Intervention Review

You have free access to this content

Extramedullary fixation implants and external fixators for extracapsular hip fractures in adults

  1. Martyn J Parker*,
  2. Avishek Das

Editorial Group: Cochrane Bone, Joint and Muscle Trauma Group

Published Online: 28 FEB 2013

Assessed as up-to-date: 20 SEP 2011

DOI: 10.1002/14651858.CD000339.pub3


How to Cite

Parker MJ, Das A. Extramedullary fixation implants and external fixators for extracapsular hip fractures in adults. Cochrane Database of Systematic Reviews 2013, Issue 2. Art. No.: CD000339. DOI: 10.1002/14651858.CD000339.pub3.

Author Information

  1. Peterborough and Stamford Hospitals NHS Foundation Trust, Department of Orthopaedics, Peterborough, Cambridgeshire, UK

*Martyn J Parker, Department of Orthopaedics, Peterborough and Stamford Hospitals NHS Foundation Trust, Peterborough City Hospital, CBU PO Box 211, Bretton Gate, Peterborough, Cambridgeshire, PE3 9GZ, UK. martyn.parker@pbh-tr.nhs.uk.

Publication History

  1. Publication Status: New search for studies and content updated (no change to conclusions)
  2. Published Online: 28 FEB 2013

SEARCH

 
Characteristics of included studies [ordered by study ID]
Bannister 1990

MethodsMethod of randomisation: not stated
Number lost to follow-up: 5 (3%)


Participants2 Orthopaedic hospitals in Bristol, UK
155 participants with a trochanteric (stable & unstable) extracapsular femoral fracture
Mean age: 80 years (range 65 to 101 years)
Percentage male: 18%


InterventionsJewett nail plate versus sliding hip screw (AO dynamic hip screw)


OutcomesLength of follow up: 12 months
Non-union
Fixation failure rate ("mechanical failure": included cut-out)
Reoperation
Superficial wound infection
Deep wound infection
Mortality (55)
Pain at follow up (hip)
Regain of mobility
(Rehabilitation rate)


NotesCorrespondence with the trialist indicated that they were unable to provide any additional information.
The information for 5 patients who were lost to follow-up and a further 55 patients who died during the study were excluded from the study. The analysis was only for the survivors.
Trial was carried out in 1981, and an abstract available in 1983.
Some outcomes (pain) expressed as % without denominators.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskQuote: 'randomly allocated'

Allocation concealment (selection bias)Unclear riskQuote: 'randomly allocated'

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskNo blinding

Blinding of outcome assessment (detection bias)
Subjective outcomes
High riskNo blinding

Blinding of outcome assessment (detection bias)
Objectives outcomes
Unclear riskNot blinded (although this is unlikely to affect outcome)

Incomplete outcome data (attrition bias)
Short-term outcomes
High riskFive patients lost to follow-up and 55 patients died during follow-up not included in the results

Incomplete outcome data (attrition bias)
Longer-term outcomes
High riskFive patients lost to follow-up and 55 patients died during follow-up not included in the results

Selective reporting (reporting bias)High riskMinimal outcomes reported

Surgeons' experience with trial operationsUnclear riskNot stated

Buciuto 1998

MethodsMethod of randomisation: not stated
Number lost to follow-up: 12 (5%)


ParticipantsOrthopaedic hospital in Linkoping, Sweden
233 participants with an unstable trochanteric extracapsular femoral fracture
Median (?) age: 82/80 years (range 47 to 97 years)
Percentage male: 27%


InterventionsRAB-plate (Rigidity Augmentation Baixauli) versus sliding hip screw (Richards compression hip screw)


OutcomesLength of follow up: 1 year minimum (1 to 3 years, mean 2 years)
Length of surgery
Operative blood loss
Cut-out
Non-union
Fixation failure rate
Reoperation
Deep infection
Leg shortening
Varus deformity
Mortality


NotesFor the 233 participants, outcomes measures other than mortality were only reported on 186 participants. Supplementary paper available on the outcome of the 20 participants who had their implant removed after fracture healing.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNot stated

Allocation concealment (selection bias)Unclear riskNot stated

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskNo blinding of participants or personnel used

Blinding of outcome assessment (detection bias)
Subjective outcomes
High riskNo blinding of participants used

Blinding of outcome assessment (detection bias)
Objectives outcomes
Unclear riskNot blinded (although this is unlikely to affect outcome)

Incomplete outcome data (attrition bias)
Short-term outcomes
High risk233 patients randomised but only 186 assessed

Incomplete outcome data (attrition bias)
Longer-term outcomes
High riskMinimal follow-up data presented

Selective reporting (reporting bias)High riskLimited outcomes reported

Surgeons' experience with trial operationsUnclear riskNot stated

Esser 1986

MethodsMethod of randomisation: computer generated random numbers
Number lost to follow-up: not stated


ParticipantsOrthopaedic hospital in Bangor, UK
100 participants, all female, with a trochanteric fracture (stable & unstable)
Mean age: 82 years (all over 60)
Percentage male: 0%


InterventionsJewett nail plate (a fixed nail plate) versus sliding hip screw (dynamic hip screw)


OutcomesLength of follow-up: 6 months
Length of surgery
Loss of fixation and cut-out rate (composite outcome: confusing data)
Length of hospital stay
Mortality
Pain at follow up
Failure to regain mobility


Notes3 Jewett nail plate group participants had change of device: 2 excluded, 1 to a dynamic hip screw (intention-to-treat problem).
Outcomes often expressed as %. Denominators deduced for analyses, based on assumption of no losses to follow-up for 98 patients.
Correspondence from trialist (Dr Jones) indicated that no additional information of the study is available.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskQuote: "a routine for generating random digits by computer"

Allocation concealment (selection bias)Unclear riskQuote: "a routine for generating random digits by computer"

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskNo blinding of participants or personnel used

Blinding of outcome assessment (detection bias)
Subjective outcomes
High riskNo blinding of personnel used

Blinding of outcome assessment (detection bias)
Objectives outcomes
Unclear riskNot blinded (although this is unlikely to affect outcome)

Incomplete outcome data (attrition bias)
Short-term outcomes
High riskLimited outcomes presented

