Skip to main content
Journal of Periodontal & Implant Science logoLink to Journal of Periodontal & Implant Science
. 2026 Sep 9;56(5):431–441. doi: 10.5051/jpis.2604380219

Publication trends and knowledge structures in implant dentistry over 25 years: a text-mining study

Goun Lee 1,2, Jeewan Chun 3, Eunhee Kang 4, Yong-Dae Kwon 2,✉
PMCID: PMC13745170  PMID: 42984269

Abstract

Purpose

This study examined changes in implant dentistry research topics from 2000 to 2024 by applying word-frequency analysis and topic modeling to abstract text.

Methods

We analyzed 10,496 PubMed abstracts published between 2000 and 2024 in 4 major implant dentistry journals: Clinical Oral Implants Research, Clinical Implant Dentistry and Related Research, International Journal of Oral & Maxillofacial Implants, and Journal of Oral Implantology. After preprocessing, changes in word frequency were calculated. To characterize the contextual progression of clinical research trends, we performed an in-depth analysis using BERTopic, a transformer-based topic-modeling technique.

Results

BERTopic analysis identified notable shifts in publication themes. Topics that were dominant early in the study period, such as “implant surface characteristics” and “preclinical animal studies,” declined steadily, suggesting consolidation of foundational knowledge. In contrast, topics related to long-term clinical stability, including “peri-implantitis” and “marginal bone loss,” increased substantially. Digital workflow topics, including “computer-guided surgery” and “digital scans,” also showed a rapid upward trend and became a core research stream.

Conclusions

Over the past 25 years, implant dentistry research has shifted from material-focused basic science toward clinical investigations emphasizing periodontal health, peri-implant tissue stability, and digital technology. This shift indicates that the field increasingly recognizes a healthy periodontium as a foundation of implant success. Extending these digital trends, future research is expected to expand in areas such as artificial intelligence-based diagnosis, patient-specific digital solutions, and the long-term management of peri-implantitis and marginal bone loss.

Keywords: Bibliometrics, Dental implants, Natural language processing, Peri-implantitis, Trends

INTRODUCTION

The introduction of osseointegration revolutionized dental rehabilitation and was followed by rapid growth in implant-related literature over the past 2 decades [1,2]. Previous bibliometric studies have provided valuable insights into publication patterns and citation networks [3,4]. However, most have relied on metadata, such as author-provided keywords, which limits their ability to track changes in research topics accurately because keywords are often inconsistent and lack detailed context.

To address these limitations, advanced text-mining and natural language processing methods, such as BERTopic [5], have emerged as useful tools for identifying latent themes with high semantic accuracy. In the present study, we applied an integrated text-mining framework to abstracts from 4 leading implant journals published between 2000 and 2024. By combining normalized word-frequency analysis with BERTopic modeling, we sought to provide a comprehensive overview of the thematic evolution of implant dentistry over the past 25 years. Although many clinicians recognize the shift toward digital workflows and peri-implant maintenance, quantitative evidence on the magnitude, timing, and specific characteristics of these changes remains limited. Therefore, this study aimed to translate broad clinical observations into measurable trends and identify emerging research areas that may guide future studies.

MATERIALS AND METHODS

Data source and selection

This study used a comprehensive collection of abstracts from 4 leading journals in implant dentistry, each of which prominently includes the term “implant” in its title: Clinical Implant Dentistry and Related Research (CIDRR), Journal of Oral Implantology (JOI), International Journal of Oral & Maxillofacial Implants (JOMI), and Clinical Oral Implants Research (COIR).

The study period was defined as 2000–2024 because 2000 marked a turning point when implant dentistry transitioned from an emerging technology to a mature, evidence-based discipline. This period was characterized by the accumulation of substantial long-term clinical data, which enabled large-scale systematic reviews, and by a rapid increase in scientific publications that helped establish implant therapy as a mainstream standard of care [6,7,8].

A systematic PubMed search was conducted using the following query: (((("Clin Implant Dent Relat Res"[Journal]) OR ("J Oral Implantol"[Journal])) OR ("Int J Oral Maxillofac Implants"[Journal])) OR ("Clin Oral Implants Res"[Journal])) AND (("2000/01/01"[Date - Publication]: "2024/12/31"[Date - Publication])). Records without available abstracts were excluded because the text-mining methodology required abstract text for analysis. The literature search and data extraction from PubMed were conducted on July 25, 2025. The initial search yielded 10,639 records. After corrections, retractions, withdrawals, and early-access articles without final publication data were excluded, 10,496 abstracts were included in the final dataset. Because this study analyzed publicly available text data from PubMed, institutional review board approval and informed consent were not required.

Data cleaning and preprocessing

All records were exported into a structured dataset containing the publication year, journal, and abstract text. Metadata, including author names, journal titles, and DOI numbers, were removed. Abstracts were standardized by converting all characters to lowercase. Lemmatization was applied to reduce inflected word forms to their base forms (e.g., studies → study; treated → treat). During text preprocessing, Scandinavian characters (e.g., å, ä, ö) were normalized to their closest ASCII equivalents with the regex library to ensure consistent tokenization. Stopwords were removed using a combined list that included the default NLTK English stopword set and an expanded custom list of non-informative academic, clinical, statistical, and publishing-related terms (e.g., objective, background, doi, volume, patient) (Supplementary Table 1). Common comparative words, such as “high,” “low,” and “different,” were intentionally retained because they are important for describing clinical results and may improve BERTopic model accuracy. The final preprocessed corpus was used for all subsequent analyses.

Word-frequency analysis and topic modeling

Word-frequency analysis

Normalized word frequency was calculated for each 5-year interval: 2000–2004, 2005–2009, 2010–2014, 2015–2019, and 2020–2024. For each period, the number of occurrences of each word was divided by the total number of tokens (words) and multiplied by 10,000 to generate normalized frequencies. This normalization accounted for differences in text length and publication volume across time periods. Temporal trends in key terms were assessed using linear regression analysis via numpy.polyfit, and the slope was used to evaluate increasing or decreasing trends across periods.

BERTopic

The BERTopic framework was used to obtain a more fine-grained thematic classification. First, document embeddings were generated using SciBERT, a transformer-based language model optimized for scientific texts. To support methodological transparency and computational reproducibility, all hyperparameter settings and package versions were explicitly documented. The dimensionality of these embeddings was reduced using Uniform Manifold Approximation and Projection with a fixed random seed (random_state=42) and the following parameters: n_components=5, n_neighbors=15, min_dist=0.0, and metric='cosine'. Documents were then clustered using the HDBSCAN algorithm with min_cluster_size=40, min_samples=1, and cluster_selection_method='leaf'.

For topic representation, a custom domain-specific stopword list comprising 236 unique terms after duplicate removal was applied during vectorization (ngram_range=[1,2], min_df=5). Keywords were extracted using a class-based term frequency–inverse document frequency (c-TF-IDF) approach, and keyword diversity was enhanced with maximal marginal relevance (diversity=0.5). After the initial modeling, the number of topics was limited to 29. Outliers were then reassigned to the most relevant topics using the c-TF-IDF strategy to improve consistency. Finally, the temporal evolution of topics was quantified by calculating topic proportions in 5-year intervals and performing linear regression analysis to identify trends.

Topic labeling was performed using a semi-automated, artificial intelligence (AI)-assisted approach. Specifically, GPT-5 (OpenAI, San Francisco, CA, USA) was used between August and November 2025. The role of this AI tool was strictly limited to generating preliminary topic-label candidates based on the top-ranked keywords and representative documents extracted from the BERTopic analysis. AI assistance was applied to the topic-labeling procedures described in the Methods section and to the topic names presented in the Results section and associated tables and figures. To minimize potential AI-induced bias, 2 authors with extensive clinical expertise in oral and maxillofacial surgery independently reviewed and manually refined these preliminary suggestions. This hybrid process improved the consistency and semantic coherence of topic naming while maintaining expert oversight.

Statistical and computational tools

A hybrid computational environment was used for the analyses. Data preprocessing and word-frequency analysis were conducted on a local workstation running Windows 10. These tasks used Python 3.11/3.13 with pandas (2.2.3), numpy (2.2.5), nltk (3.9.1), and gensim (4.3.3). The computationally intensive BERTopic modeling was performed in a Google Colab environment with Python 3.12 to leverage graphics processing unit acceleration through the cuML library. Key packages in this environment included BERTopic (0.17.3), sentence-transformers (5.1.1), umap-learn (0.5.9.post2), hdbscan (0.8.40), and scikit-learn (1.6.1). Visualizations were generated with matplotlib.

RESULTS

Overall analysis of the entire corpus

Figure 1 shows the number of articles published in the 4 selected implant journals between 2000 and 2024, grouped into 5-year intervals. The overall number of publications increased steadily from 1,202 in 2000–2004 to 2,857 in 2015–2019, followed by a modest decline to 2,113 in the most recent period, 2020–2024. In contrast, the total number of implant-related publications indexed in PubMed continued to increase over the same period. Among the 4 journals, CIDRR and COIR showed rapid growth through 2010–2019 before decreasing in 2020–2024. JOMI showed a similar pattern, whereas JOI remained relatively stable throughout the study period, with only minor fluctuations.

Figure 1. Number of articles published in the 4 implant journals—COIR, JOMI, JOI, and CIDRR—from 2000 to 2024.

Figure 1

COIR: Clinical Oral Implants Research, JOMI: International Journal of Oral & Maxillofacial Implants, JOI: Journal of Oral Implantology, CIDRR: Clinical Implant Dentistry and Related Research.

Word-frequency trends

To summarize changes in terminology in implant dentistry, normalized word frequencies were calculated for each 5-year interval between 2000 and 2024. This normalization adjusted for changes in publication volume and abstract length over time. Thus, the observed trends reflect changes in research topics rather than only changes in the number of published articles.

Overall, the frequency analysis showed a gradual but distinct transition in the thematic focus of implant-related publications over the past 2 decades. Table 1 summarizes the 30 terms with the most dynamic changes in implant-related literature between 2000 and 2024. The most rapidly increasing terms—peri, cone-beam computed tomography, deviation, zirconia, scan, implantitis, guide, abutment, and marginal bone loss/level—reflect a growing emphasis on biological complications, imaging-based diagnostics, and computer-assisted digital workflows. The inclusion of “soft,” “buccal,” and “thickness” further indicates sustained interest in peri-implant soft-tissue morphology and dimensional stability. Conversely, the most markedly decreasing terms—load, surface, titanium, cell, and system—reflect a steady decline in material- and mechanics-oriented research.

Table 1. Top 30 increasing and decreasing terms from 2000 to 2024, ranked by linear regression slopes of normalized word frequencies.

Top 30 increasing terms Top 30 decreasing terms
Rank Token β (slope)a) R2 b) Rank Token β (slope)a) R2 b)
1 Peri 12.79 0.83 1 Bone −23.91 0.95
2 High 7.12 0.98 2 Implant −20.37 0.92
3 Level 6.89 0.88 3 Load −12.58 0.99
4 Difference 5.77 0.92 4 Surface −11.95 0.86
5 CBCT 5.68 0.88 5 Mandible −6.40 0.92
6 Deviation 5.40 0.80 6 Titanium −6.35 0.96
7 Change 5.34 0.93 7 Cell −6.25 0.79
8 Zirconia 5.34 0.79 8 System −5.09 0.93
9 Scan 4.98 0.89 9 Brånemark −4.94 0.73
10 Implantitis 4.83 0.70 10 Period −4.81 0.93
11 Guide 4.69 0.76 11 Stage −4.46 0.77
12 Soft 4.64 0.95 12 Mandibular −4.46 0.90
13 Abutment 4.63 0.81 13 Graft −3.99 0.75
14 Risk 4.57 0.77 14 Contact −3.70 1.00
15 MBL 4.50 0.89 15 Coat −3.38 0.92
16 Accuracy 4.26 0.75 16 PRP −3.37 0.79
17 Complication 4.26 0.95 17 Edentulous −3.29 0.83
18 Buccal 4.04 0.92 18 Screw −3.12 0.69
19 Thickness 4.04 0.98 19 Distraction −3.10 0.85
20 Crown 3.89 0.77 20 Maxilla −2.94 0.99
21 Digital 3.47 0.89 21 Prosthesis −2.91 0.53
22 Different 3.21 0.84 22 Endosseous −2.88 0.86
23 Low 2.84 0.84 23 Fix −2.87 0.92
24 Cone 2.81 0.86 24 Beta −2.85 0.89
25 Loss 2.79 0.49 25 Form −2.84 0.97
26 Beam 2.73 0.78 26 Dog −2.83 0.78
27 Effect 2.60 0.92 27 Animal −2.80 0.77
28 Vertical 2.59 0.86 28 Machine −2.76 0.94
29 Socket 2.57 0.71 29 Defect −2.75 0.78
30 Marginal 2.50 0.91 30 Fixture −2.71 0.72

CBCT: cone-beam computed tomography, MBL: marginal bone loss/level, PRP: platelet-rich plasma.

a)β (slope): Change in normalized frequency per 5-year interval.

b)R2 : Coefficient of determination of linear regression.

Collectively, these raw frequency trends indicate that digital- and biological-related terms had the highest positive slopes, whereas material- and preclinical-related terms showed the greatest decreases in normalized frequency from 2000 to 2024.

BERTopic modeling results

BERTopic modeling was conducted using SciBERT-based contextual embeddings to identify semantically coherent topic clusters. A total of 29 distinct topics were generated. Using density-based clustering and class-based TF-IDF, BERTopic identified detailed, clinically meaningful topic groups.

To clarify these trends, the 29 topics were grouped into 5 major domains: (A) peri-implant maintenance, (B) digital workflow and guided surgery, (C) prosthetic–mechanical integrity, (D) bone regeneration and surgical technique, and (E) preclinical and materials science (Table 2). The temporal evolution and proportional distribution of these aggregated domains across 5-year intervals are shown in the grayscale heatmap in Figure 2. The temporal evolution of individual topics was then assessed using the regression slopes (β) of their proportions.

Table 2. BERTopic clusters categorized into 5 major thematic domains.

Topic label Representative keywords (slope)a) R2 b)
(A) Peri-implant maintenance
Marginal bone level & peri-implant soft tissue Implant, soft, bone, level, buccal, change, esthetic, extraction 1.60 0.93
Peri-implantitis: probing, diagnosis & risk Peri, implantitis, probe, implant, disease, bop, level, periodontitis 1.51 0.81
Titanium decontamination & biofilm removal Surface, biofilm, titanium, clean, roughness, laser, bacterial, air 0.17 0.21
Inflammatory genes/cytokines in peri-implant tissues Gene, expression, cell, inflammatory, response, surface, signal, osseointegration 0.10 0.19
Peri-implant microbiology & bacterial species Bacterial, bacteria, peri, specie[sic], microbial, gingivalis, implantitis, sulcus −0.20 0.65
(B) Digital workflow & guided surgery
Computer-guided surgery: deviation & accuracy Guide, deviation, accuracy, plan, template, position, computer, surgery 0.75 0.63
Digital impressions & intraoral scanner accuracy Impression, scan, digital, scanner, accuracy, io, cast, precision 0.36 0.77
(C) Prosthetic–mechanical integrity
Abutments, cements & zirconia crowns Abutment, cement, zirconia, crown, fracture, ceramic, titanium, restoration 0.65 0.74
Abutment–screw connection: torque & preload Abutment, screw, torque, connection, cycle, load, internal, ncm 0.14 0.29
Loading strategies in (partial) edentulism Edentulous, load, implant, partially, prosthesis, mandible, immediate, success −1.15 0.80
Finite element analysis & biomechanical stress Stress, finite, element, load, distribution, implant, strain, bone 0.29 0.42
Fixed prostheses: support design & survival Prosthesis, implant, support, survival, load, complication, fix, restoration −1.75 0.91
(D) Bone regeneration & surgical technique
Maxillary sinus floor elevation & membrane care Sinus, membrane, elevation, floor, maxillary, perforation, lateral, window 0.51 0.89
Osteotomy drilling heat & irrigation protocols Drill, temperature, irrigation, osteotomy, heat, preparation, thermal, cut 0.19 0.66
Simultaneous implant placement with augmentation Implant, bone, surgery, graft, augmentation, complication, loss, survival 0.18 0.37
Alveolar ridge augmentation & sinus grafting Graft, bone, sinus, augmentation, ridge, alveolar, autogenous, maxillary −0.29 0.24
Guided bone regeneration for bone defects (membrane-graft) Bone, defect, membrane, graft, formation, regeneration, new, collagen −1.13 0.82
(E) Preclinical & materials science
Osseointegration in animal models (rat/rabbit) Osseointegration, animal, rat, rabbit, bone, tibia, bic, implant −0.23 0.52
Cell culture & osteogenic differentiation Cell, culture, differentiation, proliferation, growth, osteoblast, platelet, scaffold −0.37 0.38
Titanium surface coatings & cell responses Surface, titanium, coat, cell, ti, implant, etch, acid −1.09 0.60

Each topic represents a semantically cohesive research domain defined by high-probability keywords and annual proportion changes. The magnitude of β reflects the rate of change over the study period. A positive β indicates an increasing temporal trend, whereas a negative β indicates a decreasing trend.

Certain terms in the representative keywords column (e.g., “peri,” “bop,” “io,” “ncm,” “bic,” “ti,” and “specie”) are exact tokens derived directly from the text-mining and lemmatization process. They are retained in their original extracted form to preserve data integrity, which may result in nonstandard abbreviations, word fragments, grammatical artifacts (indicated by [sic]), or nonstandard capitalization.

a)β (slope): Change in normalized frequency per 5-year interval.

b)R2 : Coefficient of determination of linear regression.

Figure 2. Grayscale heatmap showing the temporal evolution of five major thematic domains derived from BERTopic modeling from 2000 to 2024. Each row represents an aggregated thematic group: peri-implant maintenance, digital workflow and guided surgery, prosthetic–mechanical integrity, bone regeneration and surgical technique, or preclinical and materials science. Values indicate the mean topic proportion (%) within each 5-year interval—2000–2004, 2005–2009, 2010–2014, 2015–2019, and 2020–2024—averaged across the constituent topics of each group; darker shades denote higher proportions. Rows are ordered according to the magnitude of the 2020–2024 values for visual clarity.

Figure 2

Overall, the proportions of the peri-implant maintenance and digital workflow and guided surgery domains increased, whereas those of the prosthetic–mechanical integrity and preclinical and materials science domains declined. The steepest positive slopes underlying the domain-level increases were observed for “Marginal bone level and peri-implant soft tissue” (β=1.60, R2=0.93), “Peri-implantitis: probing, diagnosis, and risk” (β=1.51, R2=0.81), and “Computer-guided surgery: deviation and accuracy” (β=0.75, R2 =0.63).

In contrast, the prosthetic–mechanical integrity and preclinical and materials science domains showed pronounced downward trajectories. This decline was led by topics such as “Fixed prostheses: support design and survival” (β=−1.75, R2 =0.91), “Loading strategies in partial edentulism” (β=−1.15, R2 =0.80), and “Titanium surface coatings and cell responses” (β=−1.09, R2=0.60). The bone regeneration and surgical technique domain remained comparatively stable over the 25-year period. Together, these patterns indicate a sustained shift in implant research toward biological maintenance and digitally integrated care, alongside a concurrent contraction of foundational surface/material and purely mechanical topics.

Comparative analysis by journal

The temporal BERTopic analysis shown in Figure 3 illustrated distinct thematic compositions across the 4 journals. Preclinical and materials science topics, particularly “Titanium surface coatings and cell responses,” declined markedly in CIDRR (Figure 3A), COIR (Figure 3B), and JOI (Figure 3C). In all 3 journals, this decline was accompanied by a late-period increase in topics related to peri-implant maintenance. Meanwhile, JOI (Figure 3C) and JOMI (Figure 3D) maintained more consistent proportions of their core mechanical and structural themes, such as “Finite element analysis” and “Abutment–screw connection,” throughout the study period.

Figure 3. BERTopic-derived topic composition for the 4 implant journals, displayed as stacked proportions across 5-year intervals: (A) CIDRR, (B) COIR, (C) JOI, and (D) JOMI. For each journal, the top 15 topics, ranked by overall mean share, and “others” are shown; within each interval, values sum to 100%.

Figure 3

CIDRR: Clinical Implant Dentistry and Related Research, COIR: Clinical Oral Implants Research, JOI: Journal of Oral Implantology, JOMI: International Journal of Oral & Maxillofacial Implants.

DISCUSSION

The findings of this study provide a comprehensive overview of changes in implant dentistry over the past quarter-century and offer insights relevant to both clinicians and researchers. Although the total number of implant-related publications indexed in PubMed continued to increase, the proportion of such studies published in the 4 selected implant journals declined after 2020. This pattern may reflect diversification of implant-related research topics and the emergence of multidisciplinary journals that increasingly publish implant dentistry research. Thus, implant research appears to have expanded beyond traditional implant journals into broader dental, materials science, and interdisciplinary platforms. By adjusting for publication volume, this study clarified the timing and direction of these major thematic changes.

Over the past 25 years, implant dentistry has undergone a measurable thematic transition from material-based investigations to biologically oriented and digitally integrated clinical research. Word-frequency analysis showed sharp increases in terms such as “digital,” “guide,” “scan,” and “accuracy” (Table 1), while topic modeling identified domains such as “Digital workflow and guided surgery” as major emerging fields (Table 2). These findings suggest that, as 3-dimensional imaging, computer-aided design/computer-aided manufacturing, and guided surgery have become widely adopted, research has increasingly focused on validating their clinical accuracy and precision [9,10,11,12]. As the dental community overcame the initial learning curve associated with surgical and prosthetic techniques, early challenges related to mechanical failures were largely resolved, leading to high and increasingly predictable implant survival rates [7]. Consequently, the primary clinical focus has shifted from achieving initial osseointegration to maintaining implants biologically over the long term, including the prevention and management of peri-implant diseases. This shift has prompted the standardization of disease classification and evidence-graded care protocols [13,14,15].

Conversely, the decline in terms such as “surface” and “titanium” (Table 1), together with the overall contraction of preclinical and materials science topics (Table 2, Figure 2), suggests that this area has not necessarily declined but has instead reached a phase of clinical stability. With the clinical success and standardization of surface treatments such as sandblasted, large-grit, acid-etched surfaces and TiUnite, research has shifted from developing novel surfaces to examining long-term biological responses and complications associated with existing technologies [16,17]. In other words, “surface” and “osseointegration,” once standalone research topics, are now being incorporated into more complex clinical themes, such as “peri-implantitis” and “long-term stability.”

This analysis also revealed an imbalance in recent research. Diagnostic and observational studies on peri-implant diseases have increased substantially, whereas research on treatment approaches and the management of prosthetic complications remains limited. This imbalance underscores the need for periodontal expertise in developing targeted therapeutic protocols for peri-implantitis and for future studies addressing these clinically important but less extensively studied interventional areas.

The temporal analysis revealed distinct thematic distributions across the 4 journals, reflecting their complementary editorial scopes (Figure 3). Although all journals showed increasing incorporation of peri-implant maintenance topics, notable differences were observed in the pace and direction of thematic expansion. Some journals showed a more pronounced shift toward digital workflows and biologically oriented clinical outcomes, whereas others maintained a consistent and substantial focus on fundamental biomechanics, prosthetic–mechanical integrity, and structural analyses. These differences are best interpreted as reflecting each journal’s established editorial identity rather than divergent research priorities. Collectively, the 4 journals constitute a balanced research ecosystem that includes both emerging clinical applications and foundational structural inquiry, thereby advancing the field in complementary ways.

A strength of this study is its integrated use of word-frequency analysis and topic modeling, which enabled quantitative characterization of long-term thematic shifts in the implant literature. The consistent trends identified across multiple analytical approaches enhance the robustness of the findings and help visualize emerging imbalances in research attention, highlighting areas where further clinical investigation is needed.

Nevertheless, this study has several limitations. First, the analysis relied exclusively on abstract text, which may not fully represent the methodological details or contextual nuances of the included studies. Because abstracts are highly condensed summaries, they often omit detailed methodological variation and nuanced clinical discussion. Consequently, the BERTopic model was inherently weighted toward the dominant, high-frequency themes explicitly highlighted by authors, which may have limited the semantic interpretation of subtler conceptual shifts embedded in full texts. Future research could expand the dataset beyond abstracts. Full-text analysis and citation-network mapping would allow future studies to examine whether these textual trends reflect actual shifts in study design, analytical methods, or reported clinical outcomes. Second, the dataset was restricted to 4 implant-focused journals. This choice was made to maintain thematic specificity, because adding prosthodontic or periodontal journals would have introduced large volumes of non-implant literature, such as basic periodontal or prosthodontic content, and may have obscured implant-focused trends. These 4 journals were selected because they are widely recognized journals specializing in implant dentistry. As leading publications with international editorial boards, this journal set provides broad representation of global research trends while limiting regional bias. However, excluding multidisciplinary dental journals and specific periodontal and prosthodontic journals may have influenced the results. Because substantial implant research, particularly in-depth biological studies on peri-implantitis or systemic interactions, is frequently published in these broader venues, the thematic distribution observed here may reflect a concentrated focus on surgical and mechanical aspects of implantology rather than the full biological spectrum. Future research could broaden the journal set to include prosthodontic, periodontal, biomaterials, and imaging-related outlets, thereby providing a wider multidisciplinary perspective.

In conclusion, this study identified distinct temporal and thematic shifts in implant dentistry research over the past 25 years. Word-frequency and topic-modeling analyses consistently demonstrated a transition from material- and surface-oriented investigations toward biologically driven and digitally integrated clinical research. This evolution reflects a broader transformation of implant dentistry into a discipline that prioritizes soft-tissue health, peri-implant maintenance, and digital precision over purely mechanical or material performance. The successful navigation of the initial learning curve in surgical and prosthetic techniques has contributed to the stabilization of implantology as a clinical field. The discipline has moved beyond the goal of basic implant survival and now places greater emphasis on long-term biological maintenance and digital precision through evidence-based refinement. Although these areas were not directly extracted from the historical text-mining data, the rapid upward trajectory of digital workflow topics suggests that future research may expand into forward-looking areas such as AI-based diagnosis and prognosis, patient-specific digital solutions, and the control of long-term biological complications.

ACKNOWLEDGEMENTS

This study is based on part of the first author’s master’s thesis at Kyung Hee University.

Footnotes

Funding: This work was supported by a National Research Foundation of Korea (NRF) grant funded by the Korean government (NRF-2023R1A2C1003876).

Conflict of Interest: No potential conflict of interest relevant to this article was reported.

Author Contributions:
  • Conceptualization: Yong-Dae Kwon.
  • Data curation: Goun Lee.
  • Formal analysis: Goun Lee, Eunhee Kang.
  • Investigation: Goun Lee, Jeewan Chun.
  • Methodology: Goun Lee, Jeewan Chun.
  • Software: Goun Lee, Jeewan Chun.
  • Validation: Eunhee Kang, Yong-Dae Kwon.
  • Writing - original draft: Goun Lee.
  • Writing - review & editing: Jeewan Chun, Eunhee Kang, Yong-Dae Kwon.

SUPPLEMENTARY MATERIAL

Supplementary Table 1

Custom stopwords defined for the topic modeling analysis (total=236 words)

jpis-56-431-s001.xls (26KB, xls)

References

  • 1.Brånemark PI, Hansson BO, Adell R, Breine U, Lindström J, Hallén O, et al. Osseointegrated implants in the treatment of the edentulous jaw. Experience from a 10-year period. Scand J Plast Reconstr Surg Suppl. 1977;16:1–132. [PubMed] [Google Scholar]
  • 2.Moraschini V, Poubel LA, Ferreira VF, Barboza ES. Evaluation of survival and success rates of dental implants reported in longitudinal studies with a follow-up period of at least 10 years: a systematic review. Int J Oral Maxillofac Surg. 2015;44:377–388. doi: 10.1016/j.ijom.2014.10.023. [DOI] [PubMed] [Google Scholar]
  • 3.Joshi VM, Kandaswamy E. Bibliometric analysis of dental implantology journals from 1991 to 2023. Int J Oral Maxillofac Implants. 2024;39:193–202. doi: 10.11607/jomi.10758. [DOI] [PubMed] [Google Scholar]
  • 4.Sabri H, Nava P, Barootchi S, Tavelli L, Monje A, Nart J, et al. 25 years of Clinical Implant Dentistry and Related Research (CIDRR): a modern bibliometric and Altmetrics network analysis. Clin Implant Dent Relat Res. 2024;26:862–878. doi: 10.1111/cid.13335. [DOI] [PubMed] [Google Scholar]
  • 5.Grootendorst M. BERTopic: neural topic modeling with a class-based TF-IDF procedure. arXiv; 2022. Mar 11, [DOI] [Google Scholar]
  • 6.Jung RE, Pjetursson BE, Glauser R, Zembic A, Zwahlen M, Lang NP. A systematic review of the 5-year survival and complication rates of implant-supported single crowns. Clin Oral Implants Res. 2008;19:119–130. doi: 10.1111/j.1600-0501.2007.01453.x. [DOI] [PubMed] [Google Scholar]
  • 7.Pjetursson BE, Asgeirsson AG, Zwahlen M, Sailer I. Improvements in implant dentistry over the last decade: comparison of survival and complication rates in older and newer publications. Int J Oral Maxillofac Implants. 2014;29(Suppl):308–324. doi: 10.11607/jomi.2014suppl.g5.2. [DOI] [PubMed] [Google Scholar]
  • 8.Russo SP, Fiorellini JP, Weber HP, Niederman R. Benchmarking the dental implant evidence on MEDLINE. Int J Oral Maxillofac Implants. 2000;15:792–800. [PubMed] [Google Scholar]
  • 9.Dawood A, Marti Marti B, Sauret-Jackson V, Darwood A. 3D printing in dentistry. Br Dent J. 2015;219:521–529. doi: 10.1038/sj.bdj.2015.914. [DOI] [PubMed] [Google Scholar]
  • 10.Gehrke P, Rashidpour M, Sader R, Weigl P. A systematic review of factors impacting intraoral scanning accuracy in implant dentistry with emphasis on scan bodies. Int J Implant Dent. 2024;10:20. doi: 10.1186/s40729-024-00543-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Khaohoen A, Powcharoen W, Sornsuwan T, Chaijareenont P, Rungsiyakull C, Rungsiyakull P. Accuracy of implant placement with computer-aided static, dynamic, and robot-assisted surgery: a systematic review and meta-analysis of clinical trials. BMC Oral Health. 2024;24:359. doi: 10.1186/s12903-024-04033-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Shi Y, Wang J, Ma C, Shen J, Dong X, Lin D. A systematic review of the accuracy of digital surgical guides for dental implantation. Int J Implant Dent. 2023;9:38. doi: 10.1186/s40729-023-00507-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Berglundh T, Armitage G, Araujo MG, Avila-Ortiz G, Blanco J, Camargo PM, et al. Peri-implant diseases and conditions: consensus report of workgroup 4 of the 2017 World Workshop on the Classification of Periodontal and Peri-Implant Diseases and Conditions. J Clin Periodontol. 2018;45(Suppl 20):S286–S291. doi: 10.1111/jcpe.12957. [DOI] [PubMed] [Google Scholar]
  • 14.Herrera D, Berglundh T, Schwarz F, Chapple I, Jepsen S, Sculean A, et al. Prevention and treatment of peri-implant diseases-the EFP S3 level clinical practice guideline. J Clin Periodontol. 2023;50(Suppl 26):4–76. doi: 10.1111/jcpe.13823. [DOI] [PubMed] [Google Scholar]
  • 15.Schwarz F, Alcoforado G, Guerrero A, Jönsson D, Klinge B, Lang N, et al. Peri-implantitis: summary and consensus statements of group 3. The 6th EAO Consensus Conference 2021. Clin Oral Implants Res. 2021;32(Suppl 21):245–253. doi: 10.1111/clr.13827. [DOI] [PubMed] [Google Scholar]
  • 16.Albrektsson T, Wennerberg A. The impact of oral implants - past and future, 1966-2042. J Can Dent Assoc. 2005;71:327–327d. [PubMed] [Google Scholar]
  • 17.Wennerberg A, Albrektsson T. Effects of titanium surface topography on bone integration: a systematic review. Clin Oral Implants Res. 2009;20(Suppl 4):172–184. doi: 10.1111/j.1600-0501.2009.01775.x. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Table 1

Custom stopwords defined for the topic modeling analysis (total=236 words)

jpis-56-431-s001.xls (26KB, xls)

Articles from Journal of Periodontal & Implant Science are provided here courtesy of Korean Academy of Periodontology

RESOURCES