Comprehensive bibliometric analysis of pharmacotherapy for bipolar disorders: Present trends and future directions

Abstract

BACKGROUND

Bipolar disorder (BD) is a severe mental illness characterized by significant mood swings. Effective drug treatment modalities are crucial for managing BD.

AIM

To analyze the current status and future trends of global research on BD drug treatment over the last decade.

METHODS

The Web of Science Core Collection database spanning from 2015 to 2024 was utilized to retrieve literature related to BD drug treatment. A total of 2624 articles were extracted. Data visualization and analysis were conducted using CiteSpace, VOSviewer, Pajek, Scimago Graphica, and R-studio bibliometrix to identify research hotspots, key contributors, and future trends.

RESULTS

The United States, China, and the United Kingdom have made the most significant contributions to research on BD drug treatment and formed notable research collaboration networks. The University of Pittsburgh, Massachusetts General Hospital, and the University of Michigan have been identified as the major research institutions in this field. The Journal of Affective Disorders is the most influential journal. A keyword analysis revealed research hotspots related to clinical symptoms, drug efficacy, and genetic mechanisms. A citation analysis identified the management guidelines published by Yatham et al in 2018 as the most cited paper.

CONCLUSION

This study provides a detailed overview of the field of BD drug treatment, highlighting key contributors, research hotspots, and future directions. The study findings can be employed as a reference for future research and policymaking, which may enable further development and optimization of BD pharmacotherapy.

Key Words: Bipolar disorder; Drug treatment; Bibliometric analysis; Visualization; Mental disorder

Core Tip: This study employs advanced bibliometric tools, including CiteSpace, VOSviewer, and the bibliometrix package in R, to analyze the evolution of pharmacotherapy for bipolar disorder over the past decade. It identifies key contributors, research hotspots, and future trends, emphasizing significant international collaborations, particularly among institutions in the United States, China, and the United Kingdom. The findings highlight key areas such as clinical symptoms, drug efficacy, and genetic mechanisms, providing valuable insights for scholars and policymakers. This work offers clear guidance for innovation in bipolar disorder pharmacotherapy and serves as a critical reference for future research.

INTRODUCTION

Bipolar disorder (BD) is a mental illness characterized by severe mood swings, including episodes of mania and major depression, and it typically follows a recurrent course. BD is commonly categorized into two subtypes: Bipolar I disorder (BD-I) and bipolar II disorder (BD-II)[1]. In BD-I, there should be at least one manic episode, which involves elevated mood, emotional instability, impulsivity, and excessive energy, with or without one or more episodes of major depression. BD-II is characterized by at least one major depressive episode accompanied by hypomania, without full-blown manic episodes[2,3]. The lifetime prevalence of BD is approximately 1%-2%, with two peak periods of onset at ages 15-24 and 45-54[4]. Although the exact etiology of BD remains unclear, it is generally believed to result from an interplay between genetic and environmental factors. Multiple studies have shown that first-degree relatives of BD probands have a 7.9-fold higher risk and second-degree relatives have a 3.3-fold higher risk, compared to families without any probands. Additionally, twin studies indicate a heritability of 70%-80% for BD[5,6]. Social and environmental factors significantly impact the development of BD, with childhood abuse (particularly emotional abuse), stressful life events (such as disability or divorce), and substance abuse (including alcohol and cannabis) increasing the risk of developing BD[7-9].

Currently, the treatment of BD primarily involves pharmacotherapy combined with various psychological interventions, which are evaluated and administered in three phases: Manic, depressive, and preventive. The first-line medications for mania are antipsychotics, such as aripiprazole, risperidone, and quetiapine. Severe depression can be effectively treated with quetiapine, cariprazine, lurasidone, and electroconvulsive therapy. For preventing BD relapse, lithium is the most efficient first-line treatment available[10,11]. However, as the use of certain antidepressants alone may increase the risk of manic episodes, combination therapy is commonly used in clinical practice. Various studies have shown that a combination of risperidone, olanzapine, and lithium with valproate is more effective in preventing manic episodes than monotherapy[10,12]. However, pharmacotherapy often increases the risk of metabolic syndrome, a common clinical metabolic disorder characterized by abdominal obesity, hypertension, high blood glucose levels, elevated triglycerides, and low levels of high-density lipoprotein cholesterol[13,14]. The presence of metabolic syndrome presents new challenges in the treatment of BD. Hence, the development of innovative treatment modalities remains a key focus for future research.

Bibliometrics is the study of the development of scientific research and academic communication through the quantitative analysis of literature and citation data. It is an important tool to evaluate the credibility and impact of academic research[15]. As it involves quantitative, multi-dimensional analysis and long-term trend observation of data, it reflects the status of communication and collaboration within the academic community, identifies new research frontiers, hot topics, and academic trends, and provides researchers with directions and inspiration for working in their respective fields[16]. Although a substantial body of research on the pharmacological treatment of BD is available, there is a lack of bibliometric analyses specifically focused on BD and its pharmacological treatments. This study uses bibliometric methods to conduct a global analysis of the current state and future trends in BD pharmacological treatment over the last decade. It aims to identify research hotspots and future directions in the field of BD, in order to provide a more comprehensive and in-depth understanding of its pharmacological treatment.

MATERIALS AND METHODS

We conducted a comprehensive search of the literature on BP using the Web of Science Core Collection database, spanning from 2015 to 2024. The search formula was set as (((TS = (“bipolar disorder” OR “bipolar I disorder” OR “bipolar II disorder”)) AND (TS = (“medication” OR “drug treatment” OR “drug therapy” OR “pharmacotherapy”)) AND (PY = (2015-2024))). We were able to identify 3648 documents, from which we excluded 85 non-English articles and 939 documents not classified as “article”. Consequently, we ultimately included 2624 articles (Figure 1). The filtered data were exported as text files and visualized using CiteSpace (version 6.1.R6), VOSviewer (version 1.6.20), Pajek (version 32.5.18), Scimago Graphica (version 1.0.39.0), the bibliometrix package in R-studio, and Excel (Office 2021).

Figure 1 Flow diagram of the literature search and selection.

CiteSpace

CiteSpace is a visualization analysis tool that operates within a Java environment. It leverages bibliometrics, co-occurrence analysis, and cluster analysis to study and visualize hotspots and research frontiers within specific knowledge domains[17]. First, we imported the data into CiteSpace to create citation burst charts for the top 15 cited articles and keyword burst charts for the top 25 keywords appearing in the included articles. “Year” indicates the starting year of the citation or keyword appearance, “Strength” represents the intensity of the citation or keyword occurrence, “Begin” denotes the starting year of the citation or keyword burst, and “End” indicates the ending year. Each line segment represents a period, with red bars indicating the duration of a strong citation burst[18,19].

VOSviewer and Pajek

VOSviewer is an application that runs on a Java environment and can be used to create, visualize, and explore maps based on any type of network data, primarily for bibliometric network analysis[20,21]. We utilized VOSviewer to generate visualizations of author collaboration networks, institutional collaboration networks, overlay maps of research field distributions, and keyword co-occurrence networks. Pajek is a powerful tool for adjusting various complex network diagrams. To enhance the clarity of nodes, concentrate similar clusters, and improve the overall aesthetics of the scientific network diagrams, we combined Pajek with VOSviewer to adjust the visual effects of scientific network diagrams[22]. In the adjusted diagrams, we can intuitively see that nodes of the same color belong to the same cluster and exhibit a certain degree of collaboration. The size of the nodes is related to the frequency of the occurrence of the relevant content, while the lines between the nodes indicate a certain degree of correlation[23-25].

Scimago Graphica

Scimago Graphica is a user-friendly, free, no-code tool that does not rely on a Java environment and allows for the creation of complex visualizations through simple drag-and-drop interactions[26]. We used Scimago Graphica for assessing global collaborations, analyzing country citation strength, and comparing link strength. The upper right corner of the diagrams lists labels explaining the information represented by the nodes, lines, and colors, making them simple, clear, and aesthetically pleasing.

R-studio bibliometrix package

The R-studio bibliometrix package is a bibliometric analysis software tool developed by Massimo Aria and others. It inputs a code within the R language environment to run bibliometrix, an open-source tool used for a comprehensive scientific mapping analysis of scientific literature[27].

RESULTS

Analysis based on countries/regions

A global map of publications related to BD and its pharmacotherapy is presented (Figure 2A). The size of the circles indicates the volume of publications, while different colors represent various country clusters. Over the past decade, North America, East Asia, and Western Europe have made significant contributions to advances in BD pharmacotherapy. Australia has also played a notable role in this field. All countries involved in this research field are categorized into eight major clusters based on the extent of their collaboration. The green cluster, led by the United States, has the highest productivity, followed by the blue cluster, led by China. We further examined countries/regions based on the strength of their collaborative links and revealed the total citation frequency of different countries (Figure 2B). The size of the circles represents the strength of a country’s collaborative links, while the color intensity indicates the total citation frequency. A positive correlation was noted between the size of the circles and the intensity of the color. The United States leads in both the collaborative link strength and total citation frequency, followed by the United Kingdom, Germany, and Canada. In total, 20 countries were found to have a total citation frequency exceeding 31000.

Figure 2 Analysis based on countries/regions. A: Network visualization of global research collaboration; B: Network visualization of national collaboration; C: Network visualization of author collaboration; D: Network visualization of institutional collaboration. The nodes represent countries, authors, or institutions, with node size indicating the number of publications. The thickness of the connecting lines reflects the strength of collaboration between countries, authors, or institutions. Different colors denote distinct clusters.

Analysis based on authors and institutions

This study analyzes the collaboration networks and productivity of authors through co-authorship mapping (Figure 2C). The size of the circles represents the number of publications. Different colors denote different clusters of authors. A total of 133 authors with more than five publications each were grouped into over ten author clusters based on their collaborations. The dark blue cluster, led by McInnis MG and Nierenberg AA, is the primary author cluster. The red cluster led by Vieta E and the dark green cluster led by Correll CU are also identified as key author clusters in the field over the last decade. 120 institutions with more than ten publications each were grouped into seven main institutional clusters based on their collaborations (Figure 2D). The yellow cluster, led by the University of Pittsburgh, has been the dominant institutional cluster in this field over the past decade. This cluster comprises 19 institutions, including, but not limited to, Massachusetts General Hospital and the University of Michigan. The purple cluster led by the University of Toronto and the green cluster led by Harvard Medical School also demonstrated exceptional productivity.

Analysis based on journals and research areas

Table 1 lists the top ten journals in the field over the past decade based on publication volume, total citation frequency, and co-citation frequency. The publication volume indicates the productivity of a journal, the total citation frequency reflects the quality and level of attention a journal receives, and the co-citation frequency shows the relevance of journal content within the field[28]. The top three journals by publication volume are the Journal of Affective Disorders (documents = 270), Bipolar Disorders (documents = 79), and Frontiers in Psychiatry (documents = 77). The first two journals are also among the top three for total citation frequency and co-citation frequency. Over the past decade, five main research directions, represented by red, blue, green, purple, and yellow clusters, have emerged in the fields of BD and pharmacotherapy. The labels indicate the main content of these research directions. Nodes of the same color belong to sub-disciplines or research directions within that field (Figure 3A). The red cluster represents “biology and medicine”, blue represents “chemistry and physics”, green represents “psychology and social science”, purple represents “engineering & mathematics”, and yellow represents “ecology and environmental S&T”. “Biology and medicine” and “psychology and social science” are the primary research directions in this field.

Figure 3 Analysis based on journals and research areas. A: The cluster network visualization of the research field uses different colors to represent distinct clusters; B: The citation burst graph highlights periods of intense citation activity, with red segments indicating when citations are in a burst state.

Table 1 The top 10 of journals, cited journals and co-cited journals with the most counts.

