Research productivity on spontaneous intracranial hypotension: A bibliometric analysis

Introduction Spontaneous intracranial hypotension (SIH) is an important cause of devastating headaches and caused by CSF-leaks in the spine. Research question The aim of this analysis was to gain an overview of the progress of research on SIH over time. The global publication landscape relating to SIH was analyzed and comparisons between regions were made. Material and methods A bibliometric analysis was performed by searching for research articles on SIH in PubMed published between 1983 and 2022. Countries responsible for the publications were ranked by the sum of citations. An average annual growth rate was calculated and the density of SIH publications per 100 000 physicians was determined. Results We identified 974 articles. In 1983 only one SIH patient was reported; in 2021 the number of patients had increased to 4230. The average annual growth rate of SIH publications during this period was 12.7%. The most common publication type were case reports (n = 570). The most common medical specialty of the first author was neurology (n = 251) followed by neurosurgery (n = 250) and radiology (n = 191). Although most publications originated from the United States of America (USA), South Korea had the highest density of SIH investigators (37.86 publications per 100 000 medical doctors). The most cited paper (296 citations) was published in 2006 in JAMA (USA). Discussion and conclusion Research on SIH has increased exponentially over the past four decades. The international community of SIH researchers is growing, and with it the opportunities for global networks involved in research, treatment, and patient education.


Introduction
Spontaneous intracranial hypotension (SIH) is primarily caused by cerebrospinal fluid (CSF) leaks in the cervicothoracic spine.It has an estimated incidence of 4-5 cases per 100 000 population per year (Schievink, 2006;Schievink et al., 2007Schievink et al., , 2022)).SIH has a notable negative impact on the social and working life of affected patients as well as on health-related quality of life (Jesse et al., 2022).Although orthostatic headaches are the hallmark symptom of SIH, the condition gives rise to a range of clinical manifestations, frequently escaping accurate diagnosis (D'Antona et al., 2021;Schievink, 2000;Hoffmann and Goadsby, 2013;Jesse et al., 2024).
The first description of SIH dates back to 1948 (DISCUSSION ON INTRACRANIAL HYPOTENSION [Abstract], 1948).Although the disease may still be underdiagnosed, awareness about SIH in the medical community has increased.This has resulted in a surge in diagnosed cases and published literature about the condition.Some authors suggest that this is primarily attributable to advancements in medical imaging techniques.(Schievink, 2006) , (Fichtner et al., 2016) A quick search for the term SIH in the online meta database PubMed (National Library of Medicine, Bethesda, MD, USA) yielded 1154 results.Diagnostic imaging guidelines on the disease have recently been proposed (Cheema et al., 2023;Dobrocky et al., 2019;Pinto et al., 2023;Relationship of Bern Score;Brinjikji et al., 2021).The number of publications globally is increasing each year (Johnson et al.).Despite this, and the fact that extensive research on SIH is being carried out, no comprehensive bibliometric analysis of SIH articles has been conducted to date.
Objective quantification of authors and publication origins is essential for advancing coordinated research efforts and improving guidelines for diagnosis and treatment.Given that there are articles on SIH dating back more than 70 years, (DISCUSSION ON INTRACRANIAL HYPOTENSION [Abstract], 1948; Raskin, 1990) the analysis and quantification of existing research is long overdue.The aim of this article is thus to present a detailed synopsis of the current state of research articles published on this topic.With our bibliometric analysis on SIH we intend to simplify the comparison and comprehension of research on SIH from different countries.We want to stimulate the dialogue in the scientific community and facilitate efforts towards unification of guidelines.

Material and Methods
The following section headings are used in accordance with the latest PRISMA guidelines for systematic reviews (Page et al., 2021).

Study design
We conducted a bibliometric analysis on the changing trends of publications on SIH.The requirement for ethical approval was waived due to the study design.

Eligibility criteria and selection process
The inclusion and exclusion criteria were applied by author NG: Only articles available online on PubMed were included.For better comparability, and to rule out duplicates in different languages, only articles written in English were included.The following were excluded: i) articles without an abstract or access to the text, ii) article types "comment" and "erratum", and iii) articles for which access to credentials for the authors (e.g., the institutions to which they were affiliated) was not possible through the PubMed "affiliations" section or manually through full text access.Articles that were not available online on PubMed were beyond the scope of this study and were therefore excluded.

Information sources
The National Library of Medicine (NLM)'s PubMed served as the data source.We performed our search on April 1, 2022; therefore, the cutoffdate for the study was March 31, 2022.

Search strategy
We applied the following search terms: "spontaneous intracranial hypotension" or "spontaneous cerebrospinal fluid leak" or "cerebrospinal fluid hypovolemia" or "cerebrospinal fluid hypovolemia syndrome" or "hypoliquorrhea" or "spontaneous spinal cerebrospinal fluid leak".
We decided to multiply the results for the count of publications in 2022 by four to make that year's results comparable with the previous years.

