Abstract
Introduction
Fractures of the distal radius are common in the elderly population. Elderly patients have been considered to be functionally low-demand, but as the population becomes more active, expectations of outcomes may be increasing. The aim of this review was to determine whether there is any correlation between radiographic and functional outcome for elderly patients following fracture of the distal radius.
Methods
Medline, PubMed, CINAHL, Embase and AMED were searched for studies including elderly patients managed conservatively or surgically following distal radius fracture, either alone or in combination with younger patients.
Results
Eleven studies were selected for review. A wide variety of outcome measures and radiological parameters were found to be used. Of the 11 studies, 6 found no correlation between functional outcome and radiological outcome.
Discussion
The literature reviewed suggests functional outcome for elderly patients following distal radius fracture is not dependant on radiological outcome. This may give therapists better insight into potential outcomes for their patients, however further studies may be required to determine the most effective therapeutic management of these patients.
Introduction
Fractures of the distal radius (DRF) are considered to be the most common of all fractures. 1 Incidence ranges from 280 to 440 per 100,000 individuals 2 with a female to male ratio of 2.5–4.9 to 1, increasing to 7 to 1 in people over 60 years of age. 3 In the younger adult population epidemiology is different, with the female to male ratio reversing and the majority being caused by high energy trauma such as sports injuries. 3
Traditionally, elderly patients have been considered to have a lower functional demand and be a higher surgical risk; therefore management has tended to be conservative. 4 It is suggested that as the population is becoming more physically active, expectations of function following fracture may increase. 3
Management may involve reduction of the fracture fragment followed by a period of immobilisation. 5 Patients are commonly referred for physiotherapy or hand therapy for rehabilitation, and these patients can require prolonged input. However, the evidence for the role of therapy following DRF is poor.6,7
Improved knowledge of fracture outcomes may help in identifying effective interventions 5 and in shaping rehabilitation services.
Distal radius fractures
DRFs are described as those fractures occurring within 2 cm of the distal end of the radius4,8 and commonly occur as a result of a fall on an outstretched hand.9,10 Cooney et al. 11 describe a sharp palmar and a dorsally comminuted fracture line, suggesting initial tension palmarly, leading to compression forces dorsally and subsequent comminution of the dorsal cortex. This typically leads to the dorsal displacement of the distal fracture fragment first described by Abraham Colles. 12 The extent of the injury and displacement is directly related to the energy involved, direction of force and quality of the bone. Osteoporotic bone is less able to absorb energy and leads to a more comminuted fracture.11,13
Routine radiographs used to assess the wrist are postero-anterior and lateral views.11,14 Fracture displacement can be assessed and management planned. Assessment is made using a number of clinical measurements relating to the anatomy of the distal radius and ulna. The most commonly used measures are radial height, ulnar variance, radial inclination (radial angulation, radial tilt) and palmar (volar) tilt (angulation) (see Figure 1).
Radiological measures of the distal radius, including normal values.
There are also a number of fracture classification systems, including those defined by Cooney 15 and Fernandez and Jupiter 16 as well as the Muller et al.'s AO system. 17
It is generally accepted that functional outcome parallels restoration of normal anatomy. 3 The majority of DRFs are managed conservatively, often involving closed reduction of the fracture, followed by a period of immobilisation. 5 However, if this is insufficient, the fracture may be reduced and fixation obtained surgically. 18 The clinical decision on management of a DRF can be complex, taking into account the patient's health and functional demands and considering factors which may affect quality of life during recovery, such as rate of recovery and return to activities of daily living and potential complications. 19
Although criteria for acceptable reduction of DRF have been suggested, 20 these have been based mostly on studies concerning younger adults.4,8 Ranjeet and Estrella 21 highlight issues such as mid-carpal instability, ulnar impaction syndrome and osteoarthritis associated with poorly reduced DRF. There is still some disagreement over the importance of anatomical reduction following DRF in the elderly, with one study concluding that primary reduction in elderly, frail, low-demand patients was ineffective. 22 Functional outcome was not considered.
