Trends in Facial Reanimation Surgery: A Bibliometric Analysis

Abstract

Background:

Facial paralysis profoundly affects physical, psychological, and social well-being. Facial reanimation surgery (FRS) uses various static and dynamic procedures to restore function and appearance. This bibliometric analysis of the top 100 most-cited FRS articles identifies trends, gaps, and methodological quality, offering insights into the field’s evolution, guiding future research, and supporting evidence-based clinical practices to enhance patient outcomes.

Methods:

The top 100 most-cited articles related to FRS were identified through a comprehensive Web of Science search covering publications from 1976 to 2024. Data including citation count, study focus, and Oxford Centre for Evidence-Based Medicine levels of evidence (LOEs) were extracted.

Results:

The most-cited articles on FRS amassed a total of 6872 citations, involving 5601 patients, with citations per article ranging from 31 to 535. Most studies were LOE 3 (n = 52), with fewer at higher evidence levels (LOEs 1 and 2). Surgical technique was the primary focus (n = 65), whereas validated patient-reported outcome measures appeared in only 8 studies.

Conclusions:

Our findings highlight the need for better research methods and wider use of validated patient-reported outcome measures in FRS studies. Standardized tools and high-quality, multicenter research are crucial for improving patient care. This analysis offers insights into FRS evolution and recommends collaborative, interdisciplinary studies to advance the field.

Takeaways

Question: What are the key gaps and trends in facial reanimation surgery research, particularly in addressing patient-centered outcomes?

Findings: Our bibliometric analysis of the top 100 cited articles highlights a dominance of lower level evidence studies focused on surgical techniques (n = 64). Validated patient-reported outcome measures appeared in only 8 studies, revealing a lack of attention to psychosocial impacts and emotional recovery.

Meaning: Future research should shift toward holistic approaches that emphasize patient perspectives, psychological restoration, quality of life, and emotional recovery, alongside advancing surgical techniques.

INTRODUCTION

Facial palsy, also known as facial paralysis, is characterized by the loss of voluntary muscle movement on 1 or both sides of the face, imposing a significant health burden globally.¹ Its causes are diverse, spanning congenital origins, idiopathic facial palsy such as Bell palsy, trauma, tumors, and surgical complications.² This condition disrupts facial symmetry and crucial functions such as blinking, speaking, and smiling, while also carrying profound psychological and social impacts.^3,4 Individuals with facial palsy often experience self-esteem issues, social isolation, and depression, which diminish their overall quality of life (QoL).⁵ Given the central role of facial expression in human interaction, any degree of disfigurement can hinder personal and professional relationships, leading to further social and emotional challenges.⁶

Facial reanimation surgery (FRS) has developed extensively to address these physical and emotional challenges. Initially, static techniques such as eyelid weights and suspension sutures were used primarily to restore symmetry and protect the eye, without restoring movement. As the field progressed, dynamic techniques such as nerve grafts, muscle transfers, and the “babysitter” method emerged, aiming to restore active facial expressions.^7,8 These dynamic procedures have greatly enhanced both functionality and QoL for patients, allowing them to regain expressive abilities and confidence.^9,10

Bibliometric analyses in reconstructive surgery have illuminated trends and research focuses, but none have specifically addressed facial reanimation.¹¹ A targeted bibliometric analysis on this topic could highlight influential studies, identify research gaps, and outline global trends in surgical techniques and QoL outcomes. This study aims to map the broader research landscape in FRS, identifying trends, advancements, and gaps.

METHODS

A comprehensive literature review was performed to identify the 100 most highly cited publications on FRS, covering the years 1976 to 2024 from the Web of Science database (version 5.35, Clarivate Analytics, Philadelphia, PA) using keywords such as “smile reanimation,” “facial reanimation surgery,” “facial nerve reconstruction,” “temporalis tendon reconstruction,” and “gracilis transfer.” The search strategy is available in Supplemental Digital Content 1 for a more detailed view. (See appendix, Supplemental Digital Content 1, which displays the search strategy used, https://links.lww.com/PRSGO/E349.) The lower bound of 1976 was chosen because it represents the advent of modern dynamic facial reanimation techniques, specifically, the first free gracilis muscle transfer combined with cross-face nerve grafting. The search was completed on May 10, 2024. The decision to use Web of Science was based on its strong citation metrics and comprehensive journal coverage in surgical research. Any discrepancies during the review process were resolved by consensus between the reviewers, a method commonly used in systematic reviews to enhance reliability and reduce bias.¹² If no relevant citations are available, we emphasize that consensus ensures the accurate inclusion of the top 100 most relevant articles by leveraging the combined expertise of the reviewers and the senior author (Supplemental Digital Content 1, https://links.lww.com/PRSGO/E349).

The search yielded a total of 816 journal publications, which were subsequently ranked based on citation count. Publications with an equal number of citations were separated using the mean number of citations per year. To ensure that the publications were directly relevant to FRS, 2 reviewers (A.C. and A.G.) independently screened titles and abstracts until 100 journal publications were included. Discrepancies were resolved by consensus discussion with author M.A., with any remaining inconsistencies resolved through review of the publication’s full text. A total of 120 publications were screened to provide the 100 most-cited publications for inclusion. Reasons for the exclusion of the other publications are specified in the methodological flowchart (Fig. 1).

Fig. 1.

Flowchart summarizing the methodology for identifying and analyzing the 100 most-cited articles in FRS.

For the purposes of this review, FRS was defined as surgical procedures aimed at restoring dynamic facial movement, including smile reconstruction and eye closure, most commonly through techniques such as free muscle transfer, nerve grafting, or local muscle transposition. Studies pertinent to FRS that were published via a peer-reviewed process were eligible for inclusion.

Data were extracted by 2 authors (A.C. and A.G.) from full-text publications into a standardized online computer spreadsheet (Google Sheets, Google LLC, Mountain View, CA). Data extraction included the publication title, author list, corresponding author details, publication year, source journal, total citations, mean number of citations per year, geographical study setting, funding source, study design, main subject/content focus, declaration of conflict of interest, Oxford Centre for Evidence-Based Medicine levels of evidence (LOEs), and the use of validated clinical and patient-reported outcome measures (PROMs).¹³ Additionally, only articles published in English were included in the analysis, in line with common practice in bibliometric reviews.^14,15

Analyses were performed using R Studio 4.4.2.¹⁶ Categorical variables were compared using chi-square or Fisher exact tests, whereas continuous variables were assessed with Kruskal–Wallis tests. Bonferroni correction was applied to adjust for multiple comparisons. A P value less than 0.05 was considered statistically significant.

RESULTS

Distribution of Citations

The 100 most-cited articles on FRS amassed a total of 6872 citations among them, with a mean citation count of 68.72 per article. (See table, Supplemental Digital Content 2, which displays all included studies with their corresponding citation count and year of publication, ranked based on citation count, https://links.lww.com/PRSGO/E350.) The range of citations per article varied between 31 and 535. Furthermore, the average number of citations per article per year showed variability, spanning from 12.50 to 0.71. Of note, the most highly cited study is authored by Harii et al,¹⁷ who first demonstrated microvascular free gracilis muscle transfer usage in 1976 and used the deep temporal nerve as the motor nerve donor for FRS. A violin plot showcasing the distribution of citation numbers by publication year is available for viewing in Figure 2 (Supplemental Digital Content 2, https://links.lww.com/PRSGO/E350).

Fig. 2.

Violin plot showcasing the distribution of citation numbers by publication year.

Main Topics

Within this dataset, most studies (n = 65) focused on surgical techniques related to FRS, whereas only 4 studies addressed the surgical anatomy of FRS. Moreover, 23 studies covered FRS outcomes, whereas only 2 studies focused on the prognosis/risk factors of FRS. Additionally, 6 articles investigated the nonoperative management of FRS. Most studies (n = 81) used clinical outcomes, whereas only 8 studies used validated PROMs. More than half of the studies (n = 53) used cosmetic outcomes, most commonly pre- and postoperative pictures and videos. Of note, only 5 studies used the Facial Clinimetric Evaluation (FaCE) scale, and 3 used the electronic facial paralysis assessment (eFACE) score. FACIAL CLIMA, an automatic 3-dimensional quantitative analysis for facial movement, was used in 3 studies. The evolution of main publication subjects over the years can be visualized in a stacked area plot of proportions, as shown in Figure 3. The number of publications for different main subjects in FRS research can be visualized in a horizontal bar chart, as shown in Figure 4.

Fig. 3.

Stacked area plot showing the change in the publications’ main topics over the years.

Fig. 4.

Horizontal bar chart showing the number of publications for different main subjects.

LOE and Study Designs

Our analysis shows that lower level evidence studies, such as retrospective cohort studies and expert opinions, dominate FRS. Although these studies contribute valuable insights, there is a pressing need for more multicenter randomized controlled trials (RCTs). Such trials are essential to elevate the quality of evidence and guide clinical decision-making with greater confidence. A breakdown of the study designs and LOEs of the top 100 FRS studies is shown in Table 1 and Figure 5, respectively.

Table 1.

Summary of Study Designs, Including Retrospective Cohort Studies, Expert Opinions, and RCTs

Study Design	No. Articles
Retrospective cohort study	44
Expert opinion	14
Case series	13
Nonclinical study	8
Prospective cohort study	8
Narrative review	7
Case reports	3
Systematic review	3

Fig. 5.

Visualization of the LOE distribution among the top 100 articles.

Publishing Journals and Timestamps

The 100 most-cited articles were distributed across 24 journals and can be classified into 9 plastic surgery journals, 6 ENT journals, 6 maxillofacial journals, 1 neurosurgery journal, 1 pediatric surgery journal, and 1 anatomy journal. The primary contributor to the FRS literature was Plastic and Reconstructive Surgery with 38 publications, followed by the Journal of Plastic, Reconstructive, and Aesthetic Surgery with 12 publications. Other journals contributed between 1 and 8 studies. (See table, Supplemental Digital Content 3, which displays the list of journals contributing to the top 100 articles, including the number of publications per journal, https://links.lww.com/PRSGO/E351.) A scatter plot of citations per year versus publication year (Fig. 6) demonstrates a positive trend over time.

Fig. 6.

Analysis of publications by year for the most-cited articles, showing publication and citation trends over time.

Country and Author Trends

The greatest contributor to the FRS literature was the United States with 34 publications, followed by Italy and Canada with 12 and 10 publications, respectively. The United Kingdom and Japan exhibit a notable presence with 6 publications each. Spain, Austria, Australia, and Taiwan each contribute 5, 5, 4, and 4 publications, respectively. The remaining countries have between 1 and 3 studies. Terzis was the most prolific author with 9 publications, all of which were as the first author, followed by Frey and Bianchi, who each have 5 publications to their credit. Table 2 provides a more detailed breakdown of contributing countries.

Table 2.

Countries Contributing to the Top 100 Articles, With Publication Counts for Each

Country	Publication Numbers
United States	34
Italy	12
Canada	10
Japan	6
United Kingdom	6
Austria	5
Spain	5
Australia	4
Taiwan	4
Brazil	3
Belgium	2
China	2
Argentina	1
Germany	1
Greece	1
Iran	1
Mexico	1
Netherlands	1
Switzerland	1

Statistical Comparison Tests

Older studies had significantly higher citation counts (ρ = −0.295, P = 0.0028). Funded studies were cited more than nonfunded ones (χ² = 4.70, P = 0.030). Study design influenced citation counts (χ² = 19.94, P = 0.0057), with narrative reviews receiving more citations than retrospective cohort studies (P = 0.049). No other significant differences were found in citation counts based on the remaining factors. The complete statistical analysis can be viewed in Table 3.

Table 3.

Statistical Comparison Test Results Including Each Statistical Test Used for All Variables

Analysis of Citation Differences Across Variables
Variable Tested	Statistical Test	Test Statistic	P	Statistically Significant?
Year	Spearman	−0.295 (rho)	<0.01**	Yes
LOE	Spearman	0.017 (ρ)	NS	No
Funds	Kruskal–Wallis	4.701 (χ²)	<0.05*	Yes
Design	Kruskal–Wallis	19.942 (χ²)	<0.01**	Yes
Subject	Kruskal–Wallis	8.573 (χ²)	NS	No
PROM	Wilcoxon	W = 458.5	NS	No
Country	Kruskal–Wallis	23.44 (χ²)	NS	No
Funds (post hoc)	Dunn test	−2.168 (Z)	<0.05*	Yes
Design (post hoc)	Dunn test	−3.129 (Z)	<0.05*	Yes

*P < 0.05.

**P < 0.01.

DISCUSSION

This bibliometric analysis of FRS aimed to identify literature gaps and guide future research. It is also the first to classify the top 100 cited FRS articles according to the Oxford Centre for Evidence-Based Medicine’s methodological quality. Modern FRS originated from 2 key developments in the 1970s: the introduction of the cross-facial nerve graft (CFNG) by Anderl¹⁸ in 1973 and the use of free microvascular muscle transfers to restore smiles by Harii¹⁷ in 1976. By 1980, O’Brien et al^19,20 integrated these techniques, reporting outcomes in a series of 20 patients.

The 100 most-cited FRS articles gathered 6872 citations, a relatively low count compared with other craniofacial surgery fields, which often focus on congenital conditions with higher patient numbers and a longer research history.^21,22 FRS’s narrower focus and recent reliance on single-center studies may limit its citation impact. Future research should expand through the use of standardized outcome tools and larger multicenter collaborations.

RCTs could help increase the field’s overall visibility and scientific influence, as purported by Butler et al.²³ Although such studies may not always surpass innovative technique papers in citation counts, they can strengthen the evidence base and broaden the impact of FRS literature. The wide range of citations per article highlights the diverse impact of individual studies within the FRS literature. Studies focusing on pioneering techniques, such as the highest cited study on our list, by Harii et al,^17,24 which introduced the free gracilis muscle transfer for facial paralysis, and the fourth highest cited study, published by the same team, presenting an alternative approach, have garnered higher citation counts due to their broad applicability and influence.

Throughout the 1980s and 1990s, CFNG remained the dominant approach, with refinements aimed at improving nerve regeneration and optimizing donor nerve selection. During this period, staged reanimation strategies such as the babysitter technique, introduced by Terzis in 1984, gained interest as a means of providing earlier muscle reinnervation while awaiting cross-facial nerve growth.⁸ The early 2000s marked a shift toward masseteric nerve transfers, which offered faster reinnervation and improved spontaneity compared with CFNG alone. Originally described by Spira²⁵ in 1978, the masseteric nerve transfer gained widespread adoption in the 2000s due to its high axonal load, proximity to target muscles, and reduced donor site morbidity. More recently, dual-nerve techniques combining CFNG and masseteric nerve transfers have emerged as a promising approach to enhancing both volitional control and spontaneity of movement.²⁶ Our analysis reveals that microvascular gracilis muscle transfer, often in combination with CFNG, is now widely recognized as the gold standard for dynamic facial reanimation in long-standing facial paralysis.

Among the 100 most-cited articles, only 8 used validated outcome tools such as the FaCE and eFACE scores, whereas the majority continued to rely on subjective assessments such as surgeon-graded photographic and video evaluations. Although these validated instruments substantially enhance interstudy comparability and outcome reliability, their limited adoption is likely due to a combination of historical timing—FaCE emerged in 2001 and eFACE as recently as 2015—and deeper methodological and conceptual barriers.^27,28 Many clinicians remain inclined toward traditional methods such as the House–Brackmann scale or qualitative visual assessments, which are simple, familiar, and seamlessly integrated into established workflows.²⁹ By contrast, validated scales such as FaCE and eFACE introduce additional complexity; FaCE involves patient questionnaires capturing psychosocial domains, and eFACE requires detailed, multidimensional clinician scoring, which some practitioners might perceive as burdensome or of unclear additional benefit in routine clinical settings.

The significance of using validated measures, specifically the interdependence of FaCE and eFACE, is underscored by recent evidence exploring the relationship between clinician-graded facial function (eFACE) and patient-reported QoL (FaCE). A single-center logistic regression analysis recently demonstrated a moderate positive correlation (R = 0.434, P < 0.001) between eFACE scores and FaCE scores, suggesting patients with improved facial function generally report better QoL.⁵ However, a larger meta-analysis synthesizing multiple studies found that overall correlations between clinician-graded function and patient-perceived QoL were only low to moderate.³⁰ These apparently conflicting findings, rather than representing true contradictions, emphasize a critical gap in understanding patient outcomes. Specifically, although the logistic regression highlights a statistically meaningful relationship between function and QoL, the moderate correlation explains only approximately 18.8% of the variance in QoL scores. This leaves more than 80% of patient-perceived well-being unexplained by facial movement scores alone. Thus, both studies collectively underscore the complexity of QoL outcomes, suggesting that functional improvements assessed by clinicians are important but not the singular predictors of patient satisfaction.

The timing of outcome assessment likely influences the observed relationship between facial function and QoL. Single-center studies often use only 1 early postoperative measurement, implicitly assuming a stable, linear relationship between function and QoL. However, this approach overlooks the potential evolution of outcomes over time. By contrast, longitudinal studies, such as those included in the meta-analysis, can detect changes in how psychosocial factors influence QoL as patients adapt postoperatively, hence potentially explaining the weak correlation between function and QoL seen in the results. Although improved facial movement is essential, it does not automatically lead to sustained improvements in psychosocial well-being or satisfaction. Over time, factors such as self-consciousness, social interactions, compensatory muscle activation, and difficulties in speech or swallowing may increasingly affect QoL. Therefore, clinicians should integrate patient-reported QoL measures alongside objective functional assessments, while also incorporating psychological and rehabilitative interventions to comprehensively improve patient outcomes. By acting on these recommendations, future research can examine longitudinal changes in the function–QoL relationship, explore whether psychosocial interventions can bridge the function–QoL gap, and identify which nonfunctional factors best predict long-term QoL in FRS patients.

Our bibliometric analysis identifies a shift in FRS from static photographs to dynamic video assessments, as videos offer a more realistic portrayal of facial dynamics. Unlike photographs, videos capture movement, expression, and muscle function, enabling a more accurate and comprehensive evaluation of aesthetic and functional recovery.

Of the 100 most-cited FRS publications, 76 appeared in plastic surgery journals, 13 in craniofacial/maxillofacial journals, and 6 in ENT journals, spanning 24 high-impact publications. Plastic and Reconstructive Surgery was most prolific (n = 38), followed by the Journal of Plastic, Reconstructive & Aesthetic Surgery (n = 12). The highest impact journal represented was Facial Plastic Surgery & Aesthetic Medicine. This distribution underscores the central role of plastic surgery in FRS research, while also highlighting meaningful interdisciplinary involvement, especially from craniomaxillofacial specialties.

Most FRS studies are single-center retrospective cohorts (n = 44), expert opinions (n = 14), or case series (n = 13), reflecting predominantly low LOE. Although RCTs exist, ethical and practical barriers remain significant in plastic surgery, particularly due to irreversible surgical interventions and uncertainty about optimal treatments.³¹ Unlike drug trials, surgical treatments cannot be withdrawn or easily modified mid-study, limiting adaptive trial designs.³² Pretrial systematic reviews, expert consensus, and feasibility studies could enhance equipoise before randomization.³³ Bayesian and sequential designs allow early termination if 1 treatment seems inferior, reducing patient exposure to suboptimal interventions. Personalized randomization based on individual risk factors can ensure that high-risk patients receive the safest option while maintaining rigor.³⁴ Alternatively, incentives to use standardized outcome measures in multiarm nonrandomized studies with propensity matching can yield robust comparative data without requiring randomization.

The United States leads FRS literature with 34 publications, followed by Italy (n = 12) and Canada (n = 10), likely due to robust research funding and specialized institutions.³⁵ Contributions from the United Kingdom, Japan, and Austria indicate growing international interest, though most research remains regionally concentrated. Terzis was the most prolific author (n = 9), followed by Frey and Bianchi (n = 5, each), highlighting the impact of key individuals.

Despite the methodological robustness of our analysis, certain limitations are inherent to bibliometric studies. Citation choice may be subject to bias, distortion, and amplification, potentially granting undue authority to some publications.³⁶ Additionally, neither LOE nor citation count necessarily reflects study quality, though both may serve as proxies for clinical impact. Our findings indicate a substantial number of publications from non–English-speaking countries, such as Italy and Japan; however, restricting our analysis to English-language articles may have excluded relevant studies. This, along with other limitations described, may have led to underrepresentation of key contributors to FRS research in the analysis.^37–40 Furthermore, bibliometric analysis often emphasizes quantity over content depth, potentially overlooking nuanced findings within studies.³⁶ Our reliance on citation databases also introduces publication and indexing biases, as not all relevant work, especially from newer or regionally focused journals, is indexed. Additionally, as with all citation-based analyses, older publications have a natural advantage due to the accumulation of citations over time, which may skew rankings in favor of earlier works regardless of current relevance.⁴¹ Finally, lower LOE studies, even with methodological limitations, could add to systematic reviews and meta-analyses, thus enriching the evidence base in the field.

FRS continues to evolve, with advancements in microsurgical techniques and nerve grafting contributing to improved outcomes. However, this bibliometric analysis highlights persistent challenges in the field, including variability in study design, limited use of standardized outcome measures, and a lack of high-level evidence. Future research efforts would benefit from greater multicenter collaboration, broader adoption of validated PROMs, and an emphasis on long-term, patient-centered outcomes.

CONCLUSIONS

FRS has advanced significantly, yet challenges remain, including limited long-term data, inconsistent outcome reporting, and underuse of validated PROMs. Addressing these issues requires standardized methods, broader PROM adoption, and a focus on psychosocial outcomes. Our analysis reveals a predominance of lower level evidence, reflecting the need for more multicenter, high-quality studies, including RCTs where feasible. Greater collaboration and methodological rigor are essential to advancing evidence-based FRS care.

DISCLOSURE

The authors have no financial interest to declare in relation to the content of this article.