Tauber C. Morbidity and mortality after hip fracture: the impact of operative delay. Arch Orthop Trauma Surg 2004;124 (5):334–40. [20] Elliott J, Beringer T, Kee F, Marsh D, Willis C, Stevenson M. Predicting survival after treatment for fracture of the proximal femur and the effect of delays to surgery. J Clin Epidemiol 2003;56(8):788–95. [21] Bretherton CP, Parker MJ. Early surgery for patients with a fracture of the hip decreases 30-day mortality. Bone Jt J 2015;97-B(1):104–8. [22] Novack V, Jotkowitz A, Etzion O, Porath A. Does delay in surgery after hip fracture lead to worse outcomes? A multicenter survey. Int J. Qual Health Care 2007;19(3):170–6. [23] Mak JC, Cameron ID, March LM. National Health and Medical Research Council. Evidence-based guidelines for the management of hip fractures in older persons: an update. Med. J. Aust. 2010;192(1):1–37. [24] Boddaert J, Cohen-Bittan J, Khiami F, Le Manach Y, Raux M, Beinis JY, et al. Postoperative admission to a dedicated geriatric unit decreases mortality in elderly patients with hip fracture. PLoS One 2014;9(1):e83795. [25] Streubel PN, Ricci WM, Wong A, Gardner MJ. Mortality after distal femur fractures in elderly patients. Clin Orthop Relat Res 2011;469(4):1188–96. [26] Nieves JW, Bilezikian JP, Lane JM, Einhorn TA, Wang Y, Steinbuch M, et al. Fragility fractures of the hip and femur: incidence and patient characteristics. Osteoporos Int 2010;21(3):399–408. [27] Higgins TF. Distal femoral fractures. J Knee Surg 2007;20(1):56–66. [28] McLauchlan GJ, Robinson CM, Singer BR, Christie J. Results of an operative policy in the treatment of periprosthetic femoral fracture. J Orthop Trauma 1997;11(3):170–9. [29] Buttaro MA, Farfalli G, Paredes Nunez M, Comba F, Piccaluga F. Locking compression plate fixation of Vancouver type-B1 periprosthetic femoral fractures. J Bone Jt Surg Am 2007;89(9):1964–9. [30] Vallier HA, Hennessey TA, Sontich JK, Patterson BM. Failure of LCP condylar plate fixation in the distal part of the femur. A report of six cases. J Bone Jt Surg Am 2006;88(4):846–53. [31] Ehlinger M, Ducrot G, Adam P, Bonnomet F. Distal femur fractures. Surgical techniques and a review of the literature. Orthop Ttraumatol Surg Res OTSR 2013;99(3):353–60.
Limitations There are several limitations to this study. Firstly, retrospective data is always subject to collection bias. Whilst everyeffort was made to track down all patients to ensure accurate mortality data was collected there is always a risk of patients moving geographically outside of our collection areas. A prospective randomised study looking specifically at different fixation methods and post-operative rehabilitation methods would be better suited to making accurate conclusions about these issues. Conclusion Despite our results showing that delay to surgery does not affect mortality in this patient group, we would still recommend early surgical intervention following medical optimisation. This study does however provide evidence to support the fact that delay to surgery, in itself, does not appear to affect mortality. We believe it is therefore safe to treat both acute and chronic medical comorbidities to optimise patient outcomes. We found no adverse orthopaedic complications with early weight-bearing as a treatment strategy and therefore cautiously support this as a post-operative rehabilitation regime. This patient group has similar characteristics with hip fracture patients with a high overall morbidity and mortality and their treatment should be more consistently implemented. Guidelines should be developed based on this and future research to include joint orthopaedic and geriatric care as well as clear guidelines on implant choice and post-operative rehabilitation. Conflict of interest None. Acknowledgement Dr Jim Price MA, MA MB BChir, FRCP, FRCGP, FAcad, MEd, SFHEA. Principle lecturer in Medical Education, University of Brighton BN1 9PHA.
