Association of Gingival Crevicular Fluid Leptin Levels With Stages of Periodontitis: A Cross-Sectional Study

Abstract

Objectives

To evaluate the concentration of leptin in gingival crevicular fluid (GCF) across different stages of periodontitis according to the 2018 classification system and to investigate the association between GCF leptin levels and periodontitis stages.

Materials and methods

A total of 119 subjects were enrolled and allocated to 3 groups: periodontitis (P, n = 77), gingivitis (G, n = 21), and periodontal health (PH, n = 21). The P group was further classified into Stages I to IV periodontitis based on staging criteria. Clinical periodontal indexes were recorded. GCF samples were collected using absorbent paper points, and leptin concentrations in GCF were determined using enzyme-linked immunosorbent assay (ELISA). Spearman correlation, multivariable regression, and receiver operating characteristic (ROC) were evaluated.

Results

GCF leptin levels decreased progressively from the PH group to the G and P groups and declined further with increasing periodontitis stage (all P < .05). GCF leptin levels were negatively correlated with probing depth (PD) (r_s = –0.48, P < .001) and clinical attachment loss (CAL) (r_s = –0.37, P = .001). After adjustment for age, sex, and body mass index, higher GCF leptin levels were associated with a lower risk of stage III and IV periodontitis (OR = 0.74, 95% CI: 0.58 to 0.95, P = .019). The inclusion of GCF leptin improved discrimination between gingivitis and periodontitis (AUC = 0.824) and provided acceptable discrimination between stages I and II and stages III and IV periodontitis (AUC = 0.760).

Conclusions

GCF leptin levels were significantly associated with periodontitis stages and decreased progressively with advancing disease severity. GCF leptin can provide complementary information for the assessment of periodontal disease severity.

Clinical relevance

GCF leptin levels may serve as a complementary indicator of periodontal status and disease severity, supporting clinical assessment of periodontitis.

Introduction

Periodontitis is a chronic infectious disease affecting the periodontal supporting tissues, primarily initiated by plaque biofilm.¹ The inflammatory destruction of these tissues can lead to tooth mobility, displacement, and potential tooth loss. Periodontitis ranks as the second most prevalent oral health issue and the sixth most common human disease globally.² According to China’s Fourth National Oral Health Epidemiological Survey Report, the prevalence of periodontitis in China remains notably high.³ Furthermore, the 2022 World Health Organization Global Oral Health Status Report highlights that severe periodontitis contributes significantly to the global health burden.⁴ In 2018, the American Academy of Periodontology and the European Federation of Periodontology jointly introduced a new classification system for periodontal diseases, which classifies periodontitis into stages and grades.⁵ Currently, periodontitis staging relies on assessments derived from clinical periodontal examinations and imaging techniques; however, this approach possesses inherent subjectivity and limitations.

Adipose tissue serves not only as an energy storage organ but also possesses endocrine functions, secreting various adipokines that play a crucial role in maintaining the body's metabolic equilibrium and regulating immune responses. Leptin, a 16 kDa nonglycosylated peptide hormone primarily secreted by adipocytes, plays a pivotal role in regulating body weight and energy homeostasis.⁶ Additionally, leptin functions as a cytokine, exerting regulatory effects on the host’s immune defence mechanisms.^7,8 Previous studies have demonstrated that abnormal levels of leptin are associated with conditions such as obesity,⁹ type 2 diabetes,¹⁰ rheumatoid arthritis,¹¹ and reproductive dysfunction.¹²

Recently, increasing attention has been focused on the role of leptin in the pathogenesis and progression of periodontitis. Leptin is expressed in both gingival epithelial cells and connective tissues.¹³ In addition, leptin receptors are widely expressed in periodontal ligament cells, gingival epithelial cells, and connective tissues, highlighting the potential importance of leptin in regulating the periodontal microenvironment.^14,15 Biomarkers present in gingival crevicular fluid (GCF) can provide insights into the metabolic and inflammatory status of periodontal tissue.¹⁶ Previous studies have demonstrated that leptin concentration in GCF of periodontitis are significantly lower compared to those in periodontal health individuals.^17,18 However, these studies have mainly focused on comparing leptin levels between diseased and health gingiva, without specifically evaluating leptin concentrations in patients with varying severities of periodontitis. With the adoption of the 2018 classification system for periodontitis, which is now widely used in clinical practice to assess disease severity, there remains a limited of research investigating the association between GCF leptin levels and different stages of periodontitis. Therefore, the present study aims to investigate the correlation between leptin levels and the severity of periodontitis by analysing GCF leptin concentrations in patients across different periodontitis stages.

