Best Practices for Elderly Hip Fracture Patients

A Systematic Overview of the Evidence

Lauren A. Beaupre, PT, PhD,1 C. Allyson Jones, PT, PhD,2 L. Duncan Saunders, MBBCh, PhD,3
D. William C. Johnston, MD, FRCS (C),4 Jeanette Buckingham, MLIS,5
Sumit R. Majumdar, MD, MPH 6
1
Department of Surgery, Division of Orthopaedics, Capital Health, Edmonton, AB, Canada; 2Department of Physical Therapy,
University of Alberta, Edmonton, AB, Canada; 3Department of Public Health Sciences, University of Alberta, Edmonton, AB, Canada;
4
Department of Surgery, Division of Orthopaedics, University of Alberta Hospital, Capital Health, Edmonton, AB, Canada; 5John W Scott
Health Sciences Library, University of Alberta, Edmonton, AB, Canada; 6Department of Medicine, University of Alberta, Edmonton, AB,
Canada.

OBJECTIVES: To determine evidence-based best practices for elderly available health resources effectively and efficiently is relevant
hip fracture patients from the time of hospital admission to 6 months to both clinicians and policymakers.
postfracture. Standardized care, based upon current ‘‘best evidence,’’
DATA SOURCES: MEDLINE, Cochrane Library, CINAHL, Embase, PE- constitutes 1 approach to facilitate optimal outcomes and re-
Dro, Ageline, NARIC, and CIRRIE databases were searched for poten- source use. We conducted a systematic literature review of
tially eligible articles published between 1985 and 2004. management of this patient population, examining all practic-
REVIEW METHODS: Two independent reviewers determined studies es throughout the care continuum from preoperative assess-
appropriate for inclusion using standardized selection criteria, extract- ment through surgical management and subsequent
ed data, evaluated internal validity, and then rated studies according to rehabilitation. Because our systematic review examined a
levels of evidence. Only Level 1 or 2 evidence was included in our sum- broad array of treatment practices, we included not only
mary of clinical recommendations. individual studies, but also systematic reviews of specific
RESULTS: Spinal anesthesia, pressure-relieving mattresses, periop- treatment practices where available. Some of the clinical
erative antibiotics, and deep vein thromboses prophylaxes had con- areas investigated apply to elderly patients in general,
sistent evidence of benefit. Routine preoperative traction was not but are still important aspects of care for hip fracture
associated with any benefits and should be abandoned. Types of sur- patients (e.g., pressure sore prevention); thus these
gical management, postoperative wound drainage, and even ‘‘multidis-
components were also included in our review. Our intent was
ciplinary’’ care, lacked sufficient evidence to determine either benefit or
to identify those evidence-based practices that should be con-
harm. There was little evidence to either determine best subacute re-
sidered a part of routine high quality care for all hip fracture
habilitation practices or to direct ongoing medical issues (e.g., nutri-
tion). Studies conducted during the subacute recovery period were patients.
heterogeneous in terms of treatment settings, interventions, and out-
comes studied and had no clear evidence for best treatment practices.
CONCLUSIONS: The evidence for perioperative practices is relatively METHODS
robust and evidence-based perioperative treatment guidelines can be
easily established. Conversely, more evidence is required to better Data Sources
guide the care of elderly patients with hip fracture during the subacute
A detailed literature search strategy was implemented to
recovery period and convalescence.
identify potential articles published between 1985 and
KEY WORDS: systematic review; geriatrics; hip fracture; 2004 using MEDLINE, Cochrane Library, CINAHL, Embase,
postoperative care. PEDro, Ageline, NARIC, and CIRRIE databases (Appendix 1).
DOI: 10.1111/j.1525-1497.2005.0219.x Clinical practice guideline websites and reference lists of key
J GEN INTERN MED 2005; 20:1019–1025. articles were searched, and content experts questioned to cap-
ture further literature. The search strategy yielded 1,419 ab-
stracts for review; 277 abstracts were excluded from indepth
review because they did not address our study question
(Figure 1).

H ip fracture represents the second leading cause of hos-

pitalization for elderly people.1 Incidence increases sub-
stantially with age, rising from 22.5 and 23.9 per 100,000
population at age 50 to 630.2 and 1289.3 per 100,000 popu- Eligibility Criteria
lation by age 80, for men and women, respectively.2–5 Follow-
ing a hip fracture, patients have increased health service Specific eligibility criteria are described in Figure 1. Our in-
utilization for at least 1 year, with much of health care costs clusion criteria necessarily spanned a wide range of study de-
attributable to subsequent long-term care.2,6–8 Identifying signs, interventions, and outcomes. The target population was
best practices for elderly hip fracture patients while using patients over age 65 years. For surgical interventions, only
randomized study designs were considered acceptable. All
The authors have no conflicts of interest to report. studies required a comparison group or detailed and appro-
Dr. Majumdar is a Population Health Investigator of AHFMR and a New priately risk-adjusted analyses.
Investigator of the Canadian Institutes of Health Research.
Address correspondence and requests for reprints to: Dr. Beaupre: Received for publication June 2, 2005
1F1.52 WMC, 8440-112 ST, Edmonton, AB, Canada T6G 2B7 (e-mail: and in revised form June 15, 2005
[email protected]). Accepted for publication June 15, 2005
1019
1020 Beaupre et al., Best Practices for Elderly Hip Fracture Patients JGIM

Study Selection Preoperative Care

For potentially eligible abstracts (n =1,142), 2 reviewers, Three components of preoperative care were identified that
blinded to authorship and journal, independently assessed had sufficient published evidence to determine best practice:
studies for inclusion/exclusion using standardized criteria re- use of preoperative traction, pressure sore prevention, and the
garding study design, interventions, and outcomes (Figure 1). effect of surgical delay (Table 1).
Noninterventional studies describing factors affecting out-
Preoperative Traction. Preoperative skin or skeletal traction,
comes were evaluated to look for potentially modifiable risk
standard care in this patient population, consists of 5 to 10
factors (e.g., surgical delay). Observational studies examining
pounds applied to the lower leg, and is intended to
only nonmodifiable risk factors (e.g., age) were excluded. The k
decrease preoperative pain and assist with fracture reduction.
statistic, a test of interrater reliability, was ‘‘substantial’’
However, we found a systematic review of 7 high-quality
(k =0.68) for abstract selection.9
studies (1,271 participants) that reported no benefits for trac-
Following abstract evaluation, the full text article was re-
tion use.12 There were no differences in pain with or without
trieved for potentially eligible studies (n =290) and evaluated
the use of traction on either the first (RR =1.14 (0.89, 1.46)) or
by the same 2 reviewers. When the reviewers did not agree on
second (RR =1.02 (0.74, 1.41)) postoperative day.12 One trial
article selection, they undertook a third review of the article
indicated that analgesic use increased with preoperative trac-
together. Consensus was attained in all cases on the combined
tion use (RR =1.78 (1.16, 2.7)).13 Further, traction did
review. The k statistic for the full text review was ‘‘moderate’’
not improve the ease of fracture reduction (RR =1.02
(k =0.52).9
(0.34, 3.00)).12

Pressure Sore Prevention. Pressure sores are common follow-

Data Extraction
ing hip fracture, with reported postfracture incidence rates
Following study selection, the 2 reviewers independently ex- ranging from 10% to 40%.14–16 Pressure sores represent a ma-
tracted information regarding study population, design, inter- jor burden of illness as patients who develop pressure sores
ventions, and outcomes onto standardized data collection have increased risk of nosocomial infection and prolonged
sheets. hospitalization.14,17
Within a systematic review (16 RCTs) of patients requiring
Data Evaluation prolonged hospitalization, 2 trials (214 participants) examined
pressure sore prevention compared with standard care in eld-
These same reviewers independently assessed studies for in- erly orthopaedic patients.17 These 2 trials reported that foam
ternal validity using standard published criteria (Figure 1).10 and alternating pressure mattresses reduced the incidence of
Studies were rated as ‘‘Excellent’’ if they met all criteria, pressure sore development (RR =0.34 (0.14, 0.85)) and
‘‘Good’’ if they met at least 75%, and ‘‘Fair’’ if they met between (RR =0.20 (0.09, 0.45)), respectively) compared with usual
60% and 74% of validity criteria. care.17
To be included in our final summary of clinical recom-
mendations, studies had to rate as Level 1 or 2 evidence. Level Minimizing Surgical Delay. Time to surgical fixation has been
1 evidence was one or more high quality randomized clinical examined in several studies, some reporting adverse effects of
trials (RCT) or systematic reviews, while Level 2 evidence was surgical delay while others reported minimal negative conse-
one or more high quality observational studies.11 quences.18–27 Determining the effect of time to surgery on pa-
tient outcomes is difficult, because RCTs have not been
considered feasible or ethical, and only Level 2 evidence is
Data Syntheses available.
For clinical areas, where a systematic review of sound meth- In cohort studies utilizing appropriate risk adjustment, sur-
odological quality had previously been undertaken, only stud- gical delay was associated with an increase in adverse out-
ies subsequent to the review were included. To prevent comes.20–23 Hamlet et al.21 reported increased mortality at 3
duplication, we did not include individual studies contained years if surgery was delayed beyond 24 hours, even after ad-
in the systematic review, unless there was additional informa- justing for preoperative severity of illness. In contrast, in
tion not included in the review. Eighty-two RCTs were exclud- a much larger and methodologically rigorous study, Grimes
ed because their results were encapsulated within a et al.20 found no increased mortality in patients who waited
systematic review (Figure 1). Fifty articles were included in more than 96 hours for surgery compared with those who had
the syntheses. surgery within 48 hours after adjustment for comorbidities
(HR =1.07 (0.95, 1.21)), suggesting that patients who had de-
layed surgery were sicker than those not experiencing surgical
RESULTS delay.
We present our main results along the temporal continuum Delay in surgery was, however, associated with increased
from time of hospital admission through final rehabilitation. risk of decubitus ulcer (OR =2.2 (1.6, 3.1)).20 Early surgical
Data are listed as relative risks (RR), hazard ratios (HR), or intervention also significantly reduced major medical compli-
odds ratios (OR) with their respective 95% confidence intervals cations, compared with delays beyond 48 hours in the patients
(CI) to allow for comparisons between studies and across in- who were rated as most ill on hospital admission.22 Orosz
terventions. Finally, we summarized and tabulated our recom- et al.23 reported surgery performed within 24 hours of
mendations for clinical practice according to the interventions admission decreased time with severe or very severe preoper-
examined and the level of evidence provided (Table 1). Web ta- ative pain ( 0.22 days ( 0.41, 0.03)), decreased average
bles are available for more complete data review. hospital length of stay (LOS) by almost 2 days (1.94
JGIM Beaupre et al., Best Practices for Elderly Hip Fracture Patients 1021

