The management of spinal pathology has developed significantly in recent decades, an evolution perhaps best exemplified by the advances in care of patients with cervical myelopathy (CM). The field has been informed via publications from a broad variety of clinicians, institutions, and study designs that have collectively shaped our current understanding of this condition. From the pathophysiology and epidemiology to expected outcomes of surgical and nonsurgical management, understanding of this condition and the standards of care are vastly different from those employed just half a century ago. A thorough understanding of prior studies may help clinicians in their development of subsequent trials.
Citation analysis is a unique tool that has long been utilized in medical literature to quantify the effects that publications have on their respective fields (1). While direct impact on clinical practice is difficult to accurately quantify, reference of an article by other scholarly papers has been used as a proxy for data distribution and incorporation into academic discussion of a topic. Specifically, spine surgery has benefited from many bibliometric reviews that help highlight literature trends and deficiencies, as well as centers that are advancing the field (2-10). Additionally, these analyses can be used by those entering a particular field as an encompassing resource to efficiently identify some of the most discussed and shared publications. It can also be used as an imperfect gauge to measure the relative influence of publications as well as journals.
We sought to collect and review the 50 most cited articles discussing the management of CM to quantify the factors that resulted in a high degree of influence and to highlight those that have advanced the field. By describing the chronology of the field’s most significant developments, the demographics of the international cohort of contributing authors, and the quality of data published, we present a novel context with which to consider the current state of the art and to examine future advances in care for patients with CM.
Clarivate Analytics Web of Science, formerly known as Thomson Reuters Web of Science, was used to conduct a search for articles pertaining to CM. Web of Science is a search database that provides detailed citation information. Queries were conducted using varied Boolean search terms as of May 2018. The broadest search returning the greatest number of results was carried forward; the Boolean search phrases of the final search included the following: [(myelopathy OR cervical spondylotic myelopathy OR spondylosis OR laminoplasty OR JOA OR mJOA OR Nurick OR OPLL OR “ossification of the posterior longitudinal ligament”) AND (spine OR cervical OR subaxial OR C-spine OR neck)].
The search was not restricted to any particular language or data range. Articles from both medical and non-medical journals were included in the initial screen. The search results were refined to include only peer-reviewed articles, review papers, and proceedings papers. The resulting articles were sorted by descending number of citations. For inclusion, publications must have pertained to compression of the spinal cord in a medically relevant discussion in an area such as causes, diagnosis, associated signs and symptoms, treatments, anatomy, physiology, or mechanics. The articles were then screened by title and abstract with the goal of isolating 50 articles related to CM. These articles were additionally screened and verified by the first author (CJD), removing articles that did not pertain to CM specifically. For instance, an article related to cancers of the spine that briefly discusses CM would not be included.
To determine the order for articles with equal number of overall citations, the number of citations in the most recent year was used as a tiebreaker. Level of evidence (LOE), article type, and institution data were manually recorded for each article after reviewing each paper. LOE was determined based on the Oxford evidence-based medicine levels of evidence table. Categories for article types were determined from individual reviews of each manuscript. All supporting institutions for each article were recorded.
The final 50 articles were further analyzed via Web of Science for the following attributes: year of publication, country of origin, journal of publication, article type, total number of citations, average citations per year, citations in 2017, authorship frequency, LOE, and contributing institutions. A comparison of the Web of Science analyses, the Web of Science exported citation summary, and manually recorded data from web searches of the individual articles was used for final analysis.
As a secondary objective we wanted to highlight the more recent publications that may have not made the initial list due to the limited time of publication. To illustrate the more recent influential publications we used the same above parameters, but we restricted our search to articles from the past 10 years. We then ranked results by frequency of citation per year, as opposed to total citations.
The initial search yielded 8,249 preliminary results, of which we needed to review the first 3,000 to eventually identify 50 publications pertaining to the CM criteria (Figure 1). This list of the 50 most cited articles were published from 1952 to 2011. The top ranked paper was cited 776 times; the 50th ranked paper was cited 130 times (Table 1). Of these 50 articles, the average total number of citations per publication was 193. The average number of citations per year since the year of article publication was 9.2. In 2017 the mean number of citations for these 50 articles was 15.5.