Incomplete outcome data (attrition bias)
Longer-term outcomes
High riskLimited outcomes presented

Selective reporting (reporting bias)High riskLimited outcomes presented

Surgeons' experience with trial operationsUnclear riskNot mentioned

Janzing 2002

MethodsMethod of randomisation: sealed numbered envelopes
Number lost to follow-up: 9 (8%)


ParticipantsOrthopaedic hospital in Leuven, Belgium
115 participants with a trochanteric fracture (stable & unstable)
Mean age: 82.5 years
Percentage male: 17%


InterventionsGotfried percutaneous compression plate (PCCP) versus sliding hip screw (dynamic hip screw)
All participants of the SHS group had an additional stabilising screw and 17 had an additional trochanteric stabilising plate


OutcomesLength of follow-up: 12 months
Length of surgery
Units of blood transfused
Fall in haemoglobin
Haematoma (in abstract only)
Cut-out
Later fracture of the femur
Other complications of fixation
Reoperation
Mortality
Pain in first week after surgery
Use of walking aids
Residence


NotesOne person in the PCCP group had a sliding hip screw as the fracture could not be reduced adequately by closed methods.

Abstract report and full article available for 71 participants. Different fracture classification but overlap of trial periods and clearly the same trial. Another abstract reports for 100 participants.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskQuote: "drawing a sealed and numbered envelope containing the treatment modality for each patient"

Allocation concealment (selection bias)Low riskQuote: "drawing a sealed and numbered envelope containing the treatment modality for each patient"

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskNo blinding of participants or personnel

Blinding of outcome assessment (detection bias)
Subjective outcomes
High riskNo blinding of personnel

Blinding of outcome assessment (detection bias)
Objectives outcomes
Unclear riskNo blinded (although this is unlikely to affect outcome)

Incomplete outcome data (attrition bias)
Short-term outcomes
Unclear riskLimited reporting of short term outcomes

Incomplete outcome data (attrition bias)
Longer-term outcomes
Unclear riskNine patients lost to follow-up and a further 23 died during the follow-up period

Selective reporting (reporting bias)Unclear riskLimited reporting of outcomes

Surgeons' experience with trial operationsUnclear riskNot mentioned

Karn 2006

MethodsMethod of randomisation: not stated
Number lost to follow-up: not stated


ParticipantsOrthopaedic hospital in Dharan, Nepal
60 participants with a trochanteric extracapsular femoral fracture
Mean age: 67 years (range 50 to 100 years)
Percentage male: 50%


InterventionsClosed reduction and external fixation versus sliding hip screw


OutcomesLength of follow up: 24 weeks
Length of surgery
Operative blood loss
Cut-out
Non-union
Fixation failure rate
Reoperation
Wound infection
Length of hospital stay
Leg shortening
Varus deformity
Range of movement of hip and knee
Harris hip score
WMAC index
Cost of treatment


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNot stated

Allocation concealment (selection bias)Unclear riskNot stated

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskNo blinding of participants or personnel

Blinding of outcome assessment (detection bias)
Subjective outcomes
High riskNo blinding of personnel

Blinding of outcome assessment (detection bias)
Objectives outcomes
Unclear riskNo blinding of personnel

Incomplete outcome data (attrition bias)
Short-term outcomes
High riskLimited outcomes presented and loss to follow-up not mentioned

Incomplete outcome data (attrition bias)
Longer-term outcomes
High riskLimited outcomes presented and loss to follow-up not mentioned

Selective reporting (reporting bias)High riskLimited outcomes presented

Surgeons' experience with trial operationsUnclear riskNot stated

Kosygan 2002

MethodsMethod of randomisation: sealed envelopes opened in the operating theatre
Number lost-to follow up: none


ParticipantsOrthopaedic hospital in Harrogate, UK
111 participants with a trochanteric fracture (stable & unstable)
Mean age: 83 years
Percentage male: 19%


InterventionsGotfried percutaneous compression plate (PCCP) versus sliding hip screw (Richards hip screw)


OutcomesLength of follow-up: 6 months
Length of surgery
Drainage of blood
Units of blood transfused
Fall in haemoglobin
X-ray screening time during operation
Cut-out
Other complications of fixation
Reoperation
Superficial wound sepsis
Deep wound infection
Pressure infection
Deep vein thrombosis
Pulmonary embolism
Other medical complications
Length of hospital stay
Limb shortening
Mortality


NotesThree participants of the PCCP group had a sliding hip screw as the fracture could not be reduced adequately by closed methods.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskQuote: "choice of implant was determined at the induction of anaesthesia by opening of a sealed envelope"

Allocation concealment (selection bias)Unclear riskQuote: "choice of implant was determined at the induction of anaesthesia by opening of a sealed envelope"

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskNo blinding of participants or personnel

Blinding of outcome assessment (detection bias)
Subjective outcomes
High riskNo blinding of personnel

Blinding of outcome assessment (detection bias)
Objectives outcomes
Unclear riskNot blinded (although this is unlikely to affect outcome)

Incomplete outcome data (attrition bias)
Short-term outcomes
Low riskRelevant outcome reported

Incomplete outcome data (attrition bias)
Longer-term outcomes
Low riskNo patients lost to follow-up and 17 died during the follow-up period

Selective reporting (reporting bias)Low riskFull reporting of outcomes

Surgeons' experience with trial operationsLow riskQuote: "procedures were performed by three experiences surgeons who had been instructed in the use of the PCCP by its inventor". In addition, the first two cases were not included.

Lizaur Utrilla 1998

MethodsMethod of randomisation: sequence of admission
Number lost to follow-up: 8 (4%)


ParticipantsOrthopaedic unit in Alicante, Spain
200 participants with a trochanteric (stable & unstable) extracapsular femoral fracture
Mean age: 80 years (range not stated; age 60+ years)
Percentage male: 24%


InterventionsRAB-plate (Rigidity Augmentation Baixauli) versus sliding (compression) hip screw


OutcomesLength of follow-up: 6 to 12 months (mean 10 months)
Length of surgery
Operative blood loss
Need for transfusion
Radiographic screening time
Operative complications
Degree of technical difficulty
Operative fracture of the femur
Cut-out of the implant
Breakage of the implant
Loosening or disassembly of the implant
Non-union
Avascular necrosis
Reoperation
Superficial wound infection
Deep wound infection
Mortality
Urinary infection
Thrombophlebitis
Length of hospital stay
Varus deformity
Leg shortening
Hip flexion
External rotation
Thigh or hip tenderness
Time till mobilisation
Ambulation score


NotesThe results for 26 participants who died within the one year from injury and a further 8 participants who were lost to follow up were not given.