Table 1. Clinical Recommendations

Intervention Outcome(s) Assessed Recommendation Level of

Evidence

Preoperative
Preoperative traction Analgesic use Preoperative traction demonstrates no benefit 112–13
Ease of fracture reduction
Pressure sore prevention Incidence of pressure sores Pressure-reducing mattresses appear to be 117
beneficial in reducing pressure sore
development
Surgical delay Mortality Surgery should be performed once patient is 220–23
Major complications medically stable, within 24 h if possible
Decubitus ulcer
Preoperative pain
Perioperative
Conservative Nonunion, leg shortening and deformity Operative treatment is better than conservative 131
management treatment
Surgical management
Intertrochanteric Operative details (length of surgery, blood loss, Sliding hip screw fixation should be considered 128, 32–34, 38
transfusion requirements) standard of treatment
Fixation complications (nonunion, reoperations) Short femoral nails (i.e., short Gamma) should not
Anatomical restoration (limb shortening, be used (increased risk of postoperative fracture
deformity) around implant)
Function, pain, mortality Long femoral nails may be superior to sliding screw
fixation for treatment of reverse obliquity and
subtrochanteric fractures
Ender’s nails should not be used
Subcapital Operative details (length of surgery, Screws are better than pins for nondisplaced 129–30, 35–37, 39

blood loss, transfusion requirements) fractures

Fixation complications (nonunion, Cemented arthroplasties are superior to
reoperations) noncemented
Anatomical restoration (limb shortening, Bipolar implants have no advantages over unipolar
deformity) implants
function, pain, mortality Choice of hemiarthroplasty versus internal fixation
is dependent upon patient factors/surgeon
preference, but displaced fractures should be
treated with hemiarthroplasty or total hip
arthroplasty
Anesthetic Mortality Regional anesthesia should be used whenever 140–41
Morbidity possible

Deep vein thromboses Mortality DVT prophylaxis in the form of any heparin or 143–45
(DVT) prophylaxis Morbidity (DVT, pulmonary embolus) fondiparinux for 10 days postoperatively, or
mechanical pumping should be used
Vitamin K antagonists may be used for 10 days 145
postoperatively with a target international
normalized ratio of 2.5 (2.0 to 3.0 acceptable)
DVT prophylaxis should be commenced 145
preoperatively if surgery is delayed
Antibiotic prophylaxis Morbidity (wound infection, urinary and Antibiotics should be used preoperatively for all 146
respiratory tract infections) patients
Postoperative wound Morbidity (Wound infection, wound healing, Postoperative drains may not be required 147
drainage transfusions, dressing changes, reoperation)
Urinary tract Urinary retention Intermittent catheterization is superior to 150w
management indwelling catheterization
Perioperative pain Pain Analgesic use Epidural pain management may reduce myocardial 152–53
control ischemia in addition to reducing perioperative
and postoperative pain

Early postoperative (up to 7 to 10 days)

Nutritional assessment Mortality Nutrition should be assessed 157–59
and treatment Morbidity Protein supplementation should be considered for
Function malnourished patients
Maintenance of lean body mass

Multidisciplinary care Function Effectiveness of multidisciplinary care compared 1 and 260–69

Morbidity with usual care is unclear
Length of stay Presence of mild or moderate dementia should not 170,78
Mortality preclude inclusion in a rehabilitation program
Rehabilitation/discharge planning
Rehabilitation setting Function No clear benefit has been demonstrated between 1 and
Homecare Length of stay different settings for the specified outcomes 260,62,64,68,72–75,77
Subacute Institutionalization Patients can respond positively to exercise
Rural versus urban Activity of daily living mobility programs following hip fracture 171,76
Outpatient Ambulation
Level 1, at least one good quality Randomized Clinical Trial (RCT); Level 2, at least one good quality observational study.
w
Only 1 randomized trial to support that recommendation.
1022 Beaupre et al., Best Practices for Elderly Hip Fracture Patients JGIM

(1.06, 2.82)) and was associated with reduced major medical ment is recommended between hospital admission and sur-
complications (OR =0.26 (0.07, 0.95). Further, early surgical gery.45
repair was associated with earlier ambulation compared with
Antibiotic Prophylaxis. In 22 RCTs (8,307 participants) of adult
delayed surgical repair.22
patients with closed long bone fracture fixation, of which 16
trials were hip fracture patients, antibiotic prophylaxis de-
Perioperative Care
creased the incidence of deep wound infections (RR =0.36
Perioperative care was defined as the immediate preoperative (0.21, 0.65)) and urinary tract infections (RR =0.66
time through initial postoperative days to attainment of med- (0.43, 1.0)).46 A single antibiotic dose with tissue effects last-
ical stability. Evidence-based perioperative interventions in- ing greater than 12 hours (e.g., cefazolin 1 g intravenously) or
cluded surgical and anesthetic management (Table 2), deep multiple doses of antibiotics with shorter half-lives were seem-
vein thrombosis (DVT) and antibiotic prophylaxes, and other ingly equivalent.46
general medical care (e.g., wound drainage and pain control)
Postoperative Wound Drainage. Suction wound drainage, rou-
(Table 3).
tine practice in many hospitals, is implemented to promote
Surgical Management. Despite numerous clinical trials re- postoperative wound healing by preventing large hematoma
garding specific surgical techniques, best practices remain formation.47 This technique, however, has an inherent risk of
unclear, particularly for femoral neck fractures.28–39 A com- increasing postoperative infection through the creation of a
pression screw plate device is considered to be the standard of portal to deep tissues.
care for intertrochanteric or extracapsular fractures.28,32–34,38 A systematic review of 3 RCTs (333 participants) found no
Surgical management of femoral neck or intracapsular frac- significant differences reported in rates of infection (RR =0.53
tures is dependent upon patient age, activity level, health sta- (0.21, 1.35)), reoperation for wound healing problems
tus, and surgeon preference.29,30,35–37,39 Few studies have (RR =4.1 (0.47, 36.1)), or transfusions (RR =1.16 (0.84, 1.61))
examined conservative treatment because operative manage- with the use of wound drains in hip fracture patients.47
ment is considered superior if patients are medically fit for
Urinary Tract Catheterization. Following hip fracture, the inci-
surgery.31
dence rate of urinary tract infection is 23% to 25%.48,49 De-
Type of Anesthesia. Use of general anesthetics in elderly pa- spite this high rate, very few studies have examined
tient populations has been associated with increased postop- catheterization methods used in this patient population. The
erative delirium.40 A systematic review of 22 trials (2,567 1 RCT performed, reported that normal voiding pattern was
participants) demonstrated reduced risk of mortality at 1 resumed on average 4.3 (0.7, 8.0) days earlier with intermit-
month postfracture (RR =0.69 (0.50, 0.95)) and DVT tent rather than indwelling catheterization for patients with
(RR =0.64 (0.49, 0.95)) with use of regional anesthesia.41 An- postoperative urinary retention (P =.01).50
other recent systematic review of 141 studies (9,559 partici-
Perioperative Pain Control. Patients with poorly controlled
pants), demonstrated a clear benefit for regional anesthesia in
perioperative pain have reported increased hospital LOS, de-
terms of mortality (OR =0.68 (0.53, 0.88)).40 Although the re-
layed ambulation, and decreased 6-month mobility.51 Little
view included several surgical procedures, 44 (31%) of 141
evidence exists regarding appropriate analgesia for patients
study populations were elderly orthopedic patients. This re-
with a hip fracture.
view also reported a reduction in DVT (OR =0.56 (0.43, 0.72)),
Two RCTs (145 participants) reported epidural pain man-
pulmonary embolism (PE) (OR =0.45 (0.29, 0.69)), transfusion
agement reduced perioperative cardiac complications; one re-
requirements (OR =0.50 (0.39, 0.66)), and pneumonia
ported decreased intraoperative myocardial ischemia
(OR =0.61 (0.48, 0.76)) with regional compared with general
(RR =0.13 (0.02, 0.97)) 52 and the other decreased preopera-
anesthesia.
tive cardiac events (cardiac death, myocardial infarction, un-
DVT Prophylaxis. Following hip fracture surgery, patients are stable angina, heart failure, or new onset atrial fibrillation)
at increased risk of DVT with incidence rates of 27% for prox- (P =.01).53 Both trials reported decreased perioperative and/or
imal DVT.26,42 Incidence rates for fatal PE range from 1.4% to postoperative pain compared with usual analgesia. No evi-
7.5% in the first 3 months following hip fracture surgery.26 dence was found regarding specific narcotic agents, other than
Thirty-two studies (3,614 participants) of adult hip fracture an expert-based consensus statement recommending avoid-
patients consistently demonstrated that DVT prophylaxis re- ance of codeine and meperidine.54
duced the incidence of DVT and PE.43,44 The use of any he-
parin treatment versus no treatment (13 trials; 1,199 Early Postoperative Care (up to 7 to 10 days
participants; RR =0.60 (0.50, 0.71)) or a mechanical pumping Postoperative)
device versus no treatment (5 trials; 451 participants;
Following medical stabilization, the primary treatment goals
RR =0.31 (0.19, 0.51)) significantly reduced the risk of
focus on rehabilitation. Studies performed during this period
DVT.44 No significant differences were detected between un-
were extremely heterogeneous in terms of interventions un-
fractionated or fractionated heparins in this review.44 The Sev-
dertaken and outcomes measured, making it difficult to define
enth American College of Chest Physicians Conference on
best practices.
Antithrombotic and Thrombolytic Therapy currently recom-
mends the routine use of fondiparinux, or heparin of any type Optimizing Nutrition. Poor nutritional status is common in the
for at least 10 days.45 They also suggest that a vitamin K an- hip fracture population and appears to be independently
tagonist may be used for 10 days with a target international associated with increased morbidity and mortality.55,56 Inter-
normalized ratio of 2.5 (minimum–maximum 2.0 to 3.0).45 If ventions to address malnutrition following a hip fracture have
surgery is to be significantly delayed, any type of heparin treat- focused primarily on initial recovery in hospital settings.
JGIM Beaupre et al., Best Practices for Elderly Hip Fracture Patients 1023

