Abstract
Purpose The purpose of this study is to perform a citation analysis on the most frequently cited articles in the topic of cervical spine surgery and report on the top 100 most cited publication in this topic.
Methods We used the Thomson Reuters Web of Science to search citations of all articles from 1945 to 2015 relevant to cervical spine surgery and ranked them according to the number of citations. The 100 most cited articles that matched the search criteria were further analyzed by number of citations, first author, journal, year of publication, country and institution of origin.
Results The top 100 cited articles in the topic of cervical spine surgery were published from 1952-2011. The number of citations ranged from 106 times for the 100th paper to 1206 times for the top paper. The decade of 1990-1999 saw the most publications. The Journal of Spine published the most articles, followed by Journal of Bone and Joint Surgery America. Investigators from America authored the most papers and The University of California contributed the most publications. Cervical spine fusion was the most common topic published with 36 papers, followed by surgical technique and trauma.
Conclusion This article identifies the 100 most cited articles in cervical spine surgery. It has provided insight to the history and development in cervical spine surgery and many of which have shaped the way we practice today.
Introduction
Cervical spine surgery owes its advancement to key research which has shaped modern practice. Many papers are published in this area but only a few make a lasting impact to the scientific community.
A citation is recognition of an author’s work by their peers and may acknowledge the impact that their work has. Bibliometric analysis has become a popular way of providing an insight into advancements and highlighting key articles in a certain field. Several studies have evaluated the impact that articles have made in their respective field.1-4 Citation analysis involves evaluating an article based on the number of citations it receives. The list it generates may identify individuals who have contributed greatly to the scientific community.
The purpose of this study is to identify trends and characteristics which make an article highly cited in the field of cervical spine surgery. We performed a citation analysis on the most frequently cited articles in cervical spine surgery using methods validated in other similar published studies.1-4
Methods
The Institute for Scientific information has collected information on journal citations since 1945. Their current system is known as “Science Citation index” and this database was searched using the Thomson Reuters Web of Science for all articles from 1945 to 2015. The search term was “Cervical Spine” and it was limited to the English language. This search was performed in November 2015. The results were then carefully reviewed by two authors (YL and FB) and only those relevant to the cervical spine were selected. The 100 most cited articles that matched the search criteria were further analyzed by number of citations, first author, journal, year of publication, country and institution of origin and level of evidence. The total numbers of citations were compared to the average yearly citation for each article.
Results
The 100 most cited articles in the topic of cervical spine surgery were published from 1952-2011. These papers, their corresponding authors, and number of citations are shown in Table 1. The number of citations ranged from 106 times for the 100th paper to 1206 times for the top paper in a total of 15 journals. The decade of 1990-1999 saw the most publications with 36 articles, followed by 2000-2009 with 33 articles (Table 2). The oldest article was by Bailey5 published in 1952; and Carragee et al.6 published the most recent article in 2011.
The Spine Journal published the most articles (48 articles), followed by the Journal of Bone and Joint Surgery, America. The top two journals published 57% of all articles (Table 3). Investigators from America authored 52 of the 100 articles, followed by authors from Japan with 13 articles and Germany with 7 articles (Table 4). The institution that contributed the most was the University of California with 9 publications in the top 100 (Table 5). A total of 9 authors contributed ≥ 3 publications with Abumi, K and Bohlmann H, having 5 publications in the top 100 (Table 6).
Cervical spine fusion was the most common topic published with 36 articles, followed by articles on surgical technique and trauma (Table 7). When the level of evidence is reviewed, we found that the most common level of evidence was 3 with only two articles being level 1 (Figure 1). We found that only 24 articles were rated as level 1 or 2 with the remainder being rated as 3 and 4 (Table 1).
Comparison was made between the total number of citations with the average number of citations per year. There was a variation in the ranking with 7 articles ranked in the top 10 by total number of citations and also ranked in the top 10 when ranked by average (Table 8).