Traditionally, outcomes following DRF have focused on radiological outcome, objective measures of function and assessment using tools such as the Gartland and Werley, 23 Cooney et al. 24 and Saito and Shibata 25 demerit systems. These systems use a combination of subjective outcomes including pain and disability, objective outcomes and occurrence of complications. The evaluation is undertaken by the assessor not the patient. Many of these scoring systems are used in a modified form.
There is a trend towards the use of patient-rated measures of function and disability, such as the Disabilities of the Arm, Shoulder and Hand (DASH), 26 Michigan Hand Questionnaire (MHQ) 27 or the Patient-Rated Wrist and Hand Evaluation (PRWHE) 28 (also still described as the Patient-Rated Wrist Evaluation (PRWE) in some literature reviewed), which are all validated for use with DRF.2,8,29 However, limitations to the use of these tools for this patient group have been highlighted, with lack of sensitivity of the DASH 30 and the inability to differentiate between affected and unaffected limbs with the DASH and PRWE 29 noted. A more recent study 31 comparing outcome tools used following DRF, found that there was some correlation between the DASH, Gartland and Werley 23 and Cooney et al. 24 scores, however, none of these outcome measures correlated to the Sarmiento radiological score 32 used to evaluate anatomical outcome. The Sarmiento score grades fracture displacement from excellent to poor. It is a modification of a previously published scoring system.
A further aspect to be considered is patient satisfaction. Chung and Haas reviewed patient satisfaction in surgically managed DRF, compared with objective outcomes. 33 Selected elements of the MHQ pertaining to wrist motion and grip strength were used to evaluate satisfaction, and were compared against grip strength, key pinch grip strength and wrist arc of motion. It was concluded that a much greater range of motion (ROM) (95% of the uninjured wrist) is required for satisfaction, than is required for good function (25–57% of normal as described in quoted literature). Sixty-five percent grip strength was found to be required for satisfaction, which was felt to be comparable with studies assessing functional grip requirements.
To enable effective early management of DRF, it needs to be determined whether there is a correlation between the radiological findings and the patient's functional outcome.
Increasing access to radiographic images for hand therapists, with the advent of digital radiographic imaging, has the potential to give therapists working with patients following DRF a better insight into likely outcomes for their patients.
The aim of this review was to determine whether there is any correlation between radiographic and functional outcome for elderly patients following DRF.
Methods
The databases Medline (1946 to date), PubMed (1865 to date), CINAHL (1937 to date), Embase (1947 to date) and AMED (1985 to date) were searched using the terms ‘distal radius fracture’, ‘treatment outcome’, ‘outcome assessment’ and ‘radiography’. Studies were selected following a title and abstract search, and further studies of interest were selected following a manual review of references. Studies were selected if they included elderly patients either as the sole patient group or in combination with younger adults. The definition of ‘elderly’ was as defined in the studies themselves. Studies included patients managed both conservatively and surgically. Studies were excluded if not published in English or if published prior to January 2000. Of all, 11 studies were selected for review.
Results
Characteristics and results of studies reviewed.
ORIF: open reduction & internal fixation; POP: plaster cast immobilisation; DA: dorsal angulation (tilt); RA: radial angulation (tilt or inclination); UV: ulnar variance; RShift: radial shift; RS: radial shortening; VA: volar angulation (tilt); RH: radial height; UH: ulnar height; PRWE: patient rated wrist evaluation; ROM: range of movement; Jebsons: Jebson-Taylor Hand Function Test; SF-12: Medical Outcomes Short-Form 12; DASH: disabilities of the arm, shoulder & hand; GWS: Gartland and Werley Score; MHQ: Michigan Hand Outcome Questionnaire; VAS: Visual Analogue Score; MASS07: Modernised Activity Subjective Survey of 2007.