Please cite this article in press as: K. Brogan, et al., Delay to surgery does not affect survival following osteoporotic femoral fractures, Injury (2016), http://dx.doi.org/10.1016/j.injury.2016.07.003
Injury, Int. J. Care Injured xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Injury journal homepage: www.elsevier.com/locate/injury
Delay to surgery does not affect survival following osteoporotic femoral fractures Kit Brogana,* , Harold Akehurstb , Elizabeth Bondc , Chris Geea , William Poolea , Nirav N. Shahb , Steve McChesneyc , Stephen Nicola a b c
Department of Orthopaedics, Royal Sussex County Hospital, Eastern Rd, Brighton BN2 5BE, UK Department of Orthopaedics, Worthing hospital, Lyndhurst Rd, Worthing BN11 2DH, UK Department of Orthopaedics, Waikato Hospital, Selwyn Street and Pembroke Street, Hamilton 3204, New Zealand
A R T I C L E I N F O
Keywords: Osteoporotic Femur Non-NOF Fragility Distal femoral Periprosthetic Mortality
A B S T R A C T
Aims: Fragility femoral fractures occur in a similar group of patients to hip fractures but they are not routinely managed along standardised guidelines. This study looked specifically at whether delay to surgery has an impact on mortality and morbidity. Patients and methods: An international, multi-centre retrospective review was carried including all patients over 60 years with fragility femoral fractures, including most periprosthetic fractures, between December 2008–2014. Results: 243 patients met the inclusion criteria with mean follow-up 25 months. 197 (81%) were female with mean age 81 years. Median time to surgery was 2 days; 39% were operated on <24 h, 23% 24–48 h, and 37% at >48 h. 3- and 12-month mortality were 14% (95% CI: 9–18%) and 26% (20–31%) respectively. On Kaplan-Meier plotting, relationships were apparent between survival and sex, and ASA grade, but not delay to surgery or fracture type. Conclusion: Fragility femoral fractures have equivalent mortality to hip fractures but we found no link between delay to surgery and mortality. We believe it is safe to delay surgery, within reason, whilst their acute and chronic medical problems are optimised. We believe this information will help develop guidelines similar to hip fracture pathways. Crown Copyright ã 2016 Published by Elsevier Ltd. All rights reserved.
Introduction Fragility fractures are a major health problem with global numbers of hip fractures being reported as 1.3 million in 1990 and projected to reach 21 million by 2050 [1]. In general, femoral fractures occur in a bimodal distribution—young patients with high-energy fractures or elderly patients with low-energy injuries [2–4] with 85% of these fractures occurring in patients over 50 years of age. Fragility femoral fractures are approximately ten times less common than hip fractures [3] but their incidence remains likely to increase at a similar rate to NOF fractures due to the ageing population. This will present a significant burden on health care provision in the future. Given that hip fractures are the commonest cause of injury related death, it is unsurprising that so many guidelines for the management of these fractures exist [5–7]. These guidelines, and the Best Practice Tariff, have created a standardised approach to the
management of these fractures which has reduced 30-day mortality to 8.2% [7]. One of the key recommendations is that hip fracture patients should be operated on within 36 h from diagnosis to improve the clinical outcome and minimise mortality [5,8]. Non-NOF femoral fractures femoral fractures in the elderly are also recognized as a difficult problem [2,9,10] that have a high twelve-month mortality [9,11,12]. Like hip-fracture patients they often have significant medical comorbidities that do influence the ultimate success and functional outcome achieved [13] but they are currently not managed in this standardised way according to national guidelines. There has been a call for further research into these fractures in order to create guidelines with the aim of improving outcomes [14]. The aim of this study is to determine the mortality and morbidity of non-NOF femoral fractures in patients over 60 years and to assess whether delay to surgery has an impact on this. This study was classified as a clinical audit as the aim was to look at outcomes of routine care and therefore was not subject to research
* Corresponding author. E-mail address: [email protected] (K. Brogan). http://dx.doi.org/10.1016/j.injury.2016.07.003 0020-1383/Crown Copyright ã 2016 Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: K. Brogan, et al., Delay to surgery does not affect survival following osteoporotic femoral fractures, Injury (2016), http://dx.doi.org/10.1016/j.injury.2016.07.003
G Model JINJ 6799 No. of Pages 6
2
K. Brogan et al. / Injury, Int. J. Care Injured xxx (2016) xxx–xxx
ethics committee approval. However, local audit committee approval was provided by all trusts.
