Description of the condition
The scaphoid is one of the eight carpal wrist bones. It is situated on the thumb side of the hand and articulates with the distal end of the radius (one of the two forearm bones) and four carpal bones: the lunate, trapezium, trapezoid and capitate.
It has been estimated that the incidence of scaphoid fractures is five per 10,000 population (Kozin 2001) and that they account for 2% to 6% of all fractures (Hannemann 2012). Scaphoid fractures are common in young men, with a peak incidence in their second and third decades (Hove 1999).
A person with a scaphoid fracture typically presents with wrist pain following a fall on an outstretched hand. The diagnosis of a scaphoid fracture can be made following a detailed clinical history, clinical examination and by appropriate radiographic imaging. However, not all scaphoid fractures are visible on initial X-rays (Gaebler 1996), and other diagnostic modalities, such as computed tomography and magnetic resonance imaging (MRI), are often used to improve diagnosis for informing treatment (Patel 2013).
Fractures of the scaphoid are often classified by their anatomical position, whether they are displaced or undisplaced and whether they are 'stable' or 'unstable'. Herbert 1974 classified scaphoid fractures in order to determine which fractures should be managed operatively. In this classification, the fractures were divided into types A to D, with A being a stable, acute fracture involving the tuberosity or waist; B an unstable acute fracture; and C and D being delayed union and established non-union respectively. A key benefit of the classification of scaphoid fractures by their anatomical location is the relationship between location and fracture healing, particularly in the context of disruption to the blood supply to the various parts of the scaphoid. Waist fractures make up around 80% of scaphoid fractures, 15% at the proximal pole, 4% in the tuberosity of the scaphoid and 1% distal articular fractures (Leslie 1981). The healing time for these different types of fractures ranges from four to 12 weeks depending on location, with tuberosity fractures healing more quickly than distal third and waist fractures. These and 'comminuted' fractures (with a number of individual bone fragments) may take up to 12 weeks to heal (Leslie 1981).
Fractures of the scaphoid result in short-term functional disability, as well as time off work, loss of earnings and interference with recreational activities. The average time for healing of a nondisplaced scaphoid fracture in a cast is between eight and 12 weeks. This can result in a considerable loss of work time and productivity in the predominantly young and active population that sustains these fractures (van der Molen 1999). In some cases, treatment is less successful or is unsuccessful, resulting in non-union or malunion, either of which can result in pain, restricted motion and diminished grip strength, and may lead to arthritis and other long-term functional disability.
Description of the intervention
Typically, stable (undisplaced or minimally displaced) scaphoid fractures are managed conservatively, that is non-surgically, by immobilisation of the wrist in a cast or other type of splint for between four and 12 weeks. This may be done in a variety of ways, with the immobilisation of other non-injured joints and the positioning of the wrist being key differences. The main treatment options are:
- forearm cast
- forearm cast with the inclusion of the thumb (scaphoid cast)
- an above elbow cast with exclusion of the thumb
- an above elbow cast with inclusion of the thumb
- cast with the wrist in flexion (bent down), extension (bent up) or neutral (straight)
How the intervention might work
The aim of treatment of scaphoid fractures is to achieve fracture union, functional recovery, and the avoidance of complications of bone healing such as nonunion or malunion (Rhemrev 2011). The primary method of treatment for undisplaced or minimally displaced scaphoid fractures is cast immobilisation, which is generally successful and results in healing rates of between 90% to 100% (Bhat 2004; Gellman 1989). However, as described above, the extent and type of immobilisation can vary. A key decision underlying the choice of cast depends upon the joints that need to be immobilised. Some authors have advocated using casts that extend above the elbow as a means of achieving good rates of union, whilst others have suggested that above elbow casts can result in movement at the site of fracture as the normal rotation of the forearm is restricted and transmitted to the radiocarpal joint (Kuhlmann 1987). While the inclusion of the base of the proximal phalanx of the thumb in the so-called 'scaphoid' cast could enhance fracture healing through greater stabilisation, immobilisation of any joint that is not injured can result in an increase in morbidity, such as stiffness (Clay 1991; Karantana 2006). A further issue that arises from the use of forearm casts concerns the position in which the wrist is immobilised - whether it should be immobilised in flexion (Cooney 1980; Yanni 1991), or slight extension (Fisk 1970; King 1982).
There is a 'consensus' amongst orthopaedic surgeons that most stable scaphoid fractures will take between six and eight weeks to unite. However, if the injury is not managed adequately, this may have implications for fracture healing, the need for surgery and recovery time. All types of immobilisation will restrict hand function for the duration of treatment, and this may require corrective physiotherapy following removal of a cast. However, inadequate immobilisation may increase the risk for nonunion in around a third of cases (Furunes 2009; Langhoff 1988; Sjolin 1988). Thus, immobilisation for acute fractures of the scaphoid appears warranted. Conversely, there appears to be no indication for dynamic functional braces (that aid the initiation and performance of motion by the part of the body they support) in proven acute scaphoid fractures because of this high risk of nonunion (Rhemrev 2011).
Radiological follow-up is important in these cases and a change of management, such as surgery, may be needed if the fracture displaces. The healing of these fractures is a continuous process and serial imaging is required to assess progression to union and consolidation (complete repair of a fracture with no evidence of fracture on imaging). However, union and consolidation can be difficult to assess on radiographs, and poor interobserver agreement has been found in assessment of scaphoid fracture union at 12 weeks post-injury (Dias 1988). Although other imaging modalities, such as magnetic resonance imaging (MRI) and computer-tomography (CT), are also used to assess fracture union, there is variation in how these scans are interpreted (Buijze 2012; Kulkarni 1999; McNally 2000).
Why it is important to do this review
Since these common fractures occur predominantly in young adults of working age (Rhemrev 2011), and require several weeks of immobilisation while the fracture heals, they have a significant economic as well as personal impact (Ibrahim 2011). There is considerable variation in practice, as shown by the results of a survey undertaken in 2009 of British Orthopaedic Association members that asked about the use of the scaphoid cast and wrist position in the cast (Petheram 2009), which showed that 57% of surgeons were managing these fractures in scaphoid type casts that included the thumb, 40% were treating them in a below-elbow plaster with the thumb free, and the remaining 3% varied their management between the two types of plaster. In terms of wrist position, 68% of surgeons placed the wrist in a neutral position, 20% in extension and 12% in flexion. The importance of this injury and the clear variation in treatment endorses the need for a systematic review of the best quality evidence to inform on conservative treatment decisions for these fractures.
To assess the effects (benefits and harms) of conservative interventions for treating scaphoid fractures in adults.
Criteria for considering studies for this review
Types of studies
Randomised controlled trials (RCTs) and quasi-randomised controlled trials (in which the method of allocating participants to treatments is not strictly random: e.g. by date of birth, hospital record number, alternation) evaluating conservative interventions for treating scaphoid fractures in adults.
Types of participants
We will include adults with acute fractures of the scaphoid treated with conservative interventions. We will record the method of diagnosis, or its absence, in the included trials and report it in the review, but will not use it as a reason for exclusion.
Trials that include some adolescents will be included provided the proportion of adolescents was clearly under 10%, or separate data are available for them. The Tan 2009 study described the incidence of paediatric scaphoid fractures as being 11.5%, with an incidence of 15 per 100,000 (0.55% of all paediatric bone injuries referred to their unit).
Types of interventions
We will include trials that compare different conservative interventions for the definitive treatment of scaphoid fracture. The prime focus will be on comparisons of different types of cast, different positions of the wrist within the cast, different types of cast material, and different durations of immobilisation. The main guide for selection of the 'experimental' or intervention group will be that it involves less immobilisation (i.e. not involving an uninjured joint; shorter duration) than the control group.
The main comparisons will be:
- forearm cast (e.g. Colles' plaster cast) with thumb not immobilised versus forearm cast with thumb immobilised (scaphoid plaster cast);
- below elbow cast (forearm or scaphoid) versus above-elbow cast (cast with extension above the elbow);
- forearm cast with wrist in various positions (flexion versus extension versus neutral).
We will exclude trials of interventions aimed at accelerating fracture healing, such as ultrasound, which is covered in another Cochrane review (Griffin 2012).
Types of outcome measures
Where possible, primary outcome measures will be presented for follow-up times of up to one year, between one and three years, and three years and above.
- Major adverse effects including symptomatic nonunion, malunion and avascular necrosis, and complex regional pain syndrome, together with secondary treatment (e.g. other method of immobilisation, surgery).
- Other composite measures or scores of upper limb function.
- Patient-reported quality of life measures (e.g. the Short Form (36) Health Survey).
- Resumption of, and time to return to, previous activities.
- Complications of cast immobilisation (e.g. ischaemia, skin lesions, stiff non-injured joints).
- Symptomatic non-union and its potential longer-term consequences (determined in assessment of the patient during follow-up).
We will also present data for other outcomes reported in the included trials, such as time to union and range of movement, but these will not be presented in summaries of the main results of the review. The authors will also attempt to present information, where possible, on how trials determined fracture union and consolidation, and the development of non-union. Thus we will attempt to evaluate whether this assessment was done using standard radiographs (standard posterior anterior and lateral wrist views, or the scaphoid series) or other imaging such as CT scans. Where CT scans were performed, we will also check whether CT was done in all cases or only in those where there was uncertainty.