Due to the reasons previously cited, assessor-rated systems are now rarely used in isolation. However, Fujii et al. 34 solely used the Saito demerit system 25 to evaluate the outcomes of surgically managed displaced DRF in elderly patients. They noted that the functional results for patients with radial shortening of more than 6 mm were poorer, although not statistically significant.
Other studies opted to use patient-rated outcome scores, citing issues such as inconsistency between radiological parameters and objective physical variables in previous papers 35 and that pain or disability are not necessarily evaluated 36 as reasons behind their choice. Patient-rated outcomes were used either in isolation or combined with other outcome measures.
Of the studies using patient-rated scores, five used only elderly patients. Anzarut et al. 8 used a combination of three outcome measures, to study patients following conservative management of DRF. They found no relationship between radiological outcome and the functional measures used, however, dorsal/volar tilt was the only radiological measure applied. The authors concluded that lesser functional demands placed upon the upper limb by the elderly may explain this, or that their measure of radiological deformity may not have been valid, having been drawn from studies of younger patients.
Jaremko et al. 4 conducted a similar study of conservatively managed DRFs. A number of additional radiographic measures were used. They found that there was no relationship between radiological outcome and self-reported outcomes, however, it was noted that mean DASH score was slightly worse in those with radiographic deformities outside their ‘acceptable’ parameters. There was however, no discussion on whether a possible predictor of disability could be identified.
Three studies looked at conservatively and surgically managed DRFs. One study 37 included only patients aged over 70, dividing them into two groups; the surgical group being managed with open reduction and internal fixation (ORIF). They used a range of outcome measures, with follow-up ranging from 12 to 81 months, but found no statistically significant differences between the groups for any outcomes except pain. The ORIF group reported higher levels of pain, which the authors attribute to post-operative complications, such as tendon irritation. Of note is the discrepancy between the radiological measures reportedly used, and the measures used to define acceptable fracture reduction. Again, lack of correlation between radiological and other outcome measures was felt to be related to lower functional demand in the age group studied.
Synn et al. 38 also used a range of outcome measures. They stated that the inclusion of an assessor-rated score allowed comparisons with previous studies. They found no correlation between anatomic deformity, or fracture comminution and functional outcomes. Grip strength recovery was found to be greater in undisplaced fractures. In this study, the length of time to follow-up was also studied, and no correlation was found between this and functional outcome. The authors do not comment further on their findings.
One further study 39 limited their investigation to females only, aged 50–75 years. They found moderate or severe malunion in 26% patients and found no significant long-term difference between these patients and the others with regard to DASH, pain and ROM. It was noted that this group of patients was more likely to report persistent pain (pain of any severity at 1 year and 2–4 years after fracture) and that decreased supination persisted long-term. When the moderate/severe malunion subgroup were looked at in isolation, DASH, pain, grip strength and ROM were shown to improve between the 1 year and 2–4 year follow-up, although not statistically significant, but grip strength did not improve. The authors accept that their small sample size makes extrapolation of results difficult.
Five other studies using patient-rated scores looked at outcomes of DRF in adults of all ages. MacDermid et al. 36 looked to determine predictors of pain and disability following DRF. Patients were reviewed following either conservative or surgical management. Using the PRWE it was found that at 6 months, pre-reduction radial shortening was a significant predictor of pain and disability. It was felt the pre-reduction deformity may correlate to severity of injury and associated soft-tissue injury, which may then influence functional outcome. Radial shortening was the only radiological measure used, and it was noted that future studies may benefit from reviewing other additional radiological measures.
Chung et al. 2 found that 3 months following ORIF for DRF, articular incongruity and loss of normal volar tilt were associated with poorer outcomes as measured by the MHQ. However, at 1 year, neither radiographic outcome was associated with functional outcome; the authors felt this may be due to lack of large deviations from radiographic norms as a result of surgical management. They also found that increasing age tended to be associated with poorer outcomes.
Wilcke et al. 35 used the DASH to evaluate patients following DRF managed either conservatively or with external fixation. It was found that final radial shortening of ≥ 2 mm, dorsal angulation > 15 ° and radial angulation > 10 ° were significantly related to both a poor DASH score and a poor satisfaction score. Grip strength of less than 80% of the uninjured wrist was also associated with significantly poorer DASH outcomes; they found a strong correlation between DASH and satisfaction. This correlation was also seen when analysis was restricted to patients over 50 years of age.
In a large, prospective cohort study, the DASH was used alongside the SF-12 health status questionnaire, to evaluate the outcomes of conservatively or surgically managed DRFs. 40 It was found that, having adjusted the results for age, dorsal tilt and ulnar variance correlated significantly with DASH score. These findings were then used to categorise patients into no malunion and malunion groups, clarifying that patients with dorsal tilt > 10 °, and patients with ≥ 1 mm ulnar variance have a significantly worse DASH score at 1 year than those deemed to have no malunion. These findings also correlated with significantly worse SF-12 score and weaker grip strength. It was found that 1-year DASH scores were similar for both treatment groups despite better anatomical outcome in the surgically managed group. The authors highlighted that this group had more significant fracture displacement on presentation than the cast group, which may partly explain this.
Finally, Kumar et al. 41 looked at patients ranging from 22 to 94 years of age, divided into two age groups; above and below 60 years. All patients had extra-articular DRF managed conservatively. The authors concluded that satisfactory functional results, as measured by DASH and MHQ, for the older group were obtained despite unsatisfactory radiological results. Although, the authors report the use of criteria for acceptable fracture union as taken from a quoted study, they do not outline these criteria in their paper, neither are values given for acceptable DASH and MHQ outcomes.
Discussion
Although there is a plethora of literature available with regards to the management and outcomes of DRF, studies specifically looking at radiological and functional outcomes were scarcer.
There have been many attempts at defining the management of DRF, with a number of fracture classification systems available; a small number being widely used and fewer attaining validation. 42 Consequently, in research, values used to assess radiological outcome of fracture management are inconsistent. In addition, there are a number of objective and functional scores available for use, only some of which are validated for use with DRF. These factors make evaluating evidence from the available literature difficult.
Reviewing the literature, of the studies looking at elderly populations only, the majority4,8,34,37,38,41 found no significant correlation between functional outcome and radiological outcome. Studies including patients of all ages were more likely to show some correlation between radiological outcome and function, with authors2,35,36,40 citing radial shortening and dorsal/volar angulation as having the most significant effect on function.
Looking at these two groups of studies, it might be suggested that the inclusion of younger patients in the later group of studies may have biased the results, suggesting significance of radiological outcome which may only be relevant to those younger patients. Kumar et al. 41 were the only authors to divide their subjects into two further groups and found no significant correlation between radiological and functional outcomes.
In the studies reviewed, eight different terms for radiological parameters were used, some of which may be interchangeable, for example radial length and ulnar variance. Without clear, unambiguous terminology, comparison of data is difficult. Most studies defined their ‘acceptable’ ranges for radiological outcomes, but supporting evidence was not necessarily provided, and the ‘acceptable’ ranges used varied between studies. Some authors4,8 suggest that their ‘acceptable’ ranges of deformity, which were based on those used for younger adults, may need ‘fine-tuning’ to detect correlation with function in the elderly, but also comment that they feel this would have had little effect on their results.
In evaluating functional outcome, 12 different tools were employed across the 11 papers reviewed. Most employed validated patient rated tools, either in combination with other patient rated outcomes, demerit systems or other objective measures, but again this level of variation across studies does not make comparison of results easy.
Patient satisfaction as an outcome measure was used in three studies,4,8,35 each utilising different tools, including a Visual Analogue Score (VAS) and elements of a validated satisfaction score. Two studies4,8 exclusively looked at conservatively managed DRF, with the third 35 evaluating outcomes following displaced DRF. Direct comparison of findings from these studies is therefore not possible.
Most of the studies had relatively short follow-up periods of 2 years or less, which can lead to questions on the validity of their results, with some authors suggesting this as a reason for the lack of correlation found in their study. 38 However one study 37 did follow-up their participants for 5–6 years, and another 39 for an average of 39 months and although Brogren et al. 39 found that there was correlation between malunion and poorer function and pain at 1 year, no such correlation was found in either study at long-term follow-up. These results, and those of others, 2 could suggest that radiological outcomes in the short-term (3–12 months) may have some effect on function, but in the long-term these effects lose significance.
The studies reviewed included fractures managed across the spectrum of conservative and surgical management options. Of those looking at elderly patients, two studies4,8 considered only those managed conservatively; neither of these studies showing correlation between function and radiology. Similar findings were found in a study 41 looking at conservatively managed fractures in patients of all ages. Of the studies looking at both conservative and surgically managed fractures, four studies2,35,36,40 identified radiological outcomes which correlated with functional outcomes. Only Brogren et al. 40 considered the impact surgical fixation may have had on outcomes, finding those patients managed surgically achieved better radiological outcomes. None of these authors investigated the effect on functional outcomes of the treatment method or intervention used. One might expect that surgical fixation would lead to better radiological outcomes, with another author 22 highlighting the ineffectiveness of closed manipulation on fracture position in the elderly, suggesting poorer radiological outcomes in this group. Conversely, it could be argued that those patients managed conservatively, may have had fractures which were not significantly displaced initially, and consequently good radiological outcomes. It is also worth noting the potential effect of a displaced fracture on the soft tissues, 36 which in turn may influence restoration of movement and function, as well as the effect of surgery on the soft tissues, with two authors35,37 highlighting post-operative issues such as tenosynovitis, pin-tract infections and scar issues.
The majority of studies reviewed did not indicate if age played any part in the fracture management decision process. Most reported that fracture management was based on a ‘clinical’ decision, but did not elaborate further. A few studies4,8,37 did state that they excluded patients who were not living independently. Two of these4,8 were studies looking at only conservatively managed fractures. It could be argued that in some cases the ‘clinical decision’ not to operate may have taken into consideration the frailty of the patient and the demand they were likely to place on their wrist. In some studies34,37 patients were included who had made the decision themselves not to have surgery; these patients may also have been low-demand and made their decision based on their knowledge of their own needs. Some authors8,37 have acknowledged that the lack of correlation found between radiological outcome and functional outcome may be due to the lower functional demands of their elderly cohort, although it is also recognised that the elderly population is becoming more active, with greater expectations for their outcomes. 3
Conclusion
A review of available literature suggests that, for elderly patients, acceptable functional outcome following DRF is not dependant on acceptable radiological outcomes. Radial shortening and dorsal/volar tilt have been suggested as having an effect on function, but this is in studies including elderly patients in a cohort of all ages. Some correlation has also been found between radiological and functional outcome in the short term (3–12 months), but this loses significance at long-term follow-up. Hand therapists working with this group of patients can be reassured, and can reassure their patients, that despite radiological outcomes, good function can be obtained, although this may take 12 months or more. This may also suggest that prolonged or intensive one-to-one input for these patients may not be appropriate. Further studies may be beneficial to determine the most effective therapeutic management of these patients and there may be some value in investigating further the radiological parameters, particularly radial shortening and dorsal/volar tilt, aiming to define the degree of fracture displacement which does affect functional outcomes.
Footnotes
Acknowledgements
The author would like to thank Nikki Burr for giving feedback on the original manuscript submitted as part of the BAHT education process, and Bridget Ellis who gave support, encouragement and feedback throughout.
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Conflict of interest
The author declares that there is no conflict of interest.