Table 1 Demographics & descriptive statistics. n (%)/mean (SD)
Methodology This was an international multi-centre retrospective cohort study conducted between December 2008 and December 2014. Waikato Hospital, New Zealand and The Royal Sussex County Hospital (RSCH), UK are busy level 1 trauma centres, and Worthing Hospital, UK is a district general hospital. Inclusion criteria were: age over 60-years and admission with a low-energy femoral fracture, excluding NOF and subtrochanteric fractures. Exclusion criteria were pathological fracture, highenergy mechanism and periprosthetic fracture with loose components requiring revision arthroplasty. The rationale for this is that these patients require subspecialty surgeons to perform their operations and in most centres the personnel is not available on a daily basis. However, periprosthetic fractures with stable implants, suitable for open reduction and internal fixation, are often managed by general orthopaedic surgeons and therefore were included in this group. There is a growing burden of periprosthetic injuries and we felt it would be useful to include this group in the analysis [15]. Plain anteroposterior and lateral radiographs were assessed by the lead author and classified according to either the AO-OTA system [16] or the Vancouver classification, for periprosthetic fractures [17]. Chart review was also conducted and information collected on a standardised spreadsheet. Medical and surgical complications were also recorded. Statistical analysis was performed in R version 3.2.2 (R Core Team, 2015) (R Core Team, 2015). Descriptive analyses were performed using quantiles, arithmetic means and medians for continuous data, and absolute and proportional frequencies for categorical data. Mortality at 3 and 12 months was calculated with 95% confidence intervals. Potential contributing factors to mortality were evaluated firstly with Kaplan-Meier curves and the log-rank test for categorical independent variables. Multivariate analysis using Cox Proportional Hazards regression was performed to identify significant contributions simultaneously, with model optimisation using a backwards selection on AIC procedure. Complication rates were described using absolute frequencies and incidence. Relationships between rates of specific complications and potentially-influencing factors (such as age, ASA, fracture type, delay to surgery and surgical technique) were analysed using Pearson’s Chi-squared tests (with continuity correction where required), student’s t-tests or Wilcoxan rank-sum tests according to determination of normality using the Shapiro-Wilk test. A 0.05 level of significance was used throughout.
Hospital
Age (years) Female sex ASA
Time to surgery (hours)
Worthing Waikato Royal Sussex County
90 (37%) 52 (21%) 101 (42%)
1 2 3 4 5
82.2 (9.1) 196 (81%) 3 (1.2%) 56 (23%) 130 (53.5%) 32 (13.2%) 1 (0.4%)
<24 24–48 >48
Length of stay (days)
95 (39%) 57 (23%) 91 (37%) 25.1 (25)
Mortality & survival analysis Mean follow-up time was 25 months during which 102 deaths were recorded. Overall 3- and 12-month mortality were 14% (95% CI: 9–18%) and 26% (20–31%) respectively [Fig. 1]. On Kaplan-Meier plotting, relationships were apparent between survival and sex, hospital and ASA grade, but not delay to surgery or fracture site [Fig. 2]. The relationship between survival and hospital and ASA grade achieved statistical significance on log-rank testing (p = 0.017 and p < 0.001 respectively). The optimised Cox Proportional Hazards regression model retained sex, age, ASA grade and length of stay, the latter of which did not achieve statistical significance. Odds ratios for the retained variables are shown in Table 2. Complications Surgical complications data were retrieved from all patients, while complete medical complications data were retrieved from 142 patients attending the three hospitals. Medical and surgical complications occurred in 39% (95% CI: 31–47%) and 39.9% (31.9– 48.0%) of patients respectively [Table 3]/[Fig. 1]. Although Kaplan-Meier plotting was suggestive of relationships between survival and LRTI, UTI and delirium [Fig. 4], only LRTI
100%
Results
75%
Data from 243 patients admitted between December 2008 and October 2014 were included in the study. 13 patients had incomplete radiographs and were excluded from analysis, except for survival. 197 (81%) were female with a mean age of 81 years (range 60–103 years). Median length of stay was 17 days (range 2– 155 days); median time to surgery was 2 days; 39% of patients were operated on within 24 h, 23% between 24 and 48 h, and 37% at >48 h [Table 1]. 40 (17%) patients had diaphyseal (AO 32-A/B/C) fractures; 80 (35%) had distal femoral (AO 33-A/B/C) fractures; 38 (16.5%) had periprosthetic THR fractures; and 72 (31.5%) had periprosthetic TKR fractures [Table 6].
Mortality
Patient demographics
50%
25%
0% 3 months
12 months
Fig. 1. Mortality.
Please cite this article in press as: K. Brogan, et al., Delay to surgery does not affect survival following osteoporotic femoral fractures, Injury (2016), http://dx.doi.org/10.1016/j.injury.2016.07.003
G Model JINJ 6799 No. of Pages 6
K. Brogan et al. / Injury, Int. J. Care Injured xxx (2016) xxx–xxx
100%
Survival
Survival
100% 75% 50% 25%
75%
Sex
50%
F M
25%
0%
0% 0
1
2
3
4
5
0
1
Time (years)
3
4
5
100%
75%
Hospital Waikato RSCH Worthing
50% 25%
Survival
Survival
2
Time (years)
100%
ASA grade
75%
1 2 3 4 5
50% 25%
0%
0% 0
1
2
3
4
5
0
1
Time (years)
2
3
4
5
Time (years) 100%
75%
Delay to surgery (days)
50%
<24h 24−48h >48h
25%
Survival
100%
Survival
3
75%
Fracture site AO 32 AO 33 Periprosthetic hip Periprosthetic knee
50% 25%
0%
0% 0
1
2
3
4
5
0
Time (years)
1
2
3
4
5
Time (years) Fig. 2. Kaplan-Meier plots.
Medical Complications 100 75 50 25 0 None
AF
AKI
C. diff
CHF
Delirium
LRTI
MI
Pressure sore
UTI
VTE
Surgical Complications 100 75 50 25 0 None
Dislocation
Delayed union
Further fracture
Deep infection
Superficial Malunion infection
Non− union
Symp metal
Fig. 3. Complication counts.
achieved statistical significance (p < 0.001) on log-rank testing. No medical complication was significantly associated with time to surgery. Non-union, delayed union and malunion occurred in 11.9%, 1.6% and 1.2% of patients respectively. Union complications were not significantly associated with medical complications, age or ASA grade (Fig. 3). Although no individual surgical complication was significantly associated with time to surgery, a greater time to surgery was
significantly associated with absence of any surgical complication (p < 0.001) [Table 4]. Kaplan-Meier plotting was suggestive of worsened survival in periprosthetic hip fractures and, to a lesser degree, AO 32 fractures [Fig. 5], but this trend was not significant on log-rank testing (p = 0.067). There was a non-significant trend towards more frequent union complications in periprosthetic TKR fractures than AO 32, 33 or periprosthetic THR fractures [Table 5]. Within these categories
Please cite this article in press as: K. Brogan, et al., Delay to surgery does not affect survival following osteoporotic femoral fractures, Injury (2016), http://dx.doi.org/10.1016/j.injury.2016.07.003
G Model JINJ 6799 No. of Pages 6
4
K. Brogan et al. / Injury, Int. J. Care Injured xxx (2016) xxx–xxx
Table 2 Cox Proportional Hazards models.
Mortality
Variable
Odds ratio
p-value
Initial model
Sex: male Age Delay Hospital: RSCH Hospital: UK ASA Length of stay
1.793 1.054 0.990 0.716 0.869 3.043 0.994
0.020 0.000 0.670 0.189 0.731 0.000 0.395
Optimised model
Sex: male Age ASA Length of stay
1.825 1.052 3.247 0.992
0.017 0.000 0.000 0.126
Table 3 Complication rates.
Nil medical MI LRTI VTE Delirium UTI AF CHF AKI Pressure sore C. diff Nil surgical Superficial infection Deep infection Malunion Delayed union Non-union Symptomatic metal Further fracture Dislocation
Absolute
Incidence
95% CI
87 3 27 3 6 16 3 3 3 5 1 97 8 2 3 4 29 7 4 1
35.8% 1.2% 11.1% 1.2% 2.5% 6.6% 1.2% 1.2% 1.2% 2.1% 0.4% 39.9% 3.3% 0.8% 1.2% 1.6% 11.9% 2.9% 1.6% 0.4%
27.9% 0.6% 5.9% 0.6% 0.1% 2.5% 0.6% 0.6% 0.6% 0.3% 0.6% 31.9% 0.4% 0.7% 0.6% 0.4% 6.6% 0.1% 0.4% 0.6%
43.7% 3.1% 16.3% 3.1% 5.0% 10.7% 3.1% 3.1% 3.1% 4.4% 1.5% 48.0% 6.2% 2.3% 3.1% 3.7% 17.3% 5.6% 3.7% 1.5%
there were no fracture subcategories with significantly raised frequency of union complications. There was no statistically significant relationship between operation type and frequency of union complications (IM nail versus distal femoral locking plates p = 0.49). Discussion Elderly patients sustaining osteoporotic femoral fractures are a challenging group of patients and we can expect an increase in the incidence of these fractures in the future. The aim of this study was to look at the morbidity and mortality of patients sustaining fragility femoral fractures and to look at the variables that influence them. In particular, we wanted to see whether there was a link between delay to surgery and mortality. The result of this can then be used to develop guidelines for improved future management. Our study is unique in looking at patients from 3 different centres internationally and is also the largest series to date of these injuries. Waikato hospital in New Zealand receives no financial incentives for treating hip fractures within a certain time frame but still adheres to guidelines where possible [6]. As there is no financial incentive in preferentially treating hip fracture patients all femoral fracture patients are treated in a similar fashion. In contrast the two UK hospitals adhere to NICE guidelines and consequently have the Best Practice Tariff applied, which creates financial incentives to prioritise hip fracture patients.
Our study shows that the vast majority of patients with these injuries are successfully discharged from hospital and go on to union. However, this injury represents a life changing event for these patients with similar 3 and 12 month mortality rates to hip fractures [18]. Currently we mostly extrapolate data from hip fracture evidence to clinically guide the timing of surgery in these patients. However, despite extensive investigation in this area there is still much debate surrounding timing of surgery in hip fractures. There are multiple studies that have found delay beyond 48 h in hip fractures increases mortality [19,20] but they have also noted there is a significant effect from pre-existing disease severity. Parker et al. [21] found that after adjusting for these confounding factors patients operated on within 12 h had a lower mortality but they could not support the view that an operation between 12 and 48 h reduced mortality. There have been earlier studies that have also adjusted for these confounding factors and conversely found that surgical delay greater than 48 h does in fact lead to increase in 30day and 1-year mortality [8,22]. Despite this conflicting evidence, national guidelines recommend that NOF fracture fixation takes place in the first 24–36 h [5,23]. In order to achieve this, hip fracture patients have their medical comorbidities optimised rapidly so as not to delay surgery. This is often achieved with the use of a multidisciplinary team (MDT) model and a cohesive approach by all involved, which has been further shown to improve outcomes in hip fractures [24]. Despite many similarities to hip fractures, osteoporotic femoral fractures are often not currently optimised or prioritised for early surgery in this formalised MDT approach. Specifically, in non-NOF femoral fractures there is very limited evidence to guide the timing of surgery. Streubel et al. [25] did find a link between delay to surgery and mortality, although this was again linked to pre-existing morbidity was in a much smaller patient cohort, 92 patients. These medical comorbidities are a common finding in this population with 32% of patients having pre-existing cardiovascular disease, 21% with diabetes and 11% with osteoporosis [26]. In our study a delay to surgery was not associated with increased mortality, an important negative finding. In the light of this finding, the perhaps intuitive association between increased ASA grade and mortality may offer more opportunity to improve outcomes by resolving acute medical problems and optimising chronic conditions before surgery. The most significant factors influencing survival were age, ASA grade and sex. The first of these is unmodifiable but it’s confirmation, especially considered together with the high mortality demonstrated in our study, should inform discussions with the patient and their family. The influence of male sex may require further study, as it is likely to be at least partly attributable to the unequal sex distribution of osteoporosis and, especially, chronic medical conditions; this was beyond the scope of our study. We used an automated selection method to optimise our model based on the quality of information provided by putative independent variables. This process retained length of stay with a non-significant odds ratio of 0.99. While this is not straightforward to interpret, it suggests a trend towards worse survival in patients discharged earlier and warrants further study. Initial Kaplan-Meier plotting was suggestive of a relationship between survival and hospital, which was supported by statistical significance on log-rank testing. However, this effect was small and not statistically significant in the multivariate analysis, suggesting that any difference was accounted for by confounding factors (such as the age, sex and ASA grades of their respective cohorts). A
Please cite this article in press as: K. Brogan, et al., Delay to surgery does not affect survival following osteoporotic femoral fractures, Injury (2016), http://dx.doi.org/10.1016/j.injury.2016.07.003
G Model JINJ 6799 No. of Pages 6
K. Brogan et al. / Injury, Int. J. Care Injured xxx (2016) xxx–xxx
5
Survival
100% 75%
LRTI
50%
No Yes
25% 0% 0
1
2
3
4
5
Time (years)
Survival
100% 75%
UTI
50%
No Yes
25% 0% 0
1
2
3
4
5
Time (years)
Fig. 5. Kaplan-Meier plot for fracture sites. Table 5 Union complications by fracture site.
Survival
100% 75%
Delirium
50%
No Yes
25% 0% 0
1
2
3
4
Fracture
Count
%
p-valuea
AO 32 AO 33 Periprosthetic hip Periprosthetic knee
9 10 7 10
22.0 13.7 18.4 13.7
0.6103
5
Time (years) Fig. 4. Kaplan-Meier plots for medical complications.
Table 4 Surgical complications & time to surgery.
None Superficial infection Deep infection Malunion Delayed union Non-union Symptomatic metal Further fracture Dislocation a
Count
Mean delay (days)
p-valuea
97 8 2 3 4 29 7 4 1
4.06 2.13 3.00 1.67 2.00 3.24 2.43 3.75 1.00
0.00 0.88 0.86 0.59 0.27 0.62 0.98 0.41 0.34
Wilcoxan rank-sum test.
comprehensive comparison of outcomes between hospitals is beyond the scope of this study.
more union complications in patients with periprosthetic TKR fractures. This may be due to biological factors such as scarring and an effect on the metaphyseal blood supply following knee replacement surgery. It may also relate to biomechanical factors with the knee replacement changing the biomechanical load on the bone. This was not a primary research question and more basic science and clinical research would be required to answer this finding. However, it is useful information for the patient and surgeon to be aware of the higher rate of non-union and to be meticulous about preserving the blood supply and producing a biomechanically favourable construct for optimising union. Another finding in studies is that Vancouver B1 fractures have a higher rate of non-union and late failure when treated with locking plates [29], whereas we found no such association. In order to avoid these issues all centres are very careful to ensure that they accurately differentiate between B1 and B2 fractures. If there is a concern, then intra-operative screening is performed and the decision to convert to revision THR is made once the diagnosis of a B2 fracture is made. It is therefore important when managing B1 fractures that a surgeon capable of performing revision hip arthroplasty is present.
Morbidity The goals of managing these patients are to prevent further falls, manage their bone loss and to restore function and mobility. Non-operative management is very challenging, with poor results compared to operative management [27,28]. We believe it should be reserved for only the most obtunded patients. Operative fixation is challenging and is often complicated by the high frequency of periprosthetic fractures around hip or knee replacements—45% in this series. We are unable to explain the finding of fewer surgical complications in patients with longer time to surgery and the clinical relevance of this is unclear. We are reluctant to suggest that this constitutes a rationale for deliberately delaying surgery without a clear need for medical optimisation. We also found a trend towards
Table 6 Fixation method by fracture type. Fixation Classification
DHS IM Nail Distal femoral locking plate Conservative
32-A/B/C 33-A 33-B 33-C Vancouver B Vancouver C Periprosthetic TKR
2 1 0 0 0 0 0
32 9 0 0 1 0 3
6 44 6 11 7 30 69
0 1 0 0 0 0 0
Please cite this article in press as: K. Brogan, et al., Delay to surgery does not affect survival following osteoporotic femoral fractures, Injury (2016), http://dx.doi.org/10.1016/j.injury.2016.07.003
G Model JINJ 6799 No. of Pages 6
6
K. Brogan et al. / Injury, Int. J. Care Injured xxx (2016) xxx–xxx
Weight bearing
References
One of the goals of surgery is to restore function and there is no consensus on the level of post operative weight bearing in these patients. Accepted practice has historically been to protect weight bearing based on small cohort series [30]. Ehlinger et al. [31] reviewed this topic and stated that it is the quality of the surgical technique that is the most important factor in terms of outcome and therefore the optimal weight bearing strategy is still to be decided. This particular patient group often struggles to mobilise non or partial weight bearing and therefore the only real options are to allow patients to weight bear as tolerated or to non-weight bear them for a period of time. The majority of RSCH patients (84/101) were allowed to weight bear as tolerated from day one whereas patients from Worthing and Waikato were managed with protected weight bearing until signs of clinical or radiological union were seen. It is interesting to note therefore that there were no increased failures of metalwork, non-unions or further fractures in the RSCH subgroup of patients. In fact, the RSCH series has the lowest incidence of non-union. It is beyond the scope of this paper to draw absolute conclusions from this result as it was not something the study was designed to investigate. This would be better done with a prospective randomised study. However, we feel that it is probably important that all patients should have a construct that allows for immediate full weight bearing as tolerated and that there should be no detrimental effects associated with this.