We will collect economic data such as resource use, number of hospital attendances, and time off work.
Search methods for identification of studies
We will search the Cochrane Bone, Joint and Muscle Trauma Group Specialised Register (to present), the Cochrane Central Register of Controlled Trials (The Cochrane Library current issue), MEDLINE (1946 to present), and EMBASE (1980 to present). We will also search the metaRegister of Controlled Trials and the WHO International Clinical Trials Registry Platform for ongoing and recently completed studies.
We will apply no restrictions based on language or publication status.
In MEDLINE, a subject-specific strategy will be combined with the sensitivity-maximizing version of the Cochrane Highly Sensitive Search Strategy for identifying randomised trials (Lefebvre 2011) (see Appendix 1). Search strategies for the Cochrane Central Register of Controlled Trials and EMBASE are also shown in Appendix 1.
Searching other resources
We will search reference lists of selected articles to identify other RCTs. We will also review abstracts from proceedings of relevant trauma and orthopaedic conferences.
Data collection and analysis
Selection of studies
Both authors (MJM and JH) will screen the search results independently for potentially eligible trials. After full text articles of appropriate reports have been obtained, both authors will select studies independently according to the review's inclusion criteria. Any disagreement will be resolved by discussion.
Data extraction and management
Data will be extracted by MJM and checked by JH for all studies. Data will be extracted using a data collection form based on the Consolidated Standards of Reporting Trials (CONSORT) checklist (Schulz 2010). Data of baseline characteristics will include age, gender and the interval between fracture and initiation of definitive treatment. Any disagreements will be resolved by discussion. We will attempt to contact trial authors where there are incomplete details on study methods or data. As well as recording the method of diagnosis used in the trials, we will collect and report information on the methods of outcome assessment including those for fracture non-union and malunion.
Assessment of risk of bias in included studies
Both authors (MJM and JH) will assess the risk of bias of included trials using The Cochrane Collaboration's tool (Higgins 2011). This will include the following domains: random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, completeness of outcome data, selective outcome reporting and any other sources of bias. Subjective outcomes, such as pain or post-treatment function, will be reviewed separately from other outcomes, such as fracture union, when assessing blinding of outcome assessment and completeness of outcome data. We will attempt to contact the trial authors for clarification when methodological details are unclear. We will resolve differences by discussion.
For each of the above domains, the review authors will assign a judgement for risk of bias as being low, high or unclear, in line with the criteria summarised in the Cochrane Handbook (Higgins 2011).
Measures of treatment effect
Risk ratios with 95% confidence intervals will be calculated for dichotomous outcomes. Mean differences with 95% confidence intervals will be calculated for continuous data. When two or more studies present data derived from the same validated instrument of evaluation (with the same units of measure), data will be pooled as a mean difference. When primary studies express the same variables through different instruments (and different units of measure), we will use the standardised mean difference. Should time-to-event data arise, this will be analysed as dichotomous data, provided that the status of all participants in the study is known at a fixed time point.
The data will be analysed over a time period that reflects time to union, time to discharge and time to resolution of symptoms. As the rate of union can be difficult to assess, and poor interobserver agreement has been identified for union rates (Dias 1988), no specific criteria will be set for assessing short- and long-term benefit.
Unit of analysis issues
We anticipate that the unit of randomisation in the included trials will be the individual person. Occasionally, when participants are included who have sustained bilateral injuries, results may be presented in the form of fractures or limbs. Where such unit of analysis issues arise and appropriate corrections have not been made, and where the disparity between the units of analysis and randomisation is small, we will consider presenting the data for such trials. 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 to request missing data. Where appropriate, we will perform intention-to-treat analyses to include all people randomised to the intervention groups. If there is discrepancy between the randomised and analysed participants in each treatment group, we will calculate the percentage loss to follow-up in each group. Unless we can calculate missing standard deviations from standard errors, exact P values or 95% confidence intervals, we will not impute these. Where data are unavailable, as a result of participants being lost to follow-up, their outcomes will remain unknown and, therefore, will not be available for inclusion in the review.
Assessment of heterogeneity
The heterogeneity of estimate effects between the included studies will be assessed by visual inspection of the forest plot (analysis) along with consideration of the Chi² test for heterogeneity and the I² statistic.
Assessment of reporting biases
We plan to reduce reporting bias by: a) searching for published, unpublished and ongoing trials; b) placing no language restrictions on the search strategy; c) checking for multiple trial reports of the same trial; d) attempting to obtain the protocol or the trial registration document of trials; and e) contacting the authors in cases where the pre-specified primary (favourable or adverse) outcomes are not reported.
Reporting biases will be assessed using funnel plots if there are at least ten trials contributing data to a meta-analysis.
When considered appropriate, results of comparable groups of trials will be pooled using both fixed-effect and random-effects models. The choice of the model to report will be guided by a careful consideration of the extent of heterogeneity and whether it can be explained, in addition to other factors such as the number and size of studies that are included. Ninety-five per cent confidence intervals will be used throughout. We will consider not pooling data where there is considerable heterogeneity (I² greater than 75%) that cannot be explained by the diversity of methodological or clinical features among the trials. Should meta-analysis not be possible or appropriate, the data from these trials will be reported individually and data presented in forest plots.
Subgroup analysis and investigation of heterogeneity
If data allow, subgroup analyses will be undertaken to investigate heterogeneity. The following subgroup analyses will be considered; these are guided by the expectation of major differences in fracture healing potential:
- proximal third/pole fractures versus waist, tuberosity and distal third fractures (the first group are likely to have a poorer outcome);
- smokers versus non-smokers (the first group are likely to have a poorer outcome)..
We will investigate whether the results of subgroups differ significantly by inspecting the overlap of confidence intervals and performing the test for subgroup differences available in the Review Manager 5.2 software (RevMan 2012)
If data allow, sensitivity analyses will be undertaken to explore aspects of trial and review methodology, including exploring the effects of missing data; the inclusion of trials at high or unclear risk of bias from lack of allocation concealment or assessor blinding, or both; the selection of statistical model (fixed-effect versus random-effects) for meta-analysis, and the inclusion of trials only reported in conference abstracts and of trials with unclear methods of diagnosis.
'Summary of findings' tables
Where there is sufficient evidence to merit the preparation of 'Summary of findings' tables, we will develop these for the main comparisons. We shall use the GRADE approach to assess the quality of evidence related to each of the key outcomes listed in the Types of outcome measures (Chapter 12.2, Higgins 2011).
The review team would like to thank Joanne Elliot and Lindsey Elstub at the Cochrane Bone, Joint and Muscle Trauma Group for their support in writing this protocol. They would also like to thank Prof Tim Davis, Dr Nigel Hanchard and Dr Helen Handoll for their feedback at editorial and external review.
Appendix 1. Search strategies
The Cochrane Central Register of Controlled Trials (CENTRAL) (Wiley Online Library)
#1 MeSH descriptor Scaphoid Bone, this term only
#2 MeSH descriptor Carpal Bones, this term only
#4 (#1 OR #2 OR #3)
#5 MeSH descriptor Fractures, Bone explode all trees
#6 MeSH descriptor Fracture Healing, this term only
#7 MeSH descriptor Wrist Injuries, this term only
#9 (#5 OR #6 OR #7 OR #8)
#10 (#4 AND #9)
MEDLINE (Ovid online)
1 Scaphoid Bone/
2 Carpal Bones/
5 exp Fractures, Bone/
6 Fracture Healing/
7 Wrist Injuries/
11 Randomized controlled trial.pt.
12 Controlled clinical trial.pt.
15 Drug therapy.fs.
20 exp Animals/ not Humans/
21 19 not 20
EMBASE (Ovid online)
1 Scaphoid Fracture/
2 Scaphoid Bone/
3 Carpal Bone/
6 exp Fracture/
7 exp Fracture Treatment/
8 Wrist Injury/
13 Randomized Controlled Trial/
14 Clinical Trial/
15 Controlled Clinical Trial/
17 Single Blind Procedure/
18 Double Blind Procedure/
19 Crossover Procedure/
21 Prospective Study/
22 ((clinical or controlled or comparative or placebo or prospective* or randomi#ed) adj3 (trial or study)).tw.
23 (random* adj7 (allocat* or allot* or assign* or basis* or divid* or order*)).tw.
24 ((singl* or doubl* or trebl* or tripl*) adj7 (blind* or mask*)).tw.
25 (cross?over* or (cross adj1 over*)).tw.
26 ((allocat* or allot* or assign* or divid*) adj3 (condition* or experiment* or intervention* or treatment* or therap* or control* or group*)).tw.
29 Case Study/ or Abstract Report/ or Letter/
30 28 not 29
Contributions of authors
MJM drafted and revised the protocol and is the guarantor.
JH helped write the protocol and approved the final version.
Declarations of interest
MJM: none known.
JH: none known.
Sources of support
- Gateshead Health NHS Foundation Trust, QE Hospital, Sheriff Hill, Gateshead, UK.Salaries of both authors; logistical support
- No sources of support supplied