CX Cabeza Radial

Cochrane Database of Systematic Reviews
Surgical interventions for treating radial head fractures in

adults (Review)
Gao Y, Zhang W, Duan X, Yang J, Al-Qwbani M, Lv J, Xiang Z
Gao Y, Zhang W, Duan X, Yang J, Al-Qwbani M, Lv J, Xiang Z.

Surgical interventions for treating radial head fractures in adults.
Cochrane Database of Systematic Reviews 2013, Issue 5. Art. No.: CD008987.
DOI: 10.1002/14651858.CD008987.pub2.
www.cochranelibrary.com
Surgical interventions for treating radial head fractures in adults (Review)

Copyright © 2013 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
TABLE OF CONTENTS
HEADER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
PLAIN LANGUAGE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Figure 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Figure 2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Figure 3. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
AUTHORS’ CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
CHARACTERISTICS OF STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
DATA AND ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Analysis 1.1. Comparison 1 Radial head replacement versus ORIF, Outcome 1 Broberg and Morrey functional rating score
(0 to 100: best outcome) at 2 years. . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Analysis 1.2. Comparison 1 Radial head replacement versus ORIF, Outcome 2 Broberg and Morrey functional rating score
- excellent or good results. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
Analysis 1.3. Comparison 1 Radial head replacement versus ORIF, Outcome 3 Overall adverse events. . . . . . 27
Analysis 1.4. Comparison 1 Radial head replacement versus ORIF, Outcome 4 Individual adverse events. . . . . 28
Analysis 2.1. Comparison 2 Biodegradable pins versus standard metal screws, Outcome 1 Broberg and Morrey functional
rating score (0 to 100: best result) at 2 years. . . . . . . . . . . . . . . . . . . . . . . . 29
Analysis 2.2. Comparison 2 Biodegradable pins versus standard metal screws, Outcome 2 Broberg and Morrey functional
rating score - excellent or good results. . . . . . . . . . . . . . . . . . . . . . . . . . 30
Analysis 2.3. Comparison 2 Biodegradable pins versus standard metal screws, Outcome 3 Overall adverse events. . . 30
Analysis 2.4. Comparison 2 Biodegradable pins versus standard metal screws, Outcome 4 Individual adverse events. . 31
APPENDICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
CONTRIBUTIONS OF AUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
DECLARATIONS OF INTEREST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
SOURCES OF SUPPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
DIFFERENCES BETWEEN PROTOCOL AND REVIEW . . . . . . . . . . . . . . . . . . . . . 34
INDEX TERMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
Surgical interventions for treating radial head fractures in adults (Review) i

[Intervention Review]
Surgical interventions for treating radial head fractures in

adults
Yu Gao1 , Wei Zhang2 , Xin Duan3 , Jing Yang4 , Mohammed Al-Qwbani4 , Jingtong Lv4 , Zhou Xiang1
1 Department of Orthopedics, State Key Laboratory of Biotherapy, West China Hospital, Sichuan University, Chengdu, China.
2 Department of Orthopaedic Surgery, The General Hospital of the People’s Liberation Army (PLAGH), Beijing, China. 3 Department
of Orthopaedics, The Second People’s Hospital of Chengdu, Chengdu, China. 4 Department of Orthopaedic Surgery, West China
Hospital, Sichuan University, Chengdu, China
Contact address: Zhou Xiang, Department of Orthopedics, State Key Laboratory of Biotherapy, West China Hospital, Sichuan
University, No.37, Guo Xue Xiang, Chengdu, Sichuan, 610041, China. Xiangzhou15@hotmail.com.
Editorial group: Cochrane Bone, Joint and Muscle Trauma Group.

Publication status and date: New, published in Issue 5, 2013.
Review content assessed as up-to-date: 14 August 2012.
Citation: Gao Y, Zhang W, Duan X, Yang J, Al-Qwbani M, Lv J, Xiang Z. Surgical interventions for treating radial head fractures in
adults. Cochrane Database of Systematic Reviews 2013, Issue 5. Art. No.: CD008987. DOI: 10.1002/14651858.CD008987.pub2.
ABSTRACT
Background
Radial head fractures are the most common type of elbow fracture and can severely affect the function of the elbow. There is uncertainty
and controversy about when surgery is indicated as well as what type of surgical intervention is best.
Objectives
To assess the effects of surgical interventions for treating radial head fractures in adults. We aimed to compare surgical versus non-
surgical treatment, and different methods of surgical intervention.
Search methods
We searched the Cochrane Bone, Joint and Muscle Trauma Group Specialised Register (May 2012), the Cochrane Central Register of
Controlled Trials (The Cochrane Library; 2012 Issue 4), MEDLINE (1946 to May 2012), EMBASE (1980 to 2012 Week 19) and trial
registers (November 2011). No language restrictions were applied.
Selection criteria
All randomised and quasi-randomised controlled trials evaluating surgical interventions for treating radial head fractures.
Data collection and analysis
Two review authors independently selected trials, assessed risk of bias and extracted data. Where appropriate, results were pooled.
Main results
We included three randomised controlled trials, involving a total of 251 participants. All three trials were at high risk of performance
bias reflecting the fact that surgeons could not be blinded. One trial was at low risk of selection bias but was undermined by a high
attrition bias, in part resulting from post-randomisation exclusions. There were incomplete details of methodology for the other two
trials, which usually resulted in unclear risk of bias judgements.
Surgical interventions for treating radial head fractures in adults (Review) 1
Two trials compared radial head replacement with open reduction and internal fixation (ORIF) for treating Mason type III radial head
fractures. The trial authors reported outcomes at a mean of 2.8 years and 15 months respectively. There were significant differences
between the two groups in favour of radial head replacement in the Broberg and Morrey elbow scores (92.1 versus 72.4, mean difference
(MD) 19.70; 95% confidence interval (CI) 15.64 to 23.76; one trial, 45 participants), excellent or good Broberg and Morrey elbow
scores (33/36 versus 16/31, risk ratio (RR) 1.88; 95% CI 1.27 to 2.77; two trials), and overall adverse events (6/36 versus 15/31, RR
0.33; 95% CI 0.14 to 0.77; two trials). No statistically significant difference was found between the two groups in any of the reported
individual adverse events.
One trial compared biodegradable pins with standard metal screws in treating radial head fractures of AO-classification 21 B2. The
two types of fixation devices yielded similar results, with no significant between-group differences in the Broberg and Morrey scores
(93.3 versus 90.9, MD 2.40; 95% CI -0.10 to 4.90), excellent or good Broberg and Morrey elbow scores (72/74 versus 56/61, RR
1.06; 95% CI 0.97 to 1.15), and adverse events (13/82 versus 16/82, RR 0.81; 95% CI 0.42 to 1.58) at two-year follow-up.
Authors’ conclusions
Only tentative conclusions can be drawn from the available evidence in this review. Compared with ORIF, there was some evidence
that radial head replacement had better elbow function and fewer adverse events for Mason type III radial head fractures in the short
term. However, the evidence is of low quality and it is unknown whether these results would apply in the longer term or more generally.
Using biodegradable implants may be as good as metallic implants for fixation of some usually more stable fractures but more evidence
is needed to confirm this. There is a need for good quality evidence for addressing the areas of uncertainty for the surgical treatment
of radial head fractures.
PLAIN LANGUAGE SUMMARY

Surgical interventions for treating radial head fractures in adults
The radial head is located at the upper end of the radius, which is one of the two forearm bones. Radial head fractures are the most
common fractures of the elbow in adults. These can severely affect the function of the elbow. Less serious fractures are generally treated
without surgery and the more severe fractures with surgery. There are three main types of surgery. One is resection, where the fractured
radial head is removed. Another type is open reduction and internal fixation (ORIF), where the fractured bone is put back into position
and fixed in place by various devices such as screws. The third type of surgery is radial head replacement or arthroplasty, where the
radial head is replaced by a metal implant. There is uncertainty and controversy about when surgery is needed as well as what type of
surgical intervention is best.
This review includes evidence from three randomised controlled trials with a total of 251 participants. All three trials were at some risk
of bias, which means that their results may not be reliable.
Two trials compared radial head replacement with ORIF for treating highly fragmented radial head fractures. These trials showed that
after radial head replacement, patients had significantly better elbow function and fewer adverse events than those treated with ORIF
at between one and three years follow-up.
One trial compared biodegradable pins with standard metal screws in treating displaced radial head fractures. It found similar results
for the two types of materials in terms of elbow function and adverse events.
Overall, there is some evidence to support radial head replacement for treating highly fragmented radial head fractures instead of
attempting to fix the fractured bone back in place. However, the evidence is low quality and it is unknown whether these results would
apply in the longer term or more generally. Using biodegradable implants may be as good as metallic implants for fixing some usually
more stable fractures but more evidence is needed to confirm this.

BACKGROUND wires, and fixed angle/locked plates (Bano 2006). Currently, ORIF
is the most common choice of surgical intervention. Resection of
the radial head is rarely used, giving ground to replacement, using
Description of the condition metallic implants, as an alternative.
The radial head is located at the proximal (uppermost) part of the
radius, which is one of the two forearm bones. The radial head
plays a key role in stabilising the elbow. Radial head fractures are How the intervention might work
the most common fractures of the elbow in adults. They comprise Non-surgical means are generally chosen to treat undisplaced,
30% to 50% of elbow fractures, but only one per cent of fractures minimally displaced and displaced fractures with preservation of
in adults (Grant 2008). The main mechanism of injury for radial forearm rotatory motions; e.g., Mason I and some Mason II frac-
head fractures is a direct fall onto the outstretched hand with the tures. But sometimes, outcomes may be worsened due to immo-
elbow extended and the forearm pronated (palm of hand faces bilisation, and a stiff, often painful, elbow may result. Surgical
downwards) (Harrison 2007). The extent of the comminution treatment, which aims to restore pain-free elbow function, is often
(fragmentation) of the radial head is related to the energy of the fall. undertaken for more severe fractures including those with associ-
High energy trauma may result in other forearm and elbow injuries ated injuries and with damage to other elbow structures such as
such as the Essex-Lopresti injury (where there is an associated the medial and lateral collateral ligaments. There are three main
tear of the forearm interosseus membrane, a fibrous sheet between ways surgical intervention may improve outcome: improvement
the two forearm bones) and ’the terrible triad’ (involving elbow of congruency of the joint (potentially improving range of motion
dislocation and associated fracture of the coronoid process of the and preventing secondary degeneration), restoration of stability
ulna, the other forearm bone). to an unstable joint (Jensen 2003), and removal of mechanical
In 1954, a now-commonly-used classification system about radial block to function (Bano 2006). Additionally, surgery may provide
head fractures was put forth by Mason (Mason 1954). In 1962, enough stability to allow early mobilisation, hopefully avoiding
the system was modified by Johnston (Johnston 1962) to include stiffness. However, surgical intervention comes at the risk of surgi-
a Mason-Johnston Type IV (Harrison 2007). cally-related complications, such as nonunion and malunion, fix-
• Type I: marginal fractures with minimal displacement and ation failure, infection, post-traumatic osteoarthritis, heterotropic
no mechanical block to movement. ossification, and soft-tissue damage.
• Type II: marginal fractures with displacement. The type of surgery chosen will depend on the severity of the frac-
• Type III: comminuted fractures. ture and associated injuries. There will be consideration also of the
• Type IV: associated elbow dislocation. damage to other elbow structures including the medial and lat-
The main symptoms of an acute radial head fracture are pain on eral collateral ligaments that are important stabilisers of the elbow.
the outside of the elbow, swelling in the elbow joint, difficulty Removal of small fragments of bone is sometimes undertaken to
in bending or straightening the elbow accompanied by pain, and prevent these from blocking elbow movements. ORIF is generally
difficulty or inability to turn the forearm (palm up to palm down considered as a good choice for Mason II to IV fractures (Bano
or vice versa) (AAOS 2007). 2006), where there is significant displacement of a large fragment
and mechanical block to rotation. Resection or excision of the ra-
dial head is an alternative approach when the radial head cannot
Description of the intervention be repaired. Currently, resection is generally limited to older par-
ticipants who have intact medial and lateral collateral ligaments
Radial head fractures can be treated with or without surgery. Gen- (Bano 2006). However, resection can lead to late instability of the
erally, type I or undisplaced fractures, or minimally displaced frac- elbow from chronic overloading of the medial collateral ligament
tures without a rotation block are treated non-surgically (Bano resulting from the prolonged absence of the radial column support.
2006). Non-surgical interventions include immobilising the el- Thus, replacement of the radial head to restore joint mechanics
bow in a sling or cast for a week or more, followed by mobilisation by providing radial column support is an alternative to resection
exercises. Surgical interventions consist of excision of the fracture (Jensen 2003).
fragments, including resection (removal of the radial head), open
reduction (repositioning the fractured bone) and internal fixation
(ORIF) and radial head arthroplasty using usually metallic im-
plants (arthroplasty means a surgical reconstruction or replace-
Why it is important to do this review
ment of a malformed or degenerated joint). Fixation techniques Radial head fractures commonly and severely affect the function
have increased tremendously over the last three decades and now of the elbow. There is uncertainty and controversy about when
include small subarticular compression screws, interosseous head- surgery is indicated as well as what type of surgical intervention is
less screws, plates, absorbable pins, cannulated screws, threaded K- best; this is reflected by the variation in practice. There is thus a

need to inform practice by systematically reviewing the evidence We planned to include studies comparing various surgical tech-
for the use of surgery and for the different surgical interventions niques or implants (e.g. different types of arthroplasties) as well as
for treating radial head fractures. Our review will also extend and studies investigating any new surgical methods.
update the search (completed in January 2004) of a non-Cochrane We excluded studies that tested secondary surgery after failed non-
systematic review (Struijs 2007) comparing surgical versus non- surgical treatment.
surgical treatment for these fractures; this incidentally found no
randomised controlled trials for this comparison.
Types of outcome measures
OBJECTIVES Primary outcomes

1. Functional assessment, including generic and elbow-joint
To assess the effects (benefits and harms) of surgical interventions
specific validated clinical scores (e.g. the Broberg and Morrey
for treating radial head fractures in adults.
functional rating score) (Broberg 1987), patient-derived health
We aimed to compare surgical versus non-surgical (conservative) related quality of life measures (e.g. SF36) and pain measures
treatment, and different methods of surgical intervention. (e.g. Visual Analogue Scale).
2. Range of elbow motion and forearm rotation.
3. Adverse events (e.g. complications, nonunion of the radial
METHODS head fracture, late valgus laxity of the elbow, persistent instability
of the elbow, painful stiff elbow); and secondary procedures (re-
operation).
Criteria for considering studies for this review

Secondary outcomes
1. Return to activity/level of sport participation.
Types of studies
2. Pain.
Randomised and quasi-randomised (method of allocating partic- 3. Strength in flexion, extension, pronation, and supination.
ipants to a treatment which is not strictly random; e.g. by date 4. Operation time, blood loss.
of birth, hospital record number, alternation) controlled clinical
studies that evaluated surgical interventions for treating radial head
fractures in adults.
Search methods for identification of studies
Types of participants
Skeletally mature adults with a radial head fracture. Trials contain-
Electronic searches
ing participants with other elbow injuries were included provided
the primary focus was on the treatment of the radial head fracture. We searched the Cochrane Bone, Joint and Muscle Trauma Group
It was anticipated that surgery would be for primary treatment Specialised Register (accessed 14 May 2012), the Cochrane Cen-
of an acute fracture. While not formally excluded, it was antici- tral Register of Controlled Trials (The Cochrane Library 2012, Is-
pated that participants with nondisplaced or minimally displaced sue 4), MEDLINE (OvidSP; 1946 to May Week 1 2012) and
fractures without a rotation block would not be included in the EMBASE (OvidSP; 1980 to 2012 Week 19). No language restric-
eligible trials. tions were applied.
In MEDLINE, the subject-specific search strategy was combined
with the sensitivity-maximising version of the Cochrane Highly
Types of interventions Sensitive Search Strategy to identify randomised trials (Lefebvre
The main types of surgical intervention for treating radial head 2008). In EMBASE, the subject-specific strategy was combined
fractures are open reduction and internal fixation (ORIF), resec- with the Scottish Intercollegiate Guidelines Network search filter
tion and radial head replacement. Surgical intervention is com- for randomised controlled trials (SIGN 2008). All search strate-
pared with non-surgical intervention, which generally involves im- gies can be found in Appendix 1. We also searched Current
mobilisation with a plaster cast and/or sling. The main compar- Controlled Trials (Nov 2011), ClinicalTrials.gov (Nov 2011) and
isons of surgical interventions were resection versus ORIF; head the WHO International Clinical Trials Registry Platform (Nov
replacement versus ORIF; and head replacement versus resection. 2011) for unpublished and ongoing trials.

Searching other resources Measures of treatment effect
We searched reference lists of articles. We contacted experts in The treatment effects were expressed as risk ratios (RR) with
the field and companies that manufacture internal fixators and 95% confidence intervals for dichotomous outcomes (for exam-
prostheses. ple, number of participants who were pain-free). Where appropri-
ate, we planned to calculate and report numbers needed to treat
(NNTs) for outcomes where the RR was statistically significant.
The treatment effects for continuous outcomes were expressed as
mean differences (MD) with 95% confidence intervals (CI).
Data collection and analysis
Unit of analysis issues

We anticipated that the unit of analysis was the individual partic-
Selection of studies
ipant. We avoided unit of analysis problems associated with the
The search results were entered into EndNote 9, where duplicate reporting of multiple observations of the same outcome, in part
records were removed. The titles, abstracts and keywords of every by presenting data at clinically relevant time points.
record retrieved were scanned to identify potentially eligible stud-
ies. Full reports were obtained for all records that appeared likely
Dealing with missing data
to meet the inclusion criteria, including those of some doubt re-
garding eligibility. Inclusion decisions were made independently Where necessary, we contacted trial authors to request missing
by two review authors (YG and XD) according to the pre-stated data. Where possible, we performed intention-to-treat analyses to
eligibility criteria, and recorded on a specially developed form. A include all people randomised. However, where drop-outs were
third review author (ZX) was consulted when differences in opin- identified, the actual denominators of participants contributing
ions could not be resolved by discussion. Studies were blinded for data at the relevant outcome assessment were used and sensitivity
author, affiliation and source at all stages. analyses used to explore the potential effects of the missing data.
We were alert to the potential mislabelling or non identification of
standard errors and standard deviations. Unless missing standard
deviations could be derived from confidence intervals or standard
Data extraction and management errors, we did not assume values in order to present these in the
Two review authors (YG and WZ) independently extracted data analyses.
on trial and trial participants’ characteristics and results using a
piloted form. Any differences in data extraction were resolved by Assessment of heterogeneity
consensus, and by referring back to the original article. Informa-
Assessment for heterogeneity was carried out using the Chi² test
tion was sought from the authors of the primary studies when
with statistical significance being set at P < 0.1. The I² statistic was
necessary. Disagreements were resolved by discussion and, where
used to estimate the total variation across studies: an I² < 40% was
necessary, in consultation with a third review author (XD).
considered as no or little heterogeneity, 30% to 60% as moderate
level, 50% to 90% as substantial heterogeneity and 75% to 100%
as considerable heterogeneity (Higgins 2003). We also assessed the
Assessment of risk of bias in included studies heterogeneity by visual inspection of the forest plot.
The risk of bias of the included studies was independently assessed
by two review authors (YG, XD) using The Cochrane Collab- Assessment of reporting biases
oration’s ’Risk of bias’ tool (Higgins 2008). Using the updated We will investigate potential publication bias using a funnel plot
tool (Higgins 2011), we assessed the following: sequence genera- if there are 10 or more trials in a future update (Egger 1997).
tion, allocation concealment, blinding (participants and person-
nel), blinding of outcome assessment, completeness of outcome
data, selection of outcomes reported and other sources of bias. Data synthesis
The three other sources of bias assessed were that from major im- Where appropriate and possible, results of trials testing similar
balances in key baseline characteristics (e.g. age, gender, type of comparisons were pooled. Initially we used the fixed-effect model
fracture, and presence of other elbow injuries), performance bias and 95% confidence intervals. We also considered using the ran-
(e.g. differences in care provider experience and rehabilitation), dom-effects model, especially where there was unexplained hetero-
and that from conflicts of interest. A third review author (ZX) was geneity. For continuous outcomes, we used final outcome scores
asked to adjudicate on differences where consensus could not be in preference to change scores. If necessary in future updates of the
obtained by discussion. review, we will pool the results of continuous outcomes to yield

mean differences (MD), or standardised mean differences (SMD) Description of studies
where data were pooled from different scales or instruments for
See: Characteristics of included studies; Characteristics of
the same outcome, both with 95% confidence intervals (CI).
excluded studies; Characteristics of studies awaiting classification;
Characteristics of ongoing studies.
Subgroup analysis and investigation of heterogeneity
We intended to explore the following potential sources of hetero-
geneity using subgroup analyses (Higgins 2004). The main sub-
Results of the search
group analyses were by fracture type (based on the Mason classi-
fication), the presence of associated injuries, gender and age (pro- For this search (completed May 2012), we screened a total of
visionally, age up to 60 years, age 60 years and above). 357 records from the following databases: Cochrane Bone, Joint
and Muscle Trauma Group Specialised Register (155 records);
Cochrane Central Register of Controlled Trials (27), MEDLINE
Sensitivity analysis
(89), EMBASE (80), Current Controlled Trials (1), Clinicaltri-
If a sufficient number of trials had been found, we planned sensitiv- als.gov (5) and the WHO International Clinical Trials Registry
ity analyses examining various aspects of trial and review method- Platform (0). We did not identify any potentially eligible studies
ology, including study quality (specifically allocation concealment, from other sources (Sichuan University Library).
outcome assessor blinding), the effects of missing data, and use of The search resulted in the identification of 357 potentially eligible
fixed-effect and the random-effects models for pooling data where studies for which, where possible, full reports were obtained. Upon
there was substantial heterogeneity. study selection, three studies were included (Chen 2011; Helling
2006; Ruan 2009), five were excluded (Erturer 2010; Frankle
1999; Ikeda 2005; Liu 2010; Obert 2005) and one was an ongoing
study (Ascension 2009). A further study (Dabby 2006) awaits
RESULTS classification. Details of the search can be found in Figure 1.

Figure 1. Study flow diagram.

Overall, there are now three included trials, five excluded studies,
one ongoing trial and one study awaiting classification. planted from Wright Medical Technology. Then the annular liga-
ments were sutured, but not too tightly in order that the supina-
Included studies tion of the forearm was not affected. Complicating fractures and
other injuries, such as ligament injury and articulation dislocation,
For further details about the individual trials, please see the
were dealt with simultaneously. Ruan 2009 used the posterolateral
Characteristics of included studies.
approach and posterolateral capsular incision to expose the radial
head fractures. After radial head prosthesis implantation, the an-
Design nular ligament was repaired with nonabsorbable sutures. Partici-
All three included studies, which were fully reported in medical pants in the ORIF group were treated with screws and Kirschner
journals, were randomised controlled trials. wires in both Chen 2011 and Ruan 2009. Helling 2006 compared
biodegradable polylactide pins with standard mini-fragment metal
screws of ORIF. The polylactide used was a copolymer of co-L/
Sample sizes DL-lactide in a 70:30 ratio.
The three included trials recruited a total of 251 participants. Different details about postoperative rehabilitation were reported
Chen 2011 randomised 45 participants, with apparently none lost by the three trials. Chen 2011 allowed active exercises after three
to follow-up. Helling 2006 randomised 184 participants by tele- days to six weeks, depending on surgical methods and complicat-
phone by an independent agency but excluded 19 after randomi- ing injuries. Helling 2006 performed physiotherapy as a postop-
sation and lost a further 20 at two years follow-up. Ruan 2009 erative treatment for up to six weeks. In Ruan 2009, rehabilita-
randomised 22 participants with no loss to follow-up. tion started 48 hours postoperatively in the replacement group if
participants had no associated injuries, while rehabilitation started
when plasters were removed four weeks after surgery in the ORIF
Setting group.
Chen 2011 was conducted at Changhai Hospital in China. Helling
2006 was a multicentre trial, conducted in 24 centres in Germany.
Ruan 2009 was conducted at Shanghai 6th People’s Hospital in Outcomes
China. Clinical and radiological examinations were conducted and ad-
verse events reported in all three trials. Adverse events were de-
Participants fined differently in each trial. In Chen 2011, adverse events in-
cluded deep wound infection, range of motion deficit > 30°, no,
The majority of trial participants were men in all three trials (%
or delayed healing, internal fixation failure, secondary fragment
men = 76% (Chen 2011); 64% (Helling 2006); 59% (Ruan
displacement, joint stiffness and heterotopic ossification. Helling
2009)). The mean age of trial participants was 37 years in Chen
2006 gave a more complex definition of adverse events, with for-
2011, 39 in Helling 2006 and 39 years in Ruan 2009.
eign-body reactions of the soft tissues grade 2 to 4 (seroma, spon-
The mean time from injury to surgery was 4.9 days in Chen
taneous discharge, bacterial superinfection) also being included.
2011 and 5.8 days in Helling 2006. Ruan 2009 included fresh
In Ruan 2009, adverse events included heterotopic ossification,
and old fractures. All participants in Ruan 2009 who had fresh
bone nonunion and bone absorption; the latter two were reported
fractures were treated within four to 17 days after injury, and
together.
the two participants with old fractures were treated after 12 and
18 months of injury. Mason III type radial head fractures were
included in Chen 2011 (with multiple injuries) and Ruan 2009.
Excluded studies
Helling 2006 included 21 B2 type isolated fractures of the radial
head with fragment displacement > 1.0 mm or > 10°. There were five excluded studies. Two compared radial head re-
placement and ORIF (Ikeda 2005; Liu 2010). Erturer 2010 com-
pared mini-screw fixation and K-wire fixation for ORIF. Frankle
Interventions 1999 compared conservative and surgical treatment, the latter in-
Both Chen 2011 and Ruan 2009 compared radial head replace- cluded radial head replacement and ORIF. Obert 2005 compared
ment with ORIF. Helling 2006 compared two types of implants radial head resection, Swanson implant and Judet prosthesis.
for ORIF. Both Frankle 1999 and Obert 2005 were retrospective studies.
Chen 2011 used the Kocher approach to expose the radial head The other three studies were excluded because they were not ran-
fractures. Monopolar titanium radial head prostheses were im- domised or were quasi-randomised trials.

Studies awaiting classification Risk of bias in included studies
Further details of one study (Dabby 2006) are required (see the Inevitably, some risk of bias was likely due to difficulties in blinding
Characteristics of studies awaiting classification). in these studies. Requests to the trial investigators of all three
trials for clarification of study methods were unsuccessful. For
summaries of the ’Risk of bias’ assessment, please see Figure 2 and
Ongoing studies
Figure 3.
We identified one ongoing trial (Ascension 2009), details of which
can be found in the Characteristics of ongoing studies.
Figure 2. ’Risk of bias’ graph: review authors’ judgements about each risk of bias item presented as
percentages across all included studies.

Figure 3. ’Risk of bias’ summary: review authors’ judgements about each risk of bias item for each included
study.

Allocation
cases in the ORIF group were followed up for an average of 14
All participants in all three trials were randomised into two groups months (range 10 to 21 months).
but only Helling 2006 described their randomisation method,
in which an independent agency randomly allocated participants
by telephone. Chen 2011 and Ruan 2009 did not describe their Broberg and Morrey elbow scores
allocation methods and were considered at unclear risk of bias. Broberg and Morrey elbow scores were reported only in Chen
2011. At two years, the scores of participants following radial
head replacement were significantly higher than the participants
Blinding
following ORIF (see Analysis 1.1: mean difference (MD) 19.70;
Blinding of participants and personnel was not described in all 95% confidence interval (CI) 15.64 to 23.76). The outcome was
three studies. However, it was not possible to blind the operating considered to be satisfactory if the result was good or excellent and
surgeons in these trials. In Chen 2011, all reviewers were allegedly unsatisfactory if it was fair or poor (Chen 2011). Pooled results
blinded to evaluate postoperative functions by means of “a ran- from both trials for participants with excellent or good results were
domised block design”. It is not clear how this achieved blinding. significantly in favour of radial head replacement (see Analysis 1.2:
The other two trials did not report on assessor blinding and were 33/36 versus 16/31; risk ratio (RR) 1.88; 95% CI 1.27 to 2.77).
judged at high risk of detection bias.
Adverse events
Incomplete outcome data
Data on adverse events were available for both trials. There were
The inclusion criteria, outcome measures and loss of follow-up
fewer adverse events in the radial head replacement group than in
were clearly defined in all three studies. Exclusion criteria were
the ORIF group (see Analysis 1.3: 6/36 versus 15/31; RR 0.33;
clearly defined in Chen 2011 and Helling 2006. Helling 2006 was
95% CI 0.14 to 0.77). Adverse events in the two studies included
at high risk of attrition bias because of a high number of post-
range of motion deficit more than 30°, stiffness caused by prosthe-
randomisation exclusions and drop-outs.
ses, bone nonunion, an ill-defined combined outcome of “bone
and bone absorption”, secondary fragment displacement, deep
Selective reporting wound infection and heterotopic ossification. There were more
participants with range of motion more than 30°, bone nonunion,
We did not find any protocols for these trials to judge this item.
bone union and bone absorption, secondary fragment displace-
ment and deep wound infection in the ORIF group and more
Other potential sources of bias participants with stiffness caused by prostheses were in the radial
head replacement group. The two groups had a similar incidence
We judged two trials as being at unclear risk of other bias but
of heterotopic ossification. There was no statistically significant
considered Ruan 2009 at high risk of bias, given its small size and
difference between the two groups for any of the individual ad-
that two old fractures were included in the replacement group.
verse events (see Analysis 1.4).
Effects of interventions
Requests to the trial investigators of all three trials for further data Biodegradable implants versus standard metal mini-
were unsuccessful. fragment implants
The one trial in this category (Helling 2006) compared biodegrad-
able polylactide pins with standard mini-fragment metal screws of
Radial head replacement versus ORIF ORIF in 184 participants with simple radial head fractures (type
Two trials (Chen 2011 and Ruan 2009) compared radial head B2.1) and multi-fragmented radial head fractures without depres-
replacement with ORIF. Chen 2011 included 45 participants with sion (type B2.2) or with depression (type B2.3). Two-year follow-
unstable, multi-fragmented fractures of the radial head in their up data were provided for 135 participants.
analyses. The mean length of follow-up was 2.8 years (range one to
five years) with no loss of follow-up. Ruan 2009 included 22 people
with radial head fractures of Mason type III. The 14 participants Broberg and Morrey elbow scores
who had received radial head replacement were followed up for an There was little difference between the two groups in Broberg and
average of 15.9 months (range 10 to 27 months), and the eight Morrey elbow scores (see Analysis 2.1: MD 2.40; 95% CI -0.10

to 4.90). Similarly, there was no significant difference between the Helling 2006 - in 2002, several authors of Helling 2006 reported
two groups in the attainment of excellent or good Broberg and on the use of the “new Polypin C fixation pin” for radial head
Morrey scores (see Analysis 2.2: 72/74 versus 56/61; RR 1.06; 95% fractures (Prokop 2002).
CI 0.97 to 1.15). Though the inclusion of two ’old’ fractures in the study popu-
lation of Ruan 2009 was questionable, the study populations of
both replacement versus fixation trials appeared appropriate. This
Adverse events review did not find evidence to inform on when surgery is indi-
There was no statistically significant differences between the two cated but some might question whether the threshold for fracture
groups in overall adverse events (see Analysis 2.3: 13/82 versus 16/ displacement (1 mm or 10°) for study inclusion was too low in
82; RR 0.81; 95% CI 0.42 to 1.58). Adverse events in Helling Helling 2006.
2006 included range of motion deficit more than 30°, secondary All three included studies reported on loss to follow-up, which
fragment displacement, re-operation, osteolysis in grade 4 (distur- at 22% was serious only in Helling 2006. Outcome data used in
bance of fracture healing), and deep wound infection. Mone of those studies were adequately described overall but the assessment
the differences between the two groups in the individual compli- was not comprehensive. In our outcomes of interest, only func-
cations were statistically significant (see Analysis 2.4). tional assessment (the Broberg and Morrey elbow score) and ad-
verse events were reported. Other outcomes, such as range of el-
bow motion and forearm rotation, return to activity, level of sport
participation, pain, strength, operation time, and blood loss, were
DISCUSSION not listed in the three studies. The length of follow-up was too
short in the two replacement versus fixation trials, given that com-
plications relating to arthroplasty often appear later than those for
fixation.
Summary of main results
The applicability of evidence generated from single centres (as in
Two trials, with a total of 67 participants, compared radial head Chen 2011 and Ruan 2009) is also in question.
replacement with ORIF in the treatment of Mason type III radial
head fractures (Chen 2011; Ruan 2009). These trials reported
superior results in the radial head replacement group in the mean
Quality of the evidence
Broberg and Morrey elbow scores, in the numbers of participants
with excellent or good Broberg and Morrey elbow scores, and While all three included studies claimed to be randomised, details
overall adverse events at an average of 2.8 years and 15 months of the randomisation procedure were only reported by Helling
respectively. 2006. Although this trial appeared to be at low risk of selection
One trial (Helling 2006), reporting two-year outcomes for 135 bias, it was judged at high risk of attrition bias due to the high
participants, compared biodegradable pins versus standard metal loss to follow-up, including post-randomisation exclusions. The
screws for treating radial head fractures in the AO-classification other two studies were considered at unclear risk of selection bias.
21 B2 category. Helling 2006 found no significant differences The small size of Ruan 2009 and imbalances between the two
between the two groups in the Broberg and Morrey scores or overall groups in size and inclusion of old fractures makes it particularly
adverse events at two-year follow-up. susceptible to bias. All three studies did not describe blinding
of participants and personnel, perhaps because it was difficult to
blind participants to the surgical intervention and impossible to
Overall completeness and applicability of blind the operating surgeon. While Chen 2011 described blinding
evidence of outcome assessment, the effectiveness of the method used was
unknown. Thus, all three studies were at some risk of various biases
After comprehensive searching, only three studies could be in- and, linked with some reservations over applicability, the quality
cluded in our review. Two small studies compared radial head re- of the evidence is at ’low’ (further research is very likely to have an
placement with ORIF and one compared two types of prostheses important impact on our confidence in the estimate of effect and
material used in ORIF. While broadly speaking these two com- is likely to change the estimate).
parisons are relevant to current practice, there are some questions
regarding the actual prostheses and implants used. For instance,
there are many different types of radial head prostheses used in
current practice (Stuffmann 2009) and, as pointed out at external
Potential biases in the review process
review, none of the trials used “the new generation of anatomic With the exception of a few items (see Differences between protocol
implants” for internal fixation (Limb 2013). We are uncertain of and review), the review was conducted following the methods in
the availability or current use of the biodegradable pin used in the published protocol. A comprehensive database search was per-

formed but it is possible that we have missed unpublished studies surgical interventions are the most appropriate for the manage-
and study findings. We attempted to contact trial authors for fur- ment of different types of radial head fractures in adults. There
ther information but were unsuccessful. is low quality evidence that radial head replacement when com-
pared with open reduction and internal fixation had better elbow
function and fewer adverse events for Mason type III radial head
fractures in the short term. However, it is unknown whether these
Agreements and disagreements with other results would apply in the longer term or more generally, including
studies or reviews for the newer types of internal fixation implants. Using biodegrad-
A recently published narrative review, which covers all treatment able implants may be as good as metallic implants for fixation of
options for these fractures, provides a general overview of current some usually more stable fractures but more evidence is needed to
practice set in the context of evidence from various study designs confirm this.
(Duckworth 2013). Duckworth 2013 points to similar areas of
controversy as described in our review. They summarise, without Implications for research
any critical appraisal other than a comment on study design, some Well conducted and reported, multi-centre randomised controlled
of the available evidence for various treatment methods. Our re- trials involving commonly used interventions in current practice
view aimed to inform practice by identifying, evaluating and sum- are warranted to address the current lack and deficiencies of the
marising the evidence from randomised controlled trials of treat- evidence to inform key treatment decisions for these fractures. It
ment interventions for these fractures. Unfortunately, the lack of is important to precede these by obtaining a reasonable consen-
good quality evidence to inform practice limits the practical use- sus amongst orthopaedic surgeons on what are the priority ques-
fulness of our review. In areas of minimal treatment uncertainty, tions. These are likely to include whether surgery is preferable for
instructional reviews such as Duckworth 2013 help inform prac- marginally displaced fractures, and whether radial head replace-
tice but a more critical and rigorous evaluation of evidence is cru- ment versus ORIF gives superior results in the long term for those
cial when addressing areas of treatment uncertainty. unstable fractures that could be suitable for either of these meth-
The use of biodegradable materials as in Helling 2006 is linked ods. As well as the systematic collection of validated patient-re-
with concerns about foreign-body reactions. Helling 2006 found ported function and adverse effects data in all trials, follow-up
no complications, such as seroma, fistula or sinus formation result- should be at least five years in any trial evaluating radial head pros-
ing from foreign body reactions. Other investigators have also pro- theses.
vided some reassurance. Though histological examination showed
a typical nonspecific foreign-body reaction with abundant giant This Cochrane review should be maintained and updated as fur-
cells, Bostman 1990 reported that these reactions did not influ- ther randomised controlled trials become available. The authors
ence the clinical or radiographic results. Similarly, Ignatius 1996 would be pleased to receive information about any other ran-
confirmed satisfactory biocompatibility of two biodegradable ma- domised controlled trials relating to the treatment of radial head
terials. fractures.
ACKNOWLEDGEMENTS
AUTHORS’ CONCLUSIONS
We are grateful to Helen Handoll, Mario Lenza and David Limb
for valuable comments about drafts of the review; and Mario
Implications for practice Lenza, Ben Ollivere and Adam Watts for helpful comments about
Overall, there is either no, or insufficient evidence from ran- the protocol. We would also like to acknowledge Joanne Elliott for
domised or quasi-randomised controlled trials to determine which developing the search strategies, and Lindsey Elstub for support.

REFERENCES
References to studies included in this review fu chong jian wai ke za zhi = Zhongguo xiufu chongjian waike
zazhi = Chinese Journal of Reparative and Reconstructive
Chen 2011 {published data only} Surgery 2010;24(8):900–3. [EMBASE: 20839431;
Chen X, Wang SC, Cao LH, Yang GQ, Li M, Su JC. PUBMED: 20839431]
Comparison between radial head replacement and open
Obert 2005 {published data only}
reduction and internal fixation in clinical treatment
Obert L, Lepage D, Huot D, Givry F, Clappaz P,
of unstable, multi-fragmented radial head fractures.
Garbuio P, et al.Unreconstructible radial head fracture:
International Orthopaedics 2011;35(7):1071–6. [EMBASE:
resection, implant of Swanson or prosthesis? Retrospective
2011430181; MEDLINE: 20680272]
comparative study [Fracture de tete radiale non synthesable:
Helling 2006 {published data only} resection, implant de Swanson ou prothese? Etude
∗
Helling HJ, Prokop A, Schmid HU, Nagel M, Lilienthal retrospective comparative]. Chirurgie de la Main 2005;24
J, Rehm KE. Biodegradable implants versus standard metal (1):17–23. [MEDLINE: 15754706]
fixation for displaced radial head fractures. A prospective,
randomized, multicenter study. Journal of Shoulder and References to studies awaiting assessment
Elbow Surgery 2006;15(4):479–85. [PUBMED: 16831654]
Rehm KE, Helling HJ, Gatzka C. New developments in the Dabby 2006 {published data only}
application of resorbable implants [Neue Entwicklungen Dabby D, Patisch H, Blumberg N, Cohen I, Jakim I. Radial
beim Einsatz resorbierbarer Implantate]. Orthopäde 1997; head fractures - ORIF vs head prosthesis. Prospective
26(5):489–97. [MEDLINE: 9312708] randomized trial. Journal of Bone and Joint Surgery - British
Rehm KE, Nagel M, Lilienthal J, Helling HJ. Multi-centre Volume 2006;88-B (Suppl 2):346.
study radius head, metal vs. polypin - An interim report
[Multicenterstudie radiuskopfchen, metall vs polypin – Ein References to ongoing studies
zwischenbericht]. Hefte zur der Unfallchirurg 1998;265:
195–202. Ascension 2009 {published data only}
Ruan 2009 {published data only} Ascension Orthopedics Inc. Safety and effectiveness
Ruan HJ, Fan CY, Liu JJ, Zeng BF. A comparative study study of Ascension’s PyroCarbon radial head compared to
of internal fixation and prosthesis replacement for radial Ascension’s metal radial head. Accessed at clinicaltrials.gov/
head fractures of Mason type III. International Orthopaedics ct2/show/NCT00825409?term=NCT00825409&rank=1.
2009;33(1):249–53. [PUBMED: 17938924] Additional references
References to studies excluded from this review AAOS 2007
Erturer 2010 {published data only} American Academy of Orthopaedic Surgeons. Radial
Erturer E, Seckin F, Akman S, Toker S, Sari S, Ozturk I. head fractures. http://orthoinfo.aaos.org/topic.cfm?topic=
The results of open reduction and screw or K-wire fixation A00073 (accessed 03/11/10) 2007.
for isolated type II radial head fractures. Acta Orthopaedica Bano 2006
et Traumatologica Turcica 2010;44(1):20–6. [PUBMED: Bano KY, Kahlon RS. Radial head fractures - advanced
20513987] techniques in surgical management and rehabilitation.
Frankle 1999 {published data only} Journal of Hand Therapy 2006;19(2):114–35. [MEDLINE:
Frankle MA, Koval KJ, Sanders RW, Zuckerman JD. Radial 16713860]
head fractures associated with elbow dislocations treated Bostman 1990
by immediate stabilization and early motion. Journal of Böstman O, Hirvensalo E, Mäkinen J, Rokkanen P.
Shoulder and Elbow Surgery 1999;8(4):355–60. [PUBMED: Foreign-body reactions to fracture fixation implants of
10472010] biodegradable synthetic polymers. Journal of Bone and Joint
Ikeda 2005 {published data only} Surgery - British Volume 1990;72(4):592–6. [PUBMED:
Ikeda M, Sugiyama K, Kang C, Takagaki T, Oka Y. 2199452]
Comminuted fractures of the radial head. Comparison of Broberg 1987
resection and internal fixation. Journal of Bone and Joint Broberg MA, Morrey BF. Results of treatment of fracture-
Surgery - American Volume 2005;87(1):76–84. [PUBMED: dislocations of the elbow. Clinical Orthopaedics and Related
15634816] Research 1987;(216):109–19. [EMBASE: 1987108216;
Liu 2010 {published data only} PUBMED: 3102139]
Liu B, Lin Z, Cao L, Kang Y. Effectiveness comparison Duckworth 2013
between open reduction combined with internal fixation Duckworth AD, McQueen MM, Ring D. Fractures of the
and artificial radial head replacement in treating Mason radial head. Bone and Joint Journal 2013;95-B(2):151–9.
type-III comminuted fractures of radial head. Zhongguo xiu [PUBMED: 23365021]
Egger 1997 Johnston 1962
Egger M, Davey Smith G, Schneider M, Minder C. Bias Johnston GW. A follow-up of one hundred cases of fracture
in meta-analysis detected by a simple, graphical test. BMJ of the head of the radius with a review of the literature.
1997;315(7109):629–34. [MEDLINE: 9310563] Ulster Medical Journal 1962;1(31):51–56. [PUBMED:
Grant 2008 14452145]
Grant S, Sharma H. Radial head fracture management: Lefebvre 2008
current concepts and controversies. Minerva Ortopedica e Lefebvre C, Manheimer E, Glanville J. Chapter 6: Searching
Traumatologica 2008;59(1):49–60. for studies. In: Higgins JPT, Green S (editors). Cochrame
Harrison 2007 Handbook for Systematic Reviews of Interventions
Harrison JWK, Chitre A, Lammin K, Warner JG, Hodgson Version 5.0.1 [updated September 2008]. The Cochrane
SP. Radial head fractures in adults. Current Orthopaedics Collaboration, 2008. Available from www.cochrane-
2007;21(1):59–64. [EMBASE: 2007141745] handbook.org.
Higgins 2003 Limb 2013
Higgins JP, Thompson SG, Deeks JJ, Altman DG. Limb D. external referee feedback February 28 2013.
Measuring inconsistency in meta-analyses. BMJ 2003;327 Mason 1954
(7414):557–60. [MEDLINE: 12958120] Mason ML. Some observations on fractures of the head
Higgins 2004 of the radius with a review of one hundred cases. British
Higgins JP, Thompson SG. Controlling the risk of spurious Journal of Surgery 1954;42(172):123–32. [PUBMED:
findings from meta-regression. Statistics in Medicine 2004; 13209035]
23(11):1663–82. [MEDLINE: 15160401] Prokop 2002
Higgins 2008 Prokop A, Jubel A, Helling HJ, Udomkaewkanjana C,
Higgins JPT, Green S (editors). Cochrane Handbook for Brochhagen HG, Rehm KE. New biodegradable polylactide
Systematic Reviews of Interventions Version 5.0.0 [updated implants (Polypin-C) in therapy for radial head fractures
February 2008]. The Cochrane Collaboration, 2008. [Neue biodegradable Polylactid–implantate (Polypin® –C)
Available from www.cochrane-handbook.org. zur Therapie von Radiuskopffrakturen]. Der Chirurg 2002;
Higgins 2011 73(10):997–1004. [PUBMED: 12395158]
Higgins JPT, Green S (editors). Cochrane Handbook SIGN 2008
for Systematic Reviews of Interventions Version 5.1.0 Scottish Intercollegiate Guidelines Network. Search filters.
[updated March 2011]. The Cochrane Collaboration, Randomised controlled trials. Embase (accessed 19 April
2011. Available from www.cochrane-handbook.org. 2010). http://www.sign.ac.uk/methodology/filters.html#
Ignatius 1996 random.
Ignatius AA, Claes LE. In vitro biocompatibility of Struijs 2007
bioresorbable polymers: poly(L, DL-lactide) and poly(L- Struijs PA, Smit G, Steller EP. Radial head fractures:
lactide-co-glycolide). Biomaterials 1996;17(8):831–9. effectiveness of conservative treatment versus surgical
[PUBMED: 8730968] intervention. A systematic review. Archives of Orthopaedic
Jensen 2003 and Trauma Surgery 2007;127(2):125–30. [PUBMED:
Jensen SL, Deutch SR, Olsen BS, Sojbjerg JO, Sneppen O. 17066285]
Laxity of the elbow after experimental excision of the radial Stuffmann 2009
head and division of the medial collateral ligament: efficacy Stuffmann E, Gannon A, Clemente J, Baratz M. Radial
of ligament repair and radial head prosthetic replacement: head prosthesis update. Techniques in Shoulder & Elbow
a cadaver study. Journal of Bone & Joint Surgery - British Surgery 2009;10(1):31–8.
Volume 2003;85(7):1006–10. [MEDLINE: 14516036] ∗
Indicates the major publication for the study

CHARACTERISTICS OF STUDIES
Characteristics of included studies [ordered by study ID]
Chen 2011
Methods Method of randomisation: randomised.

Assessor blinding: yes.
Loss to follow-up at 2 years: 0/45.
Patients’ consent to randomisation: yes.
Number of surgeons: same surgeon performed the radial head replacement operations,
but no information on who performed ORIF
Participants Study location: Changhai Hospital, Shanghai, China.

Study period: January 2004 to January 2009.
Study participants: 45 participants with unstable, multi-fragmented fractures of the
radial head
Age range: mean age 37 years, range 19 to 68 years.
Gender: 11 women and 34 men.
Inclusion criteria:
1. Mason III type radial head fracture
2. Including multiple injuries: collateral ulnar ligament injury, ulnar coronoid process
fracture, ulnar anconeal process or ulnar proximal end fracture, interosseous ligament
injury or/and posterior cubital articulation dislocation
Exclusion criteria:
1. Mason I, II type radial head fracture
2. Severely comminuted radial head fracture
3. Old radial head fracture
4. Children with radial head fracture
5. Psychological and/social conditions with poor compliance
Interventions Interventions started: within a mean of 4.9 days after injury

1. Radial head replacement
Using a monopolar titanium radial head prostheses matched to each individual from
Wright Medical Technology.
2. ORIF
Using the AO mini steel plate interfixed system combined with Kirschner wires.
Postoperative rehabilitation:
Radial head replacement: active exercises were started after three to seven days rest in a
splint except for cases of additional elbow injuries. If complicating injuries were present,
the elbow joints were rested in a 60° flexion position for three weeks followed by reha-
bilitation exercises, with gradually increased intensity
ORIF: Postoperatively, for patients with associated injuries, long-arm cast plaster was
used to immobilise the elbow at 90°, which was changed to short-arm cast plaster after
two weeks. Meanwhile, active rotation of the forearm was allowed. After four weeks, the
plaster was removed followed by rehabilitation exercises. For patients without associated
injuries, the elbow joints were suspended at 90° for two weeks and active rotational
movement of the forearm was undertaken. Two weeks later, active flexion-extension
movement was recommended. The whole process was taken with gradually increased
intensity

Chen 2011 (Continued)
Assigned: 22/23
Completed: 22/23
Outcomes Length of follow-up: 2.8 years (range 1 to 5 years)

Broberg and Morrey functional rating scores (including range of motion, grip strength,
functional stability, and pain)
Adverse events (including deep wound infection, range of motion deficit > 30°, no or
delayed healing, internal fixation failure, secondary fragment displacement, joint stiffness
and heterotopic ossification)
Notes Postoperatively, for participants with associated injuries, long-arm cast plaster was used
to immobilise the elbow at 90°, which was changed to short-arm cast plaster after two
weeks. After four weeks, the plaster was removed followed by rehabilitation exercises.
For participants without associated injuries, the elbow joints were suspended at 90° for
two weeks and active rotational movement of the forearm was undertaken
Risk of bias
Bias Authors’ judgement Support for judgement
Random sequence generation (selection Unclear risk All 45 participants were randomly divided
bias) into the radial head replacement group (22
cases) and the ORIF group (23 cases). But
randomisation method was not described
Allocation concealment (selection bias) Unclear risk The method of concealment was not de-
scribed.
Blinding of participants and personnel High risk Surgeons performing the operations could
(performance bias) not be blinded.
All outcomes
Blinding of outcome assessment (detection Unclear risk The author wrote that “all of the review-
bias) ers were blinded to evaluate postoperative
All outcomes functions by means of a randomised block
design”, but no information to show the
success of this unusual process
Incomplete outcome data (attrition bias) Low risk No participant was lost and outcome data
All outcomes were complete.
Selective reporting (reporting bias) Unclear risk The study protocol was not found and it
was unclear if the published report included
all planned outcomes
Other bias Unclear risk The authors declared they had no conflicts
of interest.
Separate baseline characteristics were not
provided for the two groups

Chen 2011 (Continued)
No details stated of the surgeons conduct-

ing ORIF, but these seemed to experienced
Helling 2006
Methods Method of randomisation: randomised by telephone by an independent agency.

Assessor blinding: unknown.
Loss to follow-up at 2 years: 29/164 (29 loss of follow-up).
Number of surgeons: unknown.
Participants Study location: 24 centres, Germany.

Study period: April 1993 to September 1997.
Study participants: 184 participants were treated. Nineteen were excluded because of
radial head resection in 9 participants, use of biodegradable implants other than the study
material in 3, radial head prosthesis implantation in 1, non-operative treatment in 1, and
unreliable available data (i.e., the documented data were contradictory or completely
missing) in 5 participants, all of whom stemmed from 1 centre. Initial pre-operative data
were collected from 165 participants but 1 patient’s notes were lost. Two-year follow-up
data were available for 135 patients
Age: mean age 39 years, range 16 to 71 years.
Gender: 59 women, 106 men.
Inclusion criteria:
1. Isolated fracture of the radial head
2. AO-classification 21 B2 (any subgroup)
3. Fragment displacement > 1.0 mm or > 10°
4. Informed consent, written or oral in the presence of a witness
Exclusion criteria:
1. AO-classification type “A”, radial neck
2. AO-classification type “C”, additional olecranon
3. Immature skeleton
4. Implants in place from previous intervention
5. Severe chronic and/or terminal diseases
6. Psychological and/or social conditions preventing rehabilitation
7. Missing informed consent
8. Advanced osteoporosis preventing surgical treatment of fractures
Interventions Interventions started: within a mean of 5.8 days after injury

1. Biodegradable polylactide pins
Participants were fixed with 2.0-mm-diameter polylactide pins with an original length
of 35 mm. The polylactide used was a copolymer of co-L/DL-lactide in a 70:30 ratio.
At least 1 or 2 polylactide pins were inserted horizontally. The head of each pin was
countersunk below the surface of the surrounding cartilage.
2. Standard mini-fragment metal screws
Participants were treated with countersunk metal lag screws (diameters 1.5 mm to 2.0
mm) from the AO mini-fragment set
Postoperative treatment consisted of physiotherapy for up to 6 weeks. In both groups, a
temporary dorsal elbow splint at 60° flexion was applied to an equal number of partici-

Helling 2006 (Continued)
pants (22 in each group) for a maximum of 3 weeks

Postoperative rehabilitation: all patients in both groups underwent the same postop-
erative treatment
Assigned: no described
Completed: 83/82
Outcomes Length of follow-up: 2 years (average, 25 months; range 20 to 35 months), also 6 weeks
and 1 year (average, 12 months; range, 7 to 19 months)
Broberg and Morrey functional rating scores (including range of motion, grip strength,
functional stability, and pain)
Adverse events, including deep wound infection, foreign-body reactions of the soft tis-
sues grade 2 to 4 (seroma, spontaneous discharge, bacterial superinfection), range of
motion deficit of more than 30° on 1 axis, fragment displacement of more than 1 mm
compared with postoperative results, reoperation for any reason other than hardware
removal, osteolysis grade 4 at the implant site preventing or delaying fracture healing, or
a combination of these
Notes In standard mini-fragment metal implants group, additional mini-fragment plates or

Kirschner (K) wires could be used, if deemed necessary
Risk of bias
Random sequence generation (selection Low risk The participants were randomised by tele-
bias) phone by an independent agency
Allocation concealment (selection bias) Low risk The participants were randomised by tele-
phone by an independent agency
Blinding of participants and personnel High risk The surgeons conducting the surgery
(performance bias) would not have been blinded
All outcomes
Blinding of outcome assessment (detection High risk It was not described if the outcome assess-
bias) ments were blinded
All outcomes
Incomplete outcome data (attrition bias) High risk Of 184 enrolled in the trial, 19 people were
All outcomes excluded after randomisation. There was an
imbalance between the two groups in the
number of participants lost at 2-year fol-
low-up
all planned outcomes

Helling 2006 (Continued)
Other bias Unclear risk Conflicts of interest and surgeon’s expertise

with the interventions were not described.
Baseline data, provided 165 participants,
showed some comparability between the
two groups
Rehabilitation was the same for both
groups.
Ruan 2009
Methods Method of randomisation: randomised

Assessor blinding: yes
Loss to follow-up at 2 years: 0/22
Number of surgeons: not known.
Participants Study location: Shanghai 6th People’s Hospital, Shanghai, China.

Study period: April 2002 and February 2006.
Study participants: 22 participants of Mason type III.
Age range: the average age was 37.4 years in the replacement group and 40.1 years in
the ORIF group
Gender: 9 women and 13 men.
Inclusion criteria: radial head fractures of Mason type III.
Exclusion criteria: not described.
Interventions Interventions started: In the replacement group, 12 participants had fresh fractures and
two participants had old fractures. In ORIF group, all eight participants had fresh frac-
tures. All participants who had fresh fractures were treated within 4 to 17 days after
injury, and the two old fracture cases were treated after 12 and 18 months of injury.
1. radial head replacement
Using radial head prosthesis of Tornier Inc in France.
2. ORIF
With cannulated screws and K wires. And all eight participants were immobilised in
plaster for four weeks. Rehabilitation began when plasters were removed four weeks after
surgery.
Postoperative rehabilitation: same in both the groups.
Assigned: 14/8
Completed: 14/8
Outcomes Length of follow-up: The 14 participants who had received radial head replacement were
followed up for an average of 15.9 months (range 10 to 27 months). Mean follow-up of
the eight cases in the ORIF group was 14 months (range 10 to 21 months)
Broberg and Morrey functional rating scores.
Adverse events (including heterotopic ossification, bone nonunion and bone absorption)
Notes
Risk of bias

Ruan 2009 (Continued)
Random sequence generation (selection Unclear risk Twenty-two participants were randomly
bias) divided into the radial head replacement
group (14 cases) and the ORIF group (8
cases). But randomisation method was not
described
Allocation concealment (selection bias) Unclear risk The method of concealment was not de-
scribed.
Blinding of participants and personnel High risk The surgeons conducting the surgery
(performance bias) would not have been blinded
All outcomes
Blinding of outcome assessment (detection High risk It was not described if the outcome assess-
bias) ments were blinded
All outcomes
Incomplete outcome data (attrition bias) Low risk No participant was lost and outcome data
All outcomes were complete.
all expected outcomes
Other bias High risk Conflicts of interest and surgeon’s expertise

in ORIF were not described. For this small
trial, the inclusion of 2 old fractures in the
replacement group is of concern in terms
of selection bias
ORIF: open reduction and internal fixation
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Erturer 2010 This was not a randomised or quasi-randomised controlled trial
Frankle 1999 Retrospective study. Patients were treated by conservative or surgical treatment
Ikeda 2005 This was not a randomised or quasi-randomised controlled trial

(Continued)
Liu 2010 This was not a randomised or quasi-randomised controlled trial
Obert 2005 Retrospective comparative study.
Characteristics of studies awaiting assessment [ordered by study ID]
Dabby 2006
Methods Controlled trial
Participants 20 patients
Interventions Open reduction and internal fixation (10 patients) versus Corin Radial head prosthesis (10 patients)
Outcomes ROM, muscle strength, elbow stability, function ability, pain
Notes Abstract report only. Efforts continue to obtain a full report of this trial
ROM: range of motion
Characteristics of ongoing studies [ordered by study ID]
Ascension 2009
Trial name or title Safety and effectiveness study of Ascension’s pyrocarbon radial head compared to Ascension’s metal radial
head
Methods A randomised, double blind clinical trial
Participants Patients must need a radial head replacement for:

• Degenerative, post-traumatic or rheumatoid arthritis
• Primary replacement after fracture of the radial head
• Relief of symptoms after radial head resection
• Revision of failed radial head arthroplasty
• Skeletally mature individuals ages 18 - 85
Interventions Carbon modular radial head versus metal radial head
Outcomes The proportion of patient successes
Starting date January 2009
Contact information Xiaosong Yang, MSN 512-836-5001 ext 1536 xyang@ascensionortho.com

Ascension 2009 (Continued)
Notes Clinical Trials.gov Identifier: NCT00825409

DATA AND ANALYSES
Comparison 1. Radial head replacement versus ORIF
No. of No. of
Outcome or subgroup title studies participants Statistical method Effect size
1 Broberg and Morrey functional 1 Mean Difference (IV, Fixed, 95% CI) Totals not selected
rating score (0 to 100: best
outcome) at 2 years
2 Broberg and Morrey functional 2 67 Risk Ratio (M-H, Fixed, 95% CI) 1.88 [1.27, 2.77]
rating score - excellent or good
results
3 Overall adverse events 2 67 Risk Ratio (M-H, Fixed, 95% CI) 0.33 [0.14, 0.77]
4 Individual adverse events 2 Risk Ratio (M-H, Fixed, 95% CI) Subtotals only
4.1 Range of motion deficit > 1 45 Risk Ratio (M-H, Fixed, 95% CI) 0.52 [0.11, 2.57]
30°
4.2 Stiffness caused by 1 45 Risk Ratio (M-H, Fixed, 95% CI) 3.13 [0.13, 72.99]
prostheses
4.3 Secondary fragment 1 45 Risk Ratio (M-H, Fixed, 95% CI) 0.15 [0.01, 2.73]
displacement
4.4 Deep wound infection 1 45 Risk Ratio (M-H, Fixed, 95% CI) 0.35 [0.01, 8.11]
4.5 Heterotopic ossification 2 67 Risk Ratio (M-H, Fixed, 95% CI) 1.02 [0.21, 5.01]
4.6 Bone nonunion 1 45 Risk Ratio (M-H, Fixed, 95% CI) 0.35 [0.01, 8.11]
4.7 ”Bone nonunion and bone 1 22 Risk Ratio (M-H, Fixed, 95% CI) 0.07 [0.00, 1.10]
absorption”
Comparison 2. Biodegradable pins versus standard metal screws
No. of No. of
Outcome or subgroup title studies participants Statistical method Effect size
1 Broberg and Morrey functional 1 Mean Difference (IV, Fixed, 95% CI) Totals not selected
rating score (0 to 100: best
result) at 2 years
2 Broberg and Morrey functional 1 Risk Ratio (M-H, Fixed, 95% CI) Totals not selected
rating score - excellent or good
results
3 Overall adverse events 1 Risk Ratio (M-H, Fixed, 95% CI) Totals not selected
4 Individual adverse events 1 Risk Ratio (M-H, Fixed, 95% CI) Totals not selected
4.1 Range of motion deficit 1 Risk Ratio (M-H, Fixed, 95% CI) 0.0 [0.0, 0.0]
30°
4.2 Secondary fragment 1 Risk Ratio (M-H, Fixed, 95% CI) 0.0 [0.0, 0.0]
displacement
4.3 Re-operation 1 Risk Ratio (M-H, Fixed, 95% CI) 0.0 [0.0, 0.0]
4.4 Osteolysis grade 4 1 Risk Ratio (M-H, Fixed, 95% CI) 0.0 [0.0, 0.0]
(disturbance of fracture
healing)
4.5 Deep wound infection 1 Risk Ratio (M-H, Fixed, 95% CI) 0.0 [0.0, 0.0]
Analysis 1.1. Comparison 1 Radial head replacement versus ORIF, Outcome 1 Broberg and Morrey
functional rating score (0 to 100: best outcome) at 2 years.
Review: Surgical interventions for treating radial head fractures in adults
Comparison: 1 Radial head replacement versus ORIF
Outcome: 1 Broberg and Morrey functional rating score (0 to 100: best outcome) at 2 years
Radial head
replace- Mean Mean
Study or subgroup ment ORIF Difference Difference
N Mean(SD) N Mean(SD) IV,Fixed,95% CI IV,Fixed,95% CI
Chen 2011 22 92.1 (6.8) 23 72.4 (7.1) 19.70 [ 15.64, 23.76 ]
-50 -25 0 25 50
Favours ORIF Favours replacement

Analysis 1.2. Comparison 1 Radial head replacement versus ORIF, Outcome 2 Broberg and Morrey
functional rating score - excellent or good results.
Outcome: 2 Broberg and Morrey functional rating score - excellent or good results
Radial head
replace-
Study or subgroup ment ORIF Risk Ratio Weight Risk Ratio
n/N n/N M-H,Fixed,95% CI M-H,Fixed,95% CI
Chen 2011 20/22 15/23 92.0 % 1.39 [ 1.01, 1.93 ]
Ruan 2009 13/14 1/8 8.0 % 7.43 [ 1.18, 46.73 ]
Total (95% CI) 36 31 100.0 % 1.88 [ 1.27, 2.77 ]

Total events: 33 (Radial head replacement), 16 (ORIF)
Heterogeneity: Chi2 = 5.33, df = 1 (P = 0.02); I2 =81%
Test for overall effect: Z = 3.16 (P = 0.0016)
Test for subgroup differences: Not applicable
0.001 0.01 0.1 1 10 100 1000

Favours ORIF Favours replacement

Analysis 1.3. Comparison 1 Radial head replacement versus ORIF, Outcome 3 Overall adverse events.
Outcome: 3 Overall adverse events
Radial head
replace-
Chen 2011 3/22 11/23 67.9 % 0.29 [ 0.09, 0.89 ]
Ruan 2009 3/14 4/8 32.1 % 0.43 [ 0.13, 1.45 ]
Total (95% CI) 36 31 100.0 % 0.33 [ 0.14, 0.77 ]

Heterogeneity: Chi2 = 0.24, df = 1 (P = 0.63); I2 =0.0%
Test for subgroup differences: Not applicable
0.01 0.1 1 10 100

Favours replacement Favours ORIF

Analysis 1.4. Comparison 1 Radial head replacement versus ORIF, Outcome 4 Individual adverse events.
Outcome: 4 Individual adverse events
Radial head
replace-
1 Range of motion deficit > 30

Chen 2011 2/22 4/23 100.0 % 0.52 [ 0.11, 2.57 ]
Subtotal (95% CI) 22 23 100.0 % 0.52 [ 0.11, 2.57 ]

Heterogeneity: not applicable
2 Stiffness caused by prostheses
Chen 2011 1/22 0/23 100.0 % 3.13 [ 0.13, 72.99 ]
Subtotal (95% CI) 22 23 100.0 % 3.13 [ 0.13, 72.99 ]

3 Secondary fragment displacement
Chen 2011 0/22 3/23 100.0 % 0.15 [ 0.01, 2.73 ]
Subtotal (95% CI) 22 23 100.0 % 0.15 [ 0.01, 2.73 ]

4 Deep wound infection
Chen 2011 0/22 1/23 100.0 % 0.35 [ 0.01, 8.11 ]
Subtotal (95% CI) 22 23 100.0 % 0.35 [ 0.01, 8.11 ]

5 Heterotopic ossification
Chen 2011 0/22 2/23 79.7 % 0.21 [ 0.01, 4.12 ]
Ruan 2009 3/14 0/8 20.3 % 4.20 [ 0.24, 72.29 ]
Subtotal (95% CI) 36 31 100.0 % 1.02 [ 0.21, 5.01 ]

Heterogeneity: Chi2 = 2.04, df = 1 (P = 0.15); I2 =51%
6 Bone nonunion
0.001 0.01 0.1 1 10 100 1000

(Continued . . . )

(. . . Continued)
Radial head
replace-
Chen 2011 0/22 1/23 100.0 % 0.35 [ 0.01, 8.11 ]
Subtotal (95% CI) 22 23 100.0 % 0.35 [ 0.01, 8.11 ]

7 ”Bone nonunion and bone absorption”
Ruan 2009 0/14 4/8 100.0 % 0.07 [ 0.00, 1.10 ]
Subtotal (95% CI) 14 8 100.0 % 0.07 [ 0.00, 1.10 ]

Test for subgroup differences: Chi2 = 4.83, df = 6 (P = 0.57), I2 =0.0%
0.001 0.01 0.1 1 10 100 1000

Analysis 2.1. Comparison 2 Biodegradable pins versus standard metal screws, Outcome 1 Broberg and
Morrey functional rating score (0 to 100: best result) at 2 years.
Comparison: 2 Biodegradable pins versus standard metal screws
Outcome: 1 Broberg and Morrey functional rating score (0 to 100: best result) at 2 years
Mean Mean
Study or subgroup Biodegradable pins Standard screws Difference Difference
N Mean(SD) N Mean(SD) IV,Fixed,95% CI IV,Fixed,95% CI
Helling 2006 74 93.3 (7.2) 61 90.9 (7.5) 2.40 [ -0.10, 4.90 ]
-10 -5 0 5 10
Favours metal Favours biodegradable

Analysis 2.2. Comparison 2 Biodegradable pins versus standard metal screws, Outcome 2 Broberg and
Morrey functional rating score - excellent or good results.
Outcome: 2 Broberg and Morrey functional rating score - excellent or good results
Study or subgroup Biodegradable pins Standard screws Risk Ratio Risk Ratio
Helling 2006 72/74 56/61 1.06 [ 0.97, 1.15 ]
0.01 0.1 1 10 100

Favours metal Favours biodegradable
Analysis 2.3. Comparison 2 Biodegradable pins versus standard metal screws, Outcome 3 Overall adverse
events.
Outcome: 3 Overall adverse events
Helling 2006 13/82 16/82 0.81 [ 0.42, 1.58 ]
0.2 0.5 1 2 5
Favours biodegradable Favours metal

Analysis 2.4. Comparison 2 Biodegradable pins versus standard metal screws, Outcome 4 Individual
adverse events.
Outcome: 4 Individual adverse events
1 Range of motion deficit 30

Helling 2006 9/82 13/82 0.69 [ 0.31, 1.53 ]
2 Secondary fragment displacement

Helling 2006 5/82 3/82 1.67 [ 0.41, 6.75 ]
3 Re-operation
Helling 2006 2/82 3/82 0.67 [ 0.11, 3.89 ]
4 Osteolysis grade 4 (disturbance of fracture healing)

Helling 2006 1/82 1/82 1.00 [ 0.06, 15.72 ]
5 Deep wound infection

Helling 2006 1/82 0/82 3.00 [ 0.12, 72.58 ]
0.01 0.1 1 10 100

Favours biodegradable Favours metal
APPENDICES
Appendix 1. Search strategies
Cochrane Central Register of Controlled Trials (Wiley Online Library)

#1 MeSH descriptor Radius Fractures, this term only (195)
#2 MeSH descriptor Radius, this term only (127)
#3 MeSH descriptor Fractures, Bone, this term only (994)
#4 MeSH descriptor Fracture Healing, this term only (318)
#5 MeSH descriptor Fracture Fixation explode all trees (973)
#6 MeSH descriptor Orthopedic Fixation Devices explode all trees (1660)
#7 (#3 OR #4 OR #5 OR #6) (2937)
#8 (#2 AND #7) (25)
#9 (#1 OR #8) (211)
#10 (head or proximal):ti,ab,kw (12190)
#11 (#9 AND #10) (21)
#12 ((radius or radial) NEAR/3 fracture* NEAR/3 (head or proximal)):ti,ab,kw (16)
#13 (#11 OR #12) (27)
MEDLINE (OvidSP)
1 Radius Fractures/ (6408)
2 Radius/ (7074)
3 Fractures, Bone/ (43661)
4 Fracture Healing/ (8087)
5 exp Fracture Fixation/ or exp Orthopedic Fixation Devices/ (78093)
6 or/3-5 (113410)
7 and/2,6 (1056)
8 or/1,7 (7158)
9 (head or proximal).tw. (320888)
10 and/8-9 (1336)
11 ((radius or radial) adj3 fracture$ adj3 (head or proximal)).tw. (613)
12 or/10-11 (1444)
13 Randomized controlled trial.pt. (326354)
14 Controlled clinical trial.pt. (84043)
15 randomized.ab. (230515)
16 placebo.ab. (130935)
17 Drug therapy.fs. (1528972)
18 randomly.ab. (166436)
19 trial.ab. (238406)
20 groups.ab. (1094464)
21 or/13-20 (2838403)
22 exp Animals/ not Humans/ (3712810)
23 21 not 22 (2410011)
24 and/12,23 (89)
EMBASE (OvidSP)
1 Radius Head Fracture/ (618)
2 Radius Fracture/ (6554)
3 Radius/ (8076)
4 exp Fracture Fixation/ (58843)
5 Fracture/ (51985)
6 Fracture Healing/ (14154)
7 or/4-6 (110148)
8 and/3,7 (1127)
9 or/2,8 (7468)
10 (head or proximal).tw. (388384)
11 and/9-10 (1079)
12 ((radius or radial) adj3 fracture$ adj3 (head or proximal)).tw. (707)
13 or/1,11-12 (1744)
14 Clinical trial/ (865033)
15 Randomized Controlled Trial/ (321249)
16 Randomization/ (58049)
17 Single Blind Procedure/ (15834)
18 Double Blind Procedure/ (108636)
19 Crossover Procedure/ (33755)
20 Placebo/ (197741)
21 randomi?ed controlled trial$.tw. (74201)
22 rct.tw. (9108)
23 random allocation.tw. (1134)
24 randomly allocated.tw. (17009)
25 allocated randomly.tw. (1796)
26 (allocated adj2 random).tw. (705)
27 single blind$.tw. (12084)
28 double blind$.tw. (127062)
29 ((treble or triple) adj blind$).tw. (265)
30 placebo$.tw. (173569)
31 Prospective Study/ (202826)
32 or/14-31 (1242625)
33 Case Study/ (15534)
34 case report.tw. (223882)
35 Abstract Report/ or Letter/ (830646)
36 or/33-35 (1065549)
37 32 not 36 (1207845)
38 limit 37 to human (1104542)
39 and/13,38 (80)
CONTRIBUTIONS OF AUTHORS
All review authors contributed to the preparation of the review and approved the final version. Yu Gao is the guarantor.
Yu Gao: selected studies, extracted data, assessed the risk of bias of included studies and co-ordinated the review process
Wei Zhang: conceived the research question, performed statistical analysis and co-ordinated the review process
Xin Duan: selected studies, extracted data and assessed the risk of bias of included studies
Jing Yang: selected studies by searching other resources and prepared review
Mohammed Al-Qwbani: improved the language of the review
Jingtong Lv: prepared the review
Zhou Xiang: resolved differences in data selection, extraction and management, and co-ordinated the review process
DECLARATIONS OF INTEREST
None known.
SOURCES OF SUPPORT

Internal sources
• West China Hospital of Sichuan University, China.
External sources
• No sources of support supplied
DIFFERENCES BETWEEN PROTOCOL AND REVIEW

1. Assessment of risk of bias in included studies
• we split the single blinding domain of blinding of participants, personnel and outcome assessors into two domains as per
Higgins 2011: blinding of participants and personnel, and blinding of outcome assessment;
• we assessed blinding and complete outcome data for overall outcome rather than separately for patient-reported outcomes and
objective outcomes;
• in our assessment of other bias, we considered conflicts of interest.
2. Types of participants
We clarified that participants would have acute fractures, which would be displaced. In Types of interventions, we stipulated that we
would exclude studies that tested secondary surgery after failed non-surgical treatment. However, we made an exception for one of the
included trials, which included two participants with ’old fractures’ who had presumably received non-surgical treatment when their
fracture was acute.
3. Types of outcome measures
Some outcomes, such as SF36, Visual Analogue Scale, range of elbow and forearm rotation, and secondary outcomes were not reported
because of a lack of data from the included trials.
INDEX TERMS
Medical Subject Headings (MeSH)

Fracture Fixation [∗ methods]; Radius [∗ surgery]; Radius Fractures [∗ surgery]; Randomized Controlled Trials as Topic
MeSH check words

Adult; Humans


CX Cabeza Radial

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

CX Cabeza Radial

Uploaded by

Copyright:

Available Formats

Cochrane Database of Systematic Reviews

Surgical interventions for treating radial head fractures in

Gao Y, Zhang W, Duan X, Yang J, Al-Qwbani M, Lv J, Xiang Z

Gao Y, Zhang W, Duan X, Yang J, Al-Qwbani M, Lv J, Xiang Z.

Surgical interventions for treating radial head fractures in adults (Review)

Surgical interventions for treating radial head fractures in adults (Review) i

Surgical interventions for treating radial head fractures in

Editorial group: Cochrane Bone, Joint and Muscle Trauma Group.

PLAIN LANGUAGE SUMMARY

Surgical interventions for treating radial head fractures in adults (Review) 2

Surgical interventions for treating radial head fractures in adults (Review) 3

OBJECTIVES Primary outcomes

Criteria for considering studies for this review

Surgical interventions for treating radial head fractures in adults (Review) 4

Unit of analysis issues

Surgical interventions for treating radial head fractures in adults (Review) 5

Surgical interventions for treating radial head fractures in adults (Review) 6

Surgical interventions for treating radial head fractures in adults (Review) 7

Surgical interventions for treating radial head fractures in adults (Review) 8

Surgical interventions for treating radial head fractures in adults (Review) 9

Surgical interventions for treating radial head fractures in adults (Review) 10

Surgical interventions for treating radial head fractures in adults (Review) 11

Surgical interventions for treating radial head fractures in adults (Review) 12

Surgical interventions for treating radial head fractures in adults (Review) 13

Surgical interventions for treating radial head fractures in adults (Review) 15

Characteristics of included studies [ordered by study ID]

Methods Method of randomisation: randomised.

Participants Study location: Changhai Hospital, Shanghai, China.

Interventions Interventions started: within a mean of 4.9 days after injury

Surgical interventions for treating radial head fractures in adults (Review) 16

Outcomes Length of follow-up: 2.8 years (range 1 to 5 years)

Bias Authors’ judgement Support for judgement

Surgical interventions for treating radial head fractures in adults (Review) 17

No details stated of the surgeons conduct-

Methods Method of randomisation: randomised by telephone by an independent agency.

Participants Study location: 24 centres, Germany.

Interventions Interventions started: within a mean of 5.8 days after injury

Surgical interventions for treating radial head fractures in adults (Review) 18

pants (22 in each group) for a maximum of 3 weeks

Notes In standard mini-fragment metal implants group, additional mini-fragment plates or

Bias Authors’ judgement Support for judgement

Surgical interventions for treating radial head fractures in adults (Review) 19

Other bias Unclear risk Conflicts of interest and surgeon’s expertise

Methods Method of randomisation: randomised

Participants Study location: Shanghai 6th People’s Hospital, Shanghai, China.

Surgical interventions for treating radial head fractures in adults (Review) 20

Bias Authors’ judgement Support for judgement

Other bias High risk Conflicts of interest and surgeon’s expertise

ORIF: open reduction and internal fixation

Characteristics of excluded studies [ordered by study ID]

Study Reason for exclusion

Erturer 2010 This was not a randomised or quasi-randomised controlled trial

Ikeda 2005 This was not a randomised or quasi-randomised controlled trial

Surgical interventions for treating radial head fractures in adults (Review) 21

Liu 2010 This was not a randomised or quasi-randomised controlled trial

Obert 2005 Retrospective comparative study.

Characteristics of studies awaiting assessment [ordered by study ID]

Methods Controlled trial

Outcomes ROM, muscle strength, elbow stability, function ability, pain

ROM: range of motion

Characteristics of ongoing studies [ordered by study ID]

Methods A randomised, double blind clinical trial

Participants Patients must need a radial head replacement for: