Abstract Background Dyslipidemia has been implicated as a risk factor for periodontitis; however, the underlying mechanisms remain poorly understood. This study aims to investigate the correlation between blood lipid levels and periodontitis and to explore the cellular and molecular mechanisms linking dyslipidemia to periodontal disease. Methods Participants with complete data on serum lipid levels and periodontal examinations were selected from the 2009–2014 National Health and Nutrition Examination Survey (NHANES). Subgroup and logistic regression analyses were performed to assess the associations between periodontal status and lipid profiles, including triglycerides (TG), total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), and remnant cholesterol (RC). Single-cell Ribonucleic Acid (RNA) sequencing (scRNA-seq) statistical analysis, cellular metabolic pathway analysis, and CellChat were employed to examine cellular variations and intercellular communication in periodontal tissues with and without periodontitis. Results A total of 5,342 participants were included in the analysis. Subgroup analyses revealed significant positive associations between elevated levels of TG, TC, and LDL-C and the prevalence of severe periodontitis. Participants in the high TC group had a 55% higher risk of severe periodontitis compared to those in the normal TC group (OR = 1.55, 95% CI: 1.17–2.05). Similarly, the risk of severe periodontitis increased by 50% in the high LDL-C group (OR = 1.50, 95% CI: 1.09–2.06) and by 35% in the high TG group (OR = 1.35, 95% CI: 1.02–1.79). scRNA-seq analysis revealed enhanced lipid metabolism in immune cells, particularly mast cells, within the periodontitis group. These mast cells were found to modulate fibroblast activity through inflammatory signaling pathways such as Interleukin 7 (IL7), Interleukin 15 (IL15). Conclusions Elevated blood lipid levels are associated with an increased risk of severe periodontitis, potentially mediated by enhanced lipid metabolism in immune cells, particularly mast cells, and their interactions with fibroblasts via inflammatory signaling pathway. These findings suggest the importance of monitoring lipid levels in periodontitis patients with dyslipidemia and highlight potential therapeutic targets for managing periodontal disease in this population. Keywords: Periodontitis, Dyslipidemia, Total cholesterol (TC), Low-density lipoprotein cholesterol (LDL-C), Triglycerides (TG) Introduction Among adults globally, periodontitis is recognized as a chronic inflammatory condition. It’s caused by the host's immune cell response to microorganisms and biofilms, which stimulates immune cells to migrate locally and release a large amount of reactive oxygen species (ROS) [[32]1], intensifying the local inflammatory reaction and leading to periodontal tissues’ destruction [[33]2]. According to data from the NHANES conducted between 2009 and 2014, more than 42% of Americans aged 30 or older have periodontitis, with 7.8% having severe cases [[34]3]. Globally, it has been estimated that up to 11% of people suffer from severe periodontitis [[35]4]. The association between periodontitis and dyslipidemia has garnered increasing attention [[36]5–[37]8]. Changes in serum levels of TG, TC, LDL-C and HDL-C are the hallmarks of dyslipidemia [[38]9]. Accumulating evidence suggests that lipids play a critical role in modulating the severity of bone resorption-related disorders, including periodontitis, which regulate osteoclast bone resorption through specific signaling pathways and alterations in cytokine levels [[39]10–[40]12]. Lipids, including lipopolysaccharides (LPS) produced by periodontitis-associated pathogenic bacteria, have been shown to directly regulate osteoclast activity within the alveolar bone in periodontal mice [[41]13, [42]14]. Concurrently, lipid metabolism disorders, such as elevated plasma free fatty acid (FFA) levels resulting from obesity or metabolic syndrome, may act as risk factors for periodontal diseases and synergistically exacerbate the effects of LPS [[43]15–[44]18]. scRNA-seq significantly enhances our capacity to study the immunological environment of periodontal tissue by allowing for targeted investigation of cell populations at single-cell resolution [[45]19]. CellChat was applied to build cell–cell communication atlases and predict how those cells and signals coordinate [[46]20]. Therefore, the aim of this study is to investigate the potential mechanisms linking dyslipidemia, immune regulation, and periodontal bone loss, with a focus on understanding how lipid metabolism influences the inflammatory microenvironment and bone remodeling processes in periodontitis. To achieve this, we utilized the NHANES dataset (2009–2014) to examine the relationship between blood lipid levels and periodontitis in participants aged 30 years and older. Additionally, to compare the differences in the immune microenvironment and lipid metabolism between periodontitis (PD) and healthy controls (HC), we first analyzed the distribution of cell populations and their lipid metabolism levels using scRNA-seq statistical analysis and cellular metabolic pathway analysis. Subsequently, cell–cell communication networks were visualized and interpreted using CellChat. Methods Participants Figure [47]1 provides a detailed overview of the inclusion process employed in this study. Initially, we selected participants from the NHANCE database who had undergone examinations during the years 2009–2014 (other years did not include periodontal examinations), resulting in an initial cohort of 30,468 participants. Subsequently, participants without periodontal examination records were excluded, reducing the sample size to 11,752. Finally, after excluding those with incomplete lipid level data (unknown TC, HDL-C, LDL-C or TG), the final cohort comprised 5,342 participants. Fig.1. [48]Fig.1 [49]Open in a new tab Flow chart of study participants Data in this research were collected from NHANES 2009–2014 from the National Center for Health Statistics (NCHS) of the Centers for Disease Control and Prevention (CDC), and participants'consent was obtained during the survey process. The analysis of the data set is obtained in NHANES website ([50]https://www.cdc.gov/nchs/nhanes/about_nhanes.htm). Periodontal examination and classification The reference examiners for the study provided extensive training and calibration to all clinical examiners. Periodontal depth was defined as the distance from the free gingival margin (FGM) to the base of the sulcus. To measure periodontal depth, a periodontal probe was gently inserted into the gingival sulcus (the space between the tooth and the gum) at a slight angle, following the anatomical contour of the tooth. The probe was systematically walked around each tooth to record measurements at six sites per tooth: mesio-buccal, mid-buccal, disto-buccal, mesio-lingual, mid-lingual, and disto-lingual. The distance from the cemento-enamel junction (CEJ) to the free gingival margin (FGM) was measured by gently positioning the probe at the CEJ and recording the distance to the FGM. Attachment loss (AL) was then calculated by subtracting the CEJ-to-FGM measurement from the periodontal depth, using the formula: AL = periodontal depth − (CEJ-to-FGM measurement). AL