Treatment options in extra-articular distal radius fractures: a systematic review and meta-analysis

Purpose This systematic literature review aimed to make a detailed overview on the clinical and functional outcomes and to get insight into the possible superiority of a treatment method for extra-articular distal radius fractures. Methods Embase, Medline, Cochrane Library, Web of Science, and Google Scholar were searched for studies describing treatment results. Five treatment modalities were compared: plaster cast immobilization, K-wire fixation, volar plating, external fixation, and intramedullary fixation. Results Out of 7,054 screened studies, 109 were included in the analysis. Overall complication rate ranged from 9% after plaster cast treatment to 18.5% after K-wire fixation. For radiographic outcomes, only volar tilt in the plaster cast group was lower than in the other groups. Apart from better grip strength after volar plating, no clear functional differences were found across treatment groups. Conclusion Current literature does not provide uniform evidence to prove superiority of a particular treatment method when looking at complications, re-interventions, and long-term functional outcomes. Supplementary Information The online version contains supplementary material available at 10.1007/s00068-021-01679-z.


Introduction
A large variety exists in distal radius fractures; from highenergy, comminuted, intra-articular fractures, to low-energy, simple extra-articular fractures. Because of the difference in bone quality, fracture characteristics, associated soft tissue injury, and patient's needs between these groups, a different treatment approach might be necessary [1][2][3].
Fractures of the distal radius can be treated operatively or non-operatively. Closed reduction and plaster cast immobilization was traditionally preferred for both intra-and extraarticular fractures. Devices for operative treatment, such as Kirschner wires, plates, external fixators, and intramedullary implants, have been introduced over the last decades and their appropriate use yields good functional results [4][5][6]. Especially the introduction of volar locking plates has led to a significant increase in operative treatment rates, also for extra-articular fractures, because of the improvement in stability and therefore the possibility for early mobilization [7]. On the other hand, complaints and complications from tendon irritation are assumed more common and the extensive dissection of soft tissue might cause fracture and wound healing problems [7]. Other operative treatments also provide better stability than plaster casting, but have their own cons [8][9][10].
Current guidelines suggest closed reduction and plaster cast immobilization as the primary treatment for extraarticular distal radius fractures because the treatment is non-invasive and cheap [11,12]. However, it is unclear if a non-invasive method also means fewer complications and an acceptable loss of function. There is still no consensus regarding the superiority of one of these methods for treating extra-articular distal radius fractures.
The aim of this systematic literature review and metaanalysis was to make a detailed overview on the clinical and functional outcomes and to get insight into the possible superiority of a treatment method in adult patients with an extra-articular distal radius fracture.

Search strategy
A literature search was conducted June 19, 2018. Embase, Medline, The Cochrane Library, Web of Science, and Google Scholar were searched to identify relevant clinical studies that report on the outcomes of extra-articular distal radius fractures. De-duplication of studies was performed as described before [13]. The exact search strategy for the different databases is shown in Online Appendix 1. No language limits were used and any potentially eligible non-English language manuscripts were translated if possible by native speaking colleagues or using Google translate, if no native speaker was available.

Selection criteria
The studies were eligible for inclusion if they met the following criteria: (1) Patients who suffered and were treated for an extra-articular distal radius fracture; (2) age 18 years or older; (3) patients were either treated using a plaster cast, K-wire(s), plate fixation, external fixator, or an intramedullary device; (4) primary data for at least one outcome parameter had to be available.
Studies without a clear description of fracture type or intervention, or studies that lacked sufficient data for analysis were excluded. In addition, studies with incomplete registration of complications were excluded. Studies describing a not-commonly used device or technique for extra-articular fractures were excluded (i.e. dorsal plates, Epibloc system, cannulated screw, above elbow cast, etc.). If a study consisted of one eligible and one non-eligible cohort, this study was included in our study, but only if data for the eligible group was provided.
Two authors (GWVO and AA) independently screened the search results for potentially eligible studies by checking the title and abstract in phase one. Any disagreement has been resolved by consensus or consultation of a third author (TH). After the exclusion of all irrelevant studies, the full text of the remaining studies was obtained, and the eligibility was evaluated to complete selection in phase two. This again was done independently by two authors (GWVO and MRLR). Any disagreement was resolved as described above. If the full-text study was not available, the corresponding author was contacted and asked.
The study protocol has not been registered or published before.

Data extraction
In phase three, the same two authors independently extracted the following data from the included studies: author, year of publication, study design, type of treatment, number of patients, mean age of patients, number of female patients, duration of follow-up, and type of fracture classification. The outcome data that were retrieved, consisted of; complications and re-interventions, radiographic outcomes (i.e. consolidation rate, volar/dorsal tilt, radial inclination, radial height, and ulnar variance), range of motion, grip strength, and functional outcome scores (i.e. Disabilities of the Arm, Shoulder, and Hand (DASH) score, Quick-DASH score, Patient Rated Wrist Evaluation (PRWE-)score, Visual Analog Scale (VAS) for pain, Gartland and Werley score, and the Mayo Wrist Score). All outcome data were pooled for four different follow-up periods: 0-3 months, > 3-6 months, > 6-12 months, and 12 months or longer (long term).

Quality assessment
Risk of bias and methodological quality of the included studies were assessed using the Newcastle-Ottawa Quality Assessment Scale. This scale results in a score ranging from 0 to 9. Scores of 8 and 9 were defined as high-quality studies, scores of 5-7 were defined as medium quality, and scores of 1-4 were defined as low quality [14].

Data analysis
Meta-analysis for binary data was performed using MedCalc for Windows. Heterogeneity testing was performed using Cochran's Q and the I 2 statistic. Q is distributed as a chisquare statistic with k (number of studies) minus 1 degrees of freedom. Q has low power as a comprehensive test of heterogeneity [15]. The I 2 statistic describes the percentage of variation across studies that is due to heterogeneity rather than chance [16,17]. A random-effects model was used if the I 2 value was larger than 40%. When the I 2 was lower than 40% a fixed-effects model was used. Pooled estimates (i.e., proportion) are reported with their 95% confidence intervals (CI). When there was no overlap in the 95% confidence intervals, the difference was considered statistically significant. For the continuous data a pooled estimate (i.e., mean) was calculated in Microsoft Excel, using sample size as a weighting factor. Since most studies only provided the mean values but no standard deviation, 95% confidence interval, or standard error, a formal meta-analysis was not feasible for the continuous data.

Results
The primary search resulted in 14,398 hits and after deduplication 7054 studies remained. After screening these studies by title and abstract 1137 remained for full-text evaluation. A total of 109 studies, with 136 different cohorts were marked eligible for the analysis ( Fig. 1 and Supplementary Table S1). These 136 cohorts consisted of a total of 6707 patients divided over the five different treatment modalities. The pooled average age and gender of patients for each treatment method are outlined in Table 1.

Complications
The pooled overall complication rate ranged from 9% after plaster cast treatment to 19% after K-wire fixation ( Table 2). Superficial infections were most prominent and higher in the K-wire (4.7% [95% CI 2.5-7.6%]) and ExFix , respectively). In 6.1% of the K-wire group, the implant(s) failed (i.e., loosening or breakage of material). This was significantly higher than in the volar plate fixation group (0.8%). There was significantly less re-dislocation in the volar plate group when compared with plaster cast (1.0% versus 9.3%). Pooled re-intervention rates ranged from 3.8% to 5.3%. Regarding these reinterventions, only one significant difference was found: the use of antibiotics was 4.2% in the K-wire group versus 0.8% in the volar plating group. Data of all other complications and re-interventions had overlapping confidence intervals across the groups and are described in Table 2 and Table 3 and shown in Fig. 2 and Fig. 3.

Radiographic outcome
Radiographic outcomes are shown in Fig. 4 and Supplemental Table 2. Because no non-union was reported for any of the treatment methods, consolidation rates were not different. The pooled mean radial inclination, radial height, and ulnar variance were similar for all five treatments after > 12 months' follow-up. Overall, volar plate fixation shows consistent good outcome for all four radiological measurements. Especially the ulnar variance was clearly better than the other four treatment modalities with 0.3 mm, − 0.1 mm, 0.3 mm, and 0.3 mm at 0-3, 3-6, 6-12, and > 12 months after surgery, respectively. The palmar tilt in the plaster cast group was consistently lower than in other groups in all follow-up periods. The proportion of patients with a good or excellent Lidström score was reported for three treatments only: plaster cast immobilization, K-wire fixation and external fixation.

Functional clinical outcome
Range of motion and grip strength are outlined in Fig. 5 and Supplemental Table 2. For several outcome parameters, none or a small number of studies provided data for analysis. Differences in range of motion and grip strength were especially found in the short-term follow-up. For example, volar plate fixation showed a relatively good grip strength (69% of the non-injured side) and pro-/supination (81.4 and 80.0 degrees, respectively) at 0-3 months' follow-up. Both pro-and supination were above the limit of disability for the entire follow-up period. At long-term follow-up, no clear functional differences were found across the treatment groups.

Patient-reported outcomes
A large variety of patient rated outcome measures were used in literature. This resulted in small numbers of study subjects and uncertain values. Three outcome measures that were used most often, were chosen to be analyzed and are shown in Fig. 6: Disabilities of Arm, Shoulder, and Hand (DASH) score, Gartland and Werley score, and a Visual Analog Scale (VAS) for pain. Assessment of the DASH score showed low pooled means for volar plating (13.0 points) and intramedullary fixation (15.0 points) at 0-3 months' follow-up. Again, at long term, follow-up scores were similar across the treatment groups. For both the Gartland and Werley score and the VAS for pain, an improvement was seen for all treatments over time and no clear differences were found.

Discussion
This systematic review and meta-analysis aimed to evaluate complications, re-interventions, radiographic outcomes, functional outcomes, and patient-reported outcomes of the most commonly described treatment methods for extra-articular distal radius fractures. The data do not show a clear superiority or inferiority for any of the treatments evaluated.   The main treatment goals for extra-articular distal radius fractures are to regain an adequate wrist function and/or to release pain. In order to study and objectivize this, a multitude of questionnaires and outcome scores have been introduced over the last decades. However, which outcome parameters should be used for valid and accurate assessment of wrist function are still unclear. Also, the variety of treatment options for distal radius fractures impedes the ability to prove superiority or inferiority of the currently used treatment methods. Cochrane reviews and the American Academy of Orthopaedic Surgeons both labeled the evidence as "inconclusive" [124,125]; [126]; [12]. This review focused on extra-articular distal radius fractures only, where both Cochrane and the AAOS stated this inconclusiveness almost a decade ago for intra-articular fractures as well. All eligible literature has been included to show trends in techniques    used over time. Only volar plates have been subject to significant technological improvement, this probably explains the steep increase of studies in the twenty-first century (Fig. 7). Surprisingly, unintended outcome has been reported more frequently than intended outcome: 87 out of 136 cohorts in this study reported at least one type of complication. The overall complication and subsequent re-intervention rates are not significantly different for the five treatment modalities. However, not all studies reported consistently on all complications and re-intervention rates. Different definitions of complications and a possible extra focus on specific complications in certain studies, likely induced broad ranges of complication rates, and probably underreporting. For example, finger stiffness was described in only four studies with a rate up to 24% in a plaster cast immobilization cohort of Mardani et al. [18], where the other 105 studies did not even mention it.
No differences in functional and patient-reported outcomes at long-term follow-up (> 12 months) could be detected. This was also mentioned by Costa et al. and Arora et al. [5]; [77]. However, in the short term, differences in functional and patient-reported outcomes seem to exist. In all treatment groups, a lowering trend was observed after the 0-3 months' follow-up in both the DASH score and the Gartland & Werley score, which means less disability and better function. In line with these findings, also an improvement in grip strength and a decrease in pain were seen in the same time period for all treatment groups. However, volar plates and intramedullary fixation had a much faster recovery than other modalities. Also, range of motion was immediately close to, or even above the limit of disability at 0-3 months' follow-up for both treatments (Fig. 5). Pronation and supination for all follow-up moments were consistently within the normal range of motion as established by, e.g. the Guides, to the Evaluation of Permanent Impairment by the American Medical Association [127]. This early recovery of patients treated with volar plating and intramedullary fixation might be an important factor in treatment choice. In current society, one might argue that long rehabilitation periods are no longer acceptable for most patients. Moreover, a strong lobby from the industry to sell implants and a high strive to restore anatomy perfectly by surgeons may push towards surgical treatment. These arguments advocate for both short-and long-term cost-effectiveness comparison of the various treatment methods.
Radiological outcomes also seem to be in favor of volar plate fixation, as shown by a positive volar tilt, adequate radial inclination, and a negative or low ulnar variance. Radiographic parameters showed worse outcomes for plaster cast immobilization with concomitant high re-dislocation rates. Diaz-Garcia et al. published a systematic review in 2011 to examine outcomes of unstable distal radius fractures after treatment with either volar locking plate, non-bridging external fixation, bridging external fixation, percutaneous  Data are shown as sample size weighted mean. The grey line represents the lowest values that will not cause any functional impairment (disability value [127]) ◂ Kirschner-wire fixation, or plaster cast immobilization. They also found that plaster casting was associated with worse radiological outcomes than operative treatment [128]. Our systematic review reveals that since 2011 nothing has really changed.
The first thing that stood out during screening of potential studies, was the small number of studies that specifically reported on different types of distal radius fractures (AO/ OTA or other classification systems). Many studies reported treatment outcomes for distal radius fractures in general, without any fracture classification. Distinction between different fracture types is essential to provide adequate treatment. A simple extra-articular fracture requires a different approach than a comminuted intra-articular one.
This systematic review had several limitations. First, most studies that were included were retrospective or prospective observational studies, with disappointing reporting quality for continuous data. For a proper meta-analysis of continuous data the mean, a measure of dispersion (either standard deviation, standard error or confidence interval), and sample size are necessary. Many studies in this review failed to provide these data. Moreover, the number of large, well-designed RCTs on extra-articular fractures is very low, and therefore it is very well possible to have a type 2 error. Second, selection bias in the source studies might have influenced the results of the meta-analysis, since stable and simple extra-articular fractures might have been treated nonoperatively, whereas the more dislocated and comminuted fractures might have been treated by volar plating or external fixator. However, current data did not allow for metaregression analyses. Therefore, it is not entirely certain if the current whether the findings are related to the treatment type, or can be explained by a difference in indication. Third, there was high heterogeneity in type of outcome parameters and time points of measurements reported. Goldhahn et al. mentioned already in 2014 that this heterogeneity is problematic in research on distal radius fractures and suggested a set of core domains to assess outcomes as a possible solution [129]. We have not discovered any clear improvement in data consistency since this article has been published. However, the effect of this study can still be expected over the next years due to its recent character. We agree with Goldhahn et al. that minimal requirements of outcome in the domains patient history, physical examination and radiological findings should be established and journals should only publish papers that meet these criteria. Only then the value of future systematic reviews in the field of orthopedic trauma will increase.

Conclusion
Current literature does not provide enough evidence to support superiority of a particular treatment method for extraarticular distal radius fractures, when looking at complications, re-interventions, and long-term functional outcomes. A broad range of outcome parameters have been used, which makes the data partly impossible to compare. From a methodological point of view, the quality of data used in this systematic review and subsequent conclusions that can be drawn, appear to be rather weak. This paper should therefore encourage future investigators to use more sound research methods. Consensus on outcome measures and completeness of reporting is necessary to conduct high-quality studies with standardized outcome assessment. This is needed to be able to draw sound conclusions on superiority of one of the treatment methods.
Acknowledgment We thank Mr. W.M. Bramer, Biomedical information specialist at the Erasmus MC (Rotterdam, The Netherlands) for assistance with the literature search.
Author contributions All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Guido van Oijen, Esther van Lieshout, Maarten Reijnders, and Anand Appalsamy. The first draft of the manuscript was written by Guido van Oijen and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Funding No external funding was received for this study.

Availability of data and material Not applicable.
Code availability Not applicable.

Conflict of interest
The authors declare that they have no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.