Top 50 most-cited articles on craniovertebral junction surgery.

BACKGROUND: Craniovertebral junction is a complex anatomical location posing unique challenges to the surgical management of its pathologies. We aimed to identify the fifty most-cited articles that are dedicated to this field.
METHODS: A keyword search using the Thomson Reuters Web of Knowledge was conducted to identify articles relevant to the field of craniovertebral junction surgery. The articles were reviewed based on title, abstract, and methods, if necessary, and then ranked based on the total number of citations to identify the fifty most-cited articles. Characteristics of the articles were determined and analyzed.
RESULTS: The earliest top-cited article was published in 1948. When stratified by decade, 1990s was the most productive with 16 articles. The most-cited article was by Anderson and Dalonzo on a classification of odontoid fractures. By citation rate, the most-cited article was by Herms and Melcher who described Goel's technique of atlantoaxial fixation using C1 lateral mass screws and C2 pedicle screws with rod fixation. Atlantoaxial fixation was the most common topic. The United States, Barrow Neurological Institute, and VH Sonntag were the most represented country, institute, and author, respectively. The significant majority of articles were designed as case series providing level IV evidence.
CONCLUSION: Using citation analysis, we have provided a list of the most-cited articles representing important contributions of various authors from many institutions across the world to the field of craniovertebral junction surgery.

Introduction

Surgical approach to the craniovertebral junction requires intimate knowledge of the skeletal, ligamentous, neural and vascular anatomy, and the natural history of its pathologies. Our knowledge of this unique anatomical location has undergone advancements through the work of many authors.

Citation analysis is a bibliometric tool that can be utilized to quantitatively evaluate the influence of an article on a given field. The goal of this study was to select the fifty most-cited articles pertaining to the surgical management of craniovertebral junction pathologies published between 1900 and 2015.

Methods

Inclusion criteria

Following the initial search process, studies were selected for inclusion in the final list of fifty based on several inclusion criteria. Eligible studies were required to have focused primarily on the study of pathologies or interventions involving the craniovertebral junction. Studies involving nonhuman subjects and studies published in languages other than English were excluded from the final list. Studies that discussed other related conditions, such as cervical spine pathologies, were included as long as patients with craniovertebral pathologies constituted the majority of the patient population in the study.

Eligible articles were selected from all journals and databases indexed on the Thomson Reuters Web of Science at the time of this analysis.

Data collection

A two-step query of the Thomson Reuters Web of Science was conducted to assemble the final list of fifty articles.

The first stage consisted of a topic search for the term “craniovertebral.” This search yielded 2190 results, which were subsequently sorted from the most-cited articles to the least-cited articles. The top 100 relevant studies in the list were analyzed to compile a list of keywords to be used in the subsequent search phase.

Seventy-six keywords were chosen from the aforementioned articles [Table 1] based on the following four categories: Location, approach, pathology, and therapy. These keywords were then used to search the database a second time using a compiled search term [Table 2]. This phase of the search yielded 31,965 papers, which were similarly sorted in order from most to least citations. We eliminated self-citations by conducting a “cited reference search” which provides all citing articles, followed by an “author search.” Results of the latter search, which yielded all articles published by authors of the article of interest, were then subtracted from the “cited reference search.” The fifty most-cited papers in this list were selected out of the first 298 results for final inclusion in the list presented in this report.

Table 1

Keywords, in 4 categories, determined based on the first query

Table 2

Finalized search terms used in the second query

The following information was extracted from each article selected for the final list: Article title, year of publication, total number of citations, citation rate, article topic, study type, and level of evidence. Citation rate was defined as the number of citations per year since the year of publication. Article topic categories included atlantoaxial fixation, atlantoaxial instability, occipitocervical fusion, trauma, anterior approach to craniovertebral junction, mixed, and others. Study type pertains to whether a study is nonclinical or clinical in which case the study design, potentially ranging from randomized controlled trial to case report, is determined. Level of evidence is determined for clinical studies, potentially ranging I–V.

Results

Table 3 provides information on the articles including their citations' count, citation rate, article topic, article summary, article type, and level of evidence. Our final list consisted of articles published between 1948 and 2004. Publications peaked in 1985 with four articles [Figure 1];[20314940] however, the most productive decade was the 1990s with 16 articles [Table 4].[1822331530516489111074212468]

Table 3

List of top 50 most-cited articles on craniovertebral surgery

Figure 1

The number of top 50 articles published per year

Table 4

Frequency of articles per decade

The top journal was the Journal of Bone and Joint Surgery: American Volume which produced 12 out of the fifty articles. This was followed by the Journal of Neurosurgery and Spine, with nine and seven articles, respectively [Figure 2]. The authors with the most top 50 publications included VKH Sonntag[153048354034421246] (9), CA Dickman[153048354212] (6), and MN Hadley[3048403534] (5).

Figure 2

Number of top 50 articles published per journal

From Figure 3, it can be observed that atlantoaxial fixations (14) was the most common topic among the articles in the final list.

Figure 3

Frequency of article topics on the final list

Articles in the final list originated from 13 different countries, with the majority having been based in the United States (26 articles). Table 5 displays the top ten institutions affiliated with articles on the final list. The most productive institutions were Barrows Neurological Institute (six articles) and St. Luke's Hospital (three articles).

Table 5

Top 10 institutions responsible for articles on the final list

The most-cited article pertaining to the craniovertebral junction was the 1974 study by Anderson and D'Alonzo on odontoid fractures.[1] The article with the highest citation rate was Harms and Melcher's report on the use of C1 lateral mass screws and C2 pedicle screws with rod fixation for atlantoaxial fixation.[2]

Discussion

We used citation analysis to identify the most-cited articles on the topic of craniovertebral junction surgery. This list represents the authors, institutions, and countries, which have been instrumental in the advancement of the field. We characterized the most-cited research topics and the study design utilized by the authors to investigate it, providing a perspective on the trends within the field and the level of evidence available for the surgical management of this complex anatomical location.

The years of publication ranged from 1948 to 2004. The article from 1948, ranked 19th overall and authored by McGregor, provides an alternative to Chamberlain's line for the diagnosis of basilar impression. This line, which later known as McGregor line, is a line that connects the posterior edge of the hard palate to the most caudal end of the occipital bone's curve. If the tip of the odontoid process is 4.5 mm rostral to this line, the diagnosis of basilar impression is made.[19] The article in 2004, ranked 16th overall and authored by Wright,[16] is the first description of the use of bilateral crossing C2 laminar screws in rigid constructs that include C2 in occipitocervical, atlantoaxial, and subaxial fusion. The author discusses the advantage of this previously unreported technique, including the low risk of vertebral artery injury.

The distribution of the year of publication of the top 50 articles reveals interesting trends. Our search criteria began in 1900 and ended in 2015. There were a total of five articles in the first 70 years of the 20th century, with no articles in the 1950s on our list. In the 1970s, the number of articles increased to 11, followed by another increase to 14 articles in the 1980s, peaking in the 1990s with 16 articles. This was followed by a stark decrease in the 2000s to four articles. There were no articles from the 2010s that qualified for the top 50 list.

We identified the most-cited articles stratified by the decade of publication. In the 1940s, the most-cited article, and also the only article from this decade in our list, was McGregor's study, which was discussed earlier. In the 1960s, the most-cited article was a case series by Sharp and Purser[13] who reported the incidence and the diagnostic criteria of atlantoaxial subluxation in patients with rheumatoid arthritis and ankylosing spondylitis. In the 1970s, the most-cited article was by Anderson and D'Alonzo.[1] This article is also the most-cited article in our top 50 list. The authors provide their well-known classification of the dens fracture into type I, fracture of the tip of dens, type II fracture at the base of dens, and type III fracture of the dens extending to the lateral masses of the axis.[1] All the above articles are concerned primarily with the radiologic diagnosis of pathology concerning the craniovertebral junction.

The most-cited articles in the subsequent decades (1980s–2000s) are focused on surgical management of the pathologies that involve the craniovertebral junction. In 1980s, Menezes and VanGilder described their 10-year experience in using transoral approach to the craniovertebral junction in 72 patients with various pathologies.[14] Although this article is not the first to describe the anterior approach to this anatomical location, it may be considered the most-cited article that promoted this surgical technique, its efficacy and safety.[14] This article was ranked 14th overall. In the 1990s, the most-cited article, ranked 6th overall, was by Goel and Laheri who described for the first time the technique of using lateral mass screws of C1 and C2 with plates in atlantoaxial or occipitocervical fusion constructs. In the 2000s, the most-cited article was by Harms and Melcher who described the incorporation of rods, instead of plates, in connecting C1 lateral mass and C2 pedicle screws in atlantoaxial fixation. This article is ranked 2nd overall. Of note, the Harms' paper unfortunately did not cite the Goel's paper.[52]

Ranking based on citation rate revealed that the aforementioned article by Harms and Melcher was the 1st. The second-ranked article was by Dickman and Sonntag, from 1998, whereby the authors described the outcome of patients undergoing atlantoaxial transarticular fixation.[15] This article was not the first to describe the technique of transarticular atlantoaxial fixation which is attributed to Magerl.[53] This article was published in German, thus not meeting our inclusion criteria (due to the language of publication). The Dickman and Sonntag study was not the first to describe the outcome and complications of this technique either, which is credited to Grob et al.'s[11] study from 1991, ranked 11th overall and 10th based on citation rate. It should be noted that Dickman and Sonntag's study included 121 patients compared to the 38 patients in Grob et al.'s study. The 3rd ranked article based on citation rate was by Madawi et al. in 1997,[7] which, interestingly, also described the outcome of transarticular fusion of C1 and C2 in 61 patients. Thus, the top three ranked articles based on citation rate are all related to atlantoaxial fixation.

In fact, article topics were predominantly concerned with atlantoaxial fixation comprising 28% of the studies, with six of the top ten articles dedicated to this topic. Trauma and atlantoaxial instability were the second most common topics with 24% each. Among the articles on trauma, nine articles were concerned with fractures of C2. Of the 12 articles on atlantoaxial instability, eight were dedicated to patients with rheumatoid arthritis. Mixed topic articles were 10% of the top 50 cited articles in the field. All five of the mixed articles included the topic of occipitocervical fusion. Among all the fifty studies, only one was dedicated solely to this surgery as described by Abumi et al.[36] The authors described for the first time the technique of occipitocervical fusion using cervical pedicle screws and occipitocervical rod system.[36] This article was ranked 36th overall.

The most productive country, institution, and author in our list of top 50 most-cited articles were the United States of America, Barrow Neurological Institute, and Dr. Sonntag. However, as demonstrated in Figure 4 and Table 5, the advancement of the field has been a multinational and multi-institutional effort involving centers in North America, Europe, and Asia.

Figure 4

Country of origin of articles on the final list

Absence of high level of evidence among the top 50 cited articles was unambiguous. There are 46 clinical studies, all of which provide a class IV evidence with the predominant majority designed as case series. This could be attributed to the complexity, diverse, and rare pathologies affecting the craniovertebral junction, which have proved the design of higher quality studies difficult. The other four articles are in vitro biomechanical studies, to which the designation of level of evidence does not apply.

This study has limitations. Citation analysis is biased. Citation does not necessarily imply the impact of an article on the study that is citing it. Authors often provide extensive background information in the introduction section of their articles citing various previous related articles or they may cite an article without fully reading it. Another important bias of citation analysis is the issue of self-citation, whereby authors refer to their own work.[54] Although it is seldom the intention of the authors, self-citation inflates the number of citations of a given article which may be misinterpreted as the extent of its influence in the field. To eliminate this bias, we removed self-citation. On average, the total number of citations was reduced by an average of 11 citations per article. Another bias arises from the fact that the total number of citations for an article may increase merely based on how old a publication is: An older study would have a higher chance of accumulating citations than a newer study. To circumvent this bias, we also provide citation rate, the total number of citations divided by the number of years since publication. This is an attempt to take into consideration the date of publication. A review of the top ten articles in our list reflects this phenomenon, whereby four articles are from 1970s, five articles from 1990s, and only one article from 2000s. However, citation rate is not flawless either. The older the study, the higher the denominator of the equation to determine citation rate. This, in essence, “dilutes” the total number of citations of a given article. This is reflected, at least partially, in the fact that based on citation rate, the top 3 articles are from 2001, 1998, and 1997, respectively. Therefore, we believe that both the total number of citations and citation rate, when considered together, provide a more comprehensive perspective of which articles in the field of craniovertebral junction surgery are the most cited. Finally, implicit citations is another bias of citation analysis, whereby an article whose contribution to the field has been absorbed to the body of knowledge becomes cited less and less over time. As described above, we have made every attempt to consider and account for the inherent biases of citation analysis. Importantly, the flaws of citation analysis is a reminder that citation does not imply influence. Therefore, we have deliberately avoided referring to the articles in our top 50 list as “most influential.”

Conclusion

We identified the fifty most-cited articles related to the craniovertebral junction pathologies. This field has attracted the attention of various centers from various countries. Although until the 1970s, the most-cited articles were primarily focused on the diagnostic evaluation of the pathologies of this complex anatomical location, the most-cited articles thereafter are concerned with surgical management of these pathologies, with the advent of multiple new techniques, the majority of which are posterior approaches. The omnipresence of level IV evidence through case series in this list underlines the paucity of high-level evidence and the necessity to design appropriate studies toward developing such evidence.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.