Data collection process
Each article was manually screened on PubMed by NG, and the following information was extracted: Article title, date of publication, names of first author (FA), country of affiliation of FA, medical specialty of FA, institutional affiliation of FA, journal, number of citations in PubMed, publication type and the number of patients studies.Data regarding last authors (LA) is available upon request and can be found in Suppl.Tables 4 and 5 Data were entered in Microsoft Excel version 16.73, 2023 (Microsoft Corporation, Redmond, WA, USA).The number of citations was obtained using the information provided by PubMed in the section "cited by".
Additional data were downloaded externally to sort the articles further: The Journal Impact Factors were retrieved on 2 August 2022 via the Journal Citation Reports of Clarivate (Clarivate Analytics, Philadelphia, PA, USA) (Clarivate, 2022).The density of physicians per 10 000 population was derived from the World Health Organization (WHO, Geneva, Switzerland) on April 16th, 2023 (World Health Organization (WHO), 2023).The WHO only provides records of the density of physicians for each country.Therefore, the absolute number of physicians in each country was calculated by dividing the population of each country in 2021 by the number of physicians per 10 000 population reported by the WHO.The required population data were retrieved from Our World in Data, (Global Change Data Lab, England and Wales) on April 19th, 2023 (Roser et al., 2021).

Ranking of results
Authors, countries, and medical institutions were ranked according to the sum of their citation counts and occurrences on PubMed in our search.Unless otherwise stated, all our rankings apply to FAs and/or LAs.The density of publications per 100 000 physicians was calculated as follows: The count of FAs publishing articles on SIH from each country was divided by the overall population of physicians in that country.This ratio was then multiplied by 100 000 and sorted from the largest to smallest number.Data are shown as descriptive data.No additional statistical analyses were needed.

Calculation of growth rate
To calculate the average publication growth rate, we used the formula for compound annual growth rates (CAGR).
Where V(t n ) represents the sum of publications in 2021, V(t 0 ) the number of publications in 1983, and t n − t 0 the number of years.

Results
Our literature search resulted in a total of 1154 articles.After application of the above-mentioned exclusion criteria, 974 articles remained.

Analysis per year
From the data collected, a clear uptrend was evident.The first year for which a publication could be found online was 1983 and only one article was published that year.Almost forty years later this number had increased to 95 (2021).This was also the most productive year since the first publication.The uptrend is likely to continue although it is worth noting that there were less productive years too (see Fig. 1).The average yearly growth rate between 1983 and 2021 was 12.7%.
Along with the surge in publications the number of reported patients also increased.In 1983, the year of the first publication on PubMed, only one patient was studied; twenty years later that number had increased to 69.In 2013 there were 812 patients and in 2020 the number of patients had increased to 4230 (Fig. 1).

Ranking by publication type
Of the 974 publications, most (n = 570; 58.5%) were case reports (CR).Two hundred and six different journals published CR.As seen in Table 1, thirty-nine is the highest number of CR published in the same journal.Original articles (OA) accounted for 24.9% of all articles.
Ninety-four different journals published OA.Most of them (n = 28, 11.5%) were published in the American Journal of Neuroradiology (American Society of Neuroradiology, Oak Brook, IL, USA) and 14 (5.8%) each in Neurology (Wolters Kluwer on behalf of the American Academy of Neurology, Minneapolis, MN, USA), the Journal of Neurosurgery (American Association of Neurological Surgeons, Rolling Meadows, IL, USA) and Cephalalgia: An International Journal of Headache (Sage Publishing, Thousand Oaks, CA, USA).
Reviews were published in 62 different journals.The highest number of reviews published in the same journal was eight Current Pain and Headache Reports (Springer Science + Business Media, Berlin/Heidelberg, Germany).The first review on SIH was published in 1995.Since then, there has been an uptrend in the publication of reviews with the sum rising to nine in the first quarter of 2022.

Ranking by specialty
Most of the papers' first authors (FAs) affiliation was listed as a department or institute specializing in neurology (25.8%) or neurological surgery (25.7%).Radiology ranked third (19.6%).Overall, the FAs   of 9.9% of publications had a specialty that was not one of the most common ones, e.g., obstetrics and gynecology or pediatrics, or no clear allocation was possible.These cases were labeled "others" (Fig. 2).

Analysis per country
The countries of origin of the published articles were diverse with FAs' affiliations in more than 50 countries.The most represented are shown in Fig. 3.
Physicians in the USA contributed 33.8% of the total articles.The USA was responsible for 25% of all publications in 2001 and had increased its share to 43% in 2021.As shown in Fig. 3 and Suppl.Table 1, the highest number of publications originated from the USA, followed by Japan and Italy.Applying the correction for absolute number of physicians changed the distribution of publications (see Suppl.Table 2).After correction, South Korea had the highest density of published articles on SIH per physician, followed by Canada and Switzerland (Fig. 4).
An overview of the most common specializations in different countries can also be found in Suppl.Table 3.