Discussion
This study identifies the most cited papers on the cervical spine. It provides the features which make an article likely to be highly cited and highlights the journals which may allow work to be cited.
The top most cited paper with 1206 citations is the work by Frankel et al.7 on postural reduction in the management of closed injuries of the spine with paraplegia or tetraplegia. While such treatment method is less common, their work on classifying neurological impairment became the basis of the spinal assessment in spinal cord injury patients as part of the American Spinal Injury Association (ASIA) neurological impairment scale.8 In their initial description, complete and incomplete spinal cord injuries were subdivided into five categories, without detailing assessment of motor or sensory function.7 This was subsequently modified by the ASIA with the addition of key muscle function, key sensory points and presence of sacral sparing added as part of assessment to clarify zones of partial preservation.8 This allows greater precision and consistency in defining the extent of spinal cord injury amongst clinicians and researchers.
The second most cited paper is the work by Vernon and Moir,9 describing the neck disability index (NDI), which became the first instrument designed to assess self-rated disability in patients with neck pain. Vernon modelled the NDI on the Oswestry Low Back Pain Disability Questionnaire with the permission of the authors.10 Since 1991, the NDI has become the most widely used questionnaire for neck pain, with 959 citations, translation into over 20 languages, and a widely used outcome measure in clinical studies.11 The NDI has been useful for prognostication of outcome following whiplash injury, with several studies reporting a low initial NDI predicting recovery and a high score predicting chronicity.11 In addition, high NDI score (>15/50) at 3-36 months following injury strongly correlates with physiological dysfunction such as muscular dysfunction and central sensitization.11
The third most cited paper is by Smith and Robinson,12 who first describe the anterior approach to the cervical spine in the treatment of symptomatic cervical spondylosis. The authors suggested that disc degeneration, osteophyte formation and foraminal stenosis resulted in brachalgia. In their series of 14 patients who underwent anterior approach and interbody fusion with autologous iliac crest graft they reported symptomatic improvement.12 In their subsequent article on 55 patients with longer follow-up, which is also one of the top articles, Robinson continued to report good success with the anterior interbody fusion of cervical spine with low complication rates.13 This technique has since become the gold standard for the treatment of cervical spondylosis and cervical myelopathy.
In contrast to previous studies where 1980s was the most prolific decade for articles,3 the decades of 1990s saw the most publications in cervical spine surgery. This is similar to papers published in Foot and ankle surgery,4 sports medicine,111 shoulder surgery112 and spine surgery.113 These two decades coincided with the development of improved instrumentation and surgical technique for managing cervical instability and reflect the relatively young field of spinal surgery. In 1992, Jeanneret and Magerl14 (ranked 38) described the trans-articular atlantoaxial screw fixation technique for treating unstable odontoid fractures, which was alternate to the then more popular Gallie15 or Brooks16 type posterior fusion technique which lacked primary stability for treatment of unstable C2 peg fractures. In 2001, Harms and Melcher 17 (ranked 5) described another technique for treating the atlantoaxial joint instability with posterior atlas and axis screw-rod fixation and fusion. Both the Magerl and Harms methods remained popular techniques to date, demonstrating high fusion rate for treating C2 fractures. The decade of 2000-2009 saw an improvement in cervical disc arthroplasty design and increasing use for treating cervical myelopathy or radiculopathy. Goffin et al.18 in 2002 (ranked 45) described their preliminary experience with the Bryan Disc Replacement, reporting on the safety of the device and favorable results at short term follow-up. Proponents of cervical disc arthroplasty argue that maintenance of motion avoids adjacent segment degeneration. More than 15 designs of cervical disc arthroplasty are undergoing ongoing clinical trails to determine its efficacy.19
Cervical fusion was the most popular topic with 36 papers featured in the 100 articles we reviewed. Anterior fusion was the most commonly described technique and is currently the gold standard for treating mid to lower cervical spine disorder ranging from fracture-dislocation, instability and degenerative disc disease. Popularized by Smith and Robinson,12 the technique saw a trend to improve fusion rates through grafting which is the 5th most common topic. Bishop et al.20 (ranked 76) use of allograft and An et al.21 (ranked 66) used allograft plus demineralized bone matrix for anterior cervical fusion but found this not to be superior, with high rates of pseudoarthrosis and graft collapse. More recent interest has been the use of BMP and it’s association with higher complication rates. Interestingly complications as a subject ranked 4th in this study. Published in 2006 and ranked 20, Shields et al.22 reported the adverse effects associated with high-dose recombinant human bone morphogenetic protein-2 use in anterior cervical spine fusion. This was one of the first studies that reported on the adverse events associated with rhBMP-2 in cervical spine surgery, including life threatening events such as delayed haematoma, soft tissue swelling and airway compromise. This is contrary to initial industry sponsored rhBMP-2 on its use reporting a 0% adverse event rate.6
Classification systems and outcome tools have been found in other citation analysis to be prominent features. The top two articles in this analysis are describing classification systems and outcome tools respectively. This shows the importance to have a well validated outcome tool in clinical research as it allows the effect of a treatment to be assessed between studies and allows direct comparison of one technique with another.
The most popular journal for publication was the Spine Journal. This highlights its role as a journal for both neurosurgical and orthopaedic spinal surgeons to read. Nearly two thirds of the 100 most cited articles were published in just two journals. The journal of Bone and Joint Surgery, America was the second most popular destination for research. These two journals both have high impact and highlight the need to publish work in a high impact journal.
The United States was the country with the most number of articles included in the 100 articles we reviewed. This is similar to results from many previous bibliometric anaylsis papers.3,4,111,112,113 Japan featured as the second most prolific nation and this relates to their role in the development and advancement of the laminoplasty technique for the management of cervical cord compression.
All the papers which included randomized trials were published since the year 2000, highlighting an improvement in the quality of research being performed. The papers which were of lower level of evidence all tended to be published earlier. This shows that the quality of research being performed has improved as better understanding of clinical research is developed. The most common level of evidence were 3 and 4 which are similar to other orthopedic bibliometric analysis papers.111,112,113 This highlights the difficulty that we have in surgical specialties in performing good randomized control trials.
A similar bibliometric analysis has been published on the cervical spine by Ruegsegger et al.114 They performed a bibliometric analysis using the same database but a different key word search. This we believe is why our independent study results differ to theirs. Their method, we feel, will have resulted in the exclusion of some key articles we included.
Previous papers on bibliometric analysis have restricted their analysis to the journals with higher impact factor. We did not restrict our search by this, as it may create a list which excludes key articles. As a result, we feel our citation analysis is more powerful. Some question the use of citation analysis as a scientific tool. We feel these lists are beneficial for junior clinicians and researchers alike as it highlights key features of research and provides a historic account of the developments. Whilst some articles will be highly cited due to their controversial nature however we feel that these are likely to be few compared with those that advance the specialty. Previous papers have highlighted the unfair advantage that ranking by total citations can give an article which has been published for a long time. This can be seen in our comparison of articles by number of citations per year. Carragee et al.6 article has an average citation of 76.6 per year compared with Frankel et al.,7,25,68. We see articles on cervical disc arthroplasty advancing significantly when we rank by average per year. This highlight that these papers are important now and are likely to feature in bibliometric analysis in the future.
Conclusion
This article shows the journals which are likely to provide an author with high levels of readership and potentially high citation. It provides an insight to the development of modern cervical spine surgery. With continued development of surgical implant technology and research methods the features of this list will change to reflect these newer techniques and approaches. This article provides a current insight into the most cited cervical spine papers.
Disclosures & COI
The authors declare no relevant financial disclosures.
- Copyright © 2017 ISASS - This manuscript is generously published free of charge by ISASS, the International Society for the Advancement of Spine Surgery