The evolution and contributions of the Canadian Journal of Surgery: a bibliometric study

Abstract

Background:

We sought to perform a bibliometric analysis of all papers published in the Canadian Journal of Surgery (CJS) since its indexing in 1968 to identify meaningful contributions and trends in publication, and thus an overview of its influence in the field.

Methods:

Bibliometric data from all articles published in CJS between Jan. 1, 1968, and Dec. 31, 2024, were extracted from the Web of Science Core Collection database. Descriptive and visual bibliometric analysis were performed using Bibliometrix and VOSViewer software, respectively.

Results:

Excluding meeting abstracts, 6524 articles were published in CJS during the 57-year study period. The average age of these items was 28 years, with an average citation rate per item of 9.61 and an annual growth rate of 0.69%. Most of these items were original studies (n = 4594 [70.4%]) from Canadian centres (n = 3207 [49.2%]). Twenty-nine classic-cited papers were identified. Bibliometric analysis helps to identify trends in research and sharpen the focus of the readership of CJS by highlighting what makes specific articles more citable and important in advancing the literature. The characteristics of the most cited papers include novelty, generalizability across surgical disciplines, and genuine applicability to surgical practice.

Conclusion:

The most productive countries, provinces, institutions, authors, and topics were identified. CJS remains an important part of Canadian surgical discourse through its continued dissemination of quality research across Canada and beyond.

The Canadian Journal of Surgery (CJS) is currently the longest published Canadian surgical journal. CJS promotes consistent high-quality scientific reporting from study conception to translation in Canada as well as internationally.¹ Founded in 1957 by the Canadian Medical Association (CMA), the Royal College of Physicians and Surgeons of Canada (RCPSC), and departments of surgery from Canadian universities, the journal highlights evidence-based surgery and pioneering studies across surgical disciplines with an emphasis on general surgery and its subspecialties.² Its mission is to enrich the continuing medical education of Canadian surgeons, and to serve as a forum for clinical and basic science research.³ It has remained relevant under the stewardship of adept coeditors-in-chief who have shepherded the journal through tumultuous periods in its history,² such as Dr. Robert Janes, the journal’s first editor and president of the RCPSC at the time.³ CJS is also one of only a few surgical journals remaining that embraces its “generalist” surgeon audience, partly through spotlighting resident research.⁴ These characteristics contribute to its status as the longest enduring surgical journal over the 3 centuries that surgeons have been publishing their work in Canada.¹

CJS has a rich history that has paralleled and influenced the history and growth of surgery in Canada. Organized surgery was established in Canada more than 300 years ago with the mastectomy of Marie Barbier, performed by Michel Sarrazin, surgeon to the King’s troops of New France (Eastern Canada through the Gulf of Mexico).¹ In 1829, Dr. Francois-Xavier Tessier founded the Quebec Medical Society and McGill University, the nation’s first medical school; 15 years later, a rival bilingual medical school was developed, the Montreal School of Medicine and Surgery, which published the Montreal Medical Gazette, edited by Drs. Francis Badgley and William Sutherland, from 1844 until 1852.¹ Dr. Badgley then suggested that the Quebec Medical Society merge with other regional medical societies. This is considered to be the original proposal for the creation of the Canadian Medical Association (CMA), which was established in 1867.¹ In 1906, Andrew MacPhail, faculty member at McGill University, proposed the founding of the Canadian Medical Association Journal (CMAJ), which was subsequently established in 1911.¹ The large number of published surgical articles made it clear that a Canadian journal with a focus on surgery was needed; thus, CJS was established in 1957.¹

The articles published in CJS reflect the topics and breakthroughs in surgical discovery and science over the years. The 1950s elucidated challenges that came with operating on congenital and acquired cardiac conditions for resolving cardiopulmonary bypass problems. The 1960s focused on transplantation, from the problem of rejection to preserving donor organs for allografting. In 1988, there was a breakthrough in the transplantation of purified pancreatic cells and research on total parenteral nutrition (TPN), which contributed to it becoming widely accessible for clinical use. During the 1990s, advancements in the adoption of electronic digital readout monitoring for intensive care and anesthesia emerged.⁵ During this time, CJS itself went through financial and authorship challenges,² but persevered and continued to provide a forum for these many surgical innovations, as well as including papers focused on socioeconomic issues ranging from ambulatory and extended-day surgery to the surgeon’s role in cost-effectiveness.⁶ These topics represent the landscape of the significant advances discussed in CJS that have influenced the landscape of surgery and patient care outcomes in Canada today.⁵

Bibliometric analysis is a method used for discovering publication and citation trends in scientific data for a particular topic or source. It is helpful for visualizing and analyzing the accumulated scientific knowledge and its growth in fields by providing an overview and identifying gaps in the literature.⁷ Bibliometric analysis is considered a valuable tool for assessing scientific productivity; the number of studies employing this methodology is growing, with ever-increasing publication counts in most fields.⁸ In particular, surgery, clinical trials, evidence-based medicine, cancer, and epidemiology are some of the most productive and impactful areas that benefit from bibliometric analysis.⁹ Thus, CJS is a fitting source to evaluate through the lens of bibliometrics. The objective of this study was to perform a bibliometric analysis on all articles published in CJS since its indexing in 1968 in order to identify areas of particular influence and trends in its publication.

Methods

Data source and search strategy

On July 7, 2025, we queried the Web of Science (WoS) Core Collection for every item indexed in CJS between Jan. 1, 1968, and Dec. 31, 2024. We used the following advanced search string: IS = (0008–428X OR 1488–2310) AND PY = 1968 to 2024. As WoS permits a maximum export of 500 items, the corpus was downloaded as 14 consecutive “full record + cited references” plain-text files (1–500.txt … 6501–6728.txt) and concatenated without modification, thereby retaining every bibliographic field and all cited reference data.

Data set definition

Two corpora were established. The full data set from the WoS extraction and, for all descriptive statistics, mapping and network analyses, a primary data set created by removing the records tagged “meeting abstract” were used. This convention mirrors the approach adopted in earlier reported surgical bibliometric studies.¹⁰ To ensure mutually exclusive categories, records with multiple document type tags were assigned to their primary classification. Citations for “article; proceedings paper” are aggregated within the “article” category to prevent double counting citations.

Data processing and metric extraction

Parsing and metric computation were performed in Python 3.11 (pandas 2.2.2). The core indicators document age, citation and reference counts, compound annual growth rate (CAGR), and collaboration indices were verified in Bibliometrix 4.3.2 (R 4.3.1) via its Biblioshiny interface (Version 4.3.2, R Foundation for Statistical Computing, Vienna, Austria). Annual scientific production curves and mean annual citation trends were produced with the Bibliometrix functions annualGrowth() and biblioAnalysis(). Analysis was completed by 2 independent evaluators (J.W. and A.S.), and results were compared to confirm validity of the research methodology. Any discrepancies were discussed and interrogated to ensure consistency and for resolution.

Science mapping procedures

Network visualizations were generated in VOSviewer 1.6.20 (Centre for Science and Technology Studies, Leiden University, Leiden, Netherlands) using the primary data set. Unless noted otherwise, minimum inclusion thresholds were 3 documents for the country citation analysis, 20 citations for author and journal co-citation analyses, and 5 occurrences for keyword co-occurrence. Records co-authored by more than 25 countries were excluded from the country map. Keyword co-occurrence visualization used VOSviewer (unit = keywords plus; minimum occurrences = 5; overlay visualization enabled). Decade-level dual-axis graph and CAGR calculations, citation life-cycle curves, and CCP discipline graph were prepared with Microsoft Excel 365 (Microsoft Corporation, Redmond, WA, USA). Average citation growth (ACG) was calculated, and the life cycle of the paper was illustrated using the slope of best fit line from the year of its publication until Dec. 31, 2024.

Operational definitions

Classic-cited papers (CCPs) were defined in this study as CJS publications with 100 or more total citations as of Dec. 31, 2024 (TC2024).¹⁰ We determined CCP rankings by total cumulative citations through Dec. 31, 2024. Top classic-cited CJS papers (TCCPs) were defined in this study as having twice as many or more citations than CCPs (≥ 200 citations). These definitions were adapted from a previous publication but adjusted based on the current CJS impact factor (2.2 in 2024).¹⁰ The CCPs were also characterized by their total number of citations both during 2024 (C2024) and during the calendar year of publication (C0). A single-country publication lists 1 country in the first author affiliation line; a multiple-country publication (MCP) lists 2 or more. All bibliometric calculations use the integer citation counts supplied directly by WoS, with no self-citation filtering or field normalization applied.