North America and the United States
The USA and Canada accounted for 37.1% of all publications: 329 of the 974 articles originated from the USA (see Suppl.Table 1).The most frequently listed specialty regarding author affiliation in the US was neurosurgery which accounted for 33.4% of all FAs from the US publishing on SIH.Seventy-one of the FAs were based in California, eight in New York, and six in Minnesota.Eighty-one radiologists, mostly based in North Carolina (n = 24), 14 in Minnesota, and ten in California, participated in SIH research.Fifty-three neurologists did so, too.Fourteen of were based in Minnesota, five in Texas and four in Arizona.
The medical centers with which the highest numbers of FAs were affiliatedboth within the USA and globallywere Cedars-Sinai Medical Center in Los Angeles, CA, the Duke University Medical Center in Durham, NC, and the Mayo Clinic in Rochester, MN.

Asia
Twenty-five percent of the FA reported affiliations in Asian countries.Of the articles from Asia, Japan published the most.South Korea, Taiwan Republic of China (ROC), the People's Republic of China, and India followed.After rank correction, South Korea had the highest density of articles published on SIH per investigator.Japan and Singapore remained in the twenty countries with the highest publication density globally.Taiwan was ranked sixth before correction, but its rank could not be corrected as there was no data provided by the WHO.
The most common medical specialties of FAs publishing on SIH in South Korea were neurosurgery and anesthesiology (each accounted for 32.6% of all physicians).In Japan, most FAs were neurosurgeons (54.7%).In Singapore 66.7% were neurologists.In Taiwan, although not corrected, 60.5% were neurologists.

Europe
Fifty-five percent of countries in the top twenty were European.As shown in Fig. 3 and Suppl.Table 1, most publications originated from Italy, the United Kingdom (UK), France, Germany, and Switzerland.The densest European networks of SIH researchers can be found in Switzerland, Cyprus, and the UK.

Middle East
Turkey was the only Middle Eastern country in the top twenty.After correction, Turkey still led the ranking, with the highest density of SIH physicians in the Middle East.However, Oman, Qatar and Morocco were included in the corrected top twenty.Most of the authors publishing on SIH in Turkey were radiologists (47.2%).

Oceania
Australia ranked 19th.After rank correction, Australia moved up to tenth place, while New Zealand entered the top twenty in 18th place.2).

Ranking: most cited papers
The most cited publications are shown in Table 3.

Ranking: journals that carried the highest numbers of articles on SIH
The journals that published articles on SIH were sorted and ranked according to the number of papers on SIH (Table 4).Those that published more articles on SIH appear higher up in the Table .Most publications appeared in Headache: The Journal of Head and Face Pain (American Headache Society, Mount Royal, NJ, USA).

Ranking: highest impact factor (IF)
Sorting FAs by their publications in the journals with the highest Impact Factor (IF), resulted in the ranking presented in Table 5. Results are shown for journals with an IF above twelve.

Discussion
The data collected in this study confirms the exponential increase in SIH-related research.Articles on the topic date back over 70 years [9,17] and there has been a surge in publications over the past 40 years.There was an impressive increase in the average annual growth rate of 12.7% during the period from 1983 to 2021.The number of SIH patients reported has increased from single individuals to more than 4000 per year (2020).Of the many articles published on the topic, most were CR.
Articles originated from more than 50 different countries.However, more than a third of all publications originated from the USA, which accounted for most publications followed by Japan and Italy.Although, the USA was responsible for 25% of all publications in 2001, 20 years later, in 2021, almost every second publication on SIH stems from an  author affiliated with an institution in the USA.The highest density of SIH researchers was observed in South Korea, Canada, and Switzerland.
The "most productive medical centers", however, are based in the USA.Neurology, neurosurgery, and radiology were the most common specialties of FAs globally.While in Japan, the USA and Switzerland it was mostly neurosurgeons who were involved in SIH research, South Korea had equal amounts of anesthesiologists contributing.Most SIH publications from European countries had neurologists as FAs.Turkey had only radiologists.

Table 3
The ten most cited articles on SIH in descending order with names of first authors.

Table 4
The twenty journals that carried the highest numbers of articles on SIH listed with their impact factor.