[1] Gullberg B, Johnell O, Kanis JA. World-wide projections for hip fracture. Osteoporos Int 1997;7(5):407–13. [2] Ebraheim NA, Martin A, Sochacki KR, Liu J. Nonunion of distal femoral fractures: a systematic review. Orthopaed Surg 2013;5:46–50. [3] Higgins TF. Distal femoral fractures. J Knee Surg 2007;20(1):56–66. [4] Hoffman MF, Jones CB, Sietsema DL, Tornetta P, Koenig SJ. Clinical outcomes of locked plating of distal femoral fractures in a retrospective cohort. J Orthopaedic Surg Res 20138(43). [5] Committee NG, CG124 Hip fracture: the management of hip fracture in adults, 2014, 21st April 2014. [6] Zealand ON, Bone Care, 2020, A systematic approach to hip fracture care and prevention for New Zealand, 2012. [7] A. Johansen, WR, C. Boulton, F. Plant, J. Roberts, A. Williams, National Report 2013—Summary, 2013. [8] Bottle A, Aylin P. Mortality associated with delay in operation after hip fracture: observational study. BMJ 2006;332(7547):947–51. [9] Appleton PMM, Houshian S, Robinson CM. Distal femoral fractures treated by hinged total knee replacement in elderly patients. J Bone Jt Surg (Br) 2006;88B(8):1065–70. [10] Forster MC, Komarsamy B, Davison JN. Distal femoral fractures: a review of fixation methods. Injury Int J Care Injured 2006;37:97–108. [11] Butt MS, Ali MS, Ali MS. Displaced fractures of the distal femur in elderly patients. J Bone Jt Surg (Br) 1995;77-B:110–4. [12] Moore TJ, Watson T, Green SA, Garland DE, Chandler RW. Complications of surgically treated supracondylar fractures. J Trauma 1987;27(4):402–6. [13] Kammerlander C, Riedmüller P, Gosch M, Zegg M, Kammerlander-Knauer U, Schmid R, et al. Functional outcome and mortality in geriatric distal femoral fractures. Injury 2012;43(7):1096–101. [14] Smith JR, Halliday R, Aquilina AL, Morrison RJ, Yip GC, McArthur J, et al. Distal femoral fractures: the need to review the standard of care. Injury 2015;46 (6):1084–8. [15] Shields E, Behrend C, Bair J, Cram P, Kates S. Mortality and financial burden of periprosthetic fractures of the femur. Geriatr Orthop Surg Rehabil 2014;5 (4):147–53. [16] Marsh JL, Slongo TF, Agel J, Broderick JS, Creevey W, DeCoster TA, et al. Fracture and dislocation classification compendium—2007: Orthopaedic Trauma Association classification, database and outcomes committee. J Orthopaedic Trauma 2007;21(10):S1–133. [17] Masri BA, Meek RM, Duncan CP. Periprosthetic fractures evaluation and treatment. Clin Orthop Relat Res 2004;420:80–95. [18] Goldacre MJ, Roberts SE, Yeates D. Mortality after admission to hospital with fractured neck of femur: database study. BMJ 2002;325(7369):868–9. [19] Gdalevich M, Cohen D, Yosef D, Tauber C. Morbidity and mortality after hip fracture: the impact of operative delay. Arch Orthop Trauma Surg 2004;124 (5):334–40. [20] Elliott J, Beringer T, Kee F, Marsh D, Willis C, Stevenson M. Predicting survival after treatment for fracture of the proximal femur and the effect of delays to surgery. J Clin Epidemiol 2003;56(8):788–95. [21] Bretherton CP, Parker MJ. Early surgery for patients with a fracture of the hip decreases 30-day mortality. Bone Jt J 2015;97-B(1):104–8. [22] Novack V, Jotkowitz A, Etzion O, Porath A. Does delay in surgery after hip fracture lead to worse outcomes? A multicenter survey. Int J. Qual Health Care 2007;19(3):170–6. [23] Mak JC, Cameron ID, March LM. National Health and Medical Research Council. Evidence-based guidelines for the management of hip fractures in older persons: an update. Med. J. Aust. 2010;192(1):1–37. [24] Boddaert J, Cohen-Bittan J, Khiami F, Le Manach Y, Raux M, Beinis JY, et al. Postoperative admission to a dedicated geriatric unit decreases mortality in elderly patients with hip fracture. PLoS One 2014;9(1):e83795. [25] Streubel PN, Ricci WM, Wong A, Gardner MJ. Mortality after distal femur fractures in elderly patients. Clin Orthop Relat Res 2011;469(4):1188–96. [26] Nieves JW, Bilezikian JP, Lane JM, Einhorn TA, Wang Y, Steinbuch M, et al. Fragility fractures of the hip and femur: incidence and patient characteristics. Osteoporos Int 2010;21(3):399–408. [27] Higgins TF. Distal femoral fractures. J Knee Surg 2007;20(1):56–66. [28] McLauchlan GJ, Robinson CM, Singer BR, Christie J. Results of an operative policy in the treatment of periprosthetic femoral fracture. J Orthop Trauma 1997;11(3):170–9. [29] Buttaro MA, Farfalli G, Paredes Nunez M, Comba F, Piccaluga F. Locking compression plate fixation of Vancouver type-B1 periprosthetic femoral fractures. J Bone Jt Surg Am 2007;89(9):1964–9. [30] Vallier HA, Hennessey TA, Sontich JK, Patterson BM. Failure of LCP condylar plate fixation in the distal part of the femur. A report of six cases. J Bone Jt Surg Am 2006;88(4):846–53. [31] Ehlinger M, Ducrot G, Adam P, Bonnomet F. Distal femur fractures. Surgical techniques and a review of the literature. Orthop Ttraumatol Surg Res OTSR 2013;99(3):353–60.
Limitations There are several limitations to this study. Firstly, retrospective data is always subject to collection bias. Whilst everyeffort was made to track down all patients to ensure accurate mortality data was collected there is always a risk of patients moving geographically outside of our collection areas. A prospective randomised study looking specifically at different fixation methods and post-operative rehabilitation methods would be better suited to making accurate conclusions about these issues. Conclusion Despite our results showing that delay to surgery does not affect mortality in this patient group, we would still recommend early surgical intervention following medical optimisation. This study does however provide evidence to support the fact that delay to surgery, in itself, does not appear to affect mortality. We believe it is therefore safe to treat both acute and chronic medical comorbidities to optimise patient outcomes. We found no adverse orthopaedic complications with early weight-bearing as a treatment strategy and therefore cautiously support this as a post-operative rehabilitation regime. This patient group has similar characteristics with hip fracture patients with a high overall morbidity and mortality and their treatment should be more consistently implemented. Guidelines should be developed based on this and future research to include joint orthopaedic and geriatric care as well as clear guidelines on implant choice and post-operative rehabilitation. Conflict of interest None. Acknowledgement Dr Jim Price MA, MA MB BChir, FRCP, FRCGP, FAcad, MEd, SFHEA. Principle lecturer in Medical Education, University of Brighton BN1 9PHA.
Please cite this article in press as: K. Brogan, et al., Delay to surgery does not affect survival following osteoporotic femoral fractures, Injury (2016), http://dx.doi.org/10.1016/j.injury.2016.07.003