Discrepancy between text and table in the numbers of implant breakage or loosening: 4 in tables, but 5 had re-operations due to this complication.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskQuote: "assigned randomly according to the sequence of admission"

Allocation concealment (selection bias)High riskQuote: "assigned randomly according to the sequence of admission"

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskNo blinding of participants or personnel

Blinding of outcome assessment (detection bias)
Subjective outcomes
High riskNo blinding of personnel

Blinding of outcome assessment (detection bias)
Objectives outcomes
Unclear riskNot blinded (although this is unlikely to affect outcome)

Incomplete outcome data (attrition bias)
Short-term outcomes
High riskThe results for a further 26 participants who died within the one year from injury and a further 8 participants who were lost to follow-up were not given.

Incomplete outcome data (attrition bias)
Longer-term outcomes
High riskThe results for 26 participants who died within the one year from injury and a further 8 participants who were lost to follow up were not given.

Selective reporting (reporting bias)Low riskAll relevant outcomes reported

Surgeons' experience with trial operationsLow riskAll operations undertaken by four surgeons with the same degree of experiences and the first 10 cases of the RAB plate were not included

Lunsjo 1999

MethodsMethod of randomisation: sealed envelopes opened in the operating theatre
Number lost to follow-up: 10 (9.3%)


Participants8 Orthopaedic hospitals in Sweden
107 participants with a subtrochanteric fracture
Mean age: 80 years (range 21 to 99 years)
Percentage male: 20%


InterventionsMedoff sliding plate (6 hole version) versus one of three screw-plate systems (sliding hip screw: 32 cases, dynamic compression plate: 12 cases, sliding hip screw with a trochanteric stabilising plate: 8 cases) selected by operating surgeon


OutcomesLength of follow-up: 12 months minimum
Length of surgery
Operative blood loss
Total blood loss (operative and drains)
Cut-out of the implant
Non-union
Later fracture of the femur
Breakage of the implant
Reoperation
Superficial wound infection
Deep wound infection
Thromboembolic complications
Length of hospital stay
Mortality
Failure to regain mobility
Failure to return home


NotesCorrespondence from trialists (Dr Ceder) indicated that no additional information for the study is available. Separate results for each of the 3 screw-plate systems were reported but are not presented here for reasons given in the text.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskQuote: "sealed envelopes opened just starting an attempted closed reduction"

Allocation concealment (selection bias)Unclear riskQuote: "sealed envelopes opened just starting an attempted closed reduction"

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskNo blinding of participants or personnel

Blinding of outcome assessment (detection bias)
Subjective outcomes
High riskNo blinding of personnel

Blinding of outcome assessment (detection bias)
Objectives outcomes
Unclear riskNot blinded (although this is unlikely to affect outcome)

Incomplete outcome data (attrition bias)
Short-term outcomes
Low riskEarly complications fully reported

Incomplete outcome data (attrition bias)
Longer-term outcomes
High riskTen patients lost to follow-up and a further 16 died during the follow-up period

Selective reporting (reporting bias)Low riskAll main relevant outcomes reported

Surgeons' experience with trial operationsHigh riskQuote: "many of them [surgeons] had little, if any experience with this method (Medoff plate)"

Lunsjo 2001

MethodsMethod of randomisation: sealed envelopes opened in the operating theatre
Number lost to follow-up: 48 (8.4%)


Participants8 Orthopaedic hospitals in Sweden
569 participants with an unstable trochanteric fracture
Mean age: 82 years (range 42 to 99 years)
Percentage male: 27%


InterventionsMedoff sliding plate (6 hole version) versus one of three screw-plate systems (sliding hip screw: 238 cases, dynamic compression plate: 14 cases, sliding hip screw with a trochanteric stabilising plate: 49 cases) selected by operating surgeon


OutcomesLength of follow-up: 12 months minimum
Length of surgery
Operative blood loss
Total blood loss (operative and drains)
Cut-out of the implant
Non-union
Breakage of the implant
Reoperation
Superficial wound infection
Deep wound infection
Thromboembolic complications
Length of hospital stay
Mortality
Failure to regain mobility
Failure to return home


NotesSeparate results for each of the three screw-plate systems were reported but are not presented here for reasons given in the text.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskQuote: "sealed envelopes were opened just before the surgeon tried to perform closed reduction of the fracture"

Allocation concealment (selection bias)Unclear riskQuote: "sealed envelopes were opened just before the surgeon tried to perform closed reduction of the fracture"

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskNo blinding of participants or personnel

Blinding of outcome assessment (detection bias)
Subjective outcomes
High riskNo blinding of personnel

Blinding of outcome assessment (detection bias)
Objectives outcomes
Unclear riskNot blinded (although this is unlikely to affect outcome)

Incomplete outcome data (attrition bias)
Short-term outcomes
Low riskEarly complications fully reported

Incomplete outcome data (attrition bias)
Longer-term outcomes
High risk48 patients lost to follow-up and a further 132 died during the follow-up period

Selective reporting (reporting bias)Unclear riskLimited reporting of relevant outcomes reported

Surgeons' experience with trial operationsUnclear riskNot stated

McCormack 2010

MethodsMethod of randomisation: not stated
Number lost to follow-up: not stated


ParticipantsOrthopaedic hospital in British Columbia, Canada
163 participants with an unstable trochanteric extracapsular femoral fracture
Mean age: not stated (range not stated, all over 60 years)
Percentage male: not stated


InterventionsMedoff sliding plate versus sliding hip screw


OutcomesLength of follow-up: minimum 6 months
Length of surgery
Blood transfusion
Cut-out
Non-union
Reoperation
Functional outcome


NotesReported only as a conference abstract. The abstract states 163 patients were randomised but the results are presented for only 162 patients.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNot stated

Allocation concealment (selection bias)Unclear riskNot stated

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskNo blinding of participants or personnel

Blinding of outcome assessment (detection bias)
Subjective outcomes
High riskNo blinding of personnel

Blinding of outcome assessment (detection bias)
Objectives outcomes
Unclear riskNo blinding of personnel

Incomplete outcome data (attrition bias)
Short-term outcomes
High riskLimited outcomes presented and loss to follow-up not mentioned

Incomplete outcome data (attrition bias)
Longer-term outcomes
High riskLimited outcomes presented and loss to follow-up not mentioned

Selective reporting (reporting bias)High riskLimited outcomes presented

Surgeons' experience with trial operationsUnclear riskNot stated

McLaren 1991

MethodsRandomised by even or odd patient medical record number
Lost to follow-up: none


ParticipantsOrthopaedic hospital Dundee, UK
100 participants with an intertrochanteric (stable & unstable) femoral fractures. 49% unstable fractures.
Mean age: 80 years (range 66 to 97 years)
Percentage male: 19%


InterventionsPugh nail versus sliding hip screw (dynamic hip screw)


OutcomesLength of follow-up: 6 months minimum
Length of surgery
Cut-out
Fixation failure rate
Technical problems of surgery
Unsatisfactory fixations
Deep infection
Deep vein thrombosis
Length of hospital stay
Mortality
Pain at follow up
Mobility


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskQuote: "according to whether their unit numbers were od or even"

Allocation concealment (selection bias)High riskQuote: "according to whether their unit numbers were od or even"

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskNo blinding of participants or personnel

Blinding of outcome assessment (detection bias)
Subjective outcomes
High riskNo blinding of personnel

Blinding of outcome assessment (detection bias)
Objectives outcomes
Unclear riskNot blinded (although this is unlikely to affect outcome)

Incomplete outcome data (attrition bias)
Short-term outcomes
Unclear riskReporting of limited short term outcomes

Incomplete outcome data (attrition bias)
Longer-term outcomes
High riskLimited reporting of long term outcomes

Selective reporting (reporting bias)High riskLimited outcomes reported

Surgeons' experience with trial operationsUnclear riskNot stated

Moroni 2005

MethodsMethod of randomisation: computer generated list; treatment allocation revealed to surgeon after patient enrolment
Number lost to follow-up: not stated


ParticipantsOrthopaedic hospital in Bologna, Italy
40 participants with a trochanteric fracture (stable & unstable)
Mean age: 80 years
Percentage male: 0%


InterventionsPertrochanteric external fixator versus sliding hip screw
The external fixator had hydroxyapatite coated pins and was removed after three months.


OutcomesLength of follow-up: 6 months
Length of surgery
Transfusion
Mean units blood transfused
Cut-out rate
Other complications of fixation
Wound infection
Length of hospital stay
Harris hip score


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskQuote: "randomised by a computer-generated list". "The treatment assignment was disclosed to the surgeon after the patient was enrolled in the study"

Allocation concealment (selection bias)Low riskQuote: "randomised by a computer-generated list". "The treatment assignment was disclosed to the surgeon after the patient was enrolled in the study"

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskNo blinding of participants or personnel

Blinding of outcome assessment (detection bias)
Subjective outcomes
High riskNo blinding of personnel

Blinding of outcome assessment (detection bias)
Objectives outcomes
Unclear riskNot blinded (although this is unlikely to affect outcome)

Incomplete outcome data (attrition bias)
Short-term outcomes
Unclear riskLimited reporting of outcomes

Incomplete outcome data (attrition bias)
Longer-term outcomes
High riskLoss to follow-up not clear

Selective reporting (reporting bias)Unclear riskVery limited reporting of outcomes

Surgeons' experience with trial operationsUnclear riskQuote: "all operations were performed by one of the authors"

Olsson 2001

MethodsMethod of randomisation: "the drawing of an unseen card"
Number lost to follow-up: 3 excluded after randomisation, 8 were unable to attend follow-up due to poor general health


ParticipantsOrthopaedic hospital in Helsingborg, Sweden
114 participants with a trochanteric fracture
Mean age: 84 years (range 61 to 98 years)
Percentage male: 30%


InterventionsMedoff sliding plate (4 hole version) versus sliding hip screw


OutcomesLength of follow-up: 4 months
Length of surgery
Operative blood loss
Fall in haemoglobin
Mean units blood transfused
Cut-out of the implant
Reoperation
Wound infection
Thromboembolic complications
Length of hospital stay
Limb shortening
Mortality
Failure to regain mobility
Failure to return home


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskQuote: "the drawing of an unseen card"

Allocation concealment (selection bias)Unclear riskQuote: "the drawing of an unseen card"

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskNo blinding of participants or personnel

Blinding of outcome assessment (detection bias)
Subjective outcomes
High riskNo blinding of personnel

Blinding of outcome assessment (detection bias)
Objectives outcomes
Unclear riskNot blinded (although this is unlikely to affect outcome)

Incomplete outcome data (attrition bias)
Short-term outcomes
Unclear riskReporting of limited short term outcomes

Incomplete outcome data (attrition bias)
Longer-term outcomes
Unclear riskNumber of patients lost to follow-up not clearly stated

Selective reporting (reporting bias)Low riskMajor outcomes reported

Surgeons' experience with trial operationsUnclear riskOperations undertaken by 22 different surgeons

Peyser 2007

MethodsMethod of randomisation: "in the operating theatre where sequentially numbered, sealed, opaque envelopes were opened by a nurse at the time of operation"
Number lost to follow-up: 1 excluded from the PCCP as failure of closed reduction and a further one from the PCCP group as died in the peri-operative period, 69 patients assessed for functional recovery and 53 assessed for radiological outcome.


ParticipantsOrthopaedic hospital in Jerusalem, Israel
104 participants with a trochanteric fracture
Mean age: 81 years (range 62 to 95 years)
Percentage male: 33%


InterventionsGotfried percutaneous compression plate (PCCP) versus sliding hip screw (SHS)


OutcomesLength of follow-up: 12 months
Length of surgery
Operative blood loss
Fall in haemoglobin
Mean units blood transfused
Number of patients transfused
Cut-out of the implant
Avascular necrosis
Reoperation
Wound infection
Pneumonia
Thromboembolic complications
Other medical complications
Length of hospital stay
Limb shortening
Mortality
Pain
Amount of weight bearing on injured leg


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskQuote: "Computer generated random number series with blocking in groups of ten"

Allocation concealment (selection bias)Low riskQuote: "in the operating theatre where sequentially numbered, sealed, opaque envelopes were opened by a nurse at the time of operation"

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskBlinding of only the recording of blood loss and medical complications

Blinding of outcome assessment (detection bias)
Subjective outcomes
High riskNo blinding of personnel

Blinding of outcome assessment (detection bias)
Objectives outcomes
Unclear riskNot blinded (although this is unlikely to affect outcome)

Incomplete outcome data (attrition bias)
Short-term outcomes
Low riskOutcomes reported for all but one patient in the PCCP group excluded after randomisation for failure of closed reduction

Incomplete outcome data (attrition bias)
Longer-term outcomes
High risk69 out of 104 patients assessed for functional recovery

Selective reporting (reporting bias)Low riskRelevant outcomes reported

Surgeons' experience with trial operationsLow riskQuote: "all participating surgeons were experiences in both techniques"

Pitsaer 1993

MethodsRandomised study: method of randomisation not stated
Number of lost to follow-up: 8 (8%)


ParticipantsOrthopaedic hospital in Winchester, UK
100 participants with an intertrochanteric (stable & unstable) proximal femoral fracture
Mean age: not stated.
Percentage male: not stated


InterventionsMcLaughlin nail plate versus sliding hip screw (dynamic hip screw)


OutcomesLength of follow-up: 6 months
Cut-out
Breakage of implant
Shortening
Reoperation
Mortality (33)
Pain
Regain of mobility


NotesRequest for further information sent to both authors
Analysis of survivors.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNot stated

Allocation concealment (selection bias)Unclear riskNot stated

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskNo blinding

Blinding of outcome assessment (detection bias)
Subjective outcomes
High riskNo blinding

Blinding of outcome assessment (detection bias)
Objectives outcomes
Unclear riskNot blinded (although this is unlikely to affect outcome)

Incomplete outcome data (attrition bias)
Short-term outcomes
Unclear riskLimited reporting of outcomes

Incomplete outcome data (attrition bias)
Longer-term outcomes
High riskLimited reporting

Selective reporting (reporting bias)High riskLimited reporting

Surgeons' experience with trial operationsUnclear riskNot stated

Vossinakis 2002

MethodsMethod of randomisation: sealed envelopes
Number lost to follow-up: none


ParticipantsOrthopaedic hospital in Volos, Greece
100 participants with a trochanteric fracture (stable & unstable)
Mean age: 77 years
Percentage male: 26%


InterventionsPertrochanteric external fixator versus sliding hip screw


OutcomesLength of follow-up: 6 months
Length of surgery
Operative blood loss
Units of blood transfused
Fall in haemoglobin
Cut-out rate
Other complications of fixation
Reoperation
Superficial wound or pin-track infection
Deep wound infection
Length of hospital stay
Limb shortening
Varus deformity
Mortality
Pain after surgery
Failure to regain mobility
Time to walking
Failure to return home
Place of discharge from hospital
Time to fracture healing


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskQuote: "sealed envelopes containing cards, indicating the treatment for each patient"

Allocation concealment (selection bias)Unclear riskQuote: "sealed envelopes containing cards, indicating the treatment for each patient"

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskNo blinding of participants or personnel

Blinding of outcome assessment (detection bias)
Subjective outcomes
High riskNo blinding of personnel

Blinding of outcome assessment (detection bias)
Objectives outcomes
Unclear riskNot blinded (although this is unlikely to affect outcome)

Incomplete outcome data (attrition bias)
Short-term outcomes
Low riskNo patients reported to be lost to follow-up

Incomplete outcome data (attrition bias)
Longer-term outcomes
Low risk15 patients died during the follow-up period

Selective reporting (reporting bias)Low riskRelevant outcomes reported

Surgeons' experience with trial operationsUnclear riskAll operations undertaken by one of four surgeons but experience not stated

Watson 1998

MethodsMethod of randomised trial: not stated
Loss to follow-up: 12 (7%)


ParticipantsOrthopaedic hospital in Detroit, USA
178 participants with 182 intertrochanteric (stable & unstable) femoral fractures
Mean age: 76 years (range 25 to 99 years)
Percentage male: 34% male


InterventionsMedoff sliding plate (6 hole version) versus sliding hip screw (compression hip screw: 4 hole side plate)


OutcomesLength of follow-up: 6 months minimum (6 to 26 months, mean 9.5 months)
Cut-out
Non-union
Implant breakage
Fixation failure rate
Reoperation
Superficial wound infection
Deep wound infection
Thromboembolism
Mortality
Failure to regain mobility
Failure to return home


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskQuote: "according to their medical records number"

Allocation concealment (selection bias)High riskQuote: "according to their medical records number"

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskNo blinding of participants or personnel

Blinding of outcome assessment (detection bias)
Subjective outcomes
High riskNo blinding of personnel

Blinding of outcome assessment (detection bias)
Objectives outcomes
Unclear riskNo blinding of personnel

Incomplete outcome data (attrition bias)
Short-term outcomes
High riskThe outcome for 12 patients lost to follow-up and 11 patients that died were not presented

Incomplete outcome data (attrition bias)
Longer-term outcomes
High riskThe outcome for 12 patients lost to follow-up and 11 patients that died were not presented

Selective reporting (reporting bias)High riskMinimal outcomes reported

Surgeons' experience with trial operationsUnclear riskSurgery was performed by the resident staff and supervised by one of four attending staff

Yang 2011

MethodsMethod of randomisation: the randomisation order was computer generated in a 1:1 ratio on the bases of balanced blocks of ten patients. Sequentally numbered sealed envelopes containing the treatment allocation were then used.
Number lost to follow-up: 14. A further 14 patients died during the follow-up period.


ParticipantsOrthopaedic hospital in Ney York, USA
66 participants with a trochanteric fracture
Mean age: 77/76 years (range not given)
Percentage male: 29%


InterventionsGotfried percutaneous compression plate (PCCP) versus sliding hip screw (SHS)


OutcomesLength of follow-up: 12 months (but mean 10 / 9 months in supplementary data table)
Length of surgery
Operative blood loss
Length of incision
Mean units blood transfused
Number of patients transfused
Cut-out of the implant
Fracture healing complications
Mortality
Pain
Loss of functional status
Use of assistive walking at two weeks after surgery


NotesExtra information provided by lead author include details that no fracture healing complications occurred


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskQuote: "computer generated in a 1:1 ratio on the bases of balanced blocks of ten patients"

Allocation concealment (selection bias)Low riskQuote: "Sequentially numbered sealed opaque envelopes which defined the treatment assignment were opened as the subjects were enrolled"

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskNo blinding

Blinding of outcome assessment (detection bias)
Subjective outcomes
High riskNo blinding

Blinding of outcome assessment (detection bias)
Objectives outcomes
Unclear riskNo blinding

Incomplete outcome data (attrition bias)
Short-term outcomes
Unclear riskAll patients included in short term operative outcomes

Incomplete outcome data (attrition bias)
Longer-term outcomes
High risk14 patients lost to follow-up and a further 14 died during the follow-up period

Selective reporting (reporting bias)Unclear riskLimited outcomes reported

Surgeons' experience with trial operationsUnclear riskQuote: "Surgery by one of seven attending surgeons and/or residents with direct supervision by an attending surgeon". Experences of surgeons not stated

 
Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion

Enskog 1999This study was reported only as a conference abstract which gave preliminary results for 41 participants of an ongoing randomised study. The study compared a sliding hip screw versus a similar device in which the lag screw was replaced by two Olmed screws. The follow-up of 27 cases showed no notable differences between implants. No further reports of this study have been forthcoming and the study was therefore excluded due to lack of presented information.

Heyse-Moore 1983Non-randomised retrospective comparison of the Jewett nail plate with the sliding hip screw. An original randomised trial was abandoned because it was seen that the dynamic hip screw was superior. Correspondence with the authors revealed that no information on patient numbers and no results are available for the abandoned randomised trial.

Kumar 2006This study was reported as a conference abstract of a randomised trial, involving 50 patients with unstable trochanteric fractures, comparing the sliding hip screw versus the same implant supplemented with a trochanteric stabilising plate. The abstract reported that the operative blood loss and length of surgery were increased for those treated with the supplementary stabilising plate although the difference was not statistically significant. The functional outcome, limb length discrepancy and abductor muscle strength were all reported to be statistically significantly better for those treated with the stabilising plate. There were three cases of cut-out in the SHS group and none in the stabilising plate group. We were unable to contact the trial authors for this trial. The study was excluded because there was insufficient reporting of the study details and the patient numbers were not given for each group.

Lahav 2007This study of 142 patients compared the Dynamic Hip Screw with the Percutaneous Compression Plate. The study was only reported as a conference abstract and was excluded as there was no mention of randomisation of patients.

Laufer 2005Non-randomised comparison of the dynamic hip screw (routinely used in two hospitals) versus the percutaneous compression plate (routinely used in one hospital).

Li 2010The English abstract of this trial otherwise reported in Chinese describes a study of 48 patients with a trochanteric hip fracture in which treatment was "randomly assigned" between a sliding hip screw or an anatomic plate. However, perusal of the text by a Chinese speaking doctor found no reference to randomisation and there was no response to an email request to the authors for clarification on trial methodology. It thus seems very unlikely that this was actually a randomised trial.

Moroni 2004This study of 120 participants compared a standard sliding hip screw with one in which the lag screw was coated in hydroxyapatite. This comparison does not fall in the scope of this review.

Olsson 2000This study was a comparison of extracapsular proximal femoral fractures treated with either a sliding hip screw or a twin hook modification of the sliding hip screw. It was excluded because it was not a randomised study.

Ouyang 2010The English abstract of this trial otherwise reported in Chinese describes a study of 92 patients with an upper femoral shaft fracture in which patients were "randomly divided" into three groups using an interlocking intramedullary nail, compression plate or Plum nail. This does not appear to be a comparison of extramedullary fixation implants. Moreover, a translation of the Methods showed this to be described as a "retrospective study", where "92 subjects were randomly allocated into 3 groups based on their preferences". Thus this was not a randomised trial.

Peyser 2005This was a retrospective comparison of 155 patients treated with either a sliding hip screw or a percutaneous compression plate. The study was excluded as there was no randomisation of patients.

Raunest 1995In this study, 93 stable trochanteric fractures were treated with a sliding hip screw, 44 unstable trochanteric fractures were treated with a sliding hip screw and 42 unstable trochanteric fractures were treated with a 95 degree blade plate. The study was excluded because there was no randomisation of patients to the different implants.

Romero 2008This was a study of 26 patients with a trochanteric hip fracture: 13 were treated with a percutaneous compression plate (PCCP) and 13 with a dynamic hip screw. Those patients treated with the PCCP were found to have a lower surgical time and a reduced need for blood transfusion. The study was excluded as it was a comparative study with no randomisation of patients.

Weisbrot 2005This study was reported only as a conference abstract as a randomised study involving 110 patients treated with either the Percutaneous Compression Plate (PCCP) or the sliding hip screw. There was a tendency to fewer complications with the PCCP (15% versus 22%). The exact number and nature of these complications was not given in the conference abstract. The study was excluded as there was a lack of information presented within the conference abstract, which additionally gave no contact address or email.

Zhen 2010The English abstract of this trial otherwise reported in Chinese describes a study of 57 patients with a trochanteric hip fracture who were "randomly divided" to fixation into two groups with either a trapezoid compression plate (21 patients) or an anatomic plate (36 patients). These patients recruited over nine years were followed up between 5 months and 9 years and 3 months. The imbalance in numbers and variation in length of follow-up raises questions about the trial methodology. There was no response to an email request to the authors for clarification on trial methodology but it is very unlikely that this was actually a randomised trial.

 
Characteristics of studies awaiting assessment [ordered by study ID]
ChiCTR 2011

Methods"Randomized parallel trial"

Participants75 patients with an intertrochanteric fracture, aged 60+ years, with independent ambulation before their fracture

InterventionsMinimally invasive percutaneous compression plating versus dynamic hip screw

OutcomesBlood loss, postoperative complications, VAS score, Harris hip score

NotesThe trial, conducted Oct 2007 to Feb 2010 was retrospectively registered on 27.10.2011

 
Comparison 1. Fixed nail plate versus sliding hip screw (SHS)

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Fracture fixation complications2Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    1.1 Cut-out
159Risk Ratio (M-H, Fixed, 95% CI)0.63 [0.13, 3.20]

    1.2 Non-union
186Risk Ratio (M-H, Fixed, 95% CI)2.78 [0.26, 29.48]

    1.3 Implant breakage
159Risk Ratio (M-H, Fixed, 95% CI)28.96 [1.79, 469.66]

    1.4 Fixation failure rate
2145Risk Ratio (M-H, Fixed, 95% CI)4.27 [2.44, 7.45]

    1.5 Reoperation
2145Risk Ratio (M-H, Fixed, 95% CI)1.97 [0.59, 6.58]

 2 Final outcome measures1Risk Ratio (M-H, Fixed, 95% CI)Totals not selected

    2.1 Mortality
1Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

    2.2 Pain at follow up
1Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

    2.3 Impaired mobility
1Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

 
Comparison 2. RAB plate versus sliding hip screw (SHS)

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Number of patients transfused1Risk Ratio (M-H, Fixed, 95% CI)Totals not selected

 2 Fracture fixation complications2Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    2.1 Operative complications
1166Risk Ratio (M-H, Fixed, 95% CI)9.22 [0.50, 168.52]

    2.2 Breakage or disassembly of the implant
1166Risk Ratio (M-H, Fixed, 95% CI)9.22 [0.50, 168.52]

    2.3 Cut-out
2352Risk Ratio (M-H, Fixed, 95% CI)0.57 [0.22, 1.49]

    2.4 Non-union
2352Risk Ratio (M-H, Fixed, 95% CI)1.04 [0.24, 4.53]

    2.5 Avascular necrosis
1166Risk Ratio (M-H, Fixed, 95% CI)0.34 [0.01, 8.26]

    2.6 Fracture below the implant
1166Risk Ratio (M-H, Fixed, 95% CI)3.07 [0.13, 74.34]

    2.7 Fixation failure rate
2352Risk Ratio (M-H, Fixed, 95% CI)0.99 [0.56, 1.74]

    2.8 Reoperation
2352Risk Ratio (M-H, Fixed, 95% CI)0.84 [0.49, 1.43]

 3 Wound infection2Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    3.1 Superficial wound infection
1166Risk Ratio (M-H, Fixed, 95% CI)1.02 [0.15, 7.10]

    3.2 Deep infection
2352Risk Ratio (M-H, Fixed, 95% CI)0.77 [0.18, 3.41]

 4 Thromboembolism1Risk Ratio (M-H, Fixed, 95% CI)Totals not selected

 5 Anatomical deformity2Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    5.1 Varus deformity
2352Risk Ratio (M-H, Fixed, 95% CI)0.62 [0.23, 1.67]

    5.2 Leg shortening
1186Risk Ratio (M-H, Fixed, 95% CI)0.14 [0.03, 0.59]

 6 Mortality2433Risk Ratio (M-H, Fixed, 95% CI)0.73 [0.46, 1.18]

 
Comparison 3. Pugh nail versus sliding hip screw (SHS)

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Fixation failure1Risk Ratio (M-H, Fixed, 95% CI)Totals not selected

 2 Deep infection1Risk Ratio (M-H, Fixed, 95% CI)Totals not selected

 3 Deep vein thrombosis1Risk Ratio (M-H, Fixed, 95% CI)Totals not selected

 4 Final outcome measures1Risk Ratio (M-H, Fixed, 95% CI)Totals not selected

    4.1 Mortality
1Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

    4.2 Pain at follow up (severe or moderate)
1Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

 
Comparison 4. Medoff sliding plate versus sliding hip screw (SHS)

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Cut-out3436Risk Ratio (M-H, Fixed, 95% CI)0.24 [0.07, 0.82]

 2 Detachment of the plate from the femur3436Risk Ratio (M-H, Fixed, 95% CI)0.19 [0.01, 3.58]

 3 Non-union of fracture3436Risk Ratio (M-H, Fixed, 95% CI)5.65 [0.28, 115.86]

 4 Fixation failure rate3436Risk Ratio (M-H, Fixed, 95% CI)0.30 [0.11, 0.81]

    4.1 Stable trochanteric fractures
258Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

    4.2 Unstable trochanteric fractures
3378Risk Ratio (M-H, Fixed, 95% CI)0.30 [0.11, 0.81]

 5 Reoperation2276Risk Ratio (M-H, Fixed, 95% CI)0.51 [0.12, 2.23]

 6 Thromboembolic complications1Risk Ratio (M-H, Fixed, 95% CI)Totals not selected

 7 Final outcome measures1Risk Ratio (M-H, Fixed, 95% CI)Totals not selected

    7.1 Mortality
1Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

    7.2 Failure to return to living at own home
1Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

    7.3 Failure to regain independent mobility
1Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

 
Comparison 5. Medoff sliding plate versus any of three other screw-plate systems

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Cut-out2Risk Ratio (M-H, Fixed, 95% CI)Totals not selected

    1.1 Unstable trochanteric fractures
1Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

    1.2 Subtrochanteric fractures
1Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

 2 Non-union2Risk Ratio (M-H, Fixed, 95% CI)Totals not selected

    2.1 Unstable trochanteric fractures
1Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

    2.2 Subtrochanteric fractures
1Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

 3 Implant breakage or disassembly2Risk Ratio (M-H, Fixed, 95% CI)Totals not selected

    3.1 Unstable trochanteric fractures
1Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

    3.2 Subtrochanteric fractures
1Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

 4 Fixation failure2Risk Ratio (M-H, Fixed, 95% CI)Totals not selected

    4.1 Unstable trochanteric fractures
1Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

    4.2 Subtrochanteric fractures
1Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

 5 Reoperation2Risk Ratio (M-H, Fixed, 95% CI)Totals not selected

    5.1 Unstable trochanteric fractures
1Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

    5.2 Subtrochanteric fractures
1Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

 6 Superficial wound infection2Risk Ratio (M-H, Fixed, 95% CI)Totals not selected

    6.1 Unstable trochanteric fractures
1Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

    6.2 Subtrochanteric fractures
1Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

 7 Deep wound infection2Risk Ratio (M-H, Fixed, 95% CI)Totals not selected

    7.1 Unstable trochanteric fractures
1Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

    7.2 Subtrochanteric fractures
1Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

 8 Thromboembolic complications2Risk Ratio (M-H, Fixed, 95% CI)Totals not selected

    8.1 Unstable trochanteric fractures
1Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

    8.2 Subtrochanteric fractures
1Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

 9 Mortality (1 year)2Risk Ratio (M-H, Fixed, 95% CI)Totals not selected

    9.1 Unstable trochanteric fractures
1Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

    9.2 Subtrochanteric fractures
1Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

 10 Failure to return home2Risk Ratio (M-H, Fixed, 95% CI)Totals not selected

    10.1 Unstable trochanteric fractures
1Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

    10.2 Subtrochanteric fractures
1Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

 11 Failure to regain mobility2Risk Ratio (M-H, Fixed, 95% CI)Totals not selected

    11.1 Unstable trochanteric fractures
1Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

    11.2 Subtrochanteric fractures
1Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

 
Comparison 6. Gotfried percutaneous compression plate (PCCP) versus sliding hip screw (SHS)

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Length of surgery (minutes)3Mean Difference (IV, Fixed, 95% CI)Totals not selected

 2 Intraoperative blood loss (ml)3277Mean Difference (IV, Random, 95% CI)-50.00 [-103.25, 3.24]

 3 Mean fall in haemoglobin (%)2211Mean Difference (IV, Fixed, 95% CI)-0.24 [-0.60, 0.11]

 4 Mean units of blood transfused4360Mean Difference (IV, Fixed, 95% CI)-0.23 [-0.43, -0.03]

 5 Number of patients transfused2169Risk Ratio (M-H, Fixed, 95% CI)0.81 [0.65, 1.02]

 6 X-ray screening time (seconds) (approximate SDs)1Mean Difference (IV, Fixed, 95% CI)Totals not selected

 7 Fracture fixation complications4Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    7.1 Cut-out
4395Risk Ratio (M-H, Fixed, 95% CI)1.30 [0.41, 4.13]

    7.2 Avascular necrosis
1103Risk Ratio (M-H, Fixed, 95% CI)3.18 [0.13, 76.20]

    7.3 Persistant pain requiring implant removal
1103Risk Ratio (M-H, Fixed, 95% CI)3.18 [0.13, 76.20]

    7.4 All fracture healing complications
4392Risk Ratio (M-H, Fixed, 95% CI)1.43 [0.51, 4.00]

    7.5 Reoperation
3329Risk Ratio (M-H, Fixed, 95% CI)1.75 [0.59, 5.22]

 8 Wound infection2Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    8.1 Superficial wound infection
2214Risk Ratio (M-H, Fixed, 95% CI)0.35 [0.04, 3.27]

    8.2 Deep infection
2214Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

 9 Post-operative complications2Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    9.1 Pneumonia
2214Risk Ratio (M-H, Fixed, 95% CI)0.73 [0.33, 1.60]

    9.2 Pressure sores
1111Risk Ratio (M-H, Fixed, 95% CI)0.51 [0.05, 5.45]

    9.3 Deep vein thrombosis
2214Risk Ratio (M-H, Fixed, 95% CI)0.35 [0.09, 1.41]

    9.4 Pulmonary embolism
2214Risk Ratio (M-H, Fixed, 95% CI)0.52 [0.10, 2.78]

    9.5 Cardiac complications
1111Risk Ratio (M-H, Fixed, 95% CI)0.17 [0.02, 1.36]

    9.6 Cerebrovascular accident
1103Risk Ratio (M-H, Fixed, 95% CI)0.12 [0.01, 2.13]

    9.7 Gastrointestinal complications
1103Risk Ratio (M-H, Fixed, 95% CI)1.41 [0.33, 6.00]

    9.8 Renal complications
1100Risk Ratio (M-H, Fixed, 95% CI)0.86 [0.31, 2.37]

 10 Length of hospital stay (days)1Mean Difference (IV, Fixed, 95% CI)Totals not selected

 11 Final outcome measures4Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    11.1 Mortality
4395Risk Ratio (M-H, Fixed, 95% CI)0.75 [0.49, 1.14]

    11.2 Loss of independent mobility
183Risk Ratio (M-H, Fixed, 95% CI)1.13 [0.50, 2.55]

    11.3 Loss of independence
183Risk Ratio (M-H, Fixed, 95% CI)0.13 [0.01, 2.25]

 
Comparison 7. Pertrochanteric external fixator versus sliding hip screw (SHS)

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Length of surgery (minutes)2Mean Difference (IV, Fixed, 95% CI)Totals not selected

 2 Operative blood loss (ml)1Mean Difference (IV, Fixed, 95% CI)Totals not selected

 3 Fall in haemoglobin (mg/dl)1Mean Difference (IV, Fixed, 95% CI)Totals not selected

 4 Number of patients transfused2140Risk Ratio (M-H, Fixed, 95% CI)0.04 [0.01, 0.22]

 5 Fracture fixation complications3Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    5.1 Cut-out
3200Risk Ratio (M-H, Fixed, 95% CI)0.54 [0.15, 1.89]

    5.2 All technical complications of fracture healing
3200Risk Ratio (M-H, Fixed, 95% CI)0.67 [0.20, 2.26]

    5.3 Re-operation (except procedures under local or no anaesthesia)
2160Risk Ratio (M-H, Fixed, 95% CI)0.14 [0.02, 1.14]

 6 Wound infection3Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    6.1 Superficial infection
3200Risk Ratio (M-H, Fixed, 95% CI)8.25 [3.05, 22.31]

    6.2 Deep wound infection
1100Risk Ratio (M-H, Fixed, 95% CI)0.33 [0.01, 7.99]

 7 Length of hospital stay (days)2Mean Difference (IV, Fixed, 95% CI)Totals not selected

 8 Final outcome measures1Risk Ratio (M-H, Fixed, 95% CI)Totals not selected

    8.1 Mortality (6 months)
1Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

    8.2 Failure to return to same residential status (survivors)
1Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

    8.3 Reduced walking ability (survivors)
1Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]