Findings in context
The number of publications on SIH has increased exponentially in recent decades.To avoid bias, however, it is crucial to view this rate in context of the average annual growth rate in other science domains, which is 4-5% over the same period (Bornmann et al., 2021).Given this caveat, it can be concluded that the number of publications on SIH has increased disproportionately compared to the scientific literature in general.
A potential factor contributing to the increase in publications might be the shift in focus towards research on SIH.Many medical centers now have specialized teams studying this disease and publishing guidelines (Cheema et al., 2023).There are also international organizations offering support on the topic (Abstracts, 2022).Since the publication of two frequently cited reviews, SIH has gained even more attention (Schievink, 2021;Scott and Davenport, 2014;Ducros and Biousse, 2015;Dobrocky et al., 2022;Bornmann et al., 2021).
Ranking the publications might itself be controversial as there is no direct causality between the quality of an article and the number of citations.Some publications are cited significantly more often than others.One reason might be the usefulness of a specific publication to the community.Garfield Garfield and Wallin (2005) proposed that, as the number of citations increases, the possibility of the paper becoming of more practical use to the respective scientific community increases as well.Furthermore, it is important to view our results in the context of the ever-expanding global publication landscape (Johnson et al.).
While many articles are being published on SIH, most of them are CR.We assume that because research is increasing and meetings on SIH are being held globally, SIH has become more of a "hot topic" than ever and is increasingly included in the differential diagnosis list of neurologists, neuroradiologists, and neurosurgeons.Thus, more cases are being diagnosed and more cases are being reported.
Since the disease is quite new in terms of publications and guidelines, everyone involved might be eager to participate in the search for new pathophysiological interrelationships.
It is noteworthy, however, that many countries have published only a few articles on SIH and some populous countries have produced no publications on SIH at all.It is very plausible that the diagnosis is made too rarely in these countries.
Interestingly, the distribution of medical specialties of SIH researchers is not the same in every country.We can only speculate about possible reasons for these differences.One plausible hypothesis is that the investigators who originally diagnosed SIH were more likely to have had a surgical background, as the initial diagnosis of SIH may have been made clinically.After the discovery of SIH by surgeons, physicians with non-surgical specialties may have increasingly become involved in the study of SIH.In any case, it is significant that there is a great variability of specialties, and that interdisciplinary collaborations and teamwork are important.

Strengths and limitations
Although research on SIH has been changing significantly in recent years, no comprehensive bibliometric analysis has yet been conducted.Our work includes a complete survey of the publications on SIH since the  Nevertheless, our analysis has some limitations: While PubMed is a widely used platform, it is likely that some articles on SIH were outside the scope of our research.There are two reasons for this: First, we only looked at digitally available papers.While some articles were published before 1983, we did not search any physical libraries.Secondly, only articles on PubMed were studied.We consider PubMed to be the most popular library for medical publications.Therefore, we also assume that it is the most comprehensive.Nevertheless, some publications might have been published exclusively on other platforms and thus would not have been included in our research.
We excluded non-English articles, which could have led to some articles being omitted.However, as English is widely used in science, we believe this bias to be negligible.Besides this, not every journal has a Journal Impact Factor.Only 715 of the 974 articles could be linked with journals for which impact factors were available (731 of articles for the LAs ranking).
Even though most countries and regions are represented in the WHO, data on physician density are missing for some countries.This is the case for Taiwan and thus the rank of this country could not be corrected.

Conclusions
Research and publications about SIH have increased exponentially over the past 40 years.With new treatment options awaiting discovery, research on SIH shows great promise.In parallel with the growth in publications, there is a growing number of SIH researchers.This paves the way for more international cooperation on research, treatment, and patient education.With this article we would like to encourage largescale collaborations and we firmly believe that research on SIH can be both facilitated and nurtured.By providing information on the most active centers and researchers, new options for transnational partnerships arise.

Fig. 1 .
Fig. 1.Annual overview of patients included in the SIH research analyzed and count of articles published on SIH.Numbers for 2022 are estimates, based on values for the first quarter of multiplied by four.Doublings are possible.The asterisk indicates the year in which the most cited SIH paper was published.

Fig. 2 .
Fig. 2. The most common medical specialties of first authors publishing articles on SIH between 1983 and the first quarter of 2022.Radiology and neuroradiology were merged due to unclear differentiation.

Fig. 3 .
Fig. 3. Global distribution of published articles from first authors on SIH between 1983 and 2022.

Fig. 4 .
Fig. 4. Publications on SIH per 100 000 physicians.Data is based on first author-count and correction with the number of physicians per 10 000 population provided by the World Health Organization and the population count of each country provided by Our World in Data.

Table 1
Distribution of most popular publication type (case report) amongst journals.

Table 2
Institutes with at least five SIH publications listed in decreasing order of number of affiliated first authors contributing to publications between 1983 and 2022.

Table 5
SIH publications sorted according to descending order of journal impact factor with first author, journal title, and publication type.

Table 5
(continued ) Review; OA = Original Article; CR = Case Report.*Impactfactors below twelve are not shown.firstpublication on PubMed.All publications included were individually screened to ensure high data quality.As a university hospital intensively engaged in interdisciplinary research in the field of SIH, we were able to collaborate with other departments and optimize the design of our study.