Results

Overall corpus

Between 1968 and 2024 there were 6728 WoS records of CJS content. Collectively they cite 76 591 references and include 4530 WoS Keywords Plus terms. The mean document age was 28.3 years, and articles received an average of 9.30 citations (0.476 citations × article⁻¹ × year⁻¹). The journal’s output increased gradually over the 57-year period, producing a CAGR of 0.69%. Authorship was distributed across 11 239 unique individuals; 1943 papers were single-author works written by 935 surgeons or other investigators. Collaboration was modest at 3.16 coauthors per paper, and 4.45% of articles involved affiliations from more than 1 country. Removing the 204 meeting abstracts left a primary data set of 6524 papers whose descriptive metrics (mean age 28.0 years, mean citations 9.61, CAGR 0.69%, co-authors 3.16, international collaboration 4.6%) differed only minimally from those of the complete data set.

Document-type spectrum

Original research articles dominated the corpus (4594 items; 52 957 citations). “Editorial material” (932 items) and “letter” (507 items) together constituted 22% of output but attracted only 2.05 and 0.86 citations per item, respectively. Review articles (285 items) were comparatively uncommon yet achieved the greatest influence, averaging 23.97 citations each. Minor categories (e.g., notes, corrections, discussion pieces) together formed 3% of output (Table 1).

Table 1.

Summary of publication types in CJS (1968–2024) extracted from the WoS Core Collection, after excluding meeting abstracts

Publication type*	No. of publications	No. of citations†
Article	4594	52 957
Editorial material	932	1913
Letter	507	438
Review	285	6833
Note	75	462
Correction	71	27
Discussion/item about an individual/other‡	60	37
Total	6524	62 667

CJS = Canadian Journal of Surgery; WoS = Web of Science.

^*Records with multiple tags were assigned to their primary type. For example, the tag “article; proceedings paper” was aggregated into the “article” type.

^†The number of citations includes self-citations (an author citing their own earlier work).

^‡“Other” includes biographical items, meetings, reprints, and a news item.

Temporal evolution of output and impact

Annual output climbed from 75 articles in 1968 to an initial peak of 217 in 1988, then a subsequent peak of 200 articles in 2009. Mean article citation density displays a bimodal pattern: work published between 1985 and 1995 achieved the highest lifetime citation rates, whereas articles published from 2020 onward show strong early citation rates despite a more limited exposure time (Figure 1). Decade aggregation confirms the 1980s as being the most prolific (1575 documents) and most highly cited time period (about 22 900 citations in total).

Fig. 1.

Annual publication volume (bars) and mean article-citation density (red line) in the Canadian Journal of Surgery, 1968 to 2024. Citation density peaks in the late 2000s, with sustained elevation thereafter; post-2020 work shows strong early impact.

Normalized for output, the 1970s retained the largest citations-per-paper ratio (14.6), closely followed by the 1980s (14.5). Articles from the 2010s show a sharp rise in numbers but, owing to citation latency, currently register the lowest mean citation density.

Country productivity and collaboration

Authors of CJS items spanned 70 countries. Canada accounted for the large majority, with 3207 publications (49.2%), 33 325 citations, and a mean 10.39 citations per article. Only 5.8% of Canadian articles listed an international co-author. The United States ranked second (206 articles [3.2%] and 2649 citations), followed by Turkey (58 articles [0.89%] and 973 citations), the United Kingdom (57 articles [0.87%] and 347 citations), and the People’s Republic of China (51 articles [0.78%] and 630 citations). In the VOSviewer citation network, Canada is located at the centre of the North American cluster (link strength 349), the United States anchors a smaller trans-border grouping (link strength 196), and European (England, link strength 35) and Asia-Pacific countries (Australia, link strength 14, and People’s Republic of China, link strength 7) form 2 additional clusters.

Institutional contributions

The data originated from 1438 distinct organizations. The University of Toronto was the most prolific (1251 papers, 28.0%), followed by the University of British Columbia (484 papers, 10.8%) and McGill University (482 papers, 10.8%). The 10 most productive institutions were all Canadian universities or their affiliated teaching hospitals, nderscoring the journal’s domestic focus.

Author and journal co-citation

Author co-citation analysis (threshold ≥ 20 citations) identifies the current co-editor of CJS, C.G. Ball, currently affiliated with the University of Calgary, as the most influential node (link strength 219), narrowly ahead of A.W. Kirkpatrick, also affiliated with the University of Calgary (link strength 204), and H.H. Stone, who was affiliated with Emory School of Medicine before his death in 2002 (link strength 177) (Figure 2).

Fig. 2.

Author co-citation network in the Canadian Journal of Surgery (1968 to 2024), showing the most influential authors (≥ 20 citations). Colours represent groups of authors who are commonly cited together, with authors in the same colour having stronger cocitation relationships.

Journal co-citation mapping highlights Annals of Surgery as the most common source journal (3312 citations, link strength 47 624). Other important source journals are shown in Figure 3.

Fig. 3.

Journal co-citation network in the Canadian Journal of Surgery (1968 to 2024), illustrating the core literature cluster (≥ 20 citations). Colours represent groups of journals that are frequently cited together, with journals in the same colour sharing stronger cocitation relationships.

Keyword evolution

Limiting the analysis to records from 1991 onward yielded 6214 unique keywords; after frequency pruning (≥ 5 occurrences) 148 remained. The most prevalent terms were “surgery” (215 occurrences), “management” (205 occurrences), and “outcomes” (112 occurrences). A temporal overlay shows an early emphasis on procedure-specific themes such as “cholecystectomy” and “hernia,” giving way after 2010 to quality-of-care and health-services concepts, including “enhanced recovery,” “cost-effectiveness,” and “registries.”

Classic-cited papers

Twenty-nine CJS articles qualified as CCPs (Table 2), and of these 29 articles, 9 qualifed as TCCPs. The most cited CJS paper was the 2008 systematic review by Santaguida and colleagues¹¹ on prognostic factors in hip and knee arthroplasty (341 citations). General surgery topics accounted for almost half (4 of 9) of the TCCPs. Citation life-cycle curves for the top 5 TCCPs show a rapid rise in the first 3 to 5 years after publication, followed by continued positive growth with a lower amplitude tail (Figure 4). In terms of current activity (C2024), Karanicolas and colleagues¹⁴ (2010) on blinding and Santaguida and colleagues¹¹ (2008) on arthroplasty were the most frequently cited in 2024, suggesting that some of the highest-cited papers are still relevant today. The article with the most citations during its first year of publication (C0 = 8) was a retrospective study examining chest tube insertions in trauma patients to determine efficacy of residency training in chest tube placement by Ball and colleagues³¹ in 2007.

Table 2.

