Description of the condition
Fractures of the shaft of the humerus account for 1% to 3% of all fractures in adults (Court-Brown 2006). Based on a study of 401 humeral shaft fractures in Sweden, Ekholm 2006 reported an annual incidence of 14.5 per 100,000 adults with a gradually increasing age-specific incidence from the fifth decade, reaching almost 60 per 100,000 people aged 90 to 100. Ekholm 2006 reported that 90% of fractures resulted from trauma, mainly from a simple fall (68% of total). The remaining fractures were pathological (8.5%) or peri-prosthetic (1.5%) (Ekholm 2006). The majority of humeral shaft fractures are closed fractures, where the overlying skin remains intact.
The AO classification (AO classification) or the OTA classification (OTA classification) schemes are generally used to classify the various types of humeral shaft fractures. The AO classification system describes three main categories: 'simple' fractures (type A) where the humerus shaft is broken in two, 'wedge' fractures (type B) where a wedge shaped fragment is broken away, and 'complex' (type C) multiply fragmented fractures. The fracture pattern often reflects the mechanism of injury. Direct blows to the humerus generally result in transverse or a multiple fragmented fractures, and twisting injuries generally cause oblique or spiral (with or without medial wedge) fractures. Oblique and transverse fractures can also be caused by a fall on the elbow with the arm in abduction (Zafar 2007).
Patients usually present with pain and swelling of the upper arm. Assessment of motor and sensory function and a radiograph is usually sufficient for diagnosis.
Description of the intervention
Traditionally, the majority of humeral shaft fractures are treated non-surgically. Non-surgical (conservative) interventions involve immobilisation of the elbow and/or shoulder joints using the sugar tong splint, plaster splints or hanging casts; or use of functional bracing such as the Sarmiento brace where the elbow and shoulder joints are not immobilised (Sarmiento 2001).
Surgical intervention involves reduction (repositioning) of the displaced bone fragments and fixing these using either a plate secured onto the bone with screws (plate fixation) or various nails. The latter are pushed through the intramedullary canal (intramedullary nails), being inserted either through the humeral head (antegrade) or above the elbow in the humeral bone (retrograde), and sometimes locked in place with an interlocking screw. A third type of fixation, external fixation, is generally reserved for the more severe types of open fractures (Gustilo grades IIIB and IIIC).
Surgical fixation of humeral shaft fractures is generally reserved for more complex and displaced fractures, and in people with associated injuries. Typical indications for surgical treatment are: type II and III Gustilo open fractures, polytrauma or high energy trauma, ipsilateral forearm fractures (floating elbow), nerve (secondary radial nerve palsy, brachial plexus injury) and serious vascular injuries, bilateral shaft fractures, segmental fractures, fractures with axial distraction, pathological fractures, and delayed or non-unions (Zafar 2007). Surgery may also be considered for patients with Parkinson's disease or with only one functional arm, and for obese patients (Toivanen 2005).
How the intervention might work
Conservative treatment can give good functional results. There is a great capability of the glenohumeral joint (shoulder) for adjustment to malunion or shortening. It is generally accepted that up to 3 cm of shortening and 20 to 30 degrees of varus, anterior or rotational deformity of the humerus can still result in an acceptable upper limb function (Clifford 2008). However, there is still a risk of non-union, which has been reported in one in 10 conservatively treated patients (Flinkkila 2004), and of complications and functional impairments arising from immobilisation or restricted mobility of the elbow and shoulder from casts and braces.
Surgical fixation of the repositioned fractured bone keeps the bone fragments in place while the fracture heals. It should allow early mobilisation and rehabilitation to take place. However, surgery has a higher risk of complications both during and after the operation. In an article on operative treatment, Cole 2007 reported incidences of patients with post-operative complications; this included up to 7% with non-union, up to 6% with infection, and up to 5% with radial nerve palsy. Implant-related complications resulting in fixation failure also occur. For example, screws can pull out of bone or break, and rotational instability of intramedullary nails can result in non-union (Flinkkila 2004).
Why it is important to do this review
While most fractures of the humeral shaft will heal with non-surgical treatment, the results are not always satisfactory. Although there are some hard indications for surgical treatment (for which surgery is necessary), most indications are relative and there is variation in practice. Increasingly these fractures occur in older adults, for whom the indications for surgery may differ. For example, the classic non-surgical treatment with functional bracing works on the principles of active muscle contraction, but may be less effective in older people who have less muscle mass. With this review we hoped to obtain a clearer insight in the indications for conservative and surgical treatment, especially in the elderly.
To assess and compare the effects of surgical versus non-surgical interventions for treating non-pathological fractures of the humeral shaft in adults.
Criteria for considering studies for this review
Types of studies
All randomised controlled trials and quasi-randomised (method of allocating participants to a treatment which is not strictly random; e.g. by date of birth, hospital record number, alternation) controlled trials that compare surgical with non-surgical interventions for treating humeral shaft fractures in adults were considered for inclusion.
Types of participants
Adults (usually people aged at least 18 years) with a humeral shaft fracture. We excluded trials specifically focused on treating people with a primary bone disease (e.g. fibrous dysplasia), pathological fractures (from metastatic cancer) or peri-prosthetic fractures. Trials including children or patients with primary bone diseases or pathological fractures were excluded unless separate data were provided for adults with non-pathological fractures or the proportion of children or adults with primary bone diseases or pathological fractures was small (< 5%).
Types of interventions
We included surgical and non-surgical interventions for the treatment of humeral shaft fractures in adults. Surgical interventions include intramedullary nailing, plate fixation and external fixation. Conservative (non-surgical) interventions include sugar tong splint, plaster splints, hanging casts, or functional bracing such as the Sarmiento brace.
We excluded trials comparing different methods of surgical treatment alone, or different methods of non-surgical treatment alone.
Types of outcome measures
- Functional assessment, including generic and upper-limb specific validated clinical scores (e.g. DASH (Upper Extremity Collaborative Group 1996)), and patient derived health related quality of life measures (e.g. SF36)
- Serious adverse events including complications (e.g. infection, non-union, radial nerve palsy) generally requiring secondary procedures (e.g. an operation)
- Return to activities (including sport and work)
- Other complications
- Range of motion
- Pain (visual analogue scale)
- Patient satisfaction, including with cosmetic result
- Constant score (Constant 1987)
Search methods for identification of studies
We searched the Cochrane Bone, Joint and Muscle Trauma Group Specialised Register (October 2011), the Cochrane Central Register of Controlled Trials (The Cochrane Library 2011 Issue 10), MEDLINE (1948 to October week 2, 2011) and EMBASE (1980 to week 42, 2011). We also searched Current Controlled Trials, Clinicaltrials.gov and the WHO International Clinical Trials Registry for ongoing and recently completed trials. There was no restriction on the inclusion of reports based on publication language.
In MEDLINE, the subject specific search strategy was combined with the Cochrane Highly Sensitive Search Strategy for identifying randomised trials in MEDLINE: sensitivity-maximizing version (Lefebvre 2009) (see Appendix 1). This strategy was modified for use in The Cochrane Library and EMBASE (see Appendix 1).
Searching other resources
We searched the references lists of articles for relevant studies. We searched congress abstracts for relevant studies and contacted local experts.
Data collection and analysis
Selection of studies
Two authors (MG and MT) independently selected and assessed potentially eligible studies for inclusion. Any disagreement was resolved by discussion.
Data extraction and management
Should trials be included in future, two authors (MG and MT) will independently complete a pre-piloted data collection form for each included trial. One author (MG) will enter the data into RevMan. Data entry will be checked by MT.
Assessment of risk of bias in included studies
Two authors (MG and MT) will independently assess the risk of bias of individual studies using the most up-to-date version of The Cochrane Collaboration's tool (currently, Higgins 2011). Arbitration will be by a third reviewer (HJ). This tool incorporates assessment of randomisation (sequence generation and allocation concealment), blinding (participants and personnel, and outcome assessors), completeness of outcome data, selection of outcomes reported and other sources of bias, such as major imbalances in key baseline characteristics. We will assess the risk of bias associated with a) blinding and b) completeness of outcomes separately for patient-reported outcomes and objective outcomes.
Measures of treatment effect
We will calculate risk ratios with 95% confidence intervals for dichotomous outcomes. For continuous outcomes, we will calculate mean differences with 95% confidence intervals unless pooling data from studies using different scales, in which case we will use standardised mean differences with 95% confidence intervals.
Unit of analysis issues
Though the unit of randomisation in these trials is usually the individual patient, trials including people with bilateral fractures may present results for fractures or limbs rather than individual patients. Where such unit of analysis issues arise and appropriate corrections have not been made, we will consider presenting the data for such trials where the disparity between the units of analysis and randomisation is small. Where data are pooled, we will perform a sensitivity analysis to examine the effects of excluding incorrectly reported trials from the analysis.
Dealing with missing data
We will try to contact trial authors where there are missing data. Where possible, we will perform intention-to-treat analyses to include all people randomised to the intervention groups. Unless missing standard deviations can be derived from standard errors or confidence interval data, we will not assume values in order to present these in the analyses.
Assessment of heterogeneity
Heterogeneity will be assessed by visual inspection of the forest plot (analysis) along with consideration of the Chi² test for heterogeneity and the I² statistic (Higgins 2003).
Assessment of reporting biases
We will explore publication bias using a funnel plot if data from over 10 trials are available for pooling.
If considered appropriate, results of comparable groups of trials will be pooled. Initially we will use the fixed-effect model and 95% confidence intervals. Where there is clear and major heterogeneity, we will check the results using the random-effects model and consider whether presenting these results is more appropriate.
Subgroup analysis and investigation of heterogeneity
If there is heterogeneity between groups we will attempt to explore the causes using subgroup and sensitivity analysis. We plan to conduct subgroup analyses based on age (under 50 years; 50 or over); different treatment modalities (e.g. plate; nail; external fixation); and by fracture type.
We chose 50 years as a cut-off for age because this may serve to separate out two categories of patients, for which the decision to operate and the type of intervention chosen may be influenced by different factors. Factors in younger patients include open fractures, neurological and vascular injuries and complex fracture patterns. Older patients mostly have a low-energy trauma, less muscle mass and osteoporosis; factors again that may influence treatment decisions.
Where possible, we plan to perform sensitivity analyses examining various aspects of trials and review methodology, including the effects of missing data, of the inclusion of trials with inadequate allocation concealment and of trials only reported in abstracts.
Description of studies
Results of the search
We screened 532 citations up to October 2011, from the Cochrane Bone, Joint and Muscle Trauma Group's Specialised Register (95 citations), MEDLINE (132 citations), EMBASE (269 citations) and The Cochrane Central Register of Controlled Trials (The Cochrane Library) (36 citations). We found six potentially eligible completed studies, all of which were excluded.
There are no included studies in this review.
Seven studies were excluded for reasons stated in the Characteristics of excluded studies. Five studies were retrospective studies (Denard 2010; Jawa 2006; Klestil 1997; Osman 1998; Wallny 1997). One was a prospective study without randomisation (Nast-Kolb 1991), and one is an ongoing trial without randomisation (Pelet).
Risk of bias in included studies
There are no included studies in this review.
Effects of interventions
There are no included studies in this review.
Our extensive search failed to identify any completed randomised or quasi-randomised controlled trials testing the effects of surgical versus non-surgical interventions for non-pathological fractures of the humeral shaft in adults. Currently, the published literature consists of retrospective studies and one prospective non-randomised study. It may prove to be very challenging to design a randomised controlled trial comparing surgical and non-surgical treatment with enough statistical power to provide definitive evidence. However, we have identified two ongoing studies (Berry; Matsunaga) that, should these be successful, will inform this question.
Implications for practice
There is no evidence from randomised controlled trials to inform the choice between surgical and non-surgical interventions for treating humeral shaft fractures in adults.
Implications for research
There is a need for good quality evidence to inform on the use of surgery for humeral shaft fractures in adults. Humeral shaft fractures are relatively uncommon, therefore multi-centre randomised controlled trials are needed. The trials should have good follow-up and should at least record functional assessment, quality of life measures and serious adverse events including complications. While the two ongoing trials on this topic appear to be single-centre studies, It is likely that their results will help inform practice in due course.
We acknowledge the contribution of Dianne C de Visser to the development of the review.
We would like to thank Dr Helen Handoll and Mr Matthew Costa for valuable comments on drafts of the protocol, and Dr Helen Handoll for her help with the review. We are grateful to Dr Joanne Elliott for helping to develop the search strategy. We would also like to acknowledge the help of Amy Kavanagh and Lindsey Elstub from the Cochrane Bone, Joint and Muscle Trauma Group.
Data and analyses
This review has no analyses.
Appendix 1. Search strategies
1 Humeral Fractures/ (5409)
2 Humerus/ (7083)
3 Fractures, Bone/ (42969)
4 exp Fracture Fixation/ (42074)
5 Fracture Healing/ (7671)
6 or/3-5 (79293)
7 and/2,6 (548)
8 (humor* and fracture*).tw. (6198)
9 or/1,7-8 (8451)
10 (shaft or midshaft or diaphys*).tw. (19708)
11 and/9-10 (1390)
12 randomized controlled trial.pt. (320511)
13 controlled clinical trial.pt. (83807)
14 randomized.ab. (235709)
15 placebo.ab. (133714)
16 drug therapy.fs. (1509763)
17 randomly.ab. (172716)
18 trial.ab. (244314)
19 groups.ab. (1135675)
20 or/12-19 (2876815)
21 exp animals/ not humans.sh. (3705108)
22 20 not 21 (2452991)
23 and/11,22 (132)
The Cochrane Central Register of Controlled Trials (The Cochrane Library, Wiley Online Library)
#1 MeSH descriptor Humeral Fractures, this term only (55)
#2 MeSH descriptor Humerus, this term only (48)
#3 MeSH descriptor Fractures, Bone, this term only (951)
#4 MeSH descriptor Fracture Fixation explode all trees (930)
#5 MeSH descriptor Fracture Healing, this term only (295)
#6 (#3 OR #4 OR #5) (1841)
#7 (#2 AND #6) (5)
#8 (humer* and fracture*):ti,ab,kw in Clinical Trials (147)
#9 (#1 OR #7 OR #8) (168)
#10 (shaft or midshaft or diaphys*):ti,ab,kw (582)
#11 (#9 AND #10) (36)
1 humerus fracture/ (6589)
2 humerus/ (6705)
3 fracture/ (48020)
4 exp fracture treatment/ (68133)
5 fracture healing/ (13041)
6 or/3-5 (114303)
7 and/2,6 (866)
8 (humer* and fracture*).tw. (6959)
9 or/1,7-8 (9582)
10 (shaft or midshaft or diaphys*).tw. (21281)
11 and/9-10 (1572)
12 exp Randomized Controlled trial/ (291038)
13 exp Double Blind Procedure/ (101258)
14 exp Single Blind Procedure/ (14323)
15 exp Crossover Procedure/ (31003)
16 Controlled Study/ (3612833)
17 or/12-16 (3681329)
18 ((clinical or controlled or comparative or placebo or prospective$ or randomi#ed) adj3 (trial or study)).tw. (566387)
19 (random$ adj7 (allocat$ or allot$ or assign$ or basis$ or divid$ or order$)).tw. (137803)
20 ((singl$ or doubl$ or trebl$ or tripl$) adj7 (blind$ or mask$)).tw. (133137)
21 (cross?over$ or (cross adj1 over$)).tw. (56933)
22 ((allocat$ or allot$ or assign$ or divid$) adj3 (condition$ or experiment$ or intervention$ or treatment$ or therap$ or control$ or group$)).tw. (172015)
23 or/18-22 (849381)
24 or/17,23 (4089880)
25 limit 24 to human (2460476)
26 and/11,25 (269)
Protocol first published: Issue 11, 2010
Review first published: Issue 1, 2012
Contributions of authors
Review draft: MG
Independent study selection and assessment: MG and MT
Provided feedback on interim drafts and contributed to the final manuscript: JM and HJ
Guarantor of the review: MG
Declarations of interest
Differences between protocol and review
The absence of included trials meant that many of the methods detailed in the protocol could not be implemented. We stated our intention to use the most up-to-date risk of bias tool in a future update.
Medical Subject Headings (MeSH)
MeSH check words
* Indicates the major publication for the study