Protein/vitamin supplements or nasogastric feeding have dysfunction were more likely to receive some form of institu-
been examined in 15 trials (1,054 participants) and reduced tional care for prolonged periods.
long-term complications (RR =0.52 (0.32, 0.84)) but did not
PostHospitalization Rehabilitation. Binder et al.71 and Sher-
affect mortality (RR =0.92 (0.56, 1.50)).57 One RCT (62 partic-
rington et al.76 demonstrated that strengthening programs
ipants) reported patients receiving oral nutritional supplemen-
are effective following hip fracture. Patients in the intervention
tation had a reduced hospital LOS and were less likely to
groups, which consisted of strength training of various types,
experience major complications.58 Continuation of an inter-
showed modest gains in strength and function compared with
vention (oral protein supplementation plus nandrolone de-
patients following usual care (Table 4).
canoate) for 6 months following fracture increased albumin
levels and maintained lean body mass better than oral protein
supplementation only or usual care.59
DISCUSSION
We found that a number of practices in the management of hip
Multidisciplinary Care. Multidisciplinary care involves team- fracture patients have a strong evidentiary basis and should be
based management, and typically consists of a medical prac- considered part of routine high quality care (Table 1). Although
titioner and multiple other health professionals (e.g., at least some findings may seem fairly self-evident, we also found that
Nursing and Physical Therapy) who plan treatment to meet many routine practices were not supported by published lit-
patients’ complex care needs. Despite numerous studies, no erature. For example, high-quality evidence demonstrated that
clear Level 1 evidence exists that multidisciplinary care with use of preoperative traction, while common, was associated
early mobilization affords better outcomes in terms of mortal- with harms and no net clinical benefits—this is clearly a prac-
ity, morbidity, function, or service utilization than usual care. tice that ought to be abandoned. Conversely, the intuitively
Metaanalysis of such studies is difficult because of heteroge- sensible offering of multidisciplinary care with early mobiliza-
neity caused by diverse interventions, patient outcomes, and tion was not definitively associated with better outcomes than
measures used to assess the intervention effect.60–68 Some usual care. As usual care in most surgical hospitals may al-
studies suggest standardized multidisciplinary care reduced ready be ‘‘multidisciplinary’’ to some degree (Medicine, Nurs-
LOS in hospital, while others suggested it increased LOS (Ta- ing, and Physical Therapy at a minimum), the incremental
ble 4).61–70 benefit of any studied multidisciplinary intervention, over and
above usual care, may be prohibitively difficult to ascertain.
Subacute Rehabilitation and Discharge Planning Further, LOS, which is a commonly assessed outcome of ef-
Following the initial postoperative period, once the patient is fectiveness of multidisciplinary care, is highly dependent upon
medically stable, treatment focuses on discharge planning and how rehabilitation services are organized and located, making
subacute rehabilitation. Few studies have demonstrated a clear crossstudy comparisons difficult.
For the most part, evidence-based care for the hip fracture
benefit among different rehabilitation settings or timing of re-
habilitation in terms of patients’ functional outcomes and serv- population in the perioperative period is reasonably well de-
ice utilization, despite the surfeit of studies.22,60–64,67,68,71–77 fined. We, unlike others,79–81 included in our summary of prac-
tice recommendations only those practices for which Level 1 or
Although evidence is sparse, presence of dementia should
not preclude inclusion in a rehabilitation program.70,78 In a 2 evidence is available. Gaps in knowledge regarding delivery of
subgroup analysis of patients with dementia, 1 RCT (141 par- perioperative care still exist (e.g., urinary tract management)
ticipants) reported the median LOS for patients with mild and despite the many studies performed. We believe, however,
moderate dementia in a group receiving multidisciplinary in- that standardized evidence-based perioperative treatment
patient rehabilitation was significantly reduced compared with guidelines can be established for many treatment areas for
controls (Table 4).70 the typical elderly patient with a hip fracture (Table 1).
Standardization of care (medical and rehabilitation)
Rehabilitation Setting. Cameron et al.60 in a systematic review would also be expected to streamline practice and improve
of 9 RCTs (1,887 participants) comparing different formats of the quality of care, although we acknowledge this hypothesis
inpatient rehabilitation reported heterogeneous effects on ought to be more rigorously tested. Nonetheless, several clin-
costs and LOS. Using a combined index of death or depend- ical areas require much further investigation. Of particular
ency, no difference was seen between intensive rehabilitation note is the lack of evidence available in the subacute recovery
and usual care (RR =0.93 (0.83, 1.05)) (Table 4). Kiusma period, commencing after postoperative day 7 to 10, where
et al.74 reported patients discharged home with physical ther- very little research has been conducted. Summarizing the lim-
apy had better ambulation at 1 year than patients who under- ited available evidence is further hindered by heterogeneity in
went inpatient rehabilitation (P =.01) (Table 4). study settings and interventions assessed.71–77 Investigation
Tinetti et al.77 reported that multidisciplinary homecare as to type and extent of rehabilitation and nutritional services
(structured physical and occupational therapy) did not result is needed in subacute settings (e.g., long-term care, regional
in improved outcomes compared with usual homecare (phys- hospitals, homecare) as is greater consideration of secondary
ical therapy as determined by individual therapists). Similar prevention measures for recurrent fracture (e.g., falls manage-
proportions of patients regained prefracture levels in activities ment, osteoporosis treatment).
of daily living by 6 months postfracture, although the in- In summary, our systematic review found evidence to
creased homecare group showed a trend towards better gait support many facets of preoperative and perioperative care
assessed qualitatively (P =.08) (Table 4). Patient characteris- for elderly hip fracture patients. These treatment practices
tics (e.g., age, social support and prefracture function) deter- could likely be applied in most acute care settings. Standard-
mine the type of rehabilitation.72,73,75 Patients with greater ization of these practices could be expected to improve quality
of care and outcomes. Nevertheless, much work remains to
1024 Beaupre et al., Best Practices for Elderly Hip Fracture Patients JGIM

define all of the best practices for hip fracture care and deter- 24. Parker MJ, Pryor GA. The timing of surgery for proximal femoral frac-
mine how best to deliver them within a seamless health care tures. J Bone Jt Surg [Br]. 1992;74:203–5.
25. Rogers FB, Shackford SR, Keller MS. Early fixation reduces morbidity
continuum.
and mortality in elderly patients with hip fractures from low-impact falls.
J Trauma Inj Infect Crit Care. 1995;39:261–5.
26. Todd CJ, Freeman CJ, Camilleri-Ferrante C, et al. Differences in mor-
Grant Support: This project was sponsored and funded by the tality after fracture of hip: the east Anglian audit. BMJ. 1995;310:904–8.
Alberta Heritage Foundation for Medical Research and Alber- 27. Zuckerman JD, Skovron ML, Koval KJ, Aharonoff G, Frankel VH.
ta Health and Wellness. Postoperative complications and mortality associated with operative de-
lay in older patients who have a fracture of the hip. J Bone Jt Surg [Am].
1995;77:1551–6.
REFERENCES 28. Audige L, Hanson B, Swiontkowski MF. Implant-related complications
in the treatment of unstable intertrochanteric fractures: meta-analysis of
1. Wilkins K. Health care consequences of falls for seniors. Health Reports. dynamic screw-plate versus dynamic screw-intramedullary nail devices.
1999;10:47–5. Int Orthop. 2003;27:197–203.
2. Brainsky A, Glick H, Lydick E, et al. The economic cost of hip fractures 29. Bhandari M, Devereaux PJ, Swiontkowski MF, et al. Internal fixation
in community-dwelling older adults: a prospective study. J Am Geriatr compared with arthroplasty for displaced fractures of the femoral neck: a
Soc. 1997;45:281–7. meta-analysis. J Bone Jt Surg [Am]. 2003;85:1673–81.
3. Cooper C, Campion G, Melton LJ. Hip-fractures in the elderly—a world- 30. Masson M. Internal fixation versus arthroplasty for intracapsular prox-
wide projection. Osteoporosis Int. 1992;2:285–9. imal femoral fractures in adults. Cochrane Database Syst Rev. 2002;
4. Gullberg B, Johnell O, Kanis JA. World-wide projections for hip frac- Issue 4.
ture. Osteoporosis Int. 1997;7:407–13. 31. Parker MJ, Handoll HHG, Bhargara A. Conservative versus oper-
5. Maggi S, Kelsey JL, Litvak J, Heyse SP. Incidence of hip fractures in the ative treatment for hip fractures. Cochrane Database Syst Rev. 2000;
elderly: a cross-national analysis. Osteoporosis Int. 1991;1:232–41. Issue 4.
6. Donald IP, Bulpitt CJ. The prognosis of falls in elderly people living at 32. Parker MJ, Handoll HHG, Bhonsle S, Gillespie WJ. Condylocephalic
home. Age Ageing. 1999;28:121–5. nails versus extramedullary implants for extracapsular hip fractures.
7. Haentjens P, Autier P, Barette M, Boonen S. The economic cost of hip Cochrane Database Syst Rev. 1998; Issue 4.
fractures among elderly women: a one-year, prospective, observational 33. Parker MJ. Gamma and other cephalocondylic intramedullary nails ver-
cohort study with matched-pair analysis. J Bone Jt Surg [Am]. sus extramedullary implants for extracapsular hip fractures. Cochrane
2001;83:493–500. Database Syst Rev. 2004; Issue 1.
8. Wiktorowicz ME, Goeree R, Papaioannou A, Adachi JD, Papa- 34. Parker MJ, Handoll HHG, Chinoy MA. Extramedullary fixation im-
dimitropoulos E. Economic implications of hip fracture: health service plants and external fixators for extracapsular hip fractures. Cochrane
use, institutional care and cost in Canada. Osteoporosis Int. Database Syst Rev. 2002; Issue 4.
2001;12:271–8. 35. Parker MJ, Blundell C. Choice of implants for internal fixation of fem-
9. Landis JR, Koch GG. The measurement of observer agreement for cat- oral neck fractures—meta analysis of 25 randomised trials including
egorical data. Biometrics. 1977;33:159–74. 4925 patients. Acta Orthop Scand. 2002;69:138–43.
10. Hayward Ree, ed. User Guides Interactive. Chicago, Ill: JAMA Publishing 36. Parker MJ. Arthroplasties (with and without bone cement) for proximal
Group; 2002. femoral fractures in adults. Cochrane Database Syst Rev. 2004; Issue 2.
11. Wright JG, Swiontkowski MF, Heckman JD. Introducing levels of 37. Parker MJ. Internal fixation implants for intracapsular proximal femoral
evidence to the journal. J Bone Jt Surg [Am]. 2003;85:1–2. fractures in adults. Cochrane Database Syst Rev. 2001; Issue 4.
12. Parker MJ. Pre-operative traction for fractures of the proximal femur. 38. Parker MJ. Replacement arthroplasty versus internal fixation for extra-
Cochrane Database Syst Rev. 2003; Issue 3. capsular hip fractures. Cochrane Database Syst Rev. 1997; Issue 2.
13. Rosen JE, Chen FS, Hiebert R, Koval KJ. Efficacy of preoperative skin 39. Tidermark J, Ponzer S, Svensson O, Soderqvist A, Tornkvist H. In-
traction in hip fracture patients: a prospective, randomized study. J Ort- ternal fixation compared with total hip replacement for displaced femoral
hop Trauma. 2001;15:81–5. neck fractures in the elderly. A randomised, controlled trial. J Bone Jt
14. Allman RM, Goode PS, Burst N, Bartolucci AA, Thomas DR. Pressure Surg [Br]. 2003;85:380–8.
ulcers, hospital complications, and disease severity: impact on hospital 40. Rodgers A, Walker N, Schug S, McKee A, Kehlet H, vanZundert A.
costs and length of stay. Adv Wound Care. 1999;12:22–30. Reduction of postoperative mortality and morbidity with epidural or spi-
15. Baumgarten M, Margolis D, Berlin JA, et al. Risk factors for pressure nal anaesthesia: results from overview of randomised trials. BMJ.
ulcers among elderly hip fracture patients. Wound Repair Regen. 2000;321:1493–7.
2003;11:96–103. 41. Parker MJ, Handoll HH, Griffiths R. Anaesthesia for hip fracture
16. Hofman A, Geelkerken RH, Wille J, Hamming J, Hermans J, Breslau surgery in adults. Cochrane Database Syst Rev. 2004; Issue 4.
P. Pressure sores and pressure-decreasing mattresses—controlled clin- 42. Anderson FA Jr., Wheeler HB, Goldberg RJ, et al. A population-based
ical-trial. Lancet. 1994;343:568–71. perspective of the hospital incidence and case-fatality rates of deep vein
17. Nuffield Institute of Health, University of Leeds, NHS Centre for Re- thrombosis and pulmonary embolism. The Worcester DVT Study. Arch
view and Dissemination. How effective are pressure-relieving interven- Int Med. 1991;151:933–8.
tions for the prevention and treatment of pressure sores? Effective 43. Eriksson BI, Lassen MR, PENT asaccharide in HIP-FRActure Surgery
Health Care Bull. 1995;2:1–15. Plus Investigators. Duration of prophylaxi against venous thromboem-
18. Bredahl C, Nyholm B, Hindsholm KB, Mortensen JS, Olesen AS. Mor- bolism with fondaparinux after hip fracture surgery: a multicenter,
tality after hip fracture: results of operation within 12 h of admission. randomized, placebo-controlled, double-blind study. Arch Intern Med.
Injury. 1992;23:83–6. 2003;163:1337–42.
19. Dolk T. Operation in hip fracture patients—analysis of the time factor. 44. Handoll HH, Farrar MJ, McBirnie J, et al. Heparin, low molecular
Injury. 1990;21:369–72. weight heparin and physical methods for preventing deep vein thrombo-
20. Grimes JP, Gregory PM, Noveck H, Butler MS, Carson JL. The effects sis and pulmonary embolism following surgery for hip fractures. Coch-
of time-to-surgery on mortality and morbidity in patients following hip rane Database Syst Rev. 2002; Issue 4.
fracture. Am J Med. 2002;112:702–9. 45. Geerts WH, Pineo GF, Heit JA, et al. Prevention of venous thrombo-
21. Hamlet WP, Lieberman JR, Freedman EL, Dorey FJ, Fletcher A, embolism. The 7th ACCP conference on antithrombotic and thrombolytic
Johnson EE. Influence of health status and the timing of surgery on therapy. Chest. 2004;126:338–400S.
mortality in hip fracture patients. Am Orthop. 1997;26:621–7. 46. Gillespie WJ, Walenkamp G. Antibiotic prophylaxis for surgery for prox-
22. Hoenig H, Rubenstein LV, Sloane R, Horner R, Kahn K. What is imal femoral and other closed long bone fractures. Cochrane Database
the role of timing in the surgical and rehabilitative care of community- Syst Rev. 2001; Issue 1.
dwelling older persons with acute hip fracture? Arch Int Med. 1997; 47. Parker MJ. Closed suction surgical wound drainage after orthopaedic
157:513–20. surgery. Cochrane Database Syst Rev. 2001; Issue 4.
23. Orosz GM, Magaziner J, Hannan EL, et al. Association of timing 48. Johansson I, Athlin E, Frykholm L, Bolinder H, Larsson G. Intermit-
of surgery for hip fracture and patient outcomes. JAMA. 2004;291: tent versus indwelling catheters for older patients with hip fractures.
1738–43. J Clin Nurs. 2002;11:651–6.
JGIM Beaupre et al., Best Practices for Elderly Hip Fracture Patients 1025

49. Southwell-Keely JP, Russo RR, March L, Cumming R, Cameron I, 66. Naglie G, Tansey C, Kirkland JL, et al. Interdisciplinary inpatient care
Brnabic AJ. Antibiotic prophylaxis in hip fracture surgery: a metaanal- for elderly people with hip fracture: a randomized control trial. Can Med
ysis. Clin Orthop. 2004;419:179–84. Assoc J. 2002;167:25–32.
50. Skelly JM, Guyatt GH, Kalbfleisch R, Singer J, Winter L. Management 67. Roberts HC, Pickering RM, Onslow E, et al. The effectiveness of
of urinary retention after surgical repair of hip fracture. Can Med Assoc implementing a care pathway for femoral neck fracture in older people:
J. 1992;146:1185–9. a prospective controlled before and after study. Age Ageing. 2004;33:
51. Morrison RS, Magaziner J, McLaughlin MA, et al. The impact of post- 178–84.
operative pain on outcomes following hip fracture. Pain. 2003;103:303–11. 68. Swanson CE, Day GA, Yelland CE, et al. The management of elderly
52. Scheini H, Virtanen T, Kentala E, et al. Epidural infusion of bupiva- patients with femoral fractures. A randomised controlled trial of
caine and fentanyl reduces perioperative myocardial ischaemia in elderly early intervention versus standard care. Med J Austral. 1998;169:
patients with hip fracture—a randomized controlled trial. Acta An- 515–8.
aesthesiol Scand. 2000;44:1061–70. 69. Choong PF, Langford AK, Dowsey MM, Santamaria NM. Clinical path-
53. Matot I, Oppenheim-Eden A, Ratrot R, et al. Preoperative cardiac way for fractured neck of femur: a prospective, controlled study. Med J
events in elderly patients with hip fracture randomized to epidural or Austral. 2000;172:423–6.
conventional analgesia. Anesthesiology. 2003;98:156–63. 70. Huusko TM, Karppi P, Avikainen V, Kautiainen H, Sulkava R. Ran-
54. Beers MH. Explicit criteria for determining potentially inappropriate domised, clinically controlled trial of intensive geriatric rehabilitation in
medication use by the elderly. An update. Arch Intern Med. 1997; patients with hip fracture: subgroup analysis of patients with dementia.
157:1531–6. BMJ. 2000;321(34 ref):1107–11.
55. Herrmann FR, Safran C, Levkoff SE, Minaker KL. Serum albumin level 71. Binder EF, Brown M, Sinacore DR, Steger-May K, Yarasheski KE,
on admission as a predictor of death, length of stay, and readmission. Schechtman KB. Effects of extended outpatient rehabilitation after hip
Arch Int Med. 1992;152:125–30. fracture: a randomized controlled trial. JAMA. 2004;292:837–46.
56. Patterson BM, Cornell CN, Carbone B, Levine B, Chapman D. Protein 72. Koval KJ, Aharonoff GB, Su ET, Zuckerman JD. Effect of acute inpa-
depletion and metabolic stress in elderly patients who have a fracture of tient rehabilitation on outcome after fracture of the femoral neck or
the hip. J Bone Jt Surg [Am]. 1992;74:251–60. intertrochanteric fracture. J Bone Jt Surg [Am]. 1998;80:357–64.
57. Avenell A, Handoll HHG. Nutritional supplementation for hip fracture 73. Kramer AM, Steiner JF, Schlenker RE, et al. Outcomes and costs after
aftercare in the elderly. Cochrane Database Syst Rev. 2004; Issue 1. hip fracture and stroke: a comparison of rehabilitation settings. JAMA.
58. Tkatch L, Rapin C-H, Rizzoli R, et al. Benefits of oral protein supple- 1997;277:396–404.
mentation in elderly patients with fracture of the proximal femur. JAm 74. Kuisma R. A randomized, controlled comparison of home versus insti-
Coll Nutr. 1992;11:519–25. tutional rehabilitation of patients with hip fracture. Clin Rehabil.
59. Tidermark J, Ponzer S, Carlsson P, et al. Effects of protein-rich sup- 2002;16:553–61.
plementation and nandrolone in lean elderly women with femoral neck 75. Levi SJ. Posthospital setting, resource utilization, and self-care outcome
fractures. Clin Nutr. 2004;23:587–96. in older women with hip fracture. Arch Phys Med Rehabil. 1997;78:
60. Cameron I, Handoll H, Finnegan T, Madhok R, Langhorne P. Co-or- 973–9.
dinated multidisciplinary approaches for inpatient rehabilitation of older 76. Sherrington C, Lord SR, Herbert RD. A randomised trial of weight-
patients with proximal femoral fractures. Cochrane Database Syst Rev. bearing versus non-weight-bearing exercise for improving physical
2001; Issue 3. ability in inpatients after hip fracture. Austral J Physiother. 2003; 49:
61. Cameron ID, Lyle DM, Quine S. Accelerated rehabilitation after prox- 15–22.
imal femoral fracture: a randomized controlled trial. Disabil Rehabil. 77. Tinetti ME, Baker DI, Gottschalk M, et al. Home-based multicompo-
1993;15:29–34. nent rehabilitation program for older persons after hip fracture: a ran-
62. Galvard H, Samuelsson SM. Orthopedic or geriatric rehabilitation of hip domized trial. Arch Phys Med Rehabil. 1999;80:916–22.
fracture patients: a prospective, randomized, clinically controlled study 78. Rolland Y, Pillard F, Lauwers-Cances V, Busquere F, Vellas B, Lafont
in Malmo, Sweden. Aging. 1995;7:11–6. C. Rehabilitation outcome of elderly patients with hip fracture and cog-
63. Hagsten B, Svensson O, Gardulf A. Early individualized postoperative nitive impairment. Disabil Rehabil. 2004;26:425–31.
occupational therapy training in 100 patients improves ADL after hip 79. Prevention and Management of Hip Fracture in Older People.
fracture: a randomized trial. Acta Orthop Scand. 2004;75:177–83. www.sign.ac.uk/guidelines/fulltext/56/index.html. 2002. Scottish
64. Huusko TM, Karppi P, Avikainen V, Kautiainen H, Sulkava R. Inten- Intercollegiate Guidelines Network.
sive geriatric rehabilitation of hip fracture patients: a randomized, 80. March LM, Chamberlain AC, Cameron ID, et al. How best to fix a bro-
controlled trial. Acta Orthop Scand. 2002;73:425–31. ken hip. Fractured Neck of Femur Health Outcomes Project Team. Med J
65. March LM, Cameron ID, Cumming RG, et al. Mortality and morbidity Austral. 1999;170:489–94.
after hip fracture: can evidence based clinical pathways make a differ- 81. Morrison RS, Chassin MR, Siu AL. The medical consultant’s role in
ence? J Rheumatol. 2000;27:2227–31. caring for patients with hip fracture. Ann Int Med. 1998;128:1010–20.

Supplementary Material
The following supplementary material is available
for this article online:
Appendix 1. Literature Search Strategies.
Figure 1. Flowchart of Review Process.
Table 1. Preoperative Medical Management.