The earliest publication within this list is the Brain et al. paper from 1952, which reviews prior papers that documented the pathology of cervical spondylosis. Interestingly, articles from the first half of the 20th century were included in the original search, but zero articles from this time period qualified for the final list.
The most prolific time periods were the 1990s (34%) and 2000s (42%). The most prolific single year of publication was tied between 2001 and 2003, each of which yielded 6 papers (12%). No articles were published in the 1970s. A positive correlation between decade and number of articles was noted between the 1950s and the 2000s (Figure 2). Japan and the United States were the two most prolific countries contributing to these publications (46% and 42%, respectively). Interestingly, eight different countries contributed to the articles within the final list (Figure 3).
There were 10 different journals represented within this list, with Spine accounting for 26 (52%) of the publications (Table 2). The second was the J Neurosurg, which accounted for 10 (20%) publications. More than half of the papers had a LOE of III. The numbers of papers for LOE I, II, III, IV, and V was 2 (4%), 0 (0%), 30 (60%), 16 (32%), and 2 (4%), respectively (Figure 4). When indicated, a single article may contain more than one LOE.
The majority of publications were associated within the article type called clinical outcomes (n=28; 57%). An additional 9 articles (18%) were assigned the article type called Imaging. The remaining publications were classified as follows: surgical technique (n=4), general review (n=4), complications (n=3), questionnaire development (n=1), and anatomic (n=1). Interestingly, 25 authors contributed to at least 2 papers within the list, and 9 authors contributed to at least 3 papers within the list. The most prolific authors were Yonenobu K (n=9; 18%) and Ono K (n=6; 12%) (Table 3).
Additionally, in terms of institutional contribution, 6 institutions contributed to 3 or more articles on this list. Osaka University (Japan) was the institution with the greatest number of articles, contributing to 10 articles (20%). Five institutions were tied for the second greatest number of articles, each contributing to 3 articles (Table 4).
In terms of Clarivate Analytics Web of Science publication categories, 43 articles (86%) were classified as neurosciences/neurology papers while 34 articles were classified as orthopedics papers (68%). When indicated, a single article can be placed into more than one publication categories. Each of the top 50 publications were presented in English.
As a secondary objective we wanted to identify influential articles published within the past 10 years. We highlighted the 5 most frequently cited publications since 2008, based on number of citations per year as opposed to ranking by total number of citations to gather the Top 50 mentioned above. Interestingly, we found that the authors either Ames and Fehlings were 1st or last author on for the 5 most influential articles from the past 10 years (Table 5).
By performing a citation analysis of “Cervical Myelopathy” we aimed to identify the articles that have been most influential on the field. While the total number of article citations does not necessarily indicate the publication quality or active practice utilization, it does address readership and overall influence based on recognition by peers in the field (11-15). Further analysis of the qualities of each of these articles promotes a better understanding of factors associated with article importance and understanding of the demographics related to the study and management of CM.
In our analysis, we found that the most cited article pertaining to CM is the 1999 article by Hilibrand et al. (774 citations) identifying the incidence, prevalence, and progression of adjacent segment disease following anterior cervical fusion (16). Interestingly, this article also has the highest citation rate per year (38.7) as well as the highest citation rate in 2017  indicating its lasting influence in this field.
The second most cited article was the 1983 article by Hirabayashi et al. (371 citations) which introduced the open door laminoplasty as a treatment options for CM secondary to ossification of the posterior longitudinal ligament (17). Similarly, this sentinel article continues to be highly relevant with a yearly citation rate of 10.31 and 26 citations in 2017.
Evaluation of author affiliations revealed Japan and specifically Osaka University as some of the most prolific locations for CM publications. This high prevalence of CM research output by Japan is likely related to the higher incidence of OPLL within Asian population as well as the aging Japanese society, both of which predispose a patient to CM (17-19). The Japanese authors Yonenobu, Ono, and Hosona were some of the most cited. It should be noted that our methods counted an author’s contribution regardless of author order, which varied from many prior bibliometric studies that only counted the first and/or last author (7,20). This is important to highlight because Yonenobu, Ono, and Hosona practice from the same institutions and so have regularly published together with differing primary, secondary or senior author status. Their many contributions to the field distinguishes them as leaders in CM management.