	Documents, journal	Documents, counts	Total citations, journal	Total citations, counts	Co-citations, journal	Co-citations, counts
1	Journal of Affective Disorders	270	Journal of Affective Disorders	4267	Journal of Affective Disorders	5901
2	Bipolar Disorders	79	Bipolar Disorders	2380	American Journal of Psychiatry	4453
3	Frontiers in Psychiatry	77	JAMA Psychiatry	1549	Bipolar Disorders	4348
4	Psychiatry Research	67	Journal of Clinical Psychiatry	974	Journal of Clinical Psychiatry	3753
5	BMC Psychiatry	64	BMC Psychiatry	855	Archives of General Psychiatry	2987
6	Journal of Clinical Psychiatry	47	Scizophrenia Research	803	Biological Psychiatry	2735
7	Journal of Psychiatry Research	45	Psychological Medicine	789	British Journal of Psychology	2527
8	Scizophrenia Research	41	European Neuropsychopharmacology	741	Schizophrenia Bulletin	2406
9	Psychological Medicine	38	Psychiatric Services	728	Molecular Psychiatry	1841
10	ACTA Psychiatrica Scandinavica	38	Psychiatry Research	679	Psychiatry Research	1728

Analysis based on citations and literature

Deep interpretation and analysis of citations and literature can help scholars identify the most prominent research directions and future hotspots and trends in the field. Supplementary Table 1 lists the top ten most cited papers in the field over the past decade. The most frequently cited paper is the 2018 guidelines on the management of BD patients by Yatham et al[29] published in Bipolar Disorders, with a total of 933 citations. This is followed by a systematic review and meta-analysis of the field by Vancampfort et al[30] published in World Psychiatry in 2015, with a total of 790 citations. Classic papers tend to have sustained citation data, while flash-in-the-pan articles are cited frequently but only for a limited time. Therefore, understanding the flash-in-the-pan articles at different periods can enhance our understanding and knowledge of the frontiers and hotspots in a particular field[31]. We used CiteSpace to create a burst detection map of citations and identified the 15 most significant burst articles in the field (Figure 3B). The first burst article appeared in 2011, authored by Merikangas et al[32]. Most of the citation bursts occurred in 2018 and later, with the strongest citation burst associated with the guidelines by Yatham et al[29]. Currently, seven papers are still in a burst phase. These seven can be defined as the current flash-in-the-pan articles in the field.

Analysis based on keywords

Keywords reflect hotspots within a field and their interconnections[33]. This study used VOSviewer to identify 250 keywords from 2624 documents, each with an occurrence strength of at least 20, and plotted a keyword network clustering diagram (Figure 4A). The 250 keywords were divided into five clusters, with the size of the circles representing the frequency of keyword occurrences. Keywords belonging to the same research direction were marked with the same color. The red cluster represents “clinical symptoms and health research”, with high-frequency keywords, such as “depression”, “comorbidity”, and “medication adherence”. The green cluster represents “pharmacotherapy and efficacy research”, with high-frequency keywords, such as “lithium”, “medication”, and “antipsychotics”. The blue cluster represents “gene expression and mechanism research”, with representative keywords like “abnormality”, “expression”, and “inflammation”. The yellow cluster represents “age factors and differences research”, with high-frequency keywords, such as “adolescent”, “children”, and “age”. The purple cluster represents “risk factors and epidemiology research”, with representative keywords like “risk”, “mortality”, and “metabolic syndrome”.

Figure 4 Analysis based on keywords. A: Keywords clustering visualization. The circle and label together form a node, where the size of the circle is directly proportional to the frequency of the keyword’s occurrence. The thickness of the lines connecting the circles reflects the strength of the relationship between the keywords. Nodes of different colors represent distinct clusters; B: Keywords intensity visualization timing overlay. The circle and its label represent a node, with the circle’s size being positively correlated with the frequency of the keyword’s occurrence. The color of each circle is indicated by the gradient in the lower right corner, reflecting the average year of occurrence.

By overlaying time on the keyword clustering diagram, one can gain insights into changes in research directions within a field, helping scholars explore cutting-edge technology and future trends in that domain. As shown in Figure 4B, colors can reveal the average years of occurrence for different keywords, with earlier keywords appearing in blue, such as “unipolar depression”, “metabolic syndrome”, and “risperidone” and later keywords appearing in yellow, such as “psychiatry”, “genome-wide association”, and “1^st-episode psychosis”. The frequent appearance of keywords in a specific period is known as a citation burst, often indicating research inclinations and concentrated trends in a specific period within a field. Figure 5 lists the top 25 keywords with the strongest citation bursts lasting at least one year. “Primary care” is the keyword with the highest sustained attention. Keywords like “primary care”, “trend”, “weekly symptomatic status”, and “service” are currently experiencing citation bursts.

Figure 5 Top 25 keywords with the strongest citation bursts. The keyword burst chart shows periods of high activity, with red indicating when a keyword is in a burst phase.

DISCUSSION

Literature trends

The significant productivity of North America, East Asia, and Western Europe highlights their extensive involvement in this research field. Australia has also made a substantial impact in this area. To emphasize the collaborative nature of international research in this field, we grouped the countries into eight major clusters. The United States-led green cluster is the most important one, highlighting the dominance of the United States in this research field because of its highly extensive scientific resources and research infrastructure and its significant contribution in both the quantity and quality of publications on BD pharmacotherapy. There is a positive correlation between the intensity of international collaboration and citation frequency (Figure 2B). The United States is a global leader in both collaboration intensity and total citations, followed by the United Kingdom, Germany, and Canada. Hence, it can be suggested that a highly collaborative research environment often yields more impactful studies. This insight is crucial for understanding the dynamics of global research collaboration and its impact on scientific progress. It further indicates that transnational collaboration among countries not only facilitates the sharing of resources and the leveraging of complementary strengths, but also significantly expedites the creation and spread of new knowledge, thereby injecting a potent impetus into the innovation of BD treatments.

Authors and institutions play key roles in advancing research. Figure 2C shows more than ten collaboration clusters comprising 133 authors with over five publications each, with the dark blue cluster led by McInnis MG and Nierenberg AA being the most significant. These clusters highlight the key researchers driving development in this field. Similarly, an analysis of institutions showed that the yellow cluster led by the University of Pittsburgh, with Massachusetts General Hospital and the University of Michigan as the other participants, is the dominant institutional cluster (Figure 2D). This result underscores the importance of institutional collaboration in promoting research productivity and impact. Inter-institutional collaboration empowers researchers from diverse fields to collaborate closely in exploring BD treatment. By integrating multidisciplinary perspectives, it stimulates the creation of innovative treatment methods and strategies. This collaborative approach not only boosts research efficiency but also markedly enhances the quality and influence of the research outcomes.

The analysis of journals and research areas provides additional insights into the structure of the field. Table 1 lists the top journals ranked by publication volume, total citations, and co-citations. The Journal of Affective Disorders leads all these categories, underscoring its critical role in disseminating high-quality and impactful research on the pharmacotherapy of BD. Figure 3A displays five major research directions, with “biology and medicine” and “psychology and social science” being the most important ones, indicating the multidisciplinary nature of the field. This diversity suggests that extensive interdisciplinary connections and exchanges are necessary to understand and treat BD. Therefore, future research should focus on interdisciplinary collaboration, integrating insights from psychology, neuroscience, genetics, and psychiatry. This approach will drive innovation and expedite the application of research findings in clinical practice.

Citation and literature analyses further refine our understanding of key contributions and emerging trends in the field. Supplementary Table 1 lists the most cited papers of the past decade. The 2018 paper by Yatham et al[29] on management guidelines for patients with BD was identified as the most cited one. Systematic reviews and meta-analyses are also widely used in this field, as illustrated by the 2015 paper of Vancampfort et al[30], highlighting the reliance of this field on comprehensive evidence. The identification of citation bursts, especially the most notable burst related to Yatham et al’s guidelines[29], indicates shifts in the research focus and the emergence of new frontiers in BD research. These seven papers still experiencing citation bursts indicate current hot topics and areas of intense research activity. In 2018, Vieta et al[34] conducted a comprehensive discussion on BD, delving into its clinical symptoms and assessment methods and exploring both pharmacological and psychosocial treatment approaches. In the same year, Jawad et al[35] investigated the non-adherence to BD treatment medications, analyzed its causes and impacts on patients, and emphasized the necessity and methods for improving medication adherence. In 2018, the Canadian Network for Mood and Anxiety Treatments and the International Society for Bipolar Disorder collaborated to release an updated version of the guidelines for managing patients of BD, incorporating the latest advancements in treatment[29]. In 2020, McIntyre et al[36] combined epidemiological data to detail the connections and differences between BD-I and BD-II by thoroughly studying the pathogenesis and treatment strategies of BD, thereby providing a basis for clinical diagnosis. Carvalho et al[37] elucidated the genetic and neurobiological characteristics of BD. Rhee et al[38] reviewed the trends in BD pharmacotherapy from 1997 to 2016 and compared the efficacy of second-generation antipsychotics, lithium, and mood stabilizers, laying the groundwork for future drug research. In 2021, Mullins et al[39] conducted a genome-wide association study on over 40000 patients of BD and identified 64 BD-related genetic loci, deepening the understanding of the biological etiology of BD and providing new perspectives for future treatments.

Research hotspots and frontiers

We divided 250 keywords into five clusters (Figure 4A). The red cluster (clinical symptoms and health research) reveals that the prevalence of metabolic syndrome, obesity, smoking, and type 2 diabetes is higher among patients with BD, potentially leading to their premature death[40]. Research on the clinical treatment of BD is still mainly focused on pharmacotherapy. A national study on the consistency of multimodal treatment guidelines for BD indicated that, compared to pharmacotherapy, psychosocial therapy has seen only a limited implementation in daily clinical practice for BD. Hence, more effort is needed to implement non-pharmacological guideline recommendations for BD[41]. The Taiwanese Society of Biological Psychiatry and Neuropsychopharmacology provided the latest interpretations of the consensus on biological treatment for acute, maintenance, and mixed phases of BD in 2023, discussed various treatment plans for BD, evaluated the related efficacy, and summarized extensive research evidence and clinical experience to recommend grades for various clinical treatment plans of BD[42]. The International Society for International Society for Bipolar Disorder conducted an anonymous survey in 2023 on the preferences and attitudes of clinicians toward using lithium for the global maintenance treatment of BD. The survey revealed that these preferences and attitudes may be influenced by patients’ beliefs and clinical environments[43]. The formulation of various clinical guidelines and the conduct of different surveys continue to drive the development of clinical treatment modalities for BD, providing more comprehensive and advanced evidence and concepts for both clinicians and patients. We believe that more extensive clinical guidelines and clinical surveys will be carried out in the future.

The green cluster (drug therapy and efficacy research) indicates that pharmacotherapy is currently the optimal solution for patients with BD. Therefore, it has received significant attention from scholars and clinicians. Researchers in this field tend to extensively study pharmacotherapy[41]. A meta-analysis aimed at understanding BD treatment practices indicated that anticonvulsants, second-generation antipsychotics, and antidepressants are the most prescribed drugs for mood stabilization. However, because of data gaps, it is not possible to conduct cross-regional and cross-ethnic studies[44]. In the past decade, lithium and valproate remained the pharmacological foundations for BD treatment. However, the patterns of BD pharmacotherapy are continuously evolving. To evaluate the advantages and disadvantages of this evolution, Lähteenvuo et al[45] conducted a detailed cohort study on the actual effectiveness of antipsychotics and mood stabilizers in BD. They not only explored the relative efficacy and safety of antipsychotics and mood stabilizers, but also specifically studied first-episode patients, further revealing the high safety and stability of lithium treatment. According to current scientific knowledge in this field, the first-line treatment drugs for BD mainly include mood stabilizers (such as lithium), anticonvulsants (such as valproate and lamotrigine), and atypical antipsychotics (such as quetiapine, aripiprazole, asenapine, lurasidone, and cariprazine)[40]. Some novel drugs or therapies have also been reported, such as levothyroxine and neurosteroids[46,47]. However, the adverse reactions and severe side effects of these drugs still need further resolution[48].

The blue cluster (gene expression and mechanism research) shows the significant progress made in the study of gene expression and mechanisms in the pharmacotherapy of BD in recent years. Lithium, the most classic treatment drug for BD, has been shown to regulate the survival and function of nerve cells by affecting the glycogen synthase kinase 3β pathway[49]. Additionally, gene expression studies on BD revealed the specific mechanisms through which drugs act at the neuronal level. For example, antipsychotics can influence the plasticity and survival of nerve cells by regulating the synthesis and release of neurotransmitters. They can also regulate the expression of genes related to neuroinflammation, thus exerting anti-inflammatory effects[50]. Additionally, future studies should delve deeper into the relationship between gene expression and drug response, aiming to develop personalized treatment plans tailored to individual patients’ genetic profiles and specific conditions. This will optimize clinical outcomes and improve patient satisfaction with treatment.

The yellow cluster (age factors and differentiation studies) confirms that BD patients of different age groups respond differently to drugs. Young patients (under 18 years of age) show different drug efficacy and side effects from those of adult patients when using antidepressants. Research in this direction often relates to hormones and inflammation markers, as biological age and sex significantly affect gonadal hormones, stress hormones, and inflammation markers[51]. For young patients, especially those experiencing their first episode, a combination of mood stabilizers and antipsychotics is usually adopted. For older patients, monotherapy, such as lithium or lamotrigine alone, is often recommended to reduce the risk of drug interactions and side effects[52,53]. Given the differences in drug efficacy and side effects among BD patients of various age groups, future research should focus on conducting age-specific studies. This will allow for the development of tailored drug therapies and treatment models that cater to the unique needs of patients across different age brackets, ultimately enhancing clinical outcomes.

The purple cluster (risk factors and epidemiology research) shows that the risk factors for BD mainly include genetic, biological, and psychosocial factors. Among these, genetics is the most significant risk factor, with surveys indicating that BD has a high familial aggregation[5]. Biological factors are mainly associated with neurotransmitter dysfunctions (such as serotonin, norepinephrine, and dopamine) and structural changes in the brain. Additionally, life stress, stressful events, and negative life experiences can trigger or exacerbate the condition[54]. Epidemiological studies have indicated that the global lifetime prevalence of BD is approximately 2.4%, with the onset typically occurring around the age of 25 years. BD has a high relapse rate (up to 90%), especially if not properly treated after an acute episode[32,55]. Hence, we call for future epidemiological researchers in this field to focus more on interdisciplinary cooperation, including efforts from psychiatry, psychology, neuroscience, and genetics, to comprehensively understand and treat BD.

Earlier appearing keywords include “unipolar depression”, “metabolic syndrome”, and “risperidone”, while later appearing keywords include “psychiatry”, “genome-wide association”, and “1^st-episode psychosis” (Figure 4B). Hence, it can be suggested that past research tended to focus on comparing BD with unipolar depression, metabolic issues in BD patients (such as obesity and diabetes), and the application of early drugs such as risperidone[56-58]. In contrast, recent research focuses include the overall development of the impact of psychiatry on BD treatment, the identification of genes associated with BD through genome-wide association study, and first-episode psychosis patients[59-61].

Keywords such as “primary care”, “trend”, “weekly symptomatic status”, and “service” are currently experiencing a surge (Figure 5), indicating that the field is undergoing significant developments and changes. The frequent appearance of the keyword “primary care” reflects that BD treatment is increasingly being integrated into primary healthcare systems. Moreover, the integration of BD treatment into primary care systems should be a priority for future research and clinical practice. Studies should explore effective strategies for diagnosing and managing BD in primary care settings, enabling earlier identification and treatment of patients, thereby reducing the disease burden and healthcare costs[62]. The surge of “trend” and “weekly symptomatic status” suggests that researchers are closely monitoring the epidemiological trends and treatment effects of BD. Future research should focus on large-scale, long-term epidemiological studies for an improved understanding of the incidence, treatment patterns, and long-term prognosis of BD. This also highlights the importance of real-time monitoring of BD symptoms and treatment effects. In the future, researchers in this field should develop and apply digital health tools to track the symptom changes and drug responses of patients in real time. The data so collected may help in devising better treatment plans and improved treatment outcomes[45,63]. Additionally, the surge of the keyword “service” indicates that more attention is being paid to the research of integrated care models, which involves the merging of mental health services with other medical services to provide comprehensive care to patients with BD[64,65].

Pharmacological treatment of BD-I/BD-II

BD-I typically involves manic episodes, whereas BD-II involves at least one major depressive episode along with hypomania[66]. Both BD-I and BD-II have chronic courses, although the former often manifests more severely and usually requires hospitalization[67]. BD-II also presents with more severe characteristics, such as more frequent mixed episodes and more severe complications[68]. In addition, numerous studies have confirmed that, compared to BD-I, BD-II carries a higher risk of suicide, which may be related to the longer depressive cycles and mixed states in BD-II[69,70]. Therefore, there are significant differences in the pharmacological treatment of BD-I and BD-II.

For BD-I patients, the primary focus is often on treating mania, whereas for BD-II patients, antidepressant treatment is predominant[71,72]. For BD-I, lithium is the most common mood stabilizer. For BD-II depression, lamotrigine is the most common mood stabilizer, as it can also prevent depressive episodes[73]. Terao et al[67] showed that lamotrigine is more effective in stabilizing mood and preventing relapse than in alleviating mania in BD-I. It is also more effective for the long-term treatment of BD-II[67,74]. Hence, it can be suggested that lamotrigine is more suitable for the treatment of BD-II than it is for BD-I.

There are also differences in the combination of drugs used to treat BD-I and BD-II. In BD-I treatment, the combination of mood stabilizers and antipsychotics is most common, such as the use of olanzapine and lithium combined with valproate, while in BD-II, the combination of mood stabilizers and antidepressants is most common, such as the combined use of lamotrigine and quetiapine[10,75-77]. In contrast to BD-I, research on the efficacy of pharmacological treatment for BD-II is scarce. Since BD-II has a higher frequency of onset than BD-I, many drugs used for BD-I cannot be used alone for BD-II. In addition, the guidelines provide limited recommendations specifically for BD-II[29]. Swartz and Thase[78] and Datto et al[79] have evaluated the pharmacological treatment of BD-II and shown that quetiapine can effectively improve depressive and anxiety symptoms in BD-II, although the response is slower compared to that for BD-I. In addition, pramipexole and lithium show promise as second-line drugs for treating BD-II depression[80]. Hence, it is clear that current research on the pharmacological treatment of BD-II is highly limited and lacks effective clinical evidence to verify the efficacy and safety of related drugs.

Limitations

Due to differences in data formats, we only included studies from the Web of Science. Therefore, not all studies may be included. In addition, this research is limited to a single region, and therefore has only been conducted using English literature. It is possible that incorporating non-English literature or research from other regions could yield different results. Aside from this, bibliometric analysis, the quantity of key annotated bibliographies, citation normalization, and the qualitative assessment of bibliographic content, innovation, and impact are crucial but have not been the focus of further investigation. Despite its importance, this remains an area for future research.

CONCLUSION

This study utilized bibliometric methods to analyze the global research trends and current status of drug treatment for BD over the past decade, providing valuable insights for future research in BD pharmacotherapy. The United States, China, and the United Kingdom have made the most significant contributions to research in this field and have established important collaborative research networks. The University of Pittsburgh, Massachusetts General Hospital, and the University of Michigan have been identified as the leading research institutions in this area. International and inter-institutional collaboration is vital for innovating BD treatments. Going forward, we must enhance such cooperation to tackle BD treatment challenges together and provide patients with safer, more effective therapies. The Journal of Affective Disorders is recognized as the most influential journal in this field. Keyword analysis revealed research hotspots related to clinical symptoms, drug efficacy, and genetic mechanisms. Overall, this study offers a new perspective on addressing the multifaceted challenges associated with this complex disorder.