was also measured and calculated at six sites per tooth: mesio-buccal, mid-buccal, disto-buccal, mesio-lingual, mid-lingual, and disto-lingual. Based on the probing depth and loss of attachment (AL), periodontitis is identified and categorized. The Center for Disease Control and the American Academy of Periodontics (CDC/AAP) classification/case description is the basis for the periodontitis grade and severity classifications displayed in Table [51]1 [[52]21]. Table 1. Definitions of subjects according to different severity of periodontitis Subject Definition No periodontitis No evidence of mild, moderate, and severe periodontitis Mild periodontitis  ≥ two interproximal sites with attachment loss (AL) ≥ 3 mm and < 4 mm and ≥ two interproximal sites with probing depth ≥ 4 mm not on the same tooth, or one site with probing depth ≥ 5 mm Moderate periodontitis  ≥ two interproximal sites with AL ≥ 4 mm and < 6 mm not on the same tooth or ≥ two interproximal sites with probing depth ≥ 5 mm not on the same tooth Severe periodontitis  ≥ two interproximal sites with AL ≥ 6 mm not on the same tooth and ≥ one or more interproximal sites with probing depth ≥ 5 mm [53]Open in a new tab Blood lipid level measuring NHANES used a time end point method to measure plasma cholesterol concentration [[54]22]. Mayo Clinic's classification of different TC levels was used in this analysis. Participants were categorized as follows: for TC: (1) ≤200 mg/dL (normal), (2) 200–239 mg/dL (borderline), and (3) ≥240 mg/dL (high); for HDL-C: (1) ≤40 mg/dL (low), (2) 41–59 mg/dL (normal), and (3) ≥60 mg/dL (high); for LDL-C: (1) ≤100 mg/dL (low), (2) 100–159 mg/dL (normal to borderline), and (3) ≥160 mg/dL (high to very high); and for TG: (1) ≤150 mg/dL (normal), (2) 150–199 mg/dL (borderline), and (3) ≥200 mg/dL (high to very high). Besides, for RC, which was determined by TC- (HDL-C + LDL-C), participants were categorized as follows: (1) ≤15 mg/dL; (2) 15–20 mg/dL; (3) 20–30 mg/dL; (4) ≥30 mg/dL. Dyslipidemia is defined as an abnormal lipid profile characterized by elevated levels of TC, LDL-C, or TG, or reduced levels of HDL-C, as per established clinical guideline [[55]9]. Potential confounders and covariates The following were taken into account as potential confounders or covariates: (1) age; (2) gender; (3) race; (4) education (less than high school, high school or equivalent, college or above); (5) family income-to-poverty ratio (PIR, ≤ 1, 1–3, > 3); (6) body mass index (BMI, kg/m [[56]2]); (7) daily alcohol consumption; (8) glucose level; and (9) hypertension (yes or no); (10) smoking status (never, former and current smoker) [[57]21, [58]23]. Based on the data provided by NHANES, former smoker is defined as an individual who currently abstains from smoking entirely but has smoked a minimum of 100 cigarettes over the course of their lifetime [[59]21]. Statistical analyses Proportion was used to count the categorized variables, and mean ± standard deviation (SD) was used to represent continuous variables. Chi-square analyses were used to identify the study participants'general characteristics. To ascertain the variation in lipid levels between various periodontal status groups, t tests were used. To investigate the relationship between blood lipid levels and periodontitis, logistic regression analyses were performed. Model 1 unadjusted for variables, model 2 adjusted for age, gender, and race, and model 3 adjusted for age, gender, race, education level, family income-to-poverty ratio, BMI, smoking status, alcohol consumption, glucose level, and hypertension. The odds ratios (ORs) and 95% confidence intervals (CIs) were estimated. The downloaded data were visualized and analyzed using the statistical package R (The R Foundation; [60]http://www.r-project.org; version 4.2.0) and Empower-Stats 4.1 ([61]www.empowerstats.net, X&Y solutions, Inc. Boston, Massachusetts). P < 0.05 was considered statistically significant. ScRNA-seq statistical analysis We analyzed the single-cell RNA sequencing dataset ([62]GSE171213) from human chronic periodontitis and clinically healthy periodontal tissues. 