Fracture fixation in the operative management of hip ... · 15/04/2017 · Treatment of Hip fractures (FAITH) trial to examine the effect of a sliding hip screw versus cancellous

$Page 1: Fracture fixation in the operative management of hip ... · 15/04/2017 · Treatment of Hip fractures (FAITH) trial to examine the effect of a sliding hip screw versus cancellous$
Articles

www.thelancet.com Vol 389 April 15, 2017 1519

Fracture fixation in the operative management of hip fractures (FAITH): an international, multicentre, randomised controlled trialFixation using Alternative Implants for the Treatment of Hip fractures (FAITH) Investigators*

SummaryBackground Reoperation rates are high after surgery for hip fractures. We investigated the effect of a sliding hip screw versus cancellous screws on the risk of reoperation and other key outcomes.

Methods For this international, multicentre, allocation concealed randomised controlled trial, we enrolled patients aged 50 years or older with a low-energy hip fracture requiring fracture fixation from 81 clinical centres in eight countries. Patients were assigned by minimisation with a centralised computer system to receive a single large-diameter screw with a side-plate (sliding hip screw) or the present standard of care, multiple small-diameter cancellous screws. Surgeons and patients were not blinded but the data analyst, while doing the analyses, remained blinded to treatment groups. The primary outcome was hip reoperation within 24 months after initial surgery to promote fracture healing, relieve pain, treat infection, or improve function. Analyses followed the intention-to-treat principle. This study was registered with ClinicalTrials.gov, number NCT00761813.

Findings Between March 3, 2008, and March 31, 2014, we randomly assigned 1108 patients to receive a sliding hip screw (n=557) or cancellous screws (n=551). Reoperations within 24 months did not differ by type of surgical fixation in those included in the primary analysis: 107 (20%) of 542 patients in the sliding hip screw group versus 117 (22%) of 537 patients in the cancellous screws group (hazard ratio [HR] 0·83, 95% CI 0·63–1·09; p=0·18). Avascular necrosis was more common in the sliding hip screw group than in the cancellous screws group (50 patients [9%] vs 28 patients [5%]; HR 1·91, 1·06–3·44; p=0·0319). However, no significant difference was found between the number of medically related adverse events between groups (p=0·82; appendix); these events included pulmonary embolism (two patients [<1%] vs four [1%] patients; p=0·41) and sepsis (seven [1%] vs six [1%]; p=0·79).

Interpretation In terms of reoperation rates the sliding hip screw shows no advantage, but some groups of patients (smokers and those with displaced or base of neck fractures) might do better with a sliding hip screw than with cancellous screws.

Funding National Institutes of Health, Canadian Institutes of Health Research, Stichting NutsOhra, Netherlands Organisation for Health Research and Development, Physicians’ Services Incorporated.

IntroductionWorldwide, 4·5 million people per year become disabled after sustaining a hip fracture, with the number living with disability due to hip fracture expected to increase to 21 million in the next 40 years.1,2 Despite surgical intervention, the need for reoperation remains high (10·0–48·8%), has remained largely unchanged in the past 30 years,3,4 and is associated with substantial morbidity, mortality, and costs.5 The high proportion of reoperations has generated controversy about the optimum approach for fixing femoral neck fractures.6

Biomechanical and laboratory studies7 suggest that although a single large screw at a fixed angle with a side-plate (ie, a sliding hip screw) provides greater biomechanical stability, especially in displaced and unstable fracture types, multiple cancellous screws, which is the present standard of care, are less invasive and better preserve blood supply. Previous small trials6 did not find a difference in the effect of the two fixation approaches on

outcomes important to patients, particularly reoperations, leaving uncertainty among surgeons about the optimum approach for fixing femoral neck fractures.

We did the Fixation using Alternative Implants for the Treatment of Hip fractures (FAITH) trial to examine the effect of a sliding hip screw versus cancellous screws on the risk of reoperation and other key outcomes during 24 months.

MethodsStudy design and participantsFAITH was an international, multicentre, allocation concealed, randomised controlled trial assessing the effects of a sliding hip screw versus cancellous screws on reoperation rates over a 24 month follow-up in patients with a low-energy femoral neck fracture. A previous report8 details the trial objectives and methods. All participating centres obtained ethics approval.

We enrolled patients with a low-energy fracture of the hip requiring fracture fixation across 81 clinical centres

Lancet 2017; 389: 1519–27

Published Online March 2, 2017 http://dx.doi.org/10.1016/S0140-6736(17)30066-1

See Comment page 1493

*See the appendix for a full list of FAITH Trial investigators and the Contributors section for the writing committee

Correspondence to: Prof Mohit Bhandari, Division of Orthopaedic Surgery, Department of Surgery, McMaster University, Hamilton ON, L8L 8E7, Canada [email protected]

See Online for appendix

http://crossmark.crossref.org/dialog/?doi=10.1016/S0140-6736(17)30066-1&domain=pdf

Articles

1520 www.thelancet.com Vol 389 April 15, 2017

in the USA, Canada, Australia, the Netherlands, Norway, Germany, the UK, and India. Eligible patients were those aged 50 years or older with a low-energy femoral neck fracture requiring operative fixation. Complete eligibility criteria are provided in the appendix. All patients provided written informed consent.

Randomisation and maskingPatients were randomly assigned by minimisation to receive a sliding hip screw or cancellous screws, with a centralised computer system to ensure allocation concealment and balanced prognosis between inter-vention groups for fracture displacement, age, prefracture living status, prefracture function, American Society for Anesthesiologists class, and centre.9 The study computer programmer generated the minimisation algorithm. Local research personnel at each site performed randomisation by minimisation using the centralised computer system. Surgeons and patients were not masked but the data analyst, while doing the analyses, remained masked to treatment groups.

ProceduresParticipants allocated to the sliding hip screw group received a single large-diameter (8·0 mm), partly threaded screw affixed to the proximal femur with a side plate (with a minimum of two holes and a maximum of

four holes) and no supplemental fixation. Patients allocated to the cancellous screws group received multiple threaded screws, with a minimum of two screws and diameter of 6·5 mm. All participating surgeons had done at least 25 hip fracture fixation procedures during their career, and at least five fracture fixation procedures in the year before participation.

Surgeons chose the manufacturer, the reduction technique, whether to do a capsulotomy or aspiration of intracapsular haematoma, and the final screw position; injectable bone substitutes were not permitted. The protocol specified perioperative antibiotics, thrombo-prophylaxis, and weight-bearing regimens, but left to the surgeons’ discretion patient positioning, fracture reduction, and surgical exposure in the operating room. We provided surgeons with specific criteria for acceptability of postfixation radiographic fracture alignment.

Participants returned for follow-up at 1 and 10 weeks, and at 6, 9, 12, 18, and 24 months after surgery. Additional details of the trial intervention and standardisation of perioperative care, surgeon expertise, and the follow-up processes are provided in the appendix.

OutcomesThe primary endpoint was reoperation, defined as surgery that occurred subsequent to the initial procedure and within 24 months to promote fracture healing, relieve

Research in context

Evidence before this studyWe searched the computerised databases Medline and PubMed for randomised clinical trials published in English between Jan 1, 1969, and June 1, 2002, using the search terms “femoral neck fracture” AND “arthroplasty”, as well as “femoral neck fracture” AND “internal fixation”. In addition, bibliographies were searched for relevant studies. We also identified additional studies through hand searches of major orthopaedic journals, bibliographies of major orthopaedic textbooks, and personal files. Of 140 citations initially identified, 14 met all eligibility criteria. Three investigators independently graded study quality and abstracted relevant data, including information on revision and mortality rates. An international survey of orthopaedic surgeons identified that most surgeons preferred to fix femoral neck hip fractures with multiple cancellous screws rather than a sliding hip screw. Despite the popularity of cancellous screws for hip fracture fixation, biological investigations have suggested that the sliding hip screw is a more biomechanically stable construct compared with cancellous screws. Moreover, a previous Cochrane meta-analysis of small trials reported a non-significant reduction in reoperations with a sliding hip screw compared with multiple cancellous screws (relative risk 0·86, 95% CI 0·70–1·05), and that the risk of avascular necrosis with a sliding hip screw was significantly reduced. The small sample sizes and resultant imprecise estimates,

and methodological limitations, of previous work left the issue in doubt.

Added value of this studyOur trial enrolled more than 1000 patients across multiple countries, providing improved precision and generalisability to both high-income countries and low-to-middle-income countries. Our trial also addresses the dearth of health-related quality-of-life data for patients who had surgical fixation for femoral neck fractures.

Implications of all the available evidenceAlthough our findings, consistent with those of previous randomised controlled trials, did not show a difference in the likelihood of reoperation in patients randomly assigned to a sliding hip screw versus cancellous screws, health-related quality of life was similar across both interventions. However, we did show a significant increase in the likelihood of avascular necrosis with a sliding hip screw compared with cancellous screws. Nonetheless, this finding is not only inconsistent with previous results, but also did not result in an overall increase in reoperations or a decrement in health-related quality of life. Moreover, our results raise the possibility that a sliding hip screw, relative to cancellous screws, might reduce reoperations in patients with displaced fractures, in smokers, and in patients with base of neck fractures.

Articles


pain, treat infection, or improve function. An independent Central Adjudication Committee, adjudicated all primary and key secondary outcomes (mortality, fracture healing, and fracture complications, including avascular necrosis, non-union, implant failure, and infections). Health-related quality of life was measured by the Short Form-12, the EuroQol-5 Dimensions, and the Western Ontario and McMaster Universities Arthritis Index. Details of the primary outcomes, secondary outcomes, and adjudication processes are provided in the appendix.

All adverse events were reported to local ethics boards as per local requirements. An independent Data and Safety Monitoring Board monitored the safety of the trial and reviewed all serious adverse events.

Statistical analysisThe sample size calculation was based on the primary outcome of hip reoperation and used the Cox proportional hazards model as described by Collett.10 Originally we determined that enrolment of 1500 patients would give the trial 81·5% power to detect a hazard ratio (HR) of 0·75 in the sliding hip screw group, at a two-sided alpha level of 0·05, on the assumption that the percentage of the primary outcome in the cancellous screws group would be 25% (appendix).8

The Data and Safety Monitoring Board met in January, 2014, after 589 patients completed follow-up. They provided the Steering Committee with overall event rates to inform a revised power analysis that balanced feasibility of completing recruitment within an acceptable timeframe and supporting plausible hypotheses of treatment effect and baseline event rates. On the basis of the number of events in the first 589 patients, we estimated a 24 month primary event rate of 27·2%; a 24 month mortality rate of 18·2%; a 24 month 5·9% incidence of loss to follow-up; and a combined 6·8% crossover rate. With these estimates, a sample size of 500 patients per group would provide 95·7% power to detect a relative risk reduction of 35%. On the basis of these data, we targeted a sample size of 1100 patients. A methods paper8 published previously presents details of the sample size calculations and rationale.

Analyses followed the intention-to-treat principle and included all patients in the groups to which they were randomly assigned. Patients who did not complete the 24 month follow-up were censored at their last documented follow-up. After assessment of its appropriateness, the primary analysis used a Cox proportional hazards model stratified by clinical site. We report the treatment effects as HRs and 95% CIs. An analysis adjusting for death as a competing risk provided a sensitivity analysis. The analyses of treatment effect on fracture-related adverse events and mortality also relied on a stratified Cox proportional hazards regression.

We analysed the health-related quality-of-life outcomes at 24 months using a multiple linear regression model with treatment and pre-injury quality of life (obtained at

1 week) included as independent variables. Results are reported as mean differences with corresponding 95% CIs and p values.

At the trial onset we specified a single a-priori subgroup analysis investigating fracture displacement as a possible effect modifier, anticipating that sliding hip screw relative to multiple cancellous screws would do better in displaced versus non-displaced fractures.10 At the completion of the trial, but before unblinding and as described in our statistical analysis plan, we prespecified an additional five subgroup analyses that investigated a possible effect modification by location of fracture line, body-mass index, verticality of the fracture line, smoking status, and quality of fracture reduction. We did an additional post-hoc subgroup analysis assessing the possible effect

Figure 1: Trial profile

7306 patients screened

5463 ineligible

1843 eligible

1108 randomised

557 allocated to sliding hip screw 551 allocated to cancellous screws

735 excluded 146 missed (potentially eligible patients who were not screened before study surgery) 589 did not consent

15 ineligible per Adjudication Committee 9 ineligible fractures 4 had surgical treatment delayed beyond 4 days 2 not aged ≥50 years

14 ineligible per Adjudication Committee 6 ineligible fractures 2 had surgical treatment delayed beyond 4 days 2 had other major trauma 1 had surgical treatment delayed beyond 7 days 1 did not have anticipated medical optimisation 1 deemed to not have a low energy fracture by the attending surgeon 1 had retained hardware around the hip (ie, soft tissue or bone)

542 included in primary analysis 537 included in primary analysis

424 completed 24 month follow-up 420 completed 24 month follow-up

118 did not complete follow-up 73 died 23 could not be located 19 withdrew consent 3 other

117 did not complete follow-up 83 died 23 could not be located 9 withdrew consent 2 other

Articles


modification of patient age. We did tests of interaction for individual subgroups and, when three provide significant results, an analysis that simultaneously considered all their possible interactions. We used multiple criteria to consider the credibility of any possible subgroup effects.11,12 Details for postulated subgroup effects are provided in the

appendix. All analyses were done with SAS version 9.4 (Cary, NC, USA). This study was registered with ClinicalTrials.gov, number NCT00761813.

Role of the funding sourceThe funder of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication.