The 29 most cited CJS articles between January 1, 1968, and December 31, 2024

Rank (TC2024)*	Study (year)	Article title	Discipline	Avg annual TC	Rank (C0)†	Rank (C2024)†
1 (341)	Santaguida et al.11 (2008)	Patient characteristics affecting the prognosis of total hip and knee joint arthroplasty: a systematic review	Orthopedic surgery	21.3	5 (1)	2 (18)
2 (321)	Salter et al.12 (1969)	Avascular necrosis of the femoral head as a complication of treatment for congenital dislocation of the hip in young children: a clinical and experimental investigation	Orthopedic surgery	5.8	6 (0)	13 (5)
3 (318)	Maunsell et al.13 (1993)	Arm problems and psychological distress after surgery for breast cancer	General surgery	10.3	6 (0)	9 (9)
4 (312)	Karanicolas et al.14 (2010)	Blinding: Who, what, when, why, how?	Not specific	22.3	4 (2)	1 (24)
5 (306)	Rider et al.15 (1977)	Preoperative irradiation in operable cancer of rectum: report of the Toronto trial	General surgery	6.5	6 (0)	15 (3)
6 (289)	Greenberg et al.16 (2016)	Clinical practice guideline: management of acute pancreatitis	General surgery	9.6	4 (2)	11 (7)
7 (255)	Kirkpatrick et al.17 (2000)	Is clinical examination an accurate indicator of raised intra-abdominal pressure in critically injured patients?	General surgery	15.9	5 (1)	5 (14)
8 (211)	Petis et al.18 (2015)	Surgical approach in primary total hip arthroplasty: anatomy, technique and clinical outcomes	Orthopedic surgery	16.2	5 (1)	6 (12)
9 (204)	Heyland et al.19 (2001)	Total parenteral nutrition in the surgical patient: a meta-analysis	Not specific	14.6	5 (1)	7 (11)
10 (198)	Attard et al.20 (2007)	Adhesive small bowel obstruction: epidemiology, biology and prevention	General surgery	16.5	4 (2)	4 (15)
11 (187)	Singham et al.21 (2009)	Choledochal cysts part 1 of 3: classification and pathogenesis	General surgery	12.5	5 (1)	8 (10)
12 (176)	Ramzy et al.22 (2005)	Cardiac allograft vasculopathy: a review	Cardiac surgery	9.3	4 (2)	10 (8)
13 (168)	Al-Omran et al.23 (2003)	Epidemiologic features of acute appendicitis in Ontario, Canada	General surgery	8	4 (2)	12 (6)
14 (162)	Barker et al.24 (2009)	Incidence and impact of dysphagia in patients receiving prolonged endotracheal intubation after cardiac surgery	Cardiac surgery	10.8	5 (1)	9 (9)
15 (155)	Dirks et al.25 (1993)	The value of reoperation for recurrent glioblastoma	Neurosurgery	5	5 (1)	14 (4)
16 (149)	Efremidou et al.26 (2009)	The efficacy and safety of total thyroidectomy in the management of benign thyroid disease: a review of 932 cases	ENT & general surgery	9.9	4 (2)	10 (8)
17 (144)	Dumper et al.27 (2006)	Complications of Meckel’s diverticula in adults	General surgery	8	4 (2)	11 (7)
18 (138)	Nickel et al.28 (1985)	Electron microscopic study of an infected Foley catheter	Urology	3.5	4 (2)	16 (2)
19 (132)	Arenal et al.29 (2003)	Mortality associated with emergency abdominal surgery in the elderly	General surgery	6.3	5 (1)	13 (5)
20 (129)	Bourne et al.30 (1988)	In vivo comparison of four absorbable sutures: Vicryl, Dexon Plus, Maxon and PDS	Not specific	3.6	6 (0)	15 (3)
21 (124)	Ball et al.31 (2007)	Chest tube complications: How well are we training our residents?	General surgery	7.3	1 (8)	12 (6)
22 (121)	Lewis32 (2002)	Oral versus systemic antibiotic prophylaxis in elective colon surgery: a randomized study and meta-analysis send a message from the 1990s	General surgery	5.5	2 (5)	14 (4)
23 (118)	Kreder et al.33 (2003)	Provider volume and other predictors of outcome after total knee arthroplasty: a population study in Ontario	Orthopedic surgery	5.6	3 (4)	13 (5)
24 (115)	Comerota et al.34 (1993)	Thrombolytic therapy for deep venous thrombosis: a clinical review	Not specific	3.7	4 (2)	15 (3)
25 (112)	Grober et al.35 (2005)	Defining medical error	Not specific	5.9	5 (1)	14 (4)
26 (108)	Peel et al.36 (2018)	A systematic review of factors affecting choice of surgery as a career	Not specific	18	4 (2)	3 (16)
27 (104)	Temple et al.37 (1999)	A meta-analysis of laparoscopic versus open appendectomy in patients suspected of having acute appendicitis	General surgery	4.2	5 (1)	14 (4)
28 (102)	Belliveau et al.38 (1999)	Ileoanal anastomosis with reservoirs: complications and long-term results	General surgery	10.2	5 (1)	9 (9)
29 (100)	Sarkhosh et al.39 (2013)	Complications associated with laparoscopic sleeve gastrectomy for morbid obesity: a surgeon’s guide	General surgery	2.7	5 (1)	16 (2)

ENT = otolaryngology; PDS = polydioxanone; TC = total citations.

^*Articles were chosen and sequenced based on a CCP definition of TC2024 ≥ 100.

^†Ranks for C0 and C2024 were calculated using dense ranking (descending). Articles with identical citation counts share the same rank.

Fig. 4.

Citation life-cycle curves of the 5 top classic-cited papers (TCCPs) published in the Canadian Journal of Surgery (TC2024 ≥ 200). Each line represents cumulative citations since publication. Slope = average citation growth (ACG).

Among the CCPs, 4 (13.8%) were within the scope of orthopedic surgery, 14 (48.3%) within general surgery, 2 (6.9%) within cardiac surgery, 1 (3.4%) within urology, 1 (3.4%) within neurosurgery, and 7 (24.1%) were multidisciplinary (Figure 5). In terms of study type, 11 (37.9%) of the CCPs were review articles (mix of literature review, systematic review, meta-analysis), 5 (17.2%) were experimental studies, 7 (24.1%) were retrospective studies, 2 (6.9%) were guidelines, and 4 (13.8%) fell into other categories.

Fig. 5.

Distribution of classic-cited papers (CCPs) across disciplines.

Discussion

Bibliometric analysis provides an avenue by which surgeons can examine and critically appraise the ever-growing body of literature that informs our practice. CJS is no exception to this trend: the compound annual growth rate of CJS is 0.69%, leaving surgeons and surgical trainees to determine the areas of surgical literature most deserving of their attention. Bibliometric study allows surgeons to identify key papers in the larger corpus of published work to assist them with focusing and sharpening their study of what can be an overwhelming volume of published works.

Although we must acknowledge that citations do not necessarily indicate high-quality research,⁴⁰ they are useful as a valid, potentially objective method of judging the impact of articles.⁴¹ The high-citation articles in CJS focused on colorectal, orthopedic, and oncologic surgery (Table 2). Articles were also analyzed by citations in the last year of the bibliometric analysis (C2024), to determine how citations performed over time compared with during the most recent year. As expected, overall, the more recently published articles (after 2004) had a higher C2024 (average 10.4) than those published before 2004 (5.8) (Table 2). A careful review of the TCCPs was undertaken to determine what characteristics of these specific papers contributed to their status as a TCCP.

General surgery had the most TCCPs (4), and general surgery subspecialty topics included breast, colorectal, hepato-pancreato-biliary, and trauma surgery.

The third-ranked TCCP (318 citations), entitled “Arm problems and psychological distress after surgery for breast cancer” by Maunsell and colleagues¹³ from Université Laval in 1993, involved patients who received surgery for unilateral breast cancer and were interviewed at 2 time points: 3 months and 18 months postoperatively. The interviews showed that overall problems with the arm on the operated side were frequent and often persistent beyond 18 months. Although type of mastectomy did not affect the frequency of problems, axillary dissection was associated with a significantly higher frequency of problems. The authors reported that if axillary dissection could be eliminated, patients may experience improved quality of life. Now, more than 30 years later, axillary dissections have become limited to specific cases, with the importance of a more tailored approach in part supported by the work of this research group.

The fifth-ranked TCCP (306 citations) was a study by Rider and colleagues¹⁵ from the University of Toronto, published in 1977, that reported on preoperative radiation therapy for the treatment of rectal cancer. This prospective controlled trial showed that, for patients with Dukes Stage C (stage 3) rectal cancer, preoperative radiation therapy doubled survival at 5 years. This paper is interesting to the contemporary reader, as it demonstrates the changes in study procedures that have occurred over the past several decades; in that study patients were not consented for their participation as it was not standard practice at the time, which seems foundationally unethical now. The paper was seminal in that it heralded the advancement of preoperative or neoadjuvant radiation therapy for rectal cancer. Although the study did not demonstrate a difference in crude survival between control and experimental groups, the survival benefit in patients with Dukes Stage C cancer had some effect of preoperative radiation. This would later become clear in the observations made in the 1997 landmark Swedish Rectal Cancer Trial, which was the first randomized controlled trial to clearly show the benefit of neoadjuvant radiation therapy for the treatment of rectal cancer.⁴²

Greenberg and colleagues¹⁶ from the University of Toronto authored the sixth-ranked TCCP (289 citations) in CJS, a clinical practice guideline (CPG) for the treatment of pancreatitis published in 2015.¹⁶ This CPG evaluated the published evidence and guidelines for strength of evidence and strength of recommendations to provide clinicians with a clear diagnostic and therapeutic path for the treatment of both mild and more complex pancreatitis. It emerged in the context of the rapidly changing diagnostic field in terms of imaging and laboratory capabilities. Interestingly, this study was the most cited CJS paper per year since publication, with a citation rate per year of 31.67.

The seventh-ranked article (255 citations), authored by Kirkpatrick and colleagues¹⁷ at the University of Calgary, was an experimental study examining the difference between physical examination and bladder pressures for evaluating elevated intra-abdominal pressure (IAP) in trauma patients. It highlighted the challenges with physical examination and reported that, compared with bladder pressures, clinical examination was no different than random chance for predicting increased intra-abdominal pressure. The results of this study informed practice: when a clinician suspects raised intra-abdominal pressure, bladder pressure should be the preferred method of measurement.

Orthopedic surgery had the second-highest proportion of TCCPs in CJS (3 articles), claiming both the first and second place in total citations.

The top-ranked TCCP (341 citations) was a systematic review of the characteristics of patients undergoing hip and knee arthroplasty by Santaguida and colleagues from McMaster University in 2008.¹¹ It was a large systematic review that included 64 studies of 500 or more patients, examining patient characteristics and their association with revision rate after hip or knee arthroplasty. This TCCP was the first published systematic review on this important topic and presented a robust analysis of postoperative outcomes, including revision rates, mortality, functional outcomes, postoperative pain, patient satisfaction, and arthroplasty complications. Although the study found that male sex and younger age were associated with higher revision rates, the authors concluded that there were no clear population subgroup for which a restriction on performance of joint arthroplasty should be instituted. The paper provided an evidence-based reason for orthopedic surgeons to offer arthroplasty, even to patients for whom they had been considering recommending against arthroplasty based on characteristics such as age or body mass index.

The second-ranked TCCP (321 citations), authored by Salter and colleagues¹² from the University of Toronto in 1969, was a unique study that combined the clinical evaluation of human children and an experimental evaluation of piglets to better understand avascular necrosis (AVN) of the femoral head as a complication of congenital hip dislocation. The paper recognized the challenge of characterizing and identifying AVN and highlighted an incidence rate of 0% to 73% that was reported in the literature at that time. A lack of clarity on the pathogenesis and classification of AVN laid the groundwork for the basis of this study. It was included in a clinical retrospective study on 140 children (163 hips), which showed that tight adduction at the time of reduction and rigid immobilization tended to produce AVN and that this may be a consequence of increased pressure between the femoral head and the acetabulum. The results of this retrospective study were used to inform the experimental study with piglets. A series of experiments were carried out with casting in various degrees of abduction and adduction. The results mirrored those of the clinical study: tight adduction followed by casting of the immature hip in strict abduction were associated with development of AVN, while prevention of AVN could conceivably be achieved by avoiding compression of the acetabulum and the immature femoral head. The authors concluded that to prevent AVN in children, the strategy should involve continuous traction before reduction, adductor tenotomy at the time of reduction, and immobilization in a more comfortable position with flexion and slight abduction to reduce compression of acetabulum on the femoral head. The study was pivotal in the evolution of the treatment of congenital hip dislocation.

Petis and colleagues¹⁸ from Western University authored the eighth-ranked TCCP (211 citations), a literature review summarizing surgical approaches and outcomes for total hip arthroplasty. There are 3 main approaches possible for total hip arthroplasty: anterior, lateral, and posterior, each with their own advantages and disadvantages. The study examined the evidence with clinical outcomes and complications for each surgical approach, and determined that no single surgical approach was safer or more efficacious than another, and thereby recommended that surgeons perform the procedure with which they have the most experience and feel most comfortable.

Finally, 2 TCCPs did not fit into 1 specific discipline, but were more broadly applicable across surgical specialties.

Karanicolas and colleagues¹⁴ from Western University authored the fourth-ranked TCCP (312 citations) in 2010. The paper, entitled “Blinding: who, what, when, why, how?,” serves as a guide for surgeon-researchers in the development of randomized controlled trials. The paper highlights the benefits of blinding trials but also acknowledges the inherent difficulties of controlling surgical trials compared with medical trials, given the physical nature of surgery and the inability to provide a placebo. The paper also offers creative solutions to those difficulties (e.g., large dressings to hide incisions). This paper serves as a guideline of sorts that is readily citable as surgical research groups develop their controlled trials.

The ninth-ranked TCCP (204 citations) was a meta-analysis performed by Heyland and colleagues¹⁹ from Queen’s University that takes an in-depth look at the association between total parenteral nutrition (TPN) and complication and mortality rates in surgical patients. After evaluating 27 studies, the authors found no difference in mortality when comparing patients treated with TPN perioperatively with those who were not. In terms of complication rates, the study reported there were fewer complications associated with TPN before 1988, but increased complications after 1988. The authors noted that after 1988, patients who were severely malnourished had lower complication rates with TPN. Unquestionably, this article helped lay the groundwork for the well-known saying among general surgeons, “gut is best.”

Among the 9 TCCPs published in CJS, several characteristics present themselves as markers of exceptional surgical literature: strong evidence such as that found in the extensive systematic review by Santaguida and colleagues,¹¹ generalizability such as that presented in the blinding recommendations by Karanicolas and colleagues,¹⁴ and paradigm shifting findings such as those reported in the neo-adjuvant radiation study by Rider and colleagues¹⁵ that laid the groundwork for large-scale studies and practice changes in rectal cancer management. When surgeons develop research projects with passion and dedication, the results are impactful studies such as those discussed here. A uniting feature links the 9 TCCPs published in CJS: each of them laid the critical groundwork for future clinical practice and study. Examination of the life-cycle curves of the top 5 TCCPs published in CJS highlights the enduring strength of the research: each has either an increasing or stable continued cumulative citation accrual, indicating the foundational character upon which further research continues to be built (Figure 4).

Several key observations from this bibliometric study may be explained by the history surrounding CJS. The period of greatest growth in productivity for the journal was between 1968 and 1988, when one of the most significant advances in perioperative care — TPN — was established and generated many quality CJS papers.⁵ Since that time, the journal’s publications have stabilized as, in 1996 and 1997, financial constraints led to a reduction in the number of articles published in CJS. However, the average article citations per year for CJS have been rising owing to the improving quality and impact of the published papers. The peak in 2009 for publication and citation count, followed by a subsequent precipitous drop, may be attributed to the journal’s transition from print to its current digital format and the implementation of article processing charges.^2,43 While research articles were found to be the most published and cited document type, review articles had the highest average number of citations per item, which is consistent with the general literature. Other non-article document types (e.g., letters, editorials, and other minor categories) contributed fewer citations overall, consistent with their role in rapid communication. The publication numbers were lowest in the decades that were incomplete (1960s, 2020s), as there were not or have not yet been equal numbers of years compared with the other decades examined.

As expected, the most published and cited country of origin for CJS papers is Canada. Additionally, most papers are single-country publications, which suggests that international cooperative surgical research continues to be challenging.^44,45 However, even with a relatively small percentage (4.6%) of internationally co-authored papers, these global contributions are worth noting. Importantly, surgical researchers in Canada most commonly collaborate with those in the United States and England, perhaps in part because of their shared language and culture. Surprisingly, Turkey was the most cited country by average article citations, and the third most published and cited country after the United States.² Within Canada, most publications originated from Ontario, followed by British Columbia and Quebec, whereas the fewest published papers originated from the territories, which is in keeping with the population counts of, and thus productive institutions, within Canada’s provinces and territories.

Clustering is helpful to elucidate the establishment and growth of research fields.⁷ Figure 2 shows 11 clusters of authors frequently cited together; the purple cluster has the most total citations and the pink cluster the fewest but most highly cited authors. The authors in the pink cluster likely have published seminal articles in the field, which explains both the high number of citations of these authors, as well as their high rate of co-citation. One of those authors, Chad G. Ball, currently the co-editor-in-chief of CJS, is the author most frequently co-cited in CJS articles. The purple cluster appears to be composed of breast cancer researchers. One of those researchers, Bernard Fisher, is the most cocited author among all authors referenced in CJS articles. Dr. Fisher was a surgeon based out of the University of Pittsburgh who was the chair of the National Surgical Adjuvant Breast and Bowel Project, with major contributions that included preclinical studies discouraging radical mastectomies, developing breast cancer prevention trials with endocrine therapy, and introducing to the surgical field the clinical trial methodology.⁴⁶

Limitations

Though our study employed novel bibliometric methodology, such as the in-depth analysis of TCCPs and provinces, it also had limitations. Despite the WoS Core Collection being the most widely used search database, it likely omits relevant articles and may contain indexing errors or incomplete records. Likewise, as the database is continually updated, there may be a lag in the data obtained, such as from the date of extraction until present. However, this was addressed by noting the specific date of data extraction and confining extraction to that day to reduce discrepancies. No data sets were available for inclusion of CJS papers before 1968, so it is possible that some older CCPs could have been missed. Extraction or encoding errors may have happened and influenced the results, but this was mitigated by repeating all data analysis to check for consistency. Differences in the ways VOSViewer, WoS Core Collection, and Bibliometrix in R performed their analyses could have also resulted in minor differences. Finally, there are more innovative metrics that could instead be used to express a paper’s impact, such as journal downloaded and read statistics, that were not included in this analysis. These include alternative metrics that utilize online user interactions (e.g., Altmetrics, Plum X), which could supplement traditional bibliometrics. Definitions for CCPs and TCCPs have not been standardized in the literature; although we believe the citation thresholds chosen for this study were appropriate as they identified 0.44% and 0.14% of published CJS papers meeting criteria for being a CCP and TCCP, respectively, in a study evaluating or comparing journals that have higher impact factors than CJS, citation threshold definitions would likely need to be raised.¹⁰

Conclusion

This bibliometric analysis of works published in CJS over the past 57 years has highlighted 3 central findings. First, CJS has been and remains a key touchstone for surgical literature across surgical disciplines in Canada and represents the main body of surgical research in Canada. Second, through our bibliometric analysis, we identified CCPs, which represent the CJS articles with the most enduring significance for surgeons. These CCPs provided the opportunity to identify features representative of impactful surgical literature, thereby helping to provide insight into the development of further surgical research. Finally, this analysis has highlighted the development of CJS and of surgical research over the past 6 decades and traced the changes through the journal’s nascency into its maturity as a well-respected multidisciplinary journal that informs surgical practice in Canada and beyond.