By evaluating all contributing authors and institutions as opposed to simply counting the first or last author, we were able to more effectively which groups are the leading contributors to the field. Additionally, our novel methods limit potential bias of overlooking second and third authors who may have contributed in a similar capacity to the first or last author in multicenter or review studies. A critic of analyzing authorship contributions regardless of author order in the final analysis may note that this approach could result in an inflated overall impact of a given individual to the field. An association with a high-output institution may predispose individual authors to a higher rate of publications if they are included as a reviewer of colleagues’ work. In the absence of weighting or biasing author order, the potentially lesser-impact contributions of second, third and fourth authors are interpreted as equal in our analysis to those who were deemed appropriate for a first or senior author position.
There were no articles from 2012 onward, which might be expected given the nature of this analysis being based subsequent research citing prior publications. An inherent flaw in a bibliometric analysis model is that there is a time delay regard to articles gaining citations which creates a disadvantage for more recently published articles (21). To that end, this review is also unique in that we attempt to account for this selection bias through an evaluation of the more recent publications (Table 5). We limit our selection bias by evaluating the most influential articles from the past 10 years by number of citations per year. We identified the five most influential articles and interestingly noted strong similarities. Primarily, either Ames or Fehlings was the first or last author. Secondarily, 3 of these 5 recent articles pertained to cervical spine sagittal alignment. It is important for spine surgeons at every level to be familiar with the newer publications as more emphasis is placed on proper techniques, indications and radiographic parameters that were not as appreciated more than a decade ago.
The most recent article from the initial top 50 CM list (#24) is from 2011 by Coric et al. (22). This article presents the preliminary outcome results of a randomized control trial of cervical disc replacement (CDR) versus a traditional anterior cervical discetcomy and fusion (ACDF), indicating improved outcomes in the CDR group. This article has the second highest citation rate (21 citations per year) and fourth highest number of citations in 2017 (n=31) which explains the heightened position of such a relatively recent article on the list of top 50. One should expect some of the most influential publications in the future to pertain to CDR as the technology improves and follow-up time lengthens.
As opposed to less expansive bibliometric reviews, our analysis and initial search was not restricted to a preset list of certain journals. Altogether, 10 journals were represented from the 50 publications with 36 of the 50 articles identified (72%) represented by two journals, Spine and J Neurosurg. Spine accounted the majority of publications which is consistent with 2 previous bibliometric studies analyzing cervical spine surgery (7,20). Moreover, we noted that the overwhelming majority of articles pertained to clinical outcomes, which would be expected as their results would be supported or challenged in subsequent citing papers. Similarly, 18% of the top 50 papers (based on total citations) pertained to Imaging, just as 3 of the 5 more recent influential CM papers (based on citations per year) also pertained to imaging. Likely there will continue to be publications that authors deem influential pertaining to imaging as our technology advances and the community develops a further appreciation of spine balancing.
The average LOE was III, which differs from a previous bibliometric study of all “cervical spine surgery” demonstrating a plurality of level 4 evidence (7). That bibliometric analysis by Rüegsegger et al. also noted increasing levels of evidence over time, a trend that has also be noted in orthopaedic sport literature (7,23). The difference in average LOE represented by our study is likely attributed to the low number of articles prior to 1980 within our study. Interestingly, we found the most prolific decade of publication of CM articles was the 2000s. This trend mirrors that of many other fields that have highlighted a rise in publication rates possibly due to due the digital ease of manuscript drafting, increase in number of journals, emphasis on publication for career development, and increase in author collaboration (24-27).
Additionally, an overwhelming majority (82%) of the listed studies relevant to CM are concerned specifically with surgical management. Surgical treatment publications, especially those specifically pertaining to anterior cervical fusion (34%) and laminoplasty (20%), have had the long lasting impact on the field of CM. Crandall’s description of CM in 1966 and Tsuyama’s 1984 review of its pathogenesis largely stand alone as widely cited publications that are not primarily written to describe, report or compare surgical procedures. This scarcity of literature may result from a historical lack of consensus regarding optimal management, and a severity of clinical symptoms that demand surgical intervention of some sort. As the condition has long been understood to be primarily structural, the many factors impacting surgical intervention and outcomes have likely been of more pressing concern than academic curiosity regarding its pathogenesis (28). Interestingly, the most cited publication of the past 10 years is Nouri’s 2015 review of advances in understanding of the pathophysiology and natural history of CM, potentially indicating a recognized need for further understanding of the biology of this condition (29).
There are many inherent limitations with any citation analysis study. In this review, we chose to order the articles based on total citation number as opposed to citation rate in an attempt to focus on the “classic” papers in CM. The citation rate (per year) is another method to demonstrate article importance, which we did utilize to highlight an additional 5 articles more recently published. However, basing an analysis on citation rate alone could have caused “classic” papers to be over looked as overall citation rates per year of new articles are expected to increase over time due to the broad expansion of digital technology and scholarly literature. Another factor is a concept referred to as “obliteration by incorporation” which describes the incorporation of classic articles into the accepted medical knowledge therefore decreasing their citation rate going forward (7). An additional limitation to this review is that some articles may have been missed in our search criteria as we only included peer-reviewed articles. Therefore, citations that occur in chapters or non-peer reviewed literature (such as a website or news story) were not included thereby potentially underreporting the influence of certain publications.
This review highlights the centers and authors that have been at the forefront of influential advancements in the management of CM. Japan and the United States have been instrumental in producing CM related research. Studies dedicated to clinical outcomes and imaging have been the most influential, and likely will be in the future should trends continue. Interestingly, there are few high level studies regarding CM, possibly as a result of the limited ability to randomize patients into non-surgical management for a condition well treated with surgery. Knowledge of the most well cited papers provides physicians and other members of the medical community an informed context of the state of the art of a given topic and allows for a broader background from which to carry out future projects.
Conflicts of Interest: The authors have no conflicts of interest to declare.
- Merigó JM, Núñez A. Influential journals in health research: a bibliometric study. Global Health 2016;12:46. [Crossref] [PubMed]
- Badhiwala JH, Nassiri F, Witiw CD, et al. Highly Cited Works in Spinal Disorders: The Top 100 Most Cited Papers Published in Spine Journals. Spine (Phila Pa 1976) 2018;43:1746-55. [Crossref] [PubMed]
- Hohmann E, Glatt V, Tetsworth K. Worldwide orthopaedic research activity 2010-2014: Publication rates in the top 15 orthopaedic journals related to population size and gross domestic product. World J Orthop 2017;8:514-23. [Crossref] [PubMed]
- Huang W, Wang L, Wang B, et al. Top 100 Cited Articles on Back Pain Research: A Citation Analysis. Spine 2016;41:1683-92. [Crossref] [PubMed]
- Virk SS, Yu E. The Top 50 Articles on Minimally Invasive Spine Surgery. Spine 2017;42:513-9. [Crossref] [PubMed]
- De la garza-ramos R, Benvenutti-regato M, Caro-osorio E. The 100 most-cited articles in spinal oncology. J Neurosurg Spine 2016;24:810-23.
- Rüegsegger N, Ahmad SS, Benneker LM, et al. The 100 Most Influential Publications in Cervical Spine Research. Spine 2016;41:538-48. [Crossref] [PubMed]
- Ding F, Jia Z, Liu M. National representation in the spine literature: a bibliometric analysis of highly cited spine journals. Eur Spine J 2016;25:850-5. [Crossref] [PubMed]
- Gambín-botella J, Ayala M, Alfonso-beltrán J, et al. Predominance of Studies with Poor Level of Evidence Among the Top 100 Most Cited Studies on Idiopathic Scoliosis: A Bibliometric and Content Analysis. Spine Deform 2018;6:373-83. [Crossref] [PubMed]
- Sing DC, Jain D, Ouyang D. Gender trends in authorship of spine-related academic literature-a 39-year perspective. Spine J 2017;17:1749-54. [Crossref] [PubMed]
- Ioannidis JP, Boyack KW, Small H, et al. Bibliometrics: Is your most cited work your best? Nature 2014;514:561-2. [Crossref] [PubMed]
- Cheek J, Garnham B, Quan J. What's in a number? Issues in providing evidence of impact and quality of research(ers). Qual Health Res 2006;16:423-35. [Crossref] [PubMed]
- Garfield E. Citation analysis as a tool in journal evaluation. Science 1972;178:471-9. [Crossref] [PubMed]
- Lefaivre KA, Shadgan B, O'Brien PJ. 100 most cited articles in orthopaedic surgery. Clin Orthop Relat Res 2011;469:1487-97. [Crossref] [PubMed]
- Informetrics JP Q. application of informet-rics in science of science and management of science and technology. Inf Stud Theory Appl 2001;24:474-8.
- Hilibrand AS, Carlson GD, Palumbo MA, et al. Radiculopathy and myelopathy at segments adjacent to the site of a previous anterior cervical arthrodesis. J Bone Joint Surg Am 1999;81:519-28. [Crossref] [PubMed]
- Hirabayashi K, Watanabe K, Wakano K, et al. Expansive open-door laminoplasty for cervical spinal stenotic myelopathy. Spine (Phila Pa 1976) 1983;8:693-9. [Crossref] [PubMed]
- Matsunaga S, Sakou T. Ossification of the posterior longitudinal ligament of the cervical spine: etiology and natural history. Spine (Phila Pa 1976) 2012;37:E309-14. [Crossref] [PubMed]
- Aizawa T, Kokubun S, Ozawa H, et al. Increasing Incidence of Degenerative Spinal Diseases in Japan during 25 Years: The Registration System of Spinal Surgery in Tohoku University Spine Society. Tohoku J Exp Med 2016;238:153-63. [Crossref] [PubMed]
- Skovrlj B, Steinberger J, Guzman JZ, et al. The 100 Most Influential Articles in Cervical Spine Surgery. Global Spine J 2016;6:69-79. [Crossref] [PubMed]
- Callaham M, Wears RL, Weber E. Journal prestige, publication bias, and other characteristics associated with citation of published studies in peer-reviewed journals. JAMA 2002;287:2847-50. [Crossref] [PubMed]
- Coric D, Nunley PD, Guyer RD, et al. Prospective, randomized, multicenter study of cervical arthroplasty: 269 patients from the Kineflex|C artificial disc investigational device exemption study with a minimum 2-year follow-up: clinical article. J Neurosurg Spine 2011;15:348-58. [Crossref] [PubMed]
- Grant HM, Tjoumakaris FP, Maltenfort MG, et al. Levels of Evidence in the Clinical Sports Medicine Literature: Are We Getting Better Over Time? Am J Sports Med 2014;42:1738-42. [Crossref] [PubMed]
- Depasse JM, Palumbo MA, Eberson CP, et al. Academic Characteristics of Orthopaedic Surgery Residency Applicants from 2007 to 2014. J Bone Joint Surg Am 2016;98:788-95. [Crossref] [PubMed]
- Dang do AN, Munchhof AM, Terry C, et al. Research and publication trends in hospital medicine. J Hosp Med 2014;9:148-54. [Crossref] [PubMed]
- Chow DS, Ha R, Filippi CG. Increased rates of authorship in radiology publications: a bibliometric analysis of 142,576 articles published worldwide by radiologists between 1991 and 2012. AJR Am J Roentgenol 2015;204:W52-7. [Crossref] [PubMed]
- Halvorson MA, Finlay AK, Cronkite RC, et al. Ten-Year Publication Trajectories of Health Services Research Career Development Award Recipients: Collaboration, Awardee Characteristics, and Productivity Correlates. Eval Health Prof 2016;39:49-64. [Crossref] [PubMed]
- Matz PG, Anderson PA, Holly LT, et al. The natural history of cervical spondylotic myelopathy. J Neurosurg Spine 2009;11:104-11. [Crossref] [PubMed]
- Nouri A, Tetreault L, Singh A, et al. Degenerative Cervical Myelopathy: Epidemiology, Genetics, and Pathogenesis. Spine 2015;40:E675-93. [Crossref] [PubMed]