ResultsBetween March 3, 2008, and March 31, 2014, we randomly assigned 1108 patients to receive a sliding hip screw (n=557) or cancellous screws (n=551; figure 1). The Adjudication Committee determined that 29 patients were ineligible, most as a result of an ineligible fracture type, or surgical treatment delayed beyond 4 days in patients with displaced fractures, leaving 1079 patients in the final analyses. The rationale for post-randomisation exclusions is in the appendix. Of the 923 patients alive at 24 months, we achieved 24 month follow-up for 844 (91%;

Sliding hip screw (n=542)

Cancellous screws (n=537)

Total (n=1079)

Age (years) 72·2 (12·0) 72·0 (12·3) 72·1 (12·2)

Sex

Male 212/535 (40%) 210/535 (39%) 422/1070 (39%)

Female 323/535 (60%) 325/535 (61%) 648/1070 (61%)

Ethnic origin

Native 1/533 (<1%) 3/535 (1%) 4/1068 (<1%)

South Asian 65/533 (12%) 65/535 (12%) 130/1068 (12%)

East Asian 6/533 (1%) 4/535 (1%) 10/1068 (1%)

Black 22/533 (4%) 18/535 (3%) 40/1068 (4%)

Hispanic 3/533 (1%) 1/535 (<1%) 4/1068 (<1%)

White 436/533 (82%) 444/535 (83%) 880/1068 (82%)

Smoking history

Never smoked 268/533 (50%) 276/532 (52%) 544/1065 (51%)

Current smoker 101/533 (19%) 100/532 (19%) 201/1065 (19%)

Former smoker 164/533 (31%) 156/532 (29%) 320/1065 (30%)

Current drugs

None 170/535 (32%) 179/534 (34%) 349/1069 (33%)

NSAIDS 86/535 (16%) 64/534 (12%) 150/1069 (14%)

General cardiac 167/535 (31%) 167/534 (31%) 334/1069 (31%)

Opioid analgesics 43/535 (8%) 56/534 (10%) 99/1069 (9%)

Pulmonary drugs 58/535 (11%) 69/534 (13%) 127/1069 (12%)

Anti-hypertension drugs 244/535 (46%) 252/534 (47%) 496/1069 (46%)

Osteoporosis drugs 67/535 (13%) 73/534 (14%) 140/1069 (13%)

BMI

Underweight (BMI <18·5) 37/530 (7%) 33/528 (6%) 70/1058 (7%)

Normal weight (18·5–24·9) 276/530 (52%) 300/528 (57%) 576/1058 (54%)

Overweight (25–29·9) 159/530 (30%) 148/528 (28%) 307/1058 (29%)

Obese (30–39·9) 58/530 (11%) 47/528 (9%) 105/1058 (10%)

Fractured hip N=535 N=535 N=1070

Left 280/535 (52%) 281/535 (53%) 561/1070 (52%)

Right 255/535 (48%) 254/535 (47%) 509/1070 (48%)

Mechanism of injury

Fall 515/533 (97%) 521/534 (98%) 1036/1067 (97%)

Spontaneous 13/533 (2%) 6/534 (1%) 19/1067 (2%)

Other low energy trauma 5/533 (1%) 7/534 (1%) 12/1067 (1%)

History of surgery to affected hip

Yes 3/535 (1%) 0/535 (0%) 3/1070 (<1%)

No 532/535 (99%) 535/535 (100%) 1067/1070 (100%)

Additional injuries

Yes 67/535 (13%) 72/535 (13%) 139/1070 (13%)

No 468/535 (87%) 463/535 (87%) 931/1070 (87%)

Data are mean (SD) or n/N (%). For some subgroups, the numbers of patients analysed are smaller than that of the overall group number because data are missing for some variables. NSAIDS=non-steroidal anti-inflammatory drugs. BMI=body-mass index.

Table 1: Baseline characteristics



Total (n=1079)

Level of the fracture line

Subcapital 331/535 (62%) 351/536 (65%) 682/1071 (64%)

Midcervical 159/535 (30%) 154/536 (29%) 313/1071 (29%)

Basal 45/535 (8%) 31/536 (6%) 76/1071 (7%)

Garden classification13

Undisplaced 360/542 (66%) 369/537 (69%) 729/1079 (68%)

Garden I 257/542 (47%) 277/537 (52%) 534/1079 (49%)

Garden II 99/542 (18%) 92/537 (17%) 191/1079 (18%)

Displaced 182/542 (34%) 168/537 (31%) 350/1079 (32%)

Garden III 121/542 (22%) 128/537 (24%) 249/1079 (23%)

Garden IV 58/542 (11%) 39/537 (7%) 97/1079 (9%)

Pauwel’s classification14,15

Type I 59/535 (11%) 59/536 (11%) 118/1071 (11%)

Type II 398/535 (74%) 394/536 (74%) 792/1071 (74%)

Type III 78/535 (15%) 83/536 (15%) 161/1071 (15%)

Preoperative traction

Skin traction 75/535 (14%) 76/535 (14%) 151/1070 (14%)

Skeletal traction 7/535 (1%) 3/535 (1%) 10/1070 (1%)

None 453/535 (85%) 456/535 (85%) 909/1070 (85%)

Type of reduction

None 210/531 (40%) 237/528 (45%) 447/1059 (42%)

Closed 287/531 (54%) 277/528 (52%) 564/1059 (53%)

Acceptable 286/287 (100%) 275/277 (99%) 561/564 (99%)

Unacceptable 1/287 (<1%) 2/277 (1%) 3/564 (1%)

Open 34/531 (6%) 14/528 (3%) 48/1059 (5%)

Acceptable 32/34 (94%) 14/14 (100%) 46/48 (96%)

Unacceptable 2/34 (6%) 0/14 (0%) 2/48 (4%)

Data are n/N (%).

Table 2: Fracture characteristics

Articles


figure 1, appendix). The mean length of follow-up was 633 days (SD 208).

Typical patients were women aged 70–80 years who had fallen and sustained an isolated, non-displaced fracture of the femoral neck; group characteristics were similar (tables 1, 2). Acceptable reduction (open and closed) was achieved in nearly all patients (99%; table 2). In 549 patients with displaced fractures, acceptable reduction was achieved in 545 (99%). Those who underwent cancellous screw fixation typically received three parallel screws in a triangular configuration, whereas those in the sliding hip screw group received a single large compression screw in the centre–centre head position with a two-hole side plate; perioperative management was similar across groups (tables 1, 2; appendix).

The overall crossover was 2·0%, with significant differential crossovers between treatment groups (p=0·03). Surgeon compliance for the initial allocated surgical fixation approach was 97% (n=16) for sliding hip screw and 99% (n=6) for cancellous screws (appendix).

The primary study endpoint, hip reoperation within 24 months, did not differ by type of surgical fixation (table 3, figure 2). A competing risk sensitivity analysis adjusting for death yielded similar results for the primary endpoint (0·89, 0·69–1·16).

Number of deaths did not differ between groups (table 3). Avascular necrosis occurred in about 7% of patients overall and differed by fixation group, with more occurring in the sliding hip screw group than in the cancellous screws group. Of these, 54 (69%) patients required an operation: 38 patients in the sliding hip screw group and 16 in the cancellous screws group (p=0·002). Implant removal took place significantly less frequently in the sliding hip screw group than in the cancellous screws group (table 3). Implant exchange to revise to another internal fixation also occurred less frequently with sliding hip screws than cancellous screws (table 3). Alternatively, implant exchange to a total hip replacement was more common in the sliding hip group.

Non-unions, implant failures, infections, fracture shortening, and fracture healing did not differ by surgical fixation approach (table 3). Health-related quality of life did not differ between patients assigned to sliding hip screws and multiple cancellous screws at 12 month and 24 month follow-up (table 4).

Medically related adverse events did not differ by treatment group (p=0·82; table 5).

Subgroup analyses favoured sliding hip screw in patients with displaced fractures, fractures at the base of the femoral neck, and in those who were current smokers (appendix, figure 3). Only smoking status remained significant when these three subgroups were entered into a single analysis (figure 3). Sliding hip screw was superior in current smokers, but not in former or non-smokers (figure 3, appendix).

Overall (n=1079)



Hazard ratio (95% CI)

p value

Reoperation

Any reoperation 224 (21%) 107 (20%) 117 (22%) 0·83 (0·63–1·09) 0·18

Implant removal 74 (7%) 25 (5%) 49 (9%) 0·42 (0·25–0·70) 0·0009

Implant exchange: total hip arthroplasty

104 (10%) 64 (12%) 40 (7%) 1·51 (1·00–2·27) 0·0494

Implant exchange: hemiarthroplasty

55 (5%) 26 (5%) 29 (5%) 0·89 (0·52–1·51) 0·66

Implant exchange: internal 16 (1%) 2 (<1%) 14 (3%) 0·14 (0·03–0·62) 0·0024

Fixation

Implant exchange: spacer 3 (<1%) 1 (<1%) 2 (<1%) 0·50 (0·05–5·45) 0·56

Soft tissue procedure 6 (1%) 4 (1%) 2 (<1%) 1·98 (0·36–10·77) 0·42

Proximal femoral osteotomy 1 (<1%) 1 (<1%) 1 (<1%) 0·99 (0·06–15·80) 0·99

Avascular necrosis 78 (7%) 50 (9%) 28 (5%) 1·91 (1·06–3·44) 0·0319

Non-union 66 (6%) 33 (6%) 33 (6%) 0·92 (0·48–1·75) 0·80

Implant failure 87 (8%) 42 (8%) 45 (8%) 0·95 (0·61–1·48) 0·81

Infection

Any infection 19 (2%) 10 (2%) 9 (2%) 1·10 (0·45–2·69) 0·83

Superficial infection 8 (1%) 4 (1%) 4 (1%) 0·99 (0·25–3·94) 0·99

Deep infection 11 (1%) 6 (1%) 5 (1%) 1·19 (0·37–3·87) 0·77

Fracture healing*

Healed by month 24 532/795 (67%)

262/398 (66%)

270/397 (68%)

·· 0·71

Not healed by month 24 3/795 (<1%)

2/398 (1%)

1/397 (<1%)

··

Not healed at time of last visit 260/795 (33%)

134/398 (34%)

126/397 (32%)

··

Fracture shortening >5 mm (n=532)†

146/532 (27%)

69/262 (26%)

77/270 (29%)

0·92 (0·70–1·22) 0·57

Mortality 156 (14%) 73 (13%) 83 (15%) 0·81 (0·58–1·12) 0·20

Data are n (%). Relative risk was calculated where the total number of events is less than 50. *795 patients were included in the fracture healing analysis. 284 patients did not have radiograph available for fracture healing adjudication, and therefore were not included in the denominator. †532 patients were included in the shortening analysis on the basis of the number of healed fractures with shortening data.

Table 3: Study outcomes by treatment group

Figure 2: Kaplan-Meier curves for reoperation

Number at riskCancellous screws

Sliding hip screw

0 180

537542

437444

360

378388

540

315341

730

273285

Time (days)

Log-rank p=0·22570

70

80

90

100

Prob

abili

ty (%

)

Treatment Cancellous screws Sliding hip screw Censored

Articles


DiscussionWe have shown a similar risk of hip reoperation in patients with low-energy femoral neck fractures randomly assigned to sliding hip screw as in those assigned to cancellous screws at 24 months; avascular necrosis occurred more frequently in patients allocated to sliding hip screw. Subgroup analyses of low-to-moderate credibility, suggested sliding hip screws reduced reoperations in patients with displaced fractures, fractures at the base of the femoral neck, and in current smokers.

Although the frequency of reoperations was similar between the two treatment groups, there were differences in the component outcomes of reoperations. Patients in the sliding hip screw group, compared with those in the

cancellous screws group, had a lower frequency of a reoperation for an implant removal and an implant exchange based on an internal fixation approach, but had a higher frequency of reoperation for an implant exchange with a total hip arthroplasty approach. Patients undergoing total hip arthroplasty after failed internal fixation compared with those undergoing primary total hip arthroplasty could be at higher risk of complications.19 As such, the number of arthroplasties in the cancellous screws group suggests a potential advantage of the treatment over a sliding hip screw. Although the decision to choose one approach to implant exchange over another was left to the surgeon’s discretion, the choice might also indicate surgeon preference.

Looking separately at possible effect modifiers, displaced versus non-displaced fractures, fracture site, and smoking status, interaction p values all reached conventional statistical significance, all suggesting benefits of sliding hip screws over cancellous screws in a subpopulation of patients (figure 3). All were a-priori hypotheses with a biological rationale that led to a correct predicted direction of effect. In particular, the greater biomechanical stability of sliding hip screws might offer advantages in fracture with displacement and in smokers, who have greater risk of osteoporosis and diminished bone density than do non-smokers.7,20–23 Sliding hip screws remain the standard of care in patients with inter-trochanteric fractures, a region in close proximity to the base of the femoral neck,24 providing a biological rationale for the finding that sliding hip screws reduced reoperations in the subgroup of patients with base of femoral neck fractures. Furthermore, the displacement hypothesis was originally our sole hypothesis and a minimisation variable.11,12

However, when all three potential effect modifiers were considered together, only smoking retained a low, although not extremely low, p value (figure 3). Additionally, we tested multiple hypotheses, and did not find similar effects in health-related quality of life. As such, in view of all these issues, the apparent subgroup effects have only modest credibility.

Strengths of FAITH include safeguards against risk of bias (concealed randomisation, centralised and inde-pendent outcome adjudication, blinded analysis of data); high compliance with study procedures; broad inclusion criteria with a large number of centres in countries with diverse health-care systems; a focus on outcomes of importance to both patients and the health-care system (ie, reoperation, health-related quality of life); and rigorous investigation of subgroup effects, with due attention to their credibility.11,12

Our study has limitations. Surgeons and patients were not blinded. However, we did minimise the associated risk of bias with central and independent, although unblinded, radiographic adjudication of the primary endpoint. Furthermore, reoperation is an objective endpoint and a major procedure; surgeons will seldom decide to reoperate



p value

Pneumonia 23 (4%) 24 (4%) 0·86

Deep vein thrombosis 6 (1%) 8 (1%) 0·58

Pulmonary embolism 2 (<1%) 4 (1%) 0·41

Sepsis 7 (1%) 6 (1%) 0·79

Myocardial infarction 9 (2%) 7 (1%) 0·63

Congestive heart failure 7 (1%) 6 (1%) 0·79

Other cardiovascular adverse events

43 (8%) 43 (8%) 0·96

Pulmonary 24 (4%) 16 (3%) 0·21

Decreased cognitive ability 14 (3%) 23 (4%) 0·12

Neurological 10 (2%) 10 (2%) 0·98

Digestive 17 (3%) 18 (3%) 0·84

Blood 17 (3%) 12 (2%) 0·36

Renal 8 (1%) 10 (2%) 0·62

Urinary 23 (4%) 20 (4%) 0·66

Multiple organ complications 3 (1%) 0 0·08

Total 122 (23%) 124 (23%) 0·82

Table 5: Medically related adverse events

Sliding hip screw Cancellous screws n Adjusted mean difference (95% CI)

p value for differences between groups

n Mean (SD) n Mean (SD)

12 months

SF-12 PCS16 235 40·8 (11·1) 224 41·9 (10·7) 435 –0·02 (–1·79 to 1·74) 0·98

WOMAC17 240 44·69 (19·08) 226 41·32 (16·73) 438 1·98 (–1·13 to 5·09) 0·21

EQ-5D Index18

249 0·77 (0·20) 238 0·80 (0·17) 460 –0·02 (–0·05 to 0·02) 0·33

24 months

SF-12 PCS16 207 41·6 (10·9) 181 41·4 (11·8) 358 0·50 (–1·61 to 2·61) 0·64

WOMAC17 205 40·97 (16·33) 183 39·75 (17·09) 355 0·35 (–3·03 to 3·74) 0·84

EQ-5D Index18

232 0·79 (0·19) 207 0·80 (0·19) 406 –0·01 (–0·04 to 0·02) 0·51

SF-12 PCS=Short Form-12. PCS=Physical component score. WOMAC=Western Ontario and McMaster Universities Arthritis Index. EQ-5D=EuroQol-5 Dimensions.

Table 4: Health-related quality of life by treatment groups without interaction of displacement

Articles


in the absence of a compelling indication. Follow-up at 24 months was less than complete (844 [91%] of 923 patients who were alive); our success in following up patients was consistent and in most cases better than in previous smaller trials.3 Unavoidable heterogeneity related to the variables that were not standardised also existed; these variables were patient positioning, fracture reduction, surgical exposure, use of operative traction, surgical delay, type of anaesthetic, physiotherapy, and rehabilitation programmes.

In relation to previous research, a meta-analysis of small trials suggested a non-significant difference in reoperations favouring sliding hip screw (relative risk 0·86, 95% CI 0·70–1·05; p=0·13).19 An updated pooled analysis of reoperation including small trials with our FAITH results (eight trials, 1913 patients) reported a 95% CI that was narrower than our study’s but was still consistent, with no difference between fixation methods (0·91, 0·76–1·08; p=0·27, I²=7%).25–31

With respect to avascular necrosis, our results differ importantly from a previous systematic review19 of small trials, which suggested that a sliding hip screw, in comparison with cancellous screws, might reduce the risk of avascular necrosis. The addition of another small trial to a previous meta-analysis results in a significant reduction in the risk of avascular necrosis with sliding hip screws (77 events; relative risk 0·64, 95% CI 0·43–0·97; p=0·04, I²=0%).19,25

Despite the previous results, increased risk of avascular necrosis from our results has a plausible biological rationale. Hip fractures can disrupt the retinacular vessels, which are crucial for the vascular supply of the femoral head.32 A randomised trial33 of 104 patients with femoral neck fractures using bone scintigraphy showed reduced vascularity in patients receiving a sliding hip screw compared with those receiving cancellous screws (35% vs 11%, p<0·01). Furthermore, suboptimum positioning of large implants, such as sliding hip screws, risks damage to the blood supply to the femoral head.34 In terms of the importance of avascular necrosis, observational studies have shown that many patients remain asymptomatic, with only one in five requiring further surgery.35,36

Our results, in the context of previous results, suggest that the choice of procedure for internal fixation is a matter of discretion. Findings favouring cancellous screws were the non-significant difference in the likelihood of reoperation between procedures, and the higher likelihood of avascular necrosis and subsequent total arthroplasties in patients receiving a sliding hip screw. However, our finding of increased avascular necrosis with a sliding hip screw, compared with cancellous screws, is inconsistent with other studies and did not result in more operations or poorer quality of life in the total population. Moreover, findings across all trials remain consistent, with overall decreased reoperations in

Figure 3: Subgroup analyses of surgical fixation primary endpoint (reoperation)BMI=body-mass index.*Interaction p value for group comparison. †Defined at study start. ‡Defined at study close (before unblinding).

Hazard ratio (95% CI)

p value*Sliding hip screw

(n/N)

Primary subgroup analysis†Displacement Undisplaced DisplacedSecondary subgroup analysis‡BMI Normal BMI Overweight or obese BMILevel of fracture line Subcapital Midcervical BasalPauwels classification Type I Type II Type IIISmoking Current smoker Former or non-smokerQuality of reduction Acceptable UnacceptableAge (years) <60 >60 to ≤80 >80Overall study results

0·04

0·93

0·04

0·20

0·02

0·62

0·70

58/356 43/179

52/313 48/217

66/331 28/159 5/45

11/59 72/398 18/78

14/101 87/432

12/23 87/437

16/102 56/267 29/166 101/542

59/369 57/167

66/333 46/195

72/351 39/154 5/31

8/59 79/394 29/83

27/100 89/432

15/22 96/429

26/119 60/248 30/168 116/537

1·04 (0·72–1·50) 0·57 (0·38–0·87)

0·85 (0·59–1·24) 0·83 (0·55–1·27)

1·05 (0·75–1·48) 0·62 (0·37–1·03) 0·24 (0·06–0·93)

1·52 (0·59–3·94) 0·87 (0·62–1·20) 0·55 (0·29–1·04)

0·39 (0·19–0·77) 0·99 (0·73–1·33)

0·86 (0·64–1·15) 0·53 (0·23–1·59)

0·76 (0·39–1·47) 0·78 (0·54–1·14) 1·01 (0·60–1·71) 0·83 (0·63–1·09)

Cancellous screws

(n/N)

0·10·05 1·00·5

Favours cancellous screwsFavours sliding hip screw

Articles


patients receiving a sliding hip screw, and subgroup analyses of low credibility suggesting that patients with displaced fractures, smokers, and patients with base of neck fractures might do better with a sliding screw than with cancellous screws.ContributorsWriting committee: Mohit Bhandari (Chair, McMaster University), P J Devereaux (McMaster University), Gordon Guyatt (McMaster University), Lehana Thabane (McMaster University), Stephen D Walter (McMaster University), Martin J Heetveld (Spaarne Gasthuis, Haarlem), Kyle J Jeray (Greenville Health System), Susan Liew (The Alfred), Emil H Schemitsch (University of Western Ontario), Paul Tornetta III (Boston University Medical Center), Gregory J Della Rocca (Duke University), Richard E Buckley (Foothills Medical Centre), Robert McCormack (Royal Columbian Hospital/Fraser Health Authority/University of British Columbia), Todd M Oliver (Boone Hospital Center—Columbia Orthopaedic Group), Michiel J M Segers (St Antonius Ziekenhuis), Amar Rangan (The James Cook University Hospital), Martin Richardson (University of Melbourne), Sheila Sprague (McMaster University), Gerard P Slobogean (University of Maryland, Baltimore), Taryn Scott (McMaster University), Julie Agel (University of Minnesota), Alisha Garibaldi (McMaster University), Qi Zhou (McMaster University), Diane Heels-Ansdell (McMaster University), Helena Viveiros (McMaster University), Stephanie M Zielinski (Erasmus MC, University Medical Center Rotterdam), Esther M M Van Lieshout (Erasmus MC, University Medical Center Rotterdam), Herman Johal (McMaster University), Birgit C Hanusch (The James Cook University Hospital), and Marc Swiontkowski (Co-Chair, University of Minnesota)

The Clinical Advances through Research and Information Translation (CLARITY) Research Group at McMaster University coordinated the trial. The CLARITY Research Group was responsible for the trial randomisation, maintenance of the database, data validation, data analyses, and study-centre coordination. The University of Minnesota (Minneapolis, MN, USA) assisted in coordination of sites in the USA; Erasmus MC, University Medical Centre (Rotterdam, Netherlands) assisted in coordination of sites in the Netherlands; and James Cook University Hospital (North Yorkshire, UK) assisted in coordination of sites in the UK. The Steering Committee designed the trial, prespecified the statistical analysis plan, and vouch for the completeness and accuracy of the data and analyses. MB, the first author and chair of the writing committee, wrote the first draft of the manuscript; the writing committee made revisions.

Declaration of interestsMB reports grants from Canadian Institutes of Health Research, the National Institutes of Health, Stichting NutsOhra, The Netherlands Organisation for Health Research and Development, Physicians’ Services Incorporated and Stryker GmBH, and the Early Research Award Program during the conduct of the study; and personal fees from Sanofi, Pendopharm, Ferring, DJO, and Stryker, and grants from Pluristem and Amgen, outside the submitted work. MJH reports grants from Stichting NutsOhra and The Netherlands Organization for Health Research and Development during the conduct of the study. PJD reports grants from Abbott Diagnostics, Boehringer Ingelheim, Covidien, Octopharma, Roche Diagnostics, and Stryker, outside the submitted work. KJJ reports grants from the National Institutes of Health and Canadian Institutes of Health Research during the conduct of the study; and personal fees from Zimmer, outside the submitted work. EHS reports personal fees from Stryker, Smith & Nephew, Zimmer, Acumed, Amgen, Sanofi, and Bioventus, and other from Elsevier, outside the submitted work. PT III reports personal fees from Lippicott Williams & Wilkins, outside the submitted work, and has a patent (US8888824 B2) with royalties paid by Smith & Nephew. GJDR reports grants from the University of Minnesota during the conduct of the study; and grants from DePuy-Synthes, personal fees from Bioventus and DePuy-Synthes, other from Amedica, The Orthopaedic Implant Company, LuminCare, Mergenet Medical, and Intellectual ventures, outside the submitted work. RM reports personal fees from Sanofi and Pendopharm, outside the submitted work. AR reports grants from the Canadian Institutes of Health Research, during the conduct of the study; and grants from the National Institute

for Health Research (UK), Orthopaedic Research UK, and DePuy, outside the submitted work. GPS reports personal fees from Zimmer Biomet and grants from the Canadian Institutes of Health Research, outside the submitted work. SS reports other from McMaster University and Global Research Solutions, and personal fees from the University of Maryland, outside the submitted work. JA reports grants from the National Institutes of Health during the conduct of the study. SMZ reports grants from Stichting NutsOhra (SNO-T-0602-43) and The Netherlands Organisation for Health Research and Development (80-82310-97-11032), during the conduct of the study. EMMVL reports grants from Stichting NutsOhra and The Netherlands Organization for Health Research and Development during the conduct of the study. MS reports grants from the National Institutes of Health/National Institute of Arthritis and Musculoskeletal and Skin Diseases, during the conduct of the study; and other from The Journal of Bone & Joint Surgery Editor, outside the submitted work. MJMS reports personal fees from DePuy Synthes and AO Trauma Europe, outside the submitted work. GG, LT, SDW, SL, REB, TMO, MR, TS, AG, QZ, DH-A, HV, HJ, and BCH declare no competing interests.

AcknowledgmentsThis study was funded by research grants from the Canadian Institutes of Health Research (MOP-106630 and MCT-87771), National Institutes of Health (1R01AR055267-01A1), Stichting NutsOhra (SNO-T-0602-43), The Netherlands Organisation for Health Research and Development (80-82310-97-11032), Physicians’ Services Incorporated, and Stryker GmBH. MB was also funded, in part, through the Early Research Award Program, which provided funding for the present study, and by a Canada research chair in musculoskeletal trauma, which is unrelated to the present study (McMaster University, Hamilton, ON, Canada). Research reported in this publication was supported by the National Institute of Arthritis and Musculoskeletal and Skin Diseases of the National Institutes of Health under Award Number R01AR055267-01A1. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Research reported in this publication was also supported by The County Durham & Tees Valley Comprehensive Local Research Network, which operates as part of the National Institute for Health Research Comprehensive Clinical Research Network in England.

References1 Gullberg B, Johnell O, Kanis JA. World-wide projections for hip

fracture. Osteoporos Int 1997; 7: 407–13.2 Parker M, Johansen A. Hip fracture. BMJ 2006; 333: 27–30.3 Bhandari M, Devereaux PJ, Swiontkowski MF, et al. Internal fixation

compared with arthroplasty for displaced fractures of the femoral neck. A meta-analysis. J Bone Joint Surg Am 2003; 85–A: 1673–81.

4 Mundi S, Pindiprolu B, Simunovic N, Bhandari M. Similar mortality rates in hip fracture patients over the past 31 years. Acta Orthop 2014; 85: 54–59.

5 Zielinski SM, Bouwmans CA, Heetveld MJ, et al. The societal costs of femoral neck fracture patients treated with internal fixation. Osteoporos Int 2014; 25: 875–85.

6 Bhandari M, Devereaux PJ, Tornetta P 3rd, et al. Operative management of displaced femoral neck fractures in elderly patients. An international survey. J Bone Joint Surg Am 2005; 87: 2122–30.

7 Hoshino CM, O’Toole RV. Fixed angle devices versus multiple cancellous screws: what does the evidence tell us? Injury 2015; 46: 474–77.

8 FAITH Investigators. Fixation using alternative implants for the treatment of hip fractures (FAITH): design and rationale for a multi-centre randomized trial comparing sliding hip screws and cancellous screws on revision surgery rates and quality of life in the treatment of femoral neck fractures. BMC Musculoskelet Disord 2014; 15: 219.

9 Scott NW, McPherson GC, Ramsay CR, Campbell MK. The method of minimization for allocation to clinical trials. A review. Control Clin Trials 2002; 23: 662–74.

10 Collett D. Modelling survival data in medical research. 1st edn. London, UK: Chapman and Hall, 1994.

11 Sun X, Briel M, Busse JW, et al. Credibility of claims of subgroup effects in randomised controlled trials: systematic review. BMJ 2012; 344: e1553.

Articles


12 Study to Prospectively Evaluate Reamed Intramedullary Nails in Tibial Fractures (SPRINT) Investigators, Sun X, Heels-Ansdell D, et al. Is a subgroup claim believable? A user’s guide to subgroup analyses in the surgical literature. J Bone Joint Surg Am 2011; 93: e8.

13 Garden RS. Low-angle fixation in fractures of the femoral neck. J Bone Joint Surg 1961; 43B: 647.

14 Pauwels F. Der Schenkelhalsbruch. Ein mechaizisches Problem. Stuttgart: F Enke, 1935.

15 Bartonícek J. Pauwels’ classification of femoral neck fractures: correct interpretation of the original. J Orthop Trauma 2001; 15: 358–60.

16 McHorney CA, Ware JE Jr, Rogers W, et al. The validity and relative precision of MOS short- and long-form health status scales and Dartmouth COOP charts. Results from the Medical Outcomes Study. Med Care 1992; 30: MS253–65.

17 Bellamy N, Buchanan WW, Goldsmith CH, et al. Validation study of WOMAC: a health status instrument for measuring clinically important patient relevant outcomes to antirheumatic drug therapy in patients with osteoarthritis of the hip or knee. J Rheumatol 1988; 15: 1833–40.

18 Brooks R, Rabin R, de Charro F. The measurement and valuation of health status using EQ-5D: a European perspective. Dordrecht, Netherlands: Kluwer Academic Publishers, 2003.

19 Parker MJ, Gurusamy KS. Internal fixation implants for intracapsular proximal femoral fractures in adults. Cochrane Database Syst Rev 2001; 4: CD001467.

20 Deneka DA, Simonian PT, Stankewich CJ, Eckert D, Chapman JR, Tencer AF. Biomechanical comparison of internal fixation techniques for the treatment of unstable basicervical femoral neck fractures. J Orthop Trauma 1997; 11: 337–43.

21 Blair B, Koval KJ, Kummer F, Zuckerman JD. Basicervical fractures of the proximal femur. A biomechanical study of 3 internal fixation techniques. Clin Orthop Relat Res 1994; 306: 256–63.

22 Law MR, Hackshaw AK. A meta-analysis of cigarette smoking, bone mineral density and risk of hip fracture: recognition of a major effect. BMJ 1997; 315: 841–46.

23 Castillo RC, Bosse MJ, MacKenzie EJ, Patterson BM, LEAP Study Group. Impact of smoking on fracture healing and risk of complications in limb-threatening open tibia fractures. J Orthop Trauma 2005; 19: 151–57.

24 Parker MJ, Handoll HH. Gamma and other cephalocondylic intramedullary nails versus extramedullary implants for extracapsular hip fractures in adults. Cochrane Database Syst Rev 2010; 9: CD000093.

25 Watson A, Zhang Y, Beattie S, Page RS. Prospective randomized controlled trial comparing dynamic hip screw and screw fixation for undisplaced subcapital hip fractures. ANZ J Surg 2013; 83: 679–83.

26 Benterud JG, Husby T, Nordsletten L, Alho A. Fixation of displaced femoral neck fractures with a sliding screw plate and a cancellous screw or two Olmed screws. A prospective, randomized study of 225 elderly patients with a 3-year follow-up. Ann Chir Gynaecol 1997; 86: 338–42.

27 Kuokkanen H, Korkala O, Antti-Poika I, Tolonen J, Lehtimäki MY, Silvennoinen T. Three cancellous bone screws versus a screw-angle plate in the treatment of Garden I and II fractures of the femoral neck. Acta Orthop Belg 1991; 57: 53–57.

28 Madsen F, Linde F, Andersen E, Birke H, Hvass I, Poulsen TD. Fixation of displaced femoral neck fractures. A comparison between sliding screw plate and four cancellous bone screws. Acta Orthop Scand 1987; 58: 212–16.

29 Harper WM. A prospective randomised trial comparing multiple parallel cannulated screws with a sliding hip screw and plate. In: Treatment of intracapsular proximal femoral fractures. Doctoral thesis, University of Leicester, 1994.

30 Sørensen JL, Varmarken JE, Bømler J. Internal fixation of femoral neck fractures. Dynamic hip and Gouffon screws compared in 73 patients. Acta Orthop Scand 1992; 63: 288–92.

31 Paus A, Gjengedal E, Hareide A, Jørgensen JJ. Dislocated fractures of the femoral neck treated with von Bahr screws or hip compression screw. Results of a prospective, randomized study. J Oslo City Hosp 1986; 36: 55–61.

32 Swiontkowksi MF. Intracapsular fractures of the hip. J Bone Joint Surg 1994; 76A: 129–38.

33 Linde F, Andersen E, Hvass I, Madsen F, Pallesen R. Avascular femoral head necrosis following fracture fixation. Injury 1986; 17: 159–63.

34 Brodetti A. The blood supply of the femoral neck and head in relation to the damaging effects of nails and screws. J Bone Joint Surg 1960; 42B: 794–801.

35 Barnes R, Brown JT, Garden RS, Nicoll EA. Subcapital fractures of the femur. A prospective review. J Bone Joint Surg Br 1976; 58: 2–24.

36 Nikolopoulos KE, Papadakis SA, Kateros KT, et al. Long-term outcome of patients with avascular necrosis, after internal fixation of femoral neck fractures. Injury 2003; 34: 525–28.

Supplementary appendixThis appendix formed part of the original submission and has been peer reviewed. We post it as supplied by the authors.

Supplement to: Fixation using Alternative Implants for the Treatment of Hip fractures (FAITH) Investigators. Fracture fixation in the operative management of hip fractures (FAITH): an international, multicentre, randomised controlled trial. Lancet 2017; published online March 2. http://dx.doi.org/10.1016/S0140-6736(17)30066-1.

Page 1 of 21

SUPPLEMENTARY DOCUMENTS

Prepared on: January 19, 2017

Page 2 of 21

Supplementary Document Table of Contents

1.0 FAITH INVESTIGATORS ..................................................................................................................................... 3

Writing Committee .................................................................................................................................................... 3

Steering Committee ................................................................................................................................................... 3

Global Methods Centre ............................................................................................................................................. 3

United States Methods Centre ................................................................................................................................... 3

Netherlands Method Centre....................................................................................................................................... 3

United Kingdom Method Centre ............................................................................................................................... 3

Adjudication Committee ........................................................................................................................................... 3

Participating Clinical Sites ........................................................................................................................................ 3

Canada .................................................................................................................................................................. 3

United States ......................................................................................................................................................... 4

The Netherlands .................................................................................................................................................... 5

International .......................................................................................................................................................... 5

2.0 FUNDING SOURCES ............................................................................................................................................ 6

3.0 SUPPLEMENTAL APPENDICES ......................................................................................................................... 7

3.1 Eligibility Criteria ............................................................................................................................................... 7

3.2 Trial Interventions and Standardization of Peri-Operative Care ......................................................................... 7

3.3 Threshold Performance for Internal Fixation ...................................................................................................... 8

3.4 Follow-up Processes ............................................................................................................................................ 9

3.5 Outcome Definitions ........................................................................................................................................... 9

3.6 Overview of Adjudication ................................................................................................................................... 9

3.7 Sample Size Calculations .................................................................................................................................. 10

3.8 Subgroup Analyses and Hypothesized Effects .................................................................................................. 11

3.9 Rationale for Excluding Ineligible Patients from Analyses .............................................................................. 11

4.0 SUPPLEMENTARY TABLES ............................................................................................................................. 12

Table S1a: Reasons for Exclusion by Clinical Sites ................................................................................................ 12

Table S1b: Reasons for Exclusion by the Adjudication Committee ....................................................................... 13

Table S2: Surgical and Peri-Operative Management .............................................................................................. 14

Table S3: Cancellous Screw Surgical Management ................................................................................................ 15

Table S4: Sliding Hip Screw Surgical Management ............................................................................................... 16

Table S5: Study Outcomes by Treatment Group for Undisplaced Fractures .......................................................... 17

Table S6: Health-Related Quality of Life by Treatment Groups for Undisplaced Fractures .................................. 18

Table S7: Study Outcomes by Treatment Group for Displaced Fractures .............................................................. 19

Table S8: Health-Related Quality of Life by Treatment Groups For Displacement ............................................... 20

Table S9: Interaction Between Secondary Outcomes and Smoker Status ............................................................... 21

Page 3 of 21

1.0 FAITH INVESTIGATORS

Writing Committee: Mohit Bhandari (Chair, McMaster University), PJ Devereaux (McMaster University), Gordon

Guyatt (McMaster University), Lehana Thabane (McMaster University), Stephen D. Walter (McMaster University),

Martin J. Heetveld (Spaarne Gasthuis, Haarlem), Kyle J. Jeray (Greenville Health System), Susan Liew (The

Alfred), Emil H. Schemitsch (University of Western Ontario), Paul Tornetta III (Boston University Medical

Center), Gregory J. Della Rocca (Duke University), Richard E. Buckley (Foothills Medical Centre), Robert

McCormack (Royal Columbian Hospital/Fraser Health Authority/University of British Columbia), Todd M. Oliver

(Boone Hospital Center – Columbia Orthopaedic Group), Michiel J.M. Segers (St. Antonius Ziekenhuis), Amar

Rangan (The James Cook University Hospital), Martin Richardson (University of Melbourne), Sheila Sprague

(McMaster University), Gerard P. Slobogean (University of Maryland, Baltimore), Taryn Scott (McMaster

University), Julie Agel (University of Minnesota), Alisha Garibaldi (McMaster University), Qi Zhou (McMaster

University), Diane Heels-Ansdell (McMaster University), Helena Viveiros (McMaster University), Stephanie M.

Zielinski (Erasmus MC, University Medical Center Rotterdam), Esther M.M. Van Lieshout (Erasmus MC,

University Medical Center Rotterdam), Herman Johal (McMaster University), Birgit C. Hanusch (The James Cook

University Hospital), and Marc Swiontkowski (Co-Chair, University of Minnesota)

Steering Committee: Mohit Bhandari (Chair, McMaster University), Marc Swiontkowski (University of

Minnesota), PJ Devereaux (McMaster University), Gordon Guyatt (McMaster University), Martin J. Heetveld

(Spaarne Gasthuis, Haarlem), Kyle Jeray (Greenville Health System), Susan Liew (The Alfred), Martin Richardson

(University of Melbourne), Emil H. Schemitsch (University of Western Ontario), Lehana Thabane (McMaster

University), Paul Tornetta III (Boston University Medical Center), and Stephen D. Walter (McMaster University)

Global Methods Centre: Mohit Bhandari (Principal Investigator); Sheila Sprague (Research Methodologist); Paula

McKay (Manager); Taryn Scott, Alisha Garibaldi, Helena Viveiros, Marilyn Swinton, (Research Coordination);

Mark Gichuru, Sofia Bzovsky (Adjudication Coordination); Diane Heels-Ansdell, Qi Zhou (Statistical Analysis);

Lisa Buckingham, Aravin Duraikannan (Data Management); Deborah Maddock, Nicole Simunovic (Grants

Management) (McMaster University)

United States Methods Centre: Marc Swiontkowski (Principal Investigator); Julie Agel (Research Coordination)

(University of Minnesota)

Netherlands Method Centre: Martin J. Heetveld (Principal Investigator); Esther M.M. Van Lieshout (Research

Coordination); Stephanie M. Zielinski (Trial Coordination) (Erasmus MC, University Medical Center Rotterdam)

United Kingdom Method Centre: Amar Rangan (Principal Investigator), Birgit C. Hanusch, Lucksy Kottam,

Rachel Clarkson (Research Coordination) (The James Cook University Hospital)

Adjudication Committee: Gregory J. Della Rocca (Chair) (Duke University), Robert Haverlag (Onze Lieve

Vrouwe Gasthuis), Susan Liew (The Alfred), Gerard P. Slobogean (University of Maryland, Baltimore), Kyle Jeray

(Greenville Health System)

Participating Clinical Sites:

Canada:

Robert McCormack, Kelly Apostle, Dory Boyer, Farhad Moola, Bertrand Perey, Trevor Stone, Darius Viskontas, H.

Michael Lemke, Mauri Zomar, Karyn Moon, Raely Moon, Amber Oatt (Royal Columbian Hospital / Fraser

Health Authority / University of British Columbia); Richard E. Buckley, Paul Duffy, Robert Korley, Shannon

Puloski, James Powell, Kelly Johnston, Kimberly Carcary, Melissa Lorenzo, Ross McKercher (Foothills Medical

Centre); David Sanders, Mark MacLeod, Abdel-Rahman Lawendy, Christina Tieszer (London Health Sciences

Centre); David Stephen, Hans Kreder, Richard Jenkinson, Markku Nousiainen, Terry Axelrod, John Murnaghan,

Diane Nam, Veronica Wadey, Albert Yee, Katrine Milner, Monica Kunz, Wesley Ghent (Sunnybrook Health

Sciences Centre); Emil H. Schemitsch, Michael D. McKee, Jeremy A. Hall, Aaron Nauth, Henry Ahn, Daniel B.

Whelan, Milena R. Vicente, Lisa M. Wild, Ryan M. Khan, Jennifer T. Hidy (St. Michael’s Hospital); Chad Coles,

Ross Leighton, Michael Biddulph, David Johnston, Mark Glazebrook, David Alexander, Catherine Coady, Michael

Dunbar, J. David Amirault, Michael Gross, William Oxner, Gerald Reardon, C. Glen Richardson, J. Andrew

Trenholm, Ivan Wong, Kelly Trask, Shelley MacDonald, Gwendolyn Dobbin (Queen Elizabeth II Health Sciences

Page 4 of 21

Centre); Ryan Bicknell, Jeff Yach, Davide Bardana, Gavin Wood, Mark Harrison, David Yen, Sue Lambert, Fiona

Howells, Angela Ward (Human Mobility Research Centre, Queen’s University and Kingston General

Hospital); Paul Zalzal, Heather Brien, V. Naumetz, Brad Weening, Nicole Simunovic (Oakville Trafalgar

Memorial Hospital); Eugene K. Wai, Steve Papp, Wade T. Gofton, Allen Liew, Stephen P. Kingwell, Garth

Johnson, Joseph O’Neil, Darren M. Roffey, Vivian Borsella (The Ottawa Hospital); Victoria Avram (Juravinski

Hospital and Cancer Centre)

United States:

Todd M. Oliver, Vicki Jones, Michelle Vogt (Boone Hospital Center – Columbia Orthopaedic Group); Clifford

B. Jones, James R. Ringler, Terrence J. Endres, Debra L. Sietsema, Jane E. Walker (Orthopaedic Associates of

Michigan); Kyle J. Jeray, J. Scott Broderick, David R. Goetz, Thomas B. Pace, Thomas M. Schaller, Scott E.

Porter, Michael L. Beckish, John D. Adams, Benjamin B. Barden, Aaron T. Creek, Stephen H. Finley, Jonathan L.

Foret, Garland K. Gudger Jr, Richard W. Gurich Jr, Austin D. Hill, Steven M. Hollenbeck, Lyle T. Jackson, Kevin

K. Kruse, III, Wesley G. Lackey, Justin W. Langan, Julia Lee, Lauren C. Leffler, Timothy J. Miller, R. Lee Murphy,

Jr., Lawrence K. O’Malley II, Melissa E. Peters, Dustin M. Price, John A. Tanksley, Jr., Erick T. Torres, Dylan J.

Watson, Scott T. Watson, Stephanie L. Tanner, Rebecca G. Snider, Lauren A. Nastoff, Shea A. Bielby, Robert J.

Teasdall (Greenville Health System); Julie A. Switzer, Peter A. Cole, Sarah A. Anderson, Paul M. Lafferty,

Mengnai Li, Thuan V. Ly, Scott B. Marston, Amy L. Foley, Sandy Vang, David M. Wright (Regions Hospital-

University of Minnesota); Andrew J. Marcantonio, Michael S.H. Kain, Richard Iorio, Lawrence M. Specht, John F.

Tilzey, Margaret J. Lobo, John S. Garfi (Lahey Hospital & Medical Center); Heather A. Vallier, Andrea Dolenc,

Mary Breslin (MetroHealth Medical Center); Michael J. Prayson, Richard Laughlin, L. Joseph Rubino, Jedediah

May, Geoffrey Ryan Rieser, Liz Dulaney-Cripe, Chris Gayton (Miami Valley Hospital); James Shaer, Tyson

Schrickel, Barbara Hileman (St. Elizabeth Youngstown Hospital); John T. Gorczyca, Jonathan M. Gross,

Catherine A. Humphrey, Stephen Kates, John P. Ketz, Krista Noble, Allison W. McIntyre, Kaili Pecorella

(University of Rochester Medical Center); Craig A. Davis, Stuart Weinerman, Peter Weingarten, Philip Stull,

Stephen Lindenbaum, Michael Hewitt, John Schwappach, Janell K. Baker, Tori Rutherford, Heike Newman, Shane

Lieberman, Erin Finn, Kristin Robbins, Meghan Hurley, Lindsey Lyle, Khalis Mitchell, Kieran Browner, Erica

Whatley, Krystal Payton, Christina Reeves (Colorado Orthopedic Consultants); Lisa K. Cannada, David E.

Karges, Sarah A. Dawson (St. Louis University Hospital); Samir Mehta, John Esterhai, Jaimo Ahn, Derek

Donegan, Annamarie D. Horan, Patrick J. Hesketh, Evan R. Bannister (University of Pennsylvania); Jonathan P.

Keeve, Christopher G. Anderson, Michael D. McDonald, Jodi M. Hoffman (Northwest Orthopaedic Specialists);

Ivan Tarkin, Peter Siska, Gary Gruen, Andrew Evans, Dana J. Farrell, James Irrgang, Arlene Luther (University of

Pittsburgh Medical Center); William W. Cross III, Joseph R. Cass, Stephen A. Sems, Michael E. Torchia, Tyson

Scrabeck (Mayo Clinic); Mark Jenkins, Jules Dumais, Amanda W. Romero (Texas Tech University Health

Sciences Center – Lubbock); Carlos A. Sagebien, Mark S. Butler, James T. Monica, Patricia Seuffert (University

Orthopaedic Associates, LLC); Joseph R. Hsu, Daniel Stinner, James Ficke, Michael Charlton, Matthew

Napierala, Mary Fan (US Army Institute of Surgical Research); Paul Tornetta III, Chadi Tannoury, Hope Carlisle,

Heather Silva (Boston University Medical Center); Michael Archdeacon, Ryan Finnan, Toan Le, John Wyrick,

Shelley Hess (UC Health/University of Cincinnati Medical Center); Michael L. Brennan, Robert Probe, Evelyn

Kile, Kelli Mills, Lydia Clipper, Michelle Yu, Katie Erwin (Scott and White Memorial Hospital); Daniel Horwitz,

Kent Strohecker, Teresa K. Swenson (Geisinger Medical Center); Andrew H. Schmidt, Jerald R. Westberg

(Hennepin County Medical Center); Kamran Aurang, Gary Zohman, Brett Peterson, Roger B. Huff (Kaiser

Permanente); Joseph Baele, Timothy Weber, Matt Edison (OrthoIndy Trauma St. Vincent Trauma Center);

Jessica Cooper McBeth (Santa Clara Valley Medical Center); Karl Shively, Janos P. Ertl, Brian Mullis, J. Andrew

Parr, Ripley Worman, Valda Frizzell, Molly M. Moore (Indiana University – Eskenazi Health Services); Charles

J. DePaolo, Rachel Alosky, Leslie E. Shell, Lynne Hampton, Stephanie Shepard, Tracy Nanney, Claudine Cuento

(Mission Hospital Research Institute); Robert V. Cantu, Eric R. Henderson, Linda S. Eickhoff (Dartmouth-

Hitchcock Medical Center); E. Mark Hammerberg, Philip Stahel, David Hak, Cyril Mauffrey, Corey Henderson,

Hannah Gissel, Douglas Gibula (Denver Health and Hospital Authority); David P. Zamorano, Martin C. Tynan,

Deeba Pourmand, Deanna Lawson (University of California Irvine Medical Center); Gregory J. Della Rocca,

Brett D. Crist, Yvonne M. Murtha, Linda K. Anderson (University of Missouri Health Care); Colleen Linehan,

Lindsey Pilling (Covenant Healthcare of Saginaw); Courtland G. Lewis, Stephanie Caminiti, Raymond J.

Sullivan, Elizabeth Roper (Hartford Hospital); William Obremskey, Philip Kregor, Justin E. Richards, Kenya

Stringfellow (Vanderbilt University Medical Center); Michael P. Dohm, Abby Zellar (Western Slope Study

Group)

Page 5 of 21

The Netherlands:

Michiel J.M. Segers, Jacco A.C. Zijl, Bart Verhoeven, Anke B. Smits, Jean Paul P.M. de Vries, Bram Fioole, Henk

van der Hoeven, Evert B.M. Theunissen, Tammo S. de Vries Reilingh, Lonneke Govaert, Philippe Wittich, Maurits

de Brauw, Jan Wille, Peter M.N.Y.M. Go, Ewan D. Ritchie, Ronald N. Wessel, Eric R. Hammacher (St. Antonius

Ziekenhuis); Martin J. Heetveld, Gijs A. Visser, Heyn Stockmann, Rob Silvis, Jaap P. Snellen, Bram Rijbroek,

Joris J.G. Scheepers, Erik G.J. Vermeulen, Michiel P.C. Siroen, Ronald Vuylsteke, Hans L.F. Brom, Herman Rijna

(Kennemer Gasthuis); Piet A.R. de Rijcke, Cees L. Koppert, Steven E. Buijk, Richard P.R. Groenendijk, Imro

Dawson, Geert W.M. Tetteroo, Milko M.M. Bruijninckx, Pascal G. Doornebosch, Eelco J.R. de Graaf (IJsselland

Ziekenhuis); Maarten van der Elst, Carmen C. van der Pol, Martijne van ’t Riet, Tom M. Karsten, Mark R. de

Vries, Laurents P.S. Stassen, Niels W.L. Schep, G. Ben Schmidt, W.H. Hoffman (Reinier de Graaf Gasthuis);

Rudolf W. Poolman, Maarten P. Simons, Frank H.W.M. van der Heijden, W. Jaap Willems, Frank R.A.J. de

Meulemeester, Cor P. van der Hart, Kahn Turckan, Sebastiaan Festen, Frank de Nies, Robert Haverlag, Nico J.M.

Out, Jan Bosma (Onze Lieve Vrouwe Gasthuis); Albert van Kampen, Jan Biert, Arie B. van Vugt, Michael J.R.

Edwards, Taco J. Blokhuis, Jan Paul M. Frölke, Leo M.G. Geeraedts, Jean W.M. Gardeniers, Edward T.C.H. Tan,

Lodewijk M.S.J. Poelhekke, Maarten C. de Waal Malefijt, Bart Schreurs (University Medical Center St.

Radboud); Gert R Roukema, Hong A. Josaputra, Paul Keller, Peter D. de Rooij, Hans Kuiken, Han Boxma, Berry I.

Cleffken, Ronald Liem (Maasstad Ziekenhuis); Steven J. Rhemrev, Coks H.R. Bosman, Alexander de Mol van

Otterloo, Jochem Hoogendoorn, Alexander C. de Vries, Sven A.G. Meylaerts (Medisch Centrum Haaglanden);

Michiel H.J. Verhofstad, Joost Meijer, Teun van Egmond, Frank H.W.M. van der Heijden, Igor van der Brand (St.

Elisabeth Ziekenhuis); Peter Patka, Martin G. Eversdijk, Rolf Peters, Dennis Den Hartog, Oscar J.F. Van Waes,

Pim Oprel (Erasmus MC, University Medical Center Rotterdam); Harm M van der Vis, Martin Campo, Ronald

Verhagen, G.H. Robert Albers, Arthur W. Zurcher (Tergooi Ziekenhuizen); Rogier K.J. Simmermacher, Jeroen

van Mulken, Karlijn van Wessem, Taco J. Blokhuis, Steven M. van Gaalen, Luke P.H. Leenen (University Medical

Center Utrecht); Maarten W.G.A. Bronkhorst, Onno R. Guicherit (Bronovo Ziekenhuis); J. Carel Goslings,

Robert Haverlag, Kees Jan Ponsen (Academic Medical Center)

International:

Mahesh Bhatia, Vinod Arora, Vivek Tyagi (RLB Hospital and Research Centre, India); Susan Liew, Harvinder

Bedi, Ashley Carr, Hamish Curry, Andrew Chia, Steve Csongvay, Craig Donohue, Stephen Doig, Elton

Edwards, Greg Etherington, Max Esser, Andrew Gong, Arvind Jain, Doug Li, Russell Miller, Ash Moaveni,

Matthias Russ, Lu Ton, Otis Wang, Adam Dowrick, Zoe Murdoch, Claire Sage (The Alfred, Australia); Frede

Frihagen, John Clarke-Jenssen, Geir Hjorthaug, Torben Ianssen, Asgeir Amundsen, Jan Egil Brattgjerd, Tor Borch,

Berthe Bøe, Bernhard Flatøy, Sondre Hasselund, Knut Jørgen Haug, Kim Hemlock, Tor Magne Hoseth, Geir

Jomaas, Thomas Kibsgård, Tarjei Lona, Gilbert Moatshe, Oliver Müller, Marius Molund, Tor Nicolaisen, Fredrik

Nilsen, Jonas Rydinge, Morten Smedsrud, Are Stødle, Axel Trommer, Stein Ugland, Anders Karlsten, Guri Ekås,

Elise Berg Vesterhus, Anne Christine Brekke (Oslo University Hospital, Norway); Ajay Gupta, Neeraj Jain, Farah

Khan (Nirmal Hospital, India); Ateet Sharma, Amir Sanghavi, Mittal Trivedi (Satellite Orthopaedic Hospital

and Research Centre, India); Anil Rai, Subash, Kamal Rai (Highway Hospital, India); Vineet Yadav, Sanjay

Singh, Kamal Rai (Popular Hospital, India); Kevin Tetsworth, Geoff Donald, Patrick Weinrauch, Paul Pincus,

Steven Yang, Brett Halliday, Trevor Gervais, Michael Holt, Annette Flynn (Royal Brisbane and Women’s

Hospital, Australia); Amal Shankar Prasad, Vimlesh Mishra (Madhuraj Nursing Home, India); D.C. Sundaresh,

Angshuman Khanna (M.S. Rammaiah Medical College & Hospital, India); Joe Joseph Cherian, Davy J

Olakkengil, Gaurav Sharma (St John’s Medical College Hospital, India); Marinis Pirpiris, David Love, Andrew

Bucknill, Richard J Farrugia (Royal Melbourne Hospital, Australia); Hans-Christoph Pape, Matthias Knobe,

Roman Pfeifer (University of Aachen Medical Center, Germany); Peter Hull, Sophie Lewis, Simone Evans

(Cambridge University Hospitals, England); Rajesh Nanda, Rajanikanth Logishetty, Sanjeev Anand, Carol

Bowler (University Hospital of North Tees, England); Akhil Dadi, Naveen Palla, Utsav Ganguly (Sunshine

Hospital, India); B. Sachidananda Rai, Janakiraman Rajakumar (Unity Health Complex, India); Andrew

Jennings, Graham Chuter, Glynis Rose, Gillian Horner (University Hospital of North Durham and Darlington

Memorial Hospital, England); Callum Clark, Kate Eke (Wexham Park Hospital, England); Mike Reed,

Dominic Inman, Chris Herriott, Christine Dobb (Northumbria Healthcare NHS Foundation Trust, England)

Page 6 of 21

2.0 FUNDING SOURCES

Supported by research grants from the Canadian Institutes of Health Research (MOP-106630 and MCT-87771),

National Institutes of Health (1R01AR055267-01A1), Stichting NutsOhra (SNO-T-0602-43), the Netherlands

Organisation for Health Research and Development (80-82310-97-11032), Physicians’ Services Incorporated. Dr.

Bhandari was also funded, in part, through the Early Research Award Program which provided funding for the

present study as well as by a Canada Research Chair in Musculoskeletal Trauma which is unrelated to the present

study (McMaster University, Hamilton, ON, Canada).

Research reported in this publication was supported by the National Institute of Arthritis and Musculoskeletal and

Skin Diseases of the National Institutes of Health under Award Number R01AR055267-01A1. The content is solely

the responsibility of the authors and does not necessarily represent the official views of the National Institutes of

Health. Research reported in this publication was also supported by The County Durham & Tees Valley

Comprehensive Local Research Network which operates as part of the National Institute for Health Research

Comprehensive Clinical Research Network in England. The funding sources had no role in design or conduct of the

study; the collection, management, analysis, or interpretation of the data; or the preparation, review, or approval of

the manuscript.

Page 7 of 21

3.0 SUPPLEMENTAL APPENDICES

3.1 Eligibility Criteria

Inclusion Criteria

1) Adult men or women aged 50 years and older (with no upper age limit).

2) Fracture of the femoral neck confirmed with either anteroposterior and lateral hip radiographs, computed

tomography, or magnetic resonance imaging (MRI).

3) Operative treatment of displaced fractures within four days of presenting to the emergency room.

4) Operative treatment of undisplaced fractures within seven days of presenting to the emergency room.

5) Patient was ambulatory prior to fracture, though they may have used an aid such as a cane or a walker.

6) Anticipated medical optimization for operative fixation of the hip.

7) Provision of informed consent by patient or legal guardian.

8) No other major trauma (defined as an Injury Severity Score >16*).

9) Low energy fracture, in the judgment of the attending surgeon.

Exclusion Criteria

1) Patients not suitable for internal fixation (i.e., severe osteoarthritis, rheumatoid arthritis, or pathologic fracture).

2) Associated major injuries of the lower extremity (i.e., ipsilateral or contralateral fractures of the foot, ankle, tibia,

fibula, knee, or femur; dislocations of the ankle, knee, or hip; or femoral head defects or fracture).

3) Retained hardware around the affected hip.

4) Infection around the hip (i.e., soft tissue or bone).

5) Patients with disorders of known bone metabolism except osteoporosis (i.e., Paget’s disease, renal osteodystrophy,

osteomalacia).

6) Patients with a history of frank dementia that would interfere with assessment of the primary outcome (i.e., re-

operation at two years).

7) Likely problems, in the judgment of the investigators, with maintaining follow-up (i.e., patients with no fixed

address, report a plan to move out of town, or intellectually challenged patients without adequate family support).

3.2 Trial Interventions and Standardization of Peri-Operative Care

A. Multiple Cancellous Screws

Surgeons were allowed to use any threaded screw or hook pin (i.e., Gouffon, Uppsala, von Bahr, Hansson

hook pins, etc.) or cancellous threaded screw (i.e., cannulated or non-cannulated, Depuy-ACE screws,

AO/Association for the Study of Internal Fixation (ASIF), Asnis, Richards, etc.). Surgeons followed the technique

guides associated with the screw manufacturers.

Surgeons documented the following:

1. Number of screws.

2. Number of washers.

3. Manufacturer.

4. Reduction technique.

5. Decision to perform a capsulotomy or aspirate intracapsular hematoma.

6. Screw configurations, especially of the third screw (outside of two critical placements inferiorly and

posteriorly).

No injectable bone substitutes were allowed for augmentation of the implant fixation.

B. Sliding Hip Screw

Patients allocated to sliding hip screws received a single larger diameter partially threaded screw affixed to the

proximal femur with a sideplate (with a minimum of two holes and a maximum of four holes) and no supplemental

fixations. Surgeons were permitted to use any commercially available sliding hip screw implant (i.e., Stryker,

DePuy-ACE, Synthes, Smith and Nephew, Zimmer, etc.), and were to insert implants as per the manufacturers’

technical guides. The derotational kirschner wire was to penetrate the acetabulum to provide maximal resistance to

torsion. A centre-to-centre approach was recommended, while avoiding a superior and anterior approach. Spiral

Page 8 of 21

blades and helical screws were permitted, because they function similarly to the sliding hip screw. It was

documented when spiral blades were employed by the surgeon.

The use of a compression screw, manufacturer, reduction technique, decision to perform a capsulotomy, aspiration

of intracapsular hematoma, and final screw position measured by the Tip Apex Distance was documented and based

upon surgeon preference. No injectable bone substitutes were allowed for augmentation of the implant fixation.

C. Standardization of Procedures and Peri-Operative Care

Patient positioning, fracture reduction, and surgical exposure in the operating room were not standardized as these

are as these are highly variable across the world. There was, however, specific criteria for fracture reduction

acceptability. Acceptable reductions were left to the surgeons’ best judgment, while aiming at three standards for

translation in the anteroposterior and lateral fracture planes:

1. Less than 100% cortical displacement.

2. 10% or less translation of the femoral neck.

3. 3 mm of absolute translation on fluoroscopic views.

To ensure similar peri-operative regimens, it was recommended that participating centres standardize key aspects of

pre- and post-operative care.

Pre-Operative Care

1. Pre-operative antibiotic prophylaxis (i.e., cephalosporin or equivalent coverage).

2. Thromboprophylaxis (i.e., Thromboembolic Disease Stockings (TEDS), pneumatic compression boots, or

medical prophylaxis to be discontinued in sufficient time to allow surgery as guided by International

Normalized Ratio (INR)/Partial Thromboplastin Time (PTT)).

3. Medical consultation to optimize condition prior to surgery.

Post-Operative Care

1. Antibiotic prophylaxis (i.e., cephalosporin or equivalent) for 24 hours.

2. Thromboprophylaxis with unfractionated heparin, Low Molecular Weight Heparin (LMWH), warfarin,

anti-platelet agents, or intermittent pneumatic compression boots (current American College of Chest

Physicians (ACCP) guidelines recommend LMWH or adjusted dose warfarin (goal INR 2·5; range 2-3).

3. Weightbearing as tolerated was allowed as patients autoprotect the affected hip during rehabilitation. Post-

surgery, patients were encouraged to be weightbearing as tolerated and then advanced according to the

attending surgeon’s best judgment (i.e., touch weightbearing for displaced fractures was permitted and then

advanced according to the surgeon’s best judgment).

4. Calcium 600 mg by mouth (PO) daily and/or vitamin D 1000 International Units (IU) per day (provided

there were no contraindications), or further investigation and treatment of osteoporosis as recommended by

a local osteoporosis expert/consultant as necessary.

5. Appropriate nutritional assessment with administration of oral micronutrient feeds as needed.

Other Care

Due to a lack of evidence favouring a particular approach, the following was recorded but not standardized:

1. Use of pre-operative traction.

2. Surgical delay.

3. Type of anesthetic (i.e., general versus regional).

4. Physiotherapy and rehabilitation programs.

3.3 Threshold Performance for Internal Fixation

Surgeons who participated in the FAITH trial were asked to meet two criteria for expertise for either cancellous

screw fixation or sliding hip screw fixation:

1. Surgeons should have performed at least 25 procedures (of either sliding hip screws and/or cancellous

screws) in their career (including residency experience in which they assumed responsibility for the

procedure).

2. Surgeons should have continued to perform the procedure (at least five per year of either sliding hip screws

and/or cancellous screws) in the year prior to the trial start date.

Page 9 of 21

Any combination of procedures (cancellous screws and/or sliding hip screws) performed by surgeons were accepted

as expertise. Additionally, any combination of these procedures performed for any type of hip fracture (i.e., femoral

neck fracture, intertrochanteric fracture, etc.) were accepted as expertise.

3.4 Follow-up Processes

Time-point Assessments Patient Questionnaires

1 week Follow-up form, reoperations, and adverse events SF-12, EQ-5D, WOMAC*

2 weeks** Follow-up form, reoperations, and adverse events SF-12, EQ-5D, WOMAC*

10 weeks Follow-up form, radiographs, reoperations, and adverse events SF-12, EQ-5D, WOMAC*

6 months Follow-up form, radiographs (if necessary/possible), reoperations,

and adverse events

SF-12, EQ-5D, WOMAC*


and adverse events



and adverse events



and adverse events



and adverse events


SF-12=Short Form 12; EQ-5D=EuroQol 5 Dimensions; WOMAC= Western Ontario and McMaster Universities Arthritis Index

*For a sub-set of participants

**Removed to reduce participant burden

3.5 Outcome Definitions

Primary Outcomes Outcome Definition

Re-operations (study

event)

Re-operations that were classified as study events included:

o Implant removal

o Implant exchange – Total hip arthroplasty

o Implant exchange – Hemiarthroplasty

o Implant exchange – Internal fixation

o Implant exchange – Spacer

o Soft tissue procedure

o Other event as determined by the adjudication committee (proximal femoral osteotomy)

Re-operations (non-study

event)

Re-operations that were not classified as study events included:

o Irrigation and debridement

o Wound closure

o Planned re-operation

o Elective implant removal after fracture healing

Secondary Outcomes Outcome Definition

Fracture Compilations The Adjudication Committee reviewed fracture complications including avascular necrosis, nonunion, implant

failure, and infections.

Mortality Mortality was adjudicated and it was considered to be an event if it occurred within 24 month of the fracture.

Fracture Healing and

Shortening

The Adjudication Committee reviewed all available x-rays for fracture healing. In healed fractures, the

Adjudication Committee assessed how much the fracture shortened (No shortening, ≤5 mm [mild], 6-10mm

[moderate], >10mm [severe]).

Medical Adverse Events Medical adverse events, as diagnosed by physicians at the clinical sites, were documented.

Health-Related Quality of

Life

The health-related quality of life outcomes being analyzed included:

SF-12 Physical Component Score, which measures self-reported quality of life through physical

summary measures and a preference-based health utility index

EQ-5D, which is a standardized instrument for use as a measure of health outcome

WOMAC, which is self-administered and assesses the three dimensions of pain, disability and joint

stiffness in knee and hip osteoarthritis

3.6 Overview of Adjudication

Adjudication Processes

The following information was excerpted from the FAITH Adjudication Charter, which documents the

responsibilities of the Adjudication Committee and the adjudication processes for the FAITH trial.

Page 10 of 21

1) Fracture Eligibility

The Adjudication Committee adjudicated fracture eligibility in cases where eligibility was in doubt based on

available pre-surgery and post-surgery x-rays, and completed case report forms. Fracture eligibility was not affected

by the implant used, even in situations where the patient did not receive cancellous screws or a sliding hip screw (a

protocol deviation).

2) Radiographic Characteristics and Quality of Surgery

The Adjudication Committee classified the level of the femoral neck fracture as subcapital, midcervical or basal,

determined the type of fracture and classified it as being undisplaced (Garden I or II) or displaced (Garden III or IV),

and classified the fracture using the Pauwels’ classification. The Committee determined if the quality of the

reduction and the quality of implant placement were acceptable or unacceptable.

3) Fracture Healing

The Adjudication Committee began assessment of a patient’s fracture healing after the patient’s 24-month follow-

up. They adjudicated all available radiographic visits at and after the week 10 time point. Along with fracture

healing, the Committee assessed the amount of fracture shortening as follows: no shortening, mild shortening (≤ 5

mm), moderate shortening (6-10mm), or severe shortening (>10mm).

4) Re-operations

The Adjudication Committee adjudicated re-operations after each patient had completed their 24-month follow-up

(or following early withdrawal). Specifically, the Committee adjudicated all re-operations to promote fracture

healing, relieve pain, treat infection, or improve function. The Committee reviewed all available x-rays, and data

from the patient’s completed case report forms. Planned re-operations were not considered events.

5) Fracture-Related Complications The Adjudication Committee adjudicated fracture-related complications after each patient had completed their 24-

month follow-up (or following early withdrawal). The Committee reviewed all available x-rays, and data from the

patient’s completed case report forms.

6) Mortality

The Adjudication Committee adjudicated mortality as required following a patient’s early withdrawal. The

Committee reviewed all data from the patient’s completed case report forms.

3.7 Sample Size Calculations (FAITH Investigators. BMC Musculoskeletal Disorders 2014, 15:219)

Estimated study power for 750 patients per treatment arm (N = 1500) Baseline risk

(Year 1)^ Relative risk reduction

10% 15% 20% 25% 30%

20% 15·9% 30·8% 50·5% 70·5% 86·1%

25% 19·6% 38·7% 61·6% 81·5% 93·5%

30% 23·6% 46·8% 71·7% 89·3% 97·4%

35% 28·2% 55·2% 80·3% 94·5% 99·1%

40% 33·3% 63·6% 87·3% 97·5% 99·7%

*Number of patients per treatment arm, alpha = 0·05.

^Year 2 risk is 1/3 that of year 1.

Bold numbers indicate statistical power range for this study.

Estimated study power for 500 patients per treatment arm (N = 1,000) Baseline risk

(Year 1)^ Relative risk reduction

10% 15% 20% 25% 30% 35% 40%

15·5% 14·5% 28·5% 48·1% 69·3% 86·3% 95·7% 99·2%

*Number of patients per treatment arm, alpha = 0·05.

^Year 2 risk is 27·2%.

Bold numbers indicate statistical power range for this study.

Page 11 of 21

3.8 Subgroup Analyses and Hypothesized Effects

At the onset of the FAITH trial, we identified one important subgroup and planned to perform the analysis with the

primary endpoint as the outcome:

1. Displaced vs undisplaced fractures – We hypothesize that sliding hip screw will perform better in patients

with displaced fractures

At the end of the trial but prior to unblinding, we identified another five subgroup analyses and corresponding

hypotheses on their effects:

1. Normal BMI vs Overweight or Obese BMI – We hypothesize that sliding hip screw will perform better in

obese patients

2. Level of the Fracture Line – We hypothesize that sliding hip screw will perform better in base of neck

fractures (basicervical)

3. Pauwel’s Classification –We hypothesize that sliding hip screw will perform better in more vertical fracture

lines (Higher Pauwel’s class)

4. Smoking Status – We hypothesize that sliding hip screw will perform better in current smokers.

5. Quality of Reduction – We hypothesize that sliding hip screw will perform better with unacceptable

reduction. It should be noted that quality of reduction was assessed following fixation and post-

randomization; therefore any conclusions that we make based on a significant subgroup effect will be

hypothesis-generating.

After unblinding, we identified a single post hoc subgroup analysis:

1. Age – We hypothesize that there will be no significant differences in performance between sliding hip

screw and cancellous screw in patients under the age of 60. For patients between the ages of 60 and 80

years of age, we hypothesize that sliding hip screw will perform better. For patients over 80 years of age,

we hypothesize that sliding hip screw will perform better.

3.9 Rationale for Excluding Ineligible Patients from Analyses

In addressing the issue of how to manage patients who were inadvertently/incorrectly randomized despite not

satisfying all criteria for eligibility, we were guided by the recommendations of Fergusson et al. The authors of this

paper maintain that including ineligible patients in the analyses for clinical trials adds random error and therefore

decreases the power of the trial to answer the question(s) being addressed.

One must accept this added random error if it is not possible to exclude such patients without introducing bias. It is,

however, often possible to exclude ineligible patients without bias. This requires the following conditions: i) The

relevant information documenting patient ineligibility is collected soon after randomization ii) the Adjudication

Committee making the decision about excluding ineligible patients is unaware of the treatment allocation and iii) the

Adjudication Committee making the decision is unaware of the outcome of the patients who prove ineligible. In the

current study, we ensured all three conditions were met for any randomized patients excluded from the analysis.

Reference:

Fergusson D, Aaron S, Guyatt GH, Hebert P. Post-randomization exclusion of patients enrolled in clinical trials: The

intention-to-treat principle does not necessitate that all patients be analyzed. BMJ 2002;325:652-4.

Page 12 of 21

4.0 SUPPLEMENTARY TABLES

Table S1a: Reasons for Exclusion by Clinical Sites

Reason for Exclusion N=5463* N(%)

Surgeon Preference: Requires HA, THA, Prosthesis, Bipolar 2060 (37·7%)

Surgeon Preference: Unsuitable for internal fixation (arthritis/pathologic) 1738 (31·8%)

Cognitive impairment 1110 (20·3%)

Declined to consent 589 (10·8%)

Anticipated problematic follow-up 515 (9·4%)

Less than 50 years old 319 (5·8%)

Not medically optimized 305 (5·6%)

Major trauma 292 (5·3%)

High energy fracture 250 (4·6%)

Not ambulatory pre-injury 227 (4·2%)

Chronic condition or multiple comorbidities 211 (3·9%)

Fracture cannot be closed reduced 211 (3·9%)

National guidelines (for Dutch patients) 199 (3·6%)

Associated injuries 191 (3·5%)

Displaced fracture not treatable within 4 days of emergency room presentation 154 (2·8%)

Undisplaced fracture not treatable within 7 days of emergency room presentation 116 (2·1%)

Surgeon Preference: Requires specific fixation type 82 (1·5%)

Retained hardware 79 (1·4%)

Another trial participant 76 (1·4%)

Surgeon who is not participating in the study 73 (1·3%)

Language/communication barrier 70 (1·3%)

Metabolic bone disease 64 (1·2%)

Severe Parkinson's disease 54 (1·0%)

Non-operative or conservative treatment 46 (0·8%)

Infection 43 (0·8%)

Other reasons 26 (0·5%)

No femoral neck fracture 23 (0·4%)

Not seen within time frame 6 (0·1%)

Missed** 146

HA=hemiarthroplasty; THA=total hip arthroplasty;

*Please note, some patients were excluded for more than one reason

**Missed: patients who were not screened for study inclusion as result of error or lack of study staff availability

Page 13 of 21

Table S1b: Reasons for Exclusion by the Adjudication Committee

Reason for Exclusion N=29 N(%)

Ineligible fracture 15 (51·7%)

Delayed surgical treatment beyond 4 days 6 (20·7%)

Not 50 years or older 2 (6·9%)

Other major trauma 2 (6·9%)

Delayed surgical treatment beyond 7 days 1 (3·5%)

Patient did not have anticipated medical optimization 1 (3·5%)

Does not have a low energy fracture in the judgement of the attending surgeon 1 (3·5%)

Retained hardware around the hip (i.e., soft tissue or bone) 1 (3·5%)

Page 14 of 21

Table S2: Surgical and Peri-Operative Management

Sliding Hip Screw

N=542

Cancellous Screws

N=537

Total

N=1079

Time From Injury to Surgery, mean (SD) (hours) N=496

50·4 (69·5)

N=498

47·1 (61·9)

N=994

48·8 (65·8)

Length of Procedure, mean (SD) (minutes) N=531

64·6 (33·2)

N=529

55·1 (33·9)

N=1060

59·8 (33·9)

Who Performed Majority of Procedure, n (%)

Surgeon

Resident

Fellow

Registrar

N=533

292 (54·8%)

214 (40·2%)

21 (3·9%)

6 (1·1)

N=532

295 (55·5%)

214 (40·2%)

16 (3·0%)

7 (1·3%)

N=1065

587 (55·1%)

428 (4·02%)

37 (3·5%)

13 (1·2%)

Type of Anaesthesia, n (%)

General

Regional

N=533

327 (61·4%)

218 (40·9%)

N=532

316 (59·4%)

222 (41·7%)

N=1065

643 (60·4%)

440 (41·3%)

Total Blood Loss, mean (SD) (mL) 126·6 (106·9) 70·3 (95·7) 98·6 (105·2)

Capsulotomy Performed, n (%)

Yes

No

N=531

48 (9·0%)

483 (91·0%)

N=531

45 (8·5%)

486 (91·5%)

N=1062

93 (8·8%)

969 (91·2)

Aspiration of Intracapsular Hemotoma, n (%)

Yes

No

N=533

18 (3·4%)

515 (96·6%)

N=532

21 (4·0%)

511 (96·0%)

N=1065

39 (3·7%)

1026 (96·3%)

Post-operative Antibiotic Prophylaxis, n (%)

Yes

No

N=533

526 (98·7%)

7 (1·3%)

N=532

522 (98·1%)

10 (1·9%)

N=1065

1048 (98·4%)

17 (1·6%)

Thromboprophylaxis, n (%)

No

Yes

Heparin

LWMH

Warfarin

Mechanical

Other

N=533

3 (0·6%)

530 (99·4%)

46 (8·6%)

427 (80·6%)

21 (4·0%)

104 (19·6%)

39 (7·4%)

N=532

2 (0·4%)

530 (99·6%)

57 (10·8%)

426 (80·4%)

28 (5·3%)

91 (17·1%)

37 (7·0%)

N=1065

5 (0·5%)

1060 (99·5%)

103 (9·7%)

853 (80·0%)

49 (4·6%)

195 (18·4%)

76 (7·2%)

Weightbearing, n (%)

Yes, as tolerated

Non-weightbearing

Full weightbearing

N=533

503 (94·4%)

30 (5·6%)

0 (0%)

N=532

494 (92·9%)

38 (7·1%)

0 (0%)

N=1065

997 (93·6%)

68 (6·4%)

0 (0%)

Patient taking 600mg of Calcium and/or 1000mg of Vitamin D, n

(%)

Yes

No

N=533

505 (94·8%)

28 (5·2%)

N=531

493 (92·8%)

38 (7·2%)

N=1064

998 (93·8%)

66 (6·2%)

Patient Discharge Location, n (%)

Home

Rehabilitation facility

Aging and long term care facility

Other hospital

Family member’s home

Assisted living

Deceased

N=531

276 (52·0%)

169 (31·8%)

66 (12·4%)

12 (2·2%)

5 (1·0%)

1 (0·2%)

2 (0·4%)

N=528

275 (52·1%)

167 (31·5%)

71 (13·5%)

10 (1·9%)

4 (0·8%)

0 (0%)

1 (0·2%)

N=1059

551 (52·0%)

336 (31·7%)

137 (12·9%)

22 (2·1%)

9 (0·9%)

1 (0·1%)

3 (0·3%)

Patient Aids at Discharge, n (%)

Bedridden

Wheelchair

Walker

Two crutches

One crutch

Cane

Other

Person assistance

Stick

Deceased

N=530

23 (4·3%)

119 (22·5%)

401 (75·7%)

99 (18·7%)

9 (1·7%)

10 (1·9%)

14 (2·6%)

13 (2·4%)

0 (0%)

1 (0·2%)

N=528

24 (4·6%)

120 (22·7%)

384 (72·7%)

99 (18·8%)

13 (2·5%)

9 (1·7%)

12 (2·3%)

11 (2·0%)

1 (0·15%)

1 (0·15%)

N=1058

47 (4·4%)

239 (22·6%)

785 (74·2%)

198 (18·7%)

22 (2·1%)

19 (1·8%)

26 (2·5%)

24 (2·2%)

1 (0·1%)

2 (0·2%)

SD=standard deviation

Page 15 of 21

Table S3: Cancellous Screw Surgical Management

Cancellous Screws

N=537

Patients Crossed from Sliding

Hip Screw

N=16

Hook Pins Used, n (%)

Yes

No

N=508

22 (4·3%)

486 (95·7%)

0 (0%)

16 (100%)

Number of Screws Used, n (%)

1

2

3

4

5

N=509

0 (0%)

28 (5·5%)

464 (91·2%)

15 (2·9%)

2 (0·4%)

0 (0%)

1 (6·2%)

15 (93·8%)

0 (0%)

0 (0%)

Diameter of Screws, mean (SD) N=509

7·3 (3·5)

7·0 (0·4)

Partially Threaded Screws with Short Threads, n (%)

0

1

2

3

4

5

N=508

92 (18·1)

29 (5·7%)

59 (11·6%)

322 (63·4%)

5 (1·0%)

1 90·2%)

3 (18·8%)

3 (18·8%)

1 (6·2%)

9 (56·2%)

0 (0%)

0 (0%)

Partially Threaded Screws with Long Threads, n (%)

0

1

2

3

4

N=508

345 (67·9%)

40 (7·9%)

42 (8·3%)

77 (15·1%)

4 (0·8%)

9 (56·3%)

2 (12·5%)

2 (12·5%)

3 (18·7%)

0 (0%)

Formation of screws, n (%)

2 Screws Vertical

2 Screws Inferior and Posterior

3 Screws Triangle (apex at top)

3 Screws Triangle apex at bottom)

4 Screws Square

4 Screws Diamond

Other

N=509

10 (2·0%)

16 (3·1%)

147 (28·9%)

308 (60·5%)

6 (1·2%)

7 (1·4%)

15 (2·9%)

1 (6·3%)

0 (0%)

4 (25%)

11 (68·7%)

0 (0%)

0 (0%)

0 (0%)

Screw Aim, n (%)

Parallel

Converging

Diverging

N=509

442 (86·8%)

13 (2·6%)

54 (10·6%)

14 (87·5%)

0 (0%)

2 (12·5%)

Number of Washers, n (%)

0

1

2

3

4

7

N=509

283 (55·6%)

34 (6·7%)

68 (13·3%)

118 (23·2%)

5 (1·0%)

1 (0·2%)

5 (31·3%)

1 (6·3%)

3 (18·8%)

7 (43·7%)

0 (0%)

0 (0%)

Manufacturer, n (%)

Synthes

Stryker

DePuy

Smith & Nephew

Biomet

Zimmer

Hansson

Other

N=509

289 (56·8%)

22 (4·3%)

38 (7·5%)

35 (6·9%)

8 (1·5%)

39 (7·7%)

1 (0·2%)

77 (15·1%)

11 (68·8%)

0 (0%)

0 (0%)

0 (0%)

0 (0%)

1 (6·2%)

0 (0%)

4 (25·0%)


Page 16 of 21

Table S4: Sliding Hip Screw Surgical Management

Sliding Hip Screw

N=542

Patients Crossed from Cancellous

Screws

N=6

Manufacturer, n (%)

Synthes

Stryker

DePuy

Smith & Nephew

Zimmer

AO

J&J

Arthrex

Other

N=503

361 (71·8%)

21 (4·2%)

1 (0·2%)

23 (4·5%)

42 (8·4%)

0 (0%)

0 (0%)

0 (0%)

55 (10·9%)

4 (66·7%)

0 (0%)

0 (0%)

0 (0%)

0 (0%)

0 (0%)

0 (0%)

0 (0%)

2 (33·3%)

Formation of screws, n (%)

Centre-centre Position

Superior Position

Anterior Position

Inferior Position

Posterior Position

Other

N=503

368 (73·2%)

3 (0·6%)

11 (2·1%)

89 (17·7%)

19 (3·8%)

13 (2·6%)

4 (66·7%)

0 (0%)

0 (0%)

2 (33·3%)

0 (0%)

0 (0%)

Number of Holes in Sideplate, n (%)

1

2

3

4

5

6

7

N=503

1 (0·2%)

387 (76·9%)

24 (4·8%)

80 (15·9%)

1 (0·2%)

2 (0·4%)

8 (1·6%)

0 (0%)

4 (66·7%)

0(0%)

2 (33·3%)

0 (0%)

0 (0%)

0 (0%)

Bicortical Sideplate, n (%)

Yes

No

N=503

1 (0·2%)

502 (99·8%)

0 (0%)

6 (100%)

Supplemental Screws Used, n (%)

No

Yes

With washer

Without washer

N=503

388 (77·1%)

115 (22·9%)

30 (26·1%)

85 (73·9%)

4 (66·7%)

2 (33·3%)

2 (100%)

0 (0%)

Supplemental Screw Placement, n (%)

All left in place

All used and removed

Some left in place

N=115

96 (83·5%)

15 (13·0%)

4 (3·5%)

1 (50%)

0 (0%)

1 (50%)

Spiral Blades Used, n (%)

Yes

No

N=503

1 (0·2%)

502 (99·8%)

0 (0%)

6 (100%)

Helical Screws Used, n (%)

Yes

No

N=503

0 (%)

503 (100%)

0 (%)

6 (100%)

Tip to Apex Distance, mean (SD) (mm) N=503

17·3 (12·0)

26·0 (14·5)


Page 17 of 21

Table S5: Study Outcomes by Treatment Group for Undisplaced Fractures

Overall

Sliding Hip

Screw

Cancellous

Screws Hazard Ratio

p value

N=729 N=360 N=369 (95% CI)

Primary Endpoint (re-operation)

Implant Removal

Implant Exchange – THA

Implant Exchange – HA

Implant Exchange – Internal

Fixation

Implant Exchange – Spacer

Soft Tissue Procedure

Proximal Femoral Osteotomy

120 (16.5%)

31 (4.3%)

58 (8.0%)

30 (4·1%)

9 (1·2%)

2 (0·3%)

1 (0·1%)

1 (0·1%)

60 (16·7%)

6 (1.7%)

40(11·1%)

15 (4·2%)

1 (0·3%)

1 (0·3%)

0 (0%)

1 (0.3%)

60 (16.3%)

25 (6.8%)

18 (4.9%)

15 (4.1%)

8 (2·2%)

2 (0·3%)

1 (0·3%)

0 (0%)

1.00 (0·69, 1·46)

0·25(0·10, 0·59)

2.33 (1·30, 4.17)

1.03 (0·51, 2.07)

0·13 (0·02, 1.02)

1.03 (0·06, 16.33)

0.34 (0·01, 8.36)

3.07(0·13, 75.23)

0·99

0·001

0·004

0·95

0·04

0·99

0·32

0·31

Secondary Endpoints

Avascular Necrosis 56 (7·7%) 33 (9·2%) 23 (6·2%) 1·77 (0.89 , 3.51) 0·10

Nonunion 27 (3.7%) 18(5.0%) 9 (2.4%) 2.05 (0·93, 4.50) 0·07

Implant Failure 38 (5.2%) 18 (5.0%) 20 (5·4%) 0·92 (0·50, 1·71) 0·80

Infection

Superficial

Deep

7 (1.0%)

2 (0.3%)

5 (0.7%)

2 (0.6%)

0 (0%)

2 (0.6%)

5 (1.4%)

2 (0.5%)

3 (0.8%)

0.41 (0·08, 2·10)

0.21 (0.01, 4.26)

0.68 (0.11, 4.07)

0·27

0.16

0.67

Fracture Healing (N=571)*

Healed by Month 24

Not Healed by Month 24

Not Healed at Time of Last Visit

418 (73.2%)

1 (0·2%)

152 (26.6%)

198 (70.2%)

1 (0·4%)

83 (29.4%)

220 (76.1%)

0 (0%)

69 (23.9%)

0·19

Fracture Shortening >5mm (N=418)** 81 (19·4%) 32 (16.2%) 49 (22.3%) 0·73 (0·49, 1·08) 0·11

Mortality 115 (17.3%) 48 (14.6%) 67 (19.8%) 0·69 (0·47, 1·02) 0·06

THA=total hip arthroplasty; HA=hemiarthroplasty

Relative risk was calculated where the total number of events is less than 50

*571 patients were included in the fracture healing analysis. 284 patients did not have x-rays available for fracture healing adjudication, and

therefore were not included in the denominator.

**418 patients were included in the shortening analysis based on the number of healed fractures with shortening data.

Page 18 of 21

Table S6: Health-Related Quality of Life by Treatment Groups for Undisplaced Fractures

Sliding Hip Screw

Mean (SD), N

Cancellous Screws

Mean (SD), N Adjusted Mean Difference

(95% CI)

p Value for Differences

Between Groups/

12 Months

SF-12 PCS 39·9 (11·2), 158 41·6 (10·6), 157 -0·02 (-2·14, 2·10), N=298 0·99

WOMAC 45.6(19·9), 160 40·9 (16·5), 158 2·81 (-1·03, 6·64), N=296 0·15

EQ-5D Index 0·77 (0·19), 167 0·80 (0·17), 167 -0·01 (-0·05, 0·02), N=314 0·44

24 Months

SF-12 PCS 41·5(10·7), 145 40·2 (11·8), 122 1·32 (-1·27, 3·91), N=245 0·32

WOMAC 40·6 (16·1), 141 39·9 (16·9), 123 0·38 (-3·64, 4·40), N=239 0·85

EQ-5D Index 0·77 (0·20), 156 0·80 (0·18), 139 -0·02 (-0·06, 0·02), N=269 0·26

Page 19 of 21

Table S7: Study Outcomes by Treatment Group for Displaced Fractures

Overall Sliding Hip

Screw

Cancellous

Screws Hazard Ratio

p value

N=350 N=182 N=168 (95% CI)

Primary Endpoint (re-operation)

Implant Removal

Implant Exchange – THA

Implant Exchange – HA

Implant Exchange – Internal

Fixation

Implant Exchange – Spacer

Soft Tissue Procedure

Proximal Femoral Osteotomy

104 (29·7%)

43 (12·3%)

46 (13·1%)

25 (7·1%)

7 (2·0%)

1 (0·3%)

5 (1·4%)

1 (0·3%)

47 (25·8%)

19 (10·4%)

24 (13·2%)

11 (6·0%)

1 (0·6%)

0 (0%)

4 (2·2%)

0 (0%)

57(33·9%)

24 (14·3%)

22 (13·1%)

14 (8·3%)

6 (3·6%)

1 (0·6%)

1 (0·6%)

1 (0·6%)

0·53 (0·34, 0·82)

0·73 (0·42, 1·28)

1·00 (0·59, 1·73)

0·73 (0·34, 1·55)

0·15 (0·02, 1·26)

0·31 (0·01, 7·50)

3·69 (0·42, 32·70)

0·31 (0·01, 7·50)

0·005

0·27

0·98

0·41

0·06

0·30

0·21

0·30

Secondary Endpoints

Avascular Necrosis 22 (6·3%) 17 (9·3%) 5 (3·0%) 3·14 (1·18 , 8·32) 0·01

Nonunion 39 (11·1%) 15 (8·2%) 24 (14·3%) 0·58 (0·31, 1·06) 0·07

Implant Failure 49 (14·0%) 24 (13·2%) 25 (14·9%) 0·89 (0·53, 1·49) 0·65

Infection

Superficial

Deep

12 (3·4%)

6 (1·7%)

6 (1·7%)

8 (4·4%)

4 (2·2%)

4 (2·2%)

4 (2·4%)

2 (1·2%)

2 (1·2%)

1·85 (0·57, 6·02)

1·85 (0·34, 9·95)

1·85 (0·34, 9·95)

0·30

0·47

0·47

Fracture Healing (N=224)*

Healed by Month 24

Not Healed by Month 24

Not Healed at Time of Last Visit

114 (50·9%)

2 (0·9%)

108 (48·2%)

64 (55·2%)

1 (0·9%)

51 (44·0%)

50 (46·3%)

1 (0·9%)

57 (52·8%)

0·41

Fracture Shortening >5mm (N=114)** 65 (57·0%) 37 (57·8%) 28 (56·0%) 1·03 (0·75, 1·43) 0·85

Mortality 35 (13·1%) 20 (14·5%) 15 (11·5%) 1·12 (0·51, 2·46) 0·77

THA=total hip arthroplasty; HA=hemiarthroplasty

Relative risk was calculated where the total number of events is less than 50

*224 patients were included in the fracture healing analysis. 284 patients did not have x-rays available for fracture healing adjudication, and

therefore were not included in the denominator.

**114 patients were included in the shortening analysis based on the number of healed fractures with shortening data.

Page 20 of 21

Table S8: Health-Related Quality of Life by Treatment Groups For Displacement

Sliding Hip Screw

Mean (SD), N

Cancellous Screws

Mean (SD), N

Adjusted Mean Difference

(95% CI)

p Value for Differences

Between Groups/

12 Months

SF-12 PCS 42·8 (10·7), 77 42·6(11·1), 67 -0·03 (-3·29, 3·24), N=137 0·99

WOMAC 43·0 (17·4), 80 42·4 (17·3), 68 0·11 (-5·30 5·53), N=142 0·97

EQ-5D Index 0·78 (0·20), 82 0·79 (0·18), 71 -0·02 (-0·07, 0·04), N=146 0·58

24 Months

SF-12 PCS 41·8 (11·4), 62 43·8 (11·6), 59 -1·13 (-4·79, 2·53), N=113 0·54

WOMAC 41·7 (17·0), 64 39·4 (17·7), 60 0·46 (-5·88, 6·79), N=116 0·89

EQ-5D Index 0·80 (0·19), 76 0·79 (0·20), 68 0·02 (-0·04, 0·07), N=137 0·59

Page 21 of 21

Table S9: Interaction Between Secondary Outcomes and Smoker Status

Overall Sliding Hip

Screw

Cancellous

Screws Hazard Ratio

p Value

N=1079 N=542 N=537 (95% CI)

Implant Removal 74 (6·9%) 25 (4·6%) 49 (9·1%) 0·42 (0·25, 0·70) 0·001

Smoker 18 (9·0%) 5 (5·0%) 13 (13·0%) 0·17 (0·05, 0·63) 0·13 for interaction

Non-smoker 56 (6·5%) 20 (4·6%) 36 (8·3%) 0·52 (0·29, 0·92)

Implant Exchange –

Overall

169 (15·7%) 90 (16·6%) 79 (14·7%) 1·07 (0·78, 1·46) 0·67


Non-smoker 139 (16·1%) 78 (18·1%) 61 (14·1%) 1·26 (0·89, 1·79)

Avascular Necrosis 78 (7·2%) 50 (9·2%) 28 (5·2%) 1·78 (1·09 , 2·91) 0·02


Non-smoker 70 (8·1%) 46 (10·7%) 24 (5·6%) 2·01 (1·18, 3·41)

Nonunion 66 (6·1%) 33 (6·1%) 33 (6·2%) 0·86 (0·52, 1·41) 0·55


Non-smoker 50 (5·8%) 28 (6·5%) 22 (5·1%) 1·14 (0·64, 2·03)

Mortality 156 (14·5%) 73 (13·5%) 83 (15·5%) 0·81 (0·58, 1·12) 0·20

Smoker 33 (16·4%) 14 (13·9%) 19 (19%) 0·65 (0·31, 1·37) 0·47 for interaction

Non-smoker 111 (12·9%) 53 (12·3%) 58 (13·4%) 0·89 (0·61, 1·29)

Documents

Fracture fixation in the operative management of hip ... · 15/04/2017 · Treatment of Hip fractures (FAITH) trial to examine the effect of a sliding hip screw versus cancellous