51,248 periodontal cells from 5 periodontitis patients (showing attachment loss and more than 60% of the root with alveolar bone loss) and 4 healthy people (showing no attachment loss) are included in this dataset [[63]19]. The data was processed using the R package Seurat v4.0 [[64]24]. Subpar cells with a percentage of mitochondrial genes greater than 40 and fewer than 200 genes were eliminated. Harmony package was used to remove the batch effect. Run Uniform Manifold Approximation and Projection (UMAP) and Run Principal Component Analysis (PCA) were used to reduce dimensionality and visualize the clustering. Based on the differential genes in each cluster, cell types were identified. Differential gene expression and pathway enrichment analyses Significant marker genes were identified for each cell group using the FindAllMarkers function in Seurat. Gene Ontology (GO) terms pathway enrichment analysis was performed to clarify the function of these differentially expressed genes (DEGs) using the ClusterProfiler package (version 4.4.4, R) with default parameters and the org.Hs.eg.db dataset (version 3.15.0, R) [[65]25]. Cellular metabolic pathway analysis The cellular metabolism pathway is analyzed through the R package scMetabolism (version 0.2.1, R) [[66]26]. The metabolic pathway gene sets of each cell cluster were quantified to compare metabolic differences between the different cell types in periodontitis group and healthy control group. Cell–cell communication analysis using CellChat To uncover potential ligand-receptor interactions between different cell types in periodontitis group and healthy control group, CellChat (version 1.6.1, R) was employed [[67]20]. The netVisual_circle and netVisual_bubble functions were applied to visualize the network and identify the senders and receivers involved in cell–cell communication. Results Study participants A total of 5,342 participants aged ≥ 30 years with complete periodontal and laboratory examinations for TC, HLD-C, LDL-C and TG were included from the 2009–2014 NHANES. The baseline characteristics of the participants, stratified by periodontitis stages, are presented in Table [68]2. Participants were categorized into four groups: no periodontitis, mild periodontitis, moderate periodontitis, and severe periodontitis. Among the included participants, 3,135 (58.69%) were free of periodontitis, while 2,207 (41.31%) were diagnosed with periodontitis, including 206 (3.86%) with mild periodontitis, 1,442 (26.99%) with moderate periodontitis, and 559 (10.46%) with severe periodontitis. Kruskal–Wallis analysis revealed significant differences across periodontal statuses for variables including age, gender, BMI, race, education level, family poverty income ratio, hypertension status, serum glucose levels, alcohol and cigarette consumption, HDL-C, LDL-C, RC and TG (P < 0.05). Participants with poorer periodontal status were more likely to be older, smokers, have higher serum glucose levels, lower education levels, hypertension, and higher alcohol consumption (P < 0.05). Table 2. Baseline characteristics of the study population in NHANES 2009–2014 with different periodontal statuses (N = 5342) Characteristic Periodontal status P value Normal Mild periodontitis Moderate periodontitis Severe periodontitis N (%) 3135 (58.69%) 206 (3.86%) 1442 (26.99%) 559 (10.46%) TC 194.72 ± 40.31 197.81 ± 36.51 194.22 ± 39.62 197.06 ± 41.28 0.328 HDL-C 54.68 ± 15.68 52.65 ± 16.69 53.57 ± 15.81 51.86 ± 15.94  < 0.001 LDL-C 116.32 ± 35.25 121.07 ± 32.56 115.97 ± 35.16 119.36 ± 36.01 0.049 RC 23.71 ± 12.98 24.09 ± 12.42 24.67 ± 13.42 25.84 ± 14.22 0.002 TG 118.57 ± 64.83 120.48 ± 61.96 123.39 ± 67.06 129.09 ± 71.00 0.002 Age (years) 52.57 ± 15.06 46.18 ± 12.94 55.65 ± 14.63 56.49 ± 11.54  < 0.001 BMI (kg/m^2) 28.99 ± 6.77 30.25 ± 6.28 29.49 ± 6.78 29.03 ± 6.29  < 0.001 Alcoholic drinks/day 3.33 ± 28.60 3.03 ± 2.67 3.98 ± 34.51 8.85 ± 74.69  < 0.001 Glucose (mg/dl) 106.68 ± 30.53 104.80 ± 27.84 110.33 ± 33.32 116.73 ± 40.32  < 0.001 Gender  < 0.001  Male 1334 (42.55%) 113 (54.85%) 754 (52.29%) 407 (72.81%)  Female 1801 (57.45%) 93 (45.15%) 688 (47.71%) 152 (27.19%) Race  < 0.001  Mexican American 313 (9.98%) 50 (24.27%) 250 (17.34%) 111 (19.86%)  Other Hispanic 325 (10.37%) 19 (9.22%) 164 (11.37%) 51 (9.12%)  Non-Hispanic White 1587 (50.62%) 80 (38.83%) 566 (39.25%) 155 (27.73%)  Non-Hispanic Black 540 (17.22%) 44 (21.36%) 298 (20.67%) 169 (30.23%)  Other Race 370 (11.80%) 13 (6.31%) 164 (11.37%) 73 (13.06%) Education level  < 0.001  Less than high school 686 (21.91%) 55 (26.70%) 434 (30.16%) 204 (36.49%)  High school or equivalent 641 (20.47%) 55 (26.70%) 322 (22.38%) 150 (26.83%)  College or above 1804 (57.62%) 96 (46.60%) 683 (47.46%) 205 (36.67%) Family poverty income ratio < 0.001  ≤ 1 524 (18.16%) 46 (25.70%) 299 (22.53%) 139 (27.74%)  1–3 1107 (38.37%) 74 (41.34%) 592 (44.61%) 237 (47.31%)  > 3 1254 (43.47%) 59 (32.96%) 436 (32.86%) 125 (24.95%) Smoking status  < 0.001  Non-smoker 1810 (57.74%) 136 (66.02%) 778 (54.07%) 197 (35.24%)  Former-smoker 810 (25.84%) 35 (16.99%) 386 (26.82%) 169 (30.23%)  Current-smoker 515 (16.43%) 35 (16.99%) 275 (19.11%) 193 (34.53%) Hypertension  < 0.001  Yes 1206 (38.51%) 64 (31.22%) 625 (43.43%) 257 (46.06%)  No 1926 (61.49%) 141 (68.78%) 814 (56.57%) 301 (53.94%) [69]Open in a new tab Continuous variables were presented as mean and standard deviation, and categorical variables were expressed as N (%). For continuous variables, P values were analyzed by Kruskal–Wallis tests. For categorical variables, P values were analyzed by chi-square tests Abbreviations: TC total cholesterol, HDL-C high-density lipoprotein cholesterol, LDL-C low-density lipoprotein cholesterol, RC remnant cholesterol, TG triglycerides, BMI body mass index Relationship between severe periodontitis and blood lipid levels Table [70]3 presents the association between blood lipid levels (TC, HDL-C, LDL-C, RC, and TG) and the prevalence of severe periodontitis, analyzed using three logistic regression models. In the minimally adjusted model, which corrected for demographic variables (age, gender, and race), significant positive correlations were observed between elevated levels of TC, LDL-C, RC, TG, and the incidence of severe periodontitis (P = 0.0042, 0.0116, 0.0060, and 0.0068, respectively). Table 3. ORs (95% CIs) for severe periodontitis compared with normal population Blood Lipid Level (mg/dL) Crude model: OR (95%CI), P Minimally adjusted model: OR (95%CI), P Fully adjusted model: OR (95%CI), P TC 1.00 (1.00, 1.00) 0.2080 1.00 (1.00, 1.01) 0.0042 1.00 (1.00, 1.00) 0.5860 TC (classification)  ≤ 199 Reference Reference Reference  200–239 0.99 (0.81, 1.22) 0.9305 1.07 (0.86, 1.33) 0.5240 0.98 (0.72, 1.33) 0.8763  ≥ 240 1.31 (1.01, 1.71) 0.0399 1.55 (1.17, 2.05) 0.0021 1.19 (0.79, 1.78) 0.4065 P for trend 0.1022 0.0062 0.5440 HDL-C 0.99 (0.98, 0.99) < 0.0001 1.00 (0.99, 1.00) 0.5158 1.00 (0.99, 1.01) 0.5078 HDL-C (classification)  ≥ 60 Reference Reference Reference  40–59 1.38 (1.11, 1.71) 0.0034 1.05 (0.83, 1.32) 0.7116 0.93 (0.66, 1.30) 0.6580  ≤ 39 1.72 (1.31, 2.25) < 0.0001 1.19 (0.88, 1.60) 0.2668 0.96 (0.62, 1.50) 0.8607 P for trend  < 0.0001 0.2834 0.8293 LDL-C 1.00 (1.00, 1.00) 0.0616 1.00 (1.00, 1.01) 0.0116 1.00 (1.00, 1.00) 0.7794 LDL-C (classification)  ≤ 99 Reference Reference Reference  100–159 1.13 (0.92, 1.38) 0.2448 1.19 (0.96, 1.48) 0.1085 1.09 (0.81, 1.48) 0.5606  ≥ 160 1.41 (1.05, 1.90) 0.0221 1.50 (1.09, 2.06) 0.0117 1.19 (0.75, 1.88) 0.4576 P for trend 0.0274 0.0107 0.4264 RC 1.01 (1.01, 1.02) 0.0005 1.01 (1.00, 1.02) 0.0060 1.00 (0.99, 1.01) 0.8744 RC (classification)  ≤ 15 Reference Reference Reference  15–20 1.38 (1.06, 1.81) 0.0170 1.33 (1.00, 1.77) 0.0479 1.05 (0.70, 1.57) 0.8141  20–30 1.34 (1.04, 1.73) 0.0248 1.20 (0.91, 1.58) 0.1867 0.98 (0.67, 1.44) 0.9325  ≥ 30 1.60 (1.25, 2.06) 0.0002 1.52 (1.15, 1.99) 0.0029 1.09 (0.74, 1.61) 0.6717 P for trend 0.0006 0.0087 0.7544 TG 1.00 (1.00, 1.00) 0.0005 1.00 (1.00, 1.00) 0.0068 1.00 (1.00, 1.00) 0.9067 TG (classification)  ≤ 149 Reference Reference Reference  150–199 1.14 (0.88, 1.48) 0.3307 1.16 (0.88, 1.54) 0.2856 1.09 (0.75, 1.61) 0.6456  ≥ 200 1.45 (1.12, 1.88) 0.0052 1.35 (1.02, 1.79) 0.0360 0.96 (0.64, 1.45) 0.8406 P for trend 0.0051 0.0270 0.9753 [71]Open in a new tab Model 1: Non-adjusted Model 2: Adjusted for age, gender, race Model 3: Adjusted for age, gender, race, education level, family poverty income ratio, BMI, smoking status, alcohol consumption, history of hypertension and serum glucose level Abbreviations: TC total cholesterol, HDL-C high-density lipoprotein cholesterol, LDL-C low-density lipoprotein cholesterol, RC remnant cholesterol, TG triglycerides, OR odds ratio, CI confidence interval To further validate these findings, blood lipid levels were converted from continuous variables to categorical variables based on the Mayo Clinic classification for sensitivity analysis. For TC, the high-level group (≥ 240 mg/dL) exhibited a significantly higher risk of severe periodontitis compared to the normal group (≤ 199 mg/dL) (OR = 1.55, 95% CI: 1.17–2.05, P = 0.0021). Similar trends were observed for LDL-C (OR = 1.50, 95% CI: 1.09–2.06, P = 0.0117), RC (OR = 1.52, 95% CI: 1.15–1.99, P = 0.0029), and TG (OR = 1.35, 95% CI: 1.02–1.79, P = 0.0360). These results indicate that higher levels of TC, LDL-C, RC, and TG are associated with an increased risk of severe periodontitis (P for trend in the minimally adjusted model: 0.0062, 0.0107, 0.0087, and 0.0270, respectively). Association between periodontitis and lipid metabolism ScRNA-seq analysis of HC and PD UMAP analyses revealed an increased proportion of immune cells, including neutrophils, monocytes, plasma cells, and mast cell clusters, in the PD group compared to the HC group (Fig. [72]2B). Unbiased clustering of the cells identified twelve distinct clusters, categorized as follows: (1) T cells, (2) endothelial cells (ECs), (3) B cells, (4) monocytic cells, (5) neutrophils, (6) plasma cells, (7) fibroblast 1, (8) mast cells, (9) epithelial cells, (10) fibroblast 2, (11) fibroblast 3, and (12) myeloid-derived suppressor cells (MDSCs) (Fig. [73]2A). Fig.2. [74]Fig.2 [75]Open in a new tab Single-cell RNA sequencing analysis of HC and PD samples. A UMAP of all cells in four samples of HC and five samples of PD, colored by cell-type annotation. B Proportion of each cell type in HC and PD. C Metabolic pathway of cells in periodontal tissues (including four samples of HC and five samples of PD). D GO enrichment analysis of mast cell lipid metabolism related pathways. E Cell–cell interaction pathways between mast cells and fibroblasts in HC and PD samples. F Up-regulated cell–cell interaction pathways in PD samples. G Down-regulated cell–cell interaction pathways in PD samples. MDSC: Myeloid-derived suppressor cell; HC: healthy control; PD: periodontitis Cellular metabolic pathway of human periodontitis tissue Mast cells demonstrated significantly enhanced activity in fatty acid synthesis and amplification pathways, which are closely associated with lipid metabolism (Fig. [76]2C). Further GO enrichment analysis of DEGs revealed that mast cells in the PD group were significantly enriched in pathways related to fatty acid metabolism (P < 0.01) (Fig. [77]2D), indicating elevated lipid metabolism levels in mast cells during periodontitis. Cell–cell communication analysis CellChat analysis revealed that inflammatory signaling pathways between mast cells and fibroblasts were significantly activated in PD, including pathways mediated by Interleukin 7 (IL7) and Interleukin 15 (IL15) (Fig. [78]2E-F). Additionally, signaling pathways such as Cathepsin G (CTSG)-Par-3 Family Cell Polarity Regulator (PARD3), which positively regulate osteoclast differentiation, were markedly up-regulated in PD (Fig. [79]2E-F). In contrast, signaling pathways related to collagen metabolism, including Collagen Type I Alpha 1 Chain (COL1A1) and Collagen Type I Alpha 2 Chain (COL1A2), were down-regulated in PD (Fig. [80]2E, G). These findings suggest that enhanced lipid metabolism in mast cells within the PD microenvironment influences fibroblast activity through inflammatory signaling pathways, disrupting local tissue homeostasis. This imbalance promotes bone destruction while inhibiting osteogenesis, providing a potential mechanistic explanation for the close association between altered lipid markers and periodontitis in patients. Discussion The association between periodontitis and dyslipidemia has garnered increasing attention [[81]5, [82]6, [83]27]. Several studies have demonstrated a significant correlation between periodontitis and dyslipidemia [[84]5, [85]8, [86]27]. Specifically, the risk of periodontitis is substantially associated with elevated levels of TG [[87]7, [88]27, [89]28], the TG/HDL-C ratio [[90]29], TC [[91]7, [92]22, [93]27, [94]28], LDL-C [[95]28, [96]30], RC [[97]31], and reduced HDL-C [[98]7, [99]27, [100]30]. The current study reveals that elevated levels of TC, LDL-C, RC and TG are significantly associated with an increased risk of severe periodontitis. Furthermore, the proportions of various immune cell subsets, including neutrophils, monocytes, plasma cells, and mast cell clusters, were markedly higher in the periodontitis group. Additionally, the levels of lipid metabolism in immune cells, specifically mast cells, were also notably elevated. Thus, this study identified a significant positive correlation between lipid levels and periodontitis, and enhanced lipid metabolism of immune cells, particularly mast cells, as well as up-regulated interactions with fibroblasts through inflammatory signaling pathways in the periodontitis group, suggesting that abnormal lipid metabolism may exacerbate periodontal disease by modulating the metabolic and inflammatory responses of immune cells, particularly mast cells. Dyslipidemia induces systemic inflammatory burden, potentially mediated by dysregulated lipid metabolism in immune cells [[101]32, [102]33], leading to elevated levels of pro-inflammatory cytokines [[103]34]. As we have demonstrated in our findings, studies have also demonstrated an increase in proