Level of evidence in wrist ligament repair and reconstruction research: a systematic review
https://doi.org/10.1186/s40634-018-0135-7
© The Author(s). 2018
Received: 8 March 2018
Accepted: 29 May 2018
Published: 7 June 2018
Abstract
There have been numerous studies on surgery of wrist ligament injuries, but a quick assessment reveals few with a high level of evidence (LoE). The primary aim of this study was to categorize the study type and LoE of studies on repair and reconstruction of the scapholunate ligament, the lunotriquetral ligament and the triangular fibrocartilage complex by applying the LoE rating system proposed by the Oxford Centre for Evidence-Based Medicine. The secondary aims were to evaluate the journal- and geographic- distribution of the included studies.
An electronic literature search of articles published 1985–2016, in PubMed, Embase, and Cochrane Library was carried out in May 2016 and updated in April 2017. Therapeutic studies written in English were included. The PRISMA checklist guided the extraction and reporting of data.
A total of 1889 studies were analyzed, of which 362 were included. Three journals represented 40% of the included studies and American authors dominated.
Most studies (97%) had low LoE (IV-V). No studies of LoE I-II were found. There is insufficient evidence to recommend one technique over the other in terms of wrist ligament surgery in clinical practice. There is an immense lack of comparison studies with high level of evidence in the area of wrist ligament repair and reconstruction.
Keywords
Introduction
Wrist ligament injuries are known to result in persistent pain, instability, reduced grip force and range of motion, as well as degenerative osteoarthritis. Ligament repair and reconstruction has become a highly technical developing area in wrist surgery in the last years. There are still controversies in the recommendations on surgical preferences in terms of injuries to the scapholunate ligament (SLL), the lunotriquetral ligament (LTL), the triangular fibrocartilage complex (TFCC) and instability of the distal radioulnar joint (DRUJ).
Levels of evidence for primary research question on therapeutic studies
Level I | High-quality RCT with statistically significant difference or no statistically significant difference but narrow confidence intervals. | Systematic review of level I RCTs (and study results were homogenous) |
Level II | Lesser-quality RCT (e.g. 80% follow up, no blinding, or improper randomization). Prospective comparative study. | Systematic review of level II studies or level I studies with inconsistent results. |
Level III | Case-control study. Retrospective comparative study. | |
Level IV | Case series. | |
Level V | Expert opinion. Case report. Technical report. |
Considering the high number of studies, a quick assessment of the studies on wrist ligament repair and reconstruction reveals very few with a high level of evidence. Still, recommendations and guidelines in surgical treatment of SLL (Garcia-Elias et al., 2006; Kitay & Wolfe, 2012), LTL (van de Grift & Ritt, 2016) and TFCC/DRUJ injuries (Anderson et al., 2008; Zimmerman & Jupiter, 2014) are widely spread, as well as expert opinions. However, some questions are indicated. The most important of those are; How strong are in fact our treatment recommendations for patients with wrist ligament injuries? Up to this date, the patient’s needs and the surgeon’s specific skills seem to have had a great impact on choosing type of surgical treatment for SLL, LTL and TFCC/DRUJ injuries.
The recommendations in terms of diagnostics of wrist ligament injury are clear. Wrist arthroscopy is the preferred diagnostic technique with sufficient conclusive properties when it comes to wrist ligament injuries (Andersson et al., 2015; Hobby et al., 2001). In terms of surgical treatment of SLL, LTL and TFCC/DRUJ injuries, there are some strong recommendations but still many controversies – therefore this study attempts to analyse on what basis the therapeutic recommendations rests.
The primary aim of this study was to categorize the study type and level of evidence of studies on repair (suture and reinsertion) and reconstruction of SLL, LTL and TFCC/DRUJ injuries by applying the level of evidence rating system proposed by the Oxford Centre for Evidence-Based Medicine. The secondary aims were to evaluate the journal distribution and geographic distribution of the included studies and to evaluate the level of evidence over time.
Methods
Protocols
This systematic review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement (Moher et al., 2009) and follows The Cochrane Handbook for systematic Reviews of Interventions.
Eligibility criteria
Therapeutic studies that report on wrist ligament (SLL, LTL and TFCC/DRUJ) and instability repair (suture and re-insertion) and reconstruction with clinical outcome measurements related to the ligament repair or reconstruction were included. Articles with mixed cohorts with some patients operated on with ligament repair and others operated on with other methods were included. Expert opinions, case reports and technical reports on surgery for wrist ligament injuries and instability were also included. Studies on anatomical reconstruction of the TFCC and DRUJ (Adams & Berger, 2002), other ligament reconstructions and debridement plus pinning/shrinkage were included.
Studies on animals, cadavers, diagnostic tools, economics, epidemiology, imaging, diagnostic or anatomic studies without clinical outcome; rehabilitation protocols and studies on skeletally immature populations were excluded. Studies on perilunate dislocations, midcarpal instability, ulnar shortening osteotomy, bone corrective surgery, capsulodesis without concomitant ligament repair, pinning only, arthroscopic or open debridement alone without repair or shrinkage, interosseous membrane (IOM) reconstruction, salvage procedures (4-Cornerfusion, Proximal Row Carpectomy etc), partial fusions and implant arthroplasty were excluded. The purpose of the study was to analyse the level of evidence of studies on repair (suture and reinsertion) and reconstruction exclusively. Studies not written in English, comments, authors´ reply, letters to the editor, book chapters and instructional courses were also excluded.
No outcome measures were extracted from the studies, as this was not the aim of this systematic review.
Information sources and search
An electronic literature search of articles published between January 1, 1985 and May 24, 2016, in PubMed, Embase, and Cochrane Library was carried out in the end of May 2016, by an expert in electronic search strategies, at the Sahlgrenska University Hospital Library. Therapeutic studies written in English that report on wrist ligament repair and reconstruction with clinical outcome measurements related to the ligament repair or reconstruction were included, as well as LoE V publications on wrist ligament repair (suture and reinsertion) and reconstruction. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) (Moher et al., 2009) checklist guided the extraction and reporting of data. Categorization and implementation of the level of evidence and journal distribution were performed. The systematic electronic search was updated on April 28, 2017, in order to identify newly published studies that were eligible for inclusion.
No manual search of recently published articles in pertinent journals was undertaken. Corresponding authors were not contacted for additional information. The complete electronic search strategies (initial and updated) are described in the Additional file 1.
The impact factors of the journals were acquired from the Thomson Reuters Journal Citation Reports (JCR) database (ISI Web of Knowledge, http://www.webofknowledge.com). The time span in which the impact factors were able to be acquired was from 2008 to 2015/16. The Eigenfactor (Bergstrom et al., 2008) of the journals were acquired from http://www.eigenfactor.org. The time span in which the Eigenfactors were able to be acquired was from 1997 to 2013.
Study selection
The first (JKA) and second (BR) author performed the study selection. All articles, generated by the electronic search, were screened by reading the title and abstract. The articles were validated in duplicate. The first author double-checked the extracted data by processing the included studies once again, with full agreement with the prior data selection and extraction.
If initial screening failed to provide sufficient information for the purpose of inclusion or exclusion, the full text of the article was always assessed by paper copies. In terms of 35 studies, it was necessary to retrieve full text to be able to decide about inclusion or exclusion. The investigator was not blinded to the names of authors or journals during the screening and data-extraction process.
Data extraction and data items
Data extraction
Data extraction was performed according to the PRISMA (Moher et al., 2009) checklist by obtaining data on a standardized extraction sheet. Data extraction sheets with pre-determined questions were used.
Disagreements on study selection, data extraction, and assessment were resolved by discussion with the third author (JK).
Data items
Data items obtained from the included articles were as follows: participants, interventions, comparisons, outcomes, study design and setting (PICOS), publication year, authors, sample size, follow up time, journal, the journal impact factor, Eigenfactor and the level of evidence (LoE). In terms of eight papers, it was necessary to retrieve full text to be able to decide the level of evidence, according to the level of evidence rating system proposed by the Oxford Centre for Evidence-Based Medicine.
The level of evidence attributed to the study by the publishing journal was taken into account when assessing the study. However, when there was disagreement between the level of evidence as assessed by the publishing journal and the level of evidence as assessed by the researchers, the assessment by the researchers was used. Standardized, studies with ≤ four patients, were classified as case reports (LoE V) and studies with ≥8 patients, were classified as a case series (LoE IV).
Eigenfactor (Bergstrom et al., 2008) score is correlated with the impact factor and the total citation count for medical journals, but these metrics provide substantially different information. For a given number of citations, citations from more significant journals will result in a higher Eigenfactor score. To have the ability to analyse the importance of journals choosen for publications of studies in terms of surgery for wrist ligament injuries, we applied the Eigenfactor to the included studies. Originally Eigenfactor scores were measures of a journal’s importance; it has been extended to author-level. It can also be used in combination with the so-called h-index (Hirsch, 2005) to evaluate the work of individual scientists. However, we found that frequency tables for geographic distribution, combined with the level of evidence, was more illustrative in the present study.
Results
Study characteristics
Flow diagram of inclusion and exclusion (initial search May 24, 2016)
Outcome measures
Level of evidence
Distribution (%) of the level of evidence of the included studies
In general the included studies included small sample sizes (in average 23 patients) and rather short follow up (in average 37 months) – only 2% displayed long-term follow up (> 10 years) and 7% mid-term follow up (6–8 years). 91% of the included studies displayed short-term follow up (2 years). Over 80% of the LoE IV studies were retrospective.
Distribution of publications
Distribution of publications in journals (percentage)
Distribution of publications in journals (numbers, percentage) and their impact factors
Journal | Impact factor (2015) | Number (Total n = 362) | % |
|---|---|---|---|
J Hand Surg Am (JHSA) | 1.64 | 72 | 19.9 |
Hand Clin | 0.70 | 46 | 12.7 |
J Hand Surg Eur/Br (JHSE) | 1.87 | 28 | 7.7 |
Arthroscopy (A-scopy) | 3.72 | 21 | 5.8 |
Tech Hand Up Extrem Surg (THUES) | 1.36 (2010) | 20 | 5.5 |
J Wrist Surg (JWS) | No data | 17 | 4.7 |
Subtotal | 204 | 56.4 | |
Others | 158 | 43.6 |
Trends
Only 21.5% (n = 78) of the studies (66 authors) were published before 2000, in which 0% represented LoE I and II studies, 1.4% (n = 1) represented LoE III studies, 31.9% (n = 23) represented LoE IV studies and 66.7% (n = 48) represented LoE V. Almost four fifthts of the included studies were published between 2000 and 2016. There was no trend toward higher level of evidence over time. The updated electronic search performed on April 28, 2017 did not generate any new studies with high level of evidence (LoE I-II) eligible for inclusion.
The distribution (n) of the included studies on different wrist ligaments injuries and injury combinations. (SL = scapholunate ligament, LT = Lunotriquetral ligament, TFCC = Triangular fibrocartilage complex, DRUJ = Distal radioulnar joint)
a and b. The Impact factor and Eigenfactor for the most common journals. No data were available for Tech Hand Up Extrem Surg and J Wrist Surg
In general the Impact factor was relatively low, slightly increasing over years. The Eigenfactor followed the Impact factor.
Geographic distribution
Frequency table for geographic distribution combined with the level of evidence for the six most common journals
Table 3 Journal | Country | LoE III | LoE IV | LoE V | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
United States (n - %) | Japan (n) | Italy/ Spain/ France (n) | United Kingdom (n) | Switzerland/ Germany/ Belgium/ Netherlands (n) | Other Europe (n) | Other Asia etc (n) | ||||
(n-%) | (n-%) | (n-%) | ||||||||
J Hand Surg Am (n = 72) | 47 | 6 | 4 | 1 | 7 | 2 | 5 | 5 | 33 | 34 |
65% | 7% | 46% | 47% | |||||||
Hand Clin (n = 46) | 40 | 1 | 2 | 0 | 1 | 1 | 1 | 0 | 5 | 41 |
87% | 11% | 89% | ||||||||
J Hand Surg Br/Eur (n = 28) | 8 | 3 | 8 | 2 | 4 | 1 | 2 | 1 | 18 | 9 |
29% | 4% | 64% | 32% | |||||||
Arthro-scopy (n = 21) | 10 | 1 | 2 | 0 | 2 | 1 | 5 | 0 | 11 | 10 |
48% | 52% | 48% | ||||||||
Tech Hand Up Extrem Surg (n = 20) | 11 | 3 | 1 | 0 | 1 | 0 | 4 | 0 | 5 | 15 |
55% | 25% | 75% | ||||||||
J Wrist Surg (n = 17) | 4 | 3 | 4 | 0 | 3 | 1 | 2 | 0 | 14 | 3 |
24% | 82% | 18% | ||||||||
Discussion
To our knowledge, this is the first systematic review on categorizing the study type and level of evidence of studies on repair and reconstruction of wrist ligament injuries.
Level of evidence
This systematic review shows that most therapeutic studies on repair and reconstruction of wrist ligament injuries were of low level of evidence (LoE IV-V). Analysis of the study types revealed that expert opinions, level of evidence V (55.9%, n = 203) was the most frequent study type and no studies of LoE I-II were found between 1985 and April 2017.
In the present study, in general, the Impact factor was relatively low, slightly increasing over years. The Eigenfactor followed the Impact factor. Eigenfactor (Bergstrom et al., 2008) score is correlated with the impact factor and the total citation count for medical journals, but these metrics provide significantly different information. For a given number of citations, citations from more significant journals will result in a higher Eigenfactor score. The result in the present study indicates that citations were given in similar journal that publications about wrist ligament surgery are published in.
Distribution
Frequency tables for geographic distribution revealed that the United States was the country with the highest number of publications, especially in terms of the two most common journals (J Hand Surg Am – 65% and Hand Clin – 87%). Frequently recurrent authors were common. Studies with LoE V were particularly common in Hand Clin (89%) and Tech Hand Up Extrem Surg (75%).
J Hand Surg Br/Eur, J Wrist Surg and to some extent also Arthroscopy displayed a more diversified geographic frequency distribution and a relatively lower rate of LoE V studies.
To be able to recommend surgery and what kind of procedure in specific cases, the hand surgery community needs larger studies with a randomized prospective assessment. A more diversified geographic distribution in some journals could also be desirable and worthwhile.
Up to this date, the patients´ needs and the surgeon’s specific skills seem to have had a vast impact in choosing type of surgical treatment for SLL, LTL and TFCC/DRUJ injuries (Garcia-Elias et al., 2006; Kitay & Wolfe, 2012; van de Grift & Ritt, 2016; Anderson et al., 2008; Zimmerman & Jupiter, 2014). A surgical method that works in the hands of one surgeon may not work in the hands of others. Therefore studies at a high level of evidence are needed to evaluate different therapeutic methods in terms of wrist ligament injuries. Every clinician has to continuously critically review and re-evaluate his decision of surgical options, especially as we conclude that the level of evidence - in terms of our common recommendations - are really low.
Larger prospective randomized controlled trials (RCTs) are needed to evaluate the outcome of wrist ligament surgery, in terms of minimizing bias and confounding factors. Multi-center studies or non-inferiority studies could also be an option, saving time and costs in the study setting, compared to RCTs.
The literature in terms of wrist ligament repair and reconstruction area has increased dramatically over the last 15 years. Almost 80% of the included studies were published between 2000 and 2016. This systematic review showed that there was no clear trend toward higher level of evidence over time.
As 56.5% of the studies were published in six of the 74 indexed registered orthopaedic journals - and 40% in three of those journals - this could indicate a risk of publication bias.
Limitations
The search was limited to papers written in English indexed in PubMed, Embase, and Cochrane library. Studies in other languages and cited in other databases were therefore not included in this review. The data extraction was not performed in a blinded fashion, that is, by blacking out authors, title, and so on.
To our knowledge, this is the first systematic review on categorizing the study type and level of evidence of studies on repair and reconstruction of SLL, LTL and TFCC/DRUJ injuries by applying the level of evidence rating system proposed by the Oxford Centre for Evidence-Based Medicine with this extensive electronic search. This extensive search, of course, creates a large number of abstracts to analyze, which creates the possibility that some studies were classified into the wrong category. However, our results are so clear-cut, that wrong categorization in some few cases, with the highest probability, would not affect our conclusion.
Conclusion
The absolute majority of the therapeutic studies on repair and reconstruction of wrist ligament injuries were of low level of evidence (LoE IV-V; 97%, n = 351 (of 362)).
No studies of LoE I-II were found between 1985 and April 2017. There was no clear trend toward higher level of evidence over time. Prospective randomized controlled trials, multi-center studies or non-inferiority studies are needed to better evaluate the outcome of wrist ligament surgery. There is insufficient evidence to recommend one technique over the other in terms of wrist ligament surgery in clinical practice. There is an immense lack of comparison studies with high level of evidence in the area of wrist ligament repair and reconstruction.
Declarations
Acknowledgments
The authors would like to thank Therese Svanberg, librarian at the Sahlgrenska University Hospital Library for help with electronic search strategies and Professor Kristian Samuelsson, MD, PhD, for valuable advices in terms of the study setting.
Funding
This investigation conforms with the University of Gothenburg Human Research Protection Programme guidelines. Funding for this project was provided by Sahlgrenska University Hospital and the University of Gothenburg, Sweden.
This research received no other specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Authors´ contributions
All authors read and approved the final manuscript.
Competing interests
The authors declare that they have no competing interests.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.
Authors’ Affiliations
References
- Adams BD, Berger RA (2002) An anatomic reconstruction of the distal radioulnar ligaments for posttraumatic distal radioulnar joint instability. J Hand Surg Am 27:243–251View ArticlePubMedGoogle Scholar
- Anderson ML, Larson AN, Moran SL, Cooney WP, Amrami KK, Berger RA (2008) Clinical comparison of arthroscopic versus open repair of triangular fibrocartilage complex tears. J Hand Surg Am 33:675–682View ArticlePubMedGoogle Scholar
- Andersson JK, Andernord D, Karlsson J, Fridén J (2015) Efficacy of magnetic resonance imaging and clinical tests in diagnostics of wrist ligament injuries: a systematic review. Arthroscopy 31:2014–2020View ArticlePubMedGoogle Scholar
- Bergstrom CT, West JD, Wiseman MA (2008) The Eigenfactor™ metrics. J Neuroscience 28:11433–11434View ArticleGoogle Scholar
- Garcia-Elias M, Lluch AL, Stanley JK (2006) Three-ligament tenodesis for the treatment of scapholunate dissociation: indications and surgical technique. J Hand Surg Am 31:125–134View ArticlePubMedGoogle Scholar
- Hirsch JE (2005) An index to quantify an individual's scientific research output. Proc Natl Acad Sci U S A 102:16569–16572View ArticlePubMedPubMed CentralGoogle Scholar
- Hobby JL, Tom BD, Bearcroft PW, Dixon AK (2001) Magnetic resonance imaging of the wrist: diagnostic performance statistics. Clin Radiol 56:50–57View ArticlePubMedGoogle Scholar
- Kitay A, Wolfe SW (2012) Scapholunate instability: current concepts in diagnosis and management. J Hand Surg Am 37:2175–2196View ArticlePubMedGoogle Scholar
- Moher D, Liberati A, Tetzlaff J, Altman DG (2009) Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. J Clin Epidemiol 62:1006–1012View ArticlePubMedGoogle Scholar
- Sackett DL (1986) Rules of evidence and clinical recommendations on the use of antithrombotic agents. Chest 89(2 Suppl):2S–3SView ArticlePubMedGoogle Scholar
- van de Grift TC, Ritt MJ (2016) Management of lunotriquetral instability: a review of the literature. J Hand Surg Eur 41:72–85View ArticleGoogle Scholar
- Zimmerman RM, Jupiter JB (2014) Instability of the distal radioulnar joint. J Hand Surg Eur 39:727–738View ArticleGoogle Scholar