Materials and methods

Study population

From October 2021 to June 2022, a study was conducted at the Stomatological Center of the First Affiliated Hospital of Fujian Medical University, enrolling a total of 119 participants. The study was approved by the Ethics Committee of the First Affiliated Hospital of Fujian Medical University (Approval No.: MRCTA, ECFAH of FMU [2021]155), and consent was obtained from all participants prior to enrolment.

Inclusion and exclusion criteria

Han national participants aged between 18 and 70 years were included in this study. The inclusion criteria required that participants had no changes in their dietary habits or oral hygiene practices for at least 3 months prior to enrolment and had no history of periodontal treatment within the previous 6 months. Exclusion criteria included the use of antibiotics, corticosteroids, or immunomodulatory drugs within the past 6 months and the presence of systemic diseases affecting periodontal health, such as diabetes or autoimmune disease. Pregnancy, lactation, and the presence of a smoking habit were also excluded.

Diagnostic criteria

The included subjects were classified into the periodontal health (PH) group, gingivitis (G) group, and periodontitis (P) group according to the 2018 International Classification of Periodontal Diseases.¹⁹ The periodontitis group was further staged according to clinical and radiographic criteria.^5,20

The PH group included patients with clinically healthy gingivae, bleeding on probing score (BOP%) <10%, no attachment and bone loss, and probing depth (PD) ≤3 mm.

The G group included patients with BOP% ≥10%, no attachment and bone loss, and PD ≤4 mm.

Periodontitis is diagnosed when the following criteria are met: (1) interdental clinical attachment loss (CAL) is detectable at ≥2 nonadjacent teeth or (2) buccal or oral CAL ≥3 mm with pocketing ≥3mm is detectable at ≥2 teeth but the observed CAL cannot be ascribed to nonperiodontitis–related causes such as: (1) gingival recession of traumatic origin; (2) dental caries extending in the cervical area of the tooth; (3) the presence of CAL on the distal aspect of a second molar and associated with malposition or extraction of a third molar; (4) an endodontic lesion draining through the marginal periodontium; and (5) the occurrence of a vertical root fracture.

Clinical measurements

Sociodemographic and anthropometric about the subjects were collected, including age, gender, weight, height, and body mass index (BMI). Periodontal examinations were performed using a Williams periodontal probe, with six sites (mesio-buccal, buccal, disto-buccal, mesio-lingual, lingual, and disto-lingual) per tooth checked. Periodontal indexes, including PD, CAL, plaque index (PLI), bleeding on probing (BOP), and bleeding index (BI), were to be recorded in each group. All examinations were performed by a single, professionally trained periodontist to ensure consistency. Prior to the clinical periodontal examination, self-repeated measurement consistency test was evaluated with a kappa value greater than 0.95, confirming the reliability of the measurements.

Collection of GCF and leptin assay

To collect GCF, the selected tooth sites were isolated using sterile cotton rolls and air dried with a dental triple syringe. GCF samples for the PH group were collected from sites without inflammation, CAL, radiographic evidence of bone loss, and BOP. For the G group, GCF samples were obtained from sites with gingival inflammation and positive BOP, without CAL and radiographic bone loss. In the SI-P group, GCF samples were collected from sites with PD less than 5 mm, radiographic bone loss less than 15%, and positive BOP. For the SII-P group, GCF samples were collected from sites with PD ≤5 mm, radiographic bone loss of 15% to 33% confined to the coronal third of the root, and positive BOP. For the SIII-P and SIV-P groups, sites with PD greater than 5 mm, radiographic bone loss extending to the middle or apical third of the root, and positive BOP were chosen. Supragingival plaque was carefully removed without touching the gingival margin. Absorbent paper points (DIA-ISO 0.02 030, Beijing Dayading Medical Appliance Co. Ltd) were gently inserted into the sampling sites and left in place for 30 s before being removed. If a paper point was contaminated with blood or saliva, it was discarded and the sampling procedure was repeated at the same site after 10 minutes. The paper points were sealed in Eppendorf tubes and stored in a –80 °C refrigerator.

The leptin level in GCF was measured using a one-step sandwich ELISA with the human leptin ELISA kit (Shanghai Qiaodu Biotechnology Co. Ltd), with a detection sensitivity of <0.1 ng/mL. Slats were equilibrated at room temperature for 20 minutes. Subsequently, all procedures were performed according to the manufacturer’s instructions. Finally, the termination solution was added, and absorbance was measured with a microplate reader at a wavelength of 450 nm. A linear regression curve was constructed from the standard, and the concentration of each sample was calculated based on the curve equation.

Statistical analysis

All analyses were performed with the statistical package R (http://www.Rproject.org) and EmpowerStats (http://www.empowerstats.com). Continuous variables were expressed as mean ± standard deviation (SD) (normal distribution) or median (interquartile range, IQR) (skewed distribution). Categorical variables are shown as frequencies (%). One-way analysis of variance (ANOVA) test (normal distribution), the Kruskal–Wallis H test (skewed distribution), and chi-square test (categorical variables) were used for comparison among multiple groups. Holm–Bonferroni post hoc test was performed for intergroup comparisons.²¹ Spearman’s rank correlation analysis was performed to assess the associations between GCF leptin levels and clinical periodontal indexes.

Logistic regression analysis was used to evaluate the associations between exposure and outcome. Both nonadjusted and multivariate adjusted models were used. Receiver operating characteristic (ROC) curves were plotted to evaluate the diagnostic accuracy of the analyte for periodontitis. Optimal cut-off values were identified using Youden’s index derived from the ROC curves. The area under the curve (AUC) served as a summary measure of discriminative ability, with values of 0.7 to 0.8 considered acceptable, 0.8 to 0.9 considered excellent, and >0.9 considered outstanding.²² A 2-tailed P < .05 was considered to be statistically significant in all analyses.

Power analyses were performed for the one-way ANOVA based on the actual sample sizes and the observed group means and standard deviations, with the significance level set at α = 0.05. The estimated power was >0.80 for the comparison of GCF leptin levels among the PH (n = 21), G (n = 21), and P (n = 77) groups. In addition, within the P group, the estimated power was >0.80 for the analysis across periodontitis stages I to IV (n = 15, 16, 25, and 21, respectively).

Results

No statistically significant differences were observed in gender composition, age, and BMI among the PH, G, and P groups (P > .05). However, significant differences were found in periodontal indexes, including PD, CAL, PLI, and BI, among the 3 groups (P < .05) (Table 1). Statistically significant differences in age, PD, CAL, PLI, and BI were observed among the different stages of periodontitis (P < .05) (Table 2).

Table 1.

Baseline characteristics of the participants.

	PH	G	P	P-value
	(n = 21)	(n = 21)	(n = 77)
Male (n%)	6 (28.57)	10 (47.62)	27 (35.06)	⁎.415(ns)
Female (n%)	15 (71.43)	11 (52.38)	50 (64.94)
Age (years)	39.90 ± 5.83	40.05 ± 5.76	42.91 ± 9.70	†.133(ns)
BMI (kg/m2)	22.32 (1.60)	22.31 (1.74)	22.83 (1.82)	††.352(ns)
PD (mm)	2.33 (0.66)	2.33 (0.34)	4.00 (2.17)	††<.001(s)
CAL (mm)	0 (0)	0 (0)	3.00 (5.25)	††<.001(s)
PLI	1 (2)	1 (1)	2 (1.5)	††<.001(s)
BI	0 (0)	2 (1)	2 (1)	††<.001(s)

^⁎Chi-square test.

^†One-way ANOVA test.

^††Kruskal-Wallis H test.

^(s)Significant.

^(ns)Nonsignificant.

Table 2.

Baseline characteristics of the different stages of periodontitis.

	SI-P	SII-P	SIII-P	SIV-P	P-value
	(n = 15)	(n = 16)	(n = 25)	(n = 21)
Male (n%)	5 (33.3)	7 (43.8)	7 (28.0)	8 (38.1)	*.757(ns)
Female (n%)	10 (66.7)	9 (56.2)	18 (72.0)	13 (61.9)
Age (years)	37.27 ± 9.82	43.75 ± 10.49	42.52 ± 9.90	46.76 ± 7.05	†.032(s)
BMI (kg/m2)	21.48 (2.55)	22.63 (2.20)	22.83 (1.81)	23.05 (0.84)	††.071(ns)
PD (mm)	2.67 (0.84)	3.33 (0.92)	4.50 (1.84)	5.33 (2.5)	††<.001(s)
CAL (mm)	1.03 ± 0.28	1.95 ± 0.57	4.57 ± 2.14	6.84 ± 2.50	†<.001(s)
PLI	2 (1)	2 (1.5)	2 (2)	3 (0.5)	††<.001(s)
BI	2 (1)	3 (0.75)	3 (2)	4 (1)	††<.001(s)

SD, standard deviation; IQR, interquartile range; BMI, body mass index; PD, probing depth; CAL, clinical attachment loss; PLI, plaque index; BI, bleeding index.

All data (except for sex) were expressed as mean ± SD or median (IQR).

^⁎Chi-square test.

^†One-way ANOVA test.

^††Kruskal–Wallis H test.

^(s)Significant.

^(ns)Nonsignificant.

The leptin concentration was highest in the PH group, followed by the G group, and lowest in the P group, with statistically significant differences among the groups (P < .05) (Fig.1A). The leptin concentration in GCF of patients with periodontitis in stages I to IV gradually decreased, and the differences were statistically significant (P < .05). However, no statistically significant differences were observed in leptin concentrations between the SI-P and SII-P groups and between the SIII-P and SIV-P groups (P > .05) (Figure 1B). Spearman correlation analysis showed that GCF leptin levels were negatively correlated with PD (r_s = –0.48, P < .001) and CAL (r_s = –0.37, P = .001) (Supplementary Figure S1).

Fig. 1.

Comparison of GCF leptin levels among different periodontal statuses. SI-P stage I periodontitis, SII-P stage II periodontitis, SIII-P stage III periodontitis, SIV-P stage IV periodontitis, n s not significant, * P < .05, ** P < .01, and *** P < .001.

Based on the above results, SI-P and SII-P subjects were combined into a single group, while SIII-P and SIV-P subjects were grouped together for subsequent logistic regression analysis. The univariate regression analysis indicates that higher BMI in periodontitis was positively associated with the progressing to stage III and IV periodontitis (OR = 1.46, 95% CI: 1.03, 2.06, P = .031) (Table 3).

Table 3.

The association between the sociodemographic characteristics and different stages of periodontitis.

Covariate	Statistics	OR (95% CI)	P-value
Age (years)	42.91 ± 9.70	1.04 (0.99, 1.10)	0.092(ns)
Age (dichotomous)
Low (≤43 y)	37 (48.05%)	Reference
High (>43 y)	40 (51.95%)	1.97 (0.78, 4.96)	0.151(ns)
BMI (kg/m2)	22.34 ± 1.40	1.46 (1.03, 2.06)	0.031(s)
BMI (tertile)
Low (≤22.53 kg/m2)	25 (32.47%)	Reference
Middle (22.54-23.18 kg/m2)	24 (31.17%)	3.00 (0.93, 9.63)	0.065(ns)
High (>23.18 kg/m2)	28 (36.36%)	3.75 (1.19, 11.79)	0.024(s)
Sex of subjects
Female	50 (64.94%)	Reference
Male	27 (35.06%)	0.77 (0.30, 1.98)	0.583(ns)

Continuous variables were given as mean ± SD. Categorical variables were shown as n (%).

^(s)Significant.

^(ns)Nonsignificant; OR: odds ratio; CI: confidence interval; BMI: body mass index.

The association between GCF leptin levels and different stages of periodontitis are summarised in Table 4. In the crude model, without adjustment for confounding factors, a statistically significant association was observed between leptin and different stages of periodontitis (OR = 0.74, 95% CI, 0.59,0.93, P = .011). After adjusting for gender and age (Model I), and for gender, age, and BMI (Model II), the association remained statistically significant. Furthermore, when GCF leptin levels were analysed by quartiles, the results of the crude model, Models I and II consistently demonstrated that, compared with the reference group (Q1), individuals in the third quartile (Q3) and the highest quartile (Q4) had a significantly reduced risk of progressing to stage III and IV periodontitis, with all trends reaching statistical significance (P < .05).

Table 4.

The association between the levels of leptin in GCF (ng/mL) and different stages of periodontitis.

	Crude Model		Model I		Model II
	OR (95% CI)	p-value	OR (95% CI)	P-value	OR (95% CI)	P-value
Leptin	0.74 (0.59, 0.93)	.011(s)	0.73 (0.57, 0.92)	.009(s)	0.74 (0.58, 0.95)	.019(s)
Leptin (quartile)
Q1	Reference		Reference		Reference
Q2	0.32 (0.07, 1.51)	.150(ns)	0.31 (0.06, 1.49)	.143(ns)	0.36 (0.07, 1.83)	.216(ns)
Q3	0.17 (0.04, 0.78)	.022(s)	0.16 (0.03, 0.76)	.021(s)	0.18 (0.04, 0.87)	.033(s)
Q4	0.15 (0.03, 0.70)	.015(s)	0.14 (0.03, 0.68)	.015(s)	0.16 (0.03, 0.83)	.029(s)
P for trend	0.56 (0.36, 0.87)	.010(s)	0.54 (0.34, 0.85)	.009(s)	0.56 (0.34, 0.90)	.017(s)

Model I adjusted for sex and age, Model II adjusted for age, sex, and BMI.

^(s)Significant.

^(ns)Nonsignificant; OR: odds ratio; CI: confidence interval; BMI: body mass index.

The ROC analysis showed that GCF leptin levels had outstanding discriminatory ability between periodontitis and periodontal health (AUC = 0.912). When distinguishing periodontitis from gingivitis, GCF leptin levels demonstrated acceptable discriminatory performance (AUC = 0.797) (Table 5, Figure 2).

Table 5.

ROC curves and cut-off values for the GCF leptin levels to identify different periodontal statuses.

Analytes	AUC	95% CI	Cut-off	P-value	Sensitivity (%)	Specificity (%)
P group vs PH group
Leptin	0.912	0.856-0.968	12.45	<.001(s)	95.2	76.6
P group vs G group
Leptin	0.797	0.705-0.888	11.33	<.001(s)	90.5	64.9
SI-P + SII-P groups vs SIII-P + SIV-P groups
Leptin	0.689	0.571-0.808	11.39	0.005(s)	54.8	80.4

PH, periodontal healthy; G, gingivitis; P, periodontitis; SI-P, stage I periodontitis; SII-P, stage II periodontitis; SIII-P, stage III periodontitis; SIV-P, stage IV periodontitis.

^(s)Significant.

Fig. 2.

The figure of ROC curves for the GCF leptin levels to identify different periodontal statuses. PH, periodontal healthy; G, gingivitis; P, periodontitis; SI-P, stage I periodontitis; SII-P, stage II periodontitis; SIII-P, stage III periodontitis; SIV-P, stage IV periodontitis.

To assess the added value of GCF leptin beyond baseline demographic factors and BMI, multivariable ROC was performed. In the comparison between G and P groups, the baseline model including age, sex, and BMI (Model I) showed limited discriminatory performance. After the inclusion of GCF leptin, model performance improved, with Model II achieving the highest AUC of 0.824 (Figure 3A).

Fig. 3.

Multivariable ROC for the discrimination of different periodontal statuses. (A) Discrimination between gingivitis and periodontitis. (B) Discrimination between stage I-II periodontitis (SI-P and SII-P) and stage III-IV periodontitis (SIII-P and SIV-P). Model I: include age, sex, and body mass index (BMI) Model II: include age, sex, BMI, and gingival crevicular fluid (GCF) leptin levels.

When comparing SI-P and SII-P groups with SIII-P and SIV-P groups, Model I demonstrated acceptable performance. The addition of GCF leptin further improved model performance, with Model II yielding the highest AUC (AUC = 0.760) (Figure 3B).

Discussion

This cross-sectional study aimed to evaluate variations in GCF leptin levels across different periodontal statuses and to explore their association with stages of periodontitis. The findings demonstrated that, after adjustment for potential confounding factors, GCF leptin levels were inversely associated with stages of periodontitis. In additional sensitivity analyses, consistent results were observed when leptin was modelled as categorical and ordinal variables after covariate adjustment, suggesting a certain degree of robustness of the study findings. Periodontal indexes, including PD, AL, and BI, serve as critical indicators for assessing the severity of periodontal tissue destruction and the extent of local inflammation in clinical practice.^23,24 The 2018 classification of periodontal diseases implements a staging system for periodontitis based on multiple variables, including CAL, PD, bone loss amount and percentage, and tooth loss due to periodontitis.²⁰ Although increasing attention has been paid to the potential role of biomarkers in supporting periodontal risk assessment and staging, no specific biomarkers have yet been widely accepted or formally incorporated into the current classification framework.

Previous studies have identified a range of biomarkers associated with periodontal inflammation and tissue degradation. Pro-inflammatory cytokines, such as interleukin-6 (IL-6) and tumour necrosis factor-α, have been reported to be present at higher levels in GCF and saliva of individuals with periodontitis compared with those with periodontal health.^25,26 Vascular endothelial growth factors, which are key regulators of angiogenesis and vasculogenesis,²⁷ have also been shown to be elevated in GCF at sites of periodontal progression and are closely associated with inflammatory activity and vascular responses within periodontal tissues.²⁸ In addition, matrix metalloproteinases, particularly MMP-8 and MMP-9, play a central role in extracellular matrix degradation and periodontal tissue breakdown and are abundantly present in gingival tissues, GCF, and saliva, with levels closely associated with disease severity and progression.²⁹

In this context, leptin, a cytokine with dual immunomodulatory and metabolic functions, has gained increasing attention in periodontal research and has been proposed as a complementary and exploratory biomarker. A recent microbial profiling study further highlighted the interaction between microbial dysbiosis and host immunoregulatory mechanisms in periodontitis.³⁰ Structurally and functionally, leptin shares similarities with IL-6 and is capable of enhancing the activity of multiple innate immune cells, including dendritic cells, monocytes, neutrophils, and natural killer cells.³¹ Within the periodontal microenvironment, both human periodontal ligament cells and gingival fibroblasts have been shown to express leptin receptors, providing a cellular basis for the direct involvement of leptin in periodontal tissue regulation. Studies have further demonstrated that leptin receptor-positive periodontal ligament stem cells possess multilineage differentiation potential and play a critical role in periodontal tissue regeneration and repair.³² Accumulating experimental evidence supports the importance of leptin signalling in maintaining periodontal tissue homeostasis. In leptin-deficient mouse models, a significantly greater number of osteoclasts and disorganisation of the periodontal ligament structure have been observed in the alveolar bone compared with wild-type mice, suggesting that leptin deficiency may exacerbate periodontal tissue breakdown.³³ More recently, single-cell RNA sequencing analyses have shown that disruption of leptin signalling directly impairs gingival barrier function, thereby compromising the periodontal tissues’ ability to respond to inflammatory challenges.³⁴ With respect to inflammatory regulation, leptin deficiency has been associated with marked upregulation of MMP-9,interleukin-1β, and transforming growth factor-β1.³³ In vitro studies have shown that leptin binding to its receptor on human gingival fibroblasts modulates the expression of matrix metalloproteinases, thereby participating in the remodelling of the extracellular matrix in gingival connective tissues.³⁵ Furthermore, under hypoxic conditions, increased leptin levels have been shown to suppress hypoxia-induced apoptosis in human periodontal ligament cells, suggesting a potential role in regulating cell survival within the chronic inflammatory microenvironment of periodontitis.³⁶ With respect to bone metabolism, leptin has also been reported to promote the osteogenic differentiation of periodontal ligament stem cells.³⁷

Previous studies used ELISA to estimate leptin concentrations in healthy and diseased gingiva and found that leptin was present in all groups, with significantly higher concentrations observed in healthy gingival tissues compared to diseased tissues.³⁸ Additionally, the distribution of leptin and leptin receptors in healthy and diseased gingival tissues was analysed using immunohistochemistry, revealing that both leptin and its receptors were consistently present in the gingival epithelium of all groups.³⁹ GCF is a complex mixture derived from serum and periodontal tissues, containing various cytokines and inflammatory mediators, and serves as a critical microenvironmental factor in the gingival crevicular region. Changes in the components of GCF are associated with different states of periodontal tissues, providing measurable and quantifiable indicators for assessing periodontal health and disease status.⁴⁰ Compared with invasive gingival tissue sampling, GCF collection is simple and non-invasive, and has therefore been widely applied in periodontal biomarker research. Previous studies have reported that leptin levels are highest in the GCF of periodontally healthy individuals and lowest in those with periodontitis.^18,41 The findings of the present study are consistent with these observations and further support the potential value of leptin in reflecting local immune and metabolic activity.

To the best of our knowledge, evidence regarding the association between leptin levels in GCF and stages of periodontitis remains limited. In the present study, GCF leptin levels were further compared across different stages of periodontitis, and the results showed higher leptin levels in the SI-P group than in the SIII-P and SIV-P groups, although differences between some adjacent stages did not reach statistical significance. The observed tendency toward decreasing GCF leptin levels with increasing disease severity may be related to inflammation-induced vascular and epithelial alterations. For example, inflammation-induced alterations in the periodontal vascular network may increase local leptin consumption,³⁸ while upregulation of leptin receptor expression under inflammatory conditions may further promote the formation of leptin-receptor complexes,⁴² thereby reducing detectable leptin levels in GCF.

In addition, logistic regression analysis demonstrated that an increase in BMI was significantly associated with a higher risk of stage III and stage IV periodontitis. Consistent with these findings, previous epidemiological studies have demonstrated that a higher BMI is associated with an increased likelihood of stage III and stage IV periodontitis.⁴³ Furthermore, a positive correlation has been observed between BMI and periodontal clinical indexes, including CAL and PD.⁴³ This association may be attributed to the heightened inflammatory response induced by obesity, as pro-inflammatory cytokines secreted by adipose tissue can exacerbate the destruction of periodontal tissues.⁴⁴ Accordingly, BMI was included as a covariate in multivariable regression models to more accurately evaluate the association between GCF leptin levels and periodontal disease severity.

ROC analyses showed that GCF leptin levels provided outstanding discriminatory performance between periodontitis and periodontal health. Incorporation of GCF leptin into baseline models, including age, sex, and BMI, further improved model discrimination, particularly in differentiating gingivitis from periodontitis. For comparisons between stage I-II and stage III-IV periodontitis, the baseline demographic model demonstrated acceptable performance, and the addition of GCF leptin was associated with a further increase in AUC, suggesting a complementary role in reflecting disease severity.

Several limitations of the present study should be acknowledged. First, as a cross-sectional study, the findings reflect associations between GCF leptin levels and periodontitis staging at a single time point and do not allow causal inferences to be drawn regarding the role of leptin in periodontal disease progression. Second, all participants were recruited from a Chinese population, resulting in limited geographic and ethnic diversity, which may restrict the generalisability of the findings to other populations. In addition, only a single inflammatory biomarker was assessed, which may not fully capture the complexity of the inflammatory milieu in the GCF of patients with periodontitis.

In light of these limitations, several directions for future research can be proposed. Prospective longitudinal studies are warranted to dynamically assess changes in GCF leptin levels during the onset and progression of periodontal disease. Studies comparing GCF leptin levels before and after periodontal treatment may further clarify its association with inflammatory burden. In addition, integrating multiple inflammatory mediators, additional biomarkers, or proteomic approaches could allow a more comprehensive evaluation of biomarker performance and provide deeper insights into host immune-metabolic interactions in periodontitis. Finally, multicentre studies involving larger and more diverse populations are needed to enhance the generalisability of these findings.

Conclusion

In conclusion, this cross-sectional study demonstrated that GCF leptin levels decreased with increasing severity of periodontitis and were negatively correlated with key periodontal clinical indexes. After adjustment for potential confounding factors, GCF leptin remained inversely associated with periodontitis stage. Moreover, GCF leptin demonstrated acceptable discriminatory ability in distinguishing periodontitis from gingivitis, providing complementary information for the assessment of periodontal disease severity.

Author contributions

All authors contributed to the conception and design of the study. Most experiments and data analysis were performed by Liting Xie and Zhiyou Guo. Liting Xie and Jinhuai Lv drafted the manuscript. All authors revised the manuscript and approved the final version. Yun Wu, Ling Chen and Chongwu Liu jointly supervised the project. Liting Xie, Zhiyou Guo, and Jinhuai Lv contributed equally to this study and are co-first authors. Yun Wu and Chongwu Liu are co-corresponding authors. Yun Wu is the primary contact for correspondence.

Ethical approval

Approved by the Ethics Committee of the First Affiliated Hospital of Fujian Medical University (Approval No.: MRCTA, ECFAH of FMU [2021]155).

Funding

This work was supported by the Science and Technology Innovation Joint Foundation of Fujian Province, China (Grant No. 2024Y9136) and Fujian Provincial Health Technology Project, China (Grant No. 2022CXB013).

Declaration of competing interest

The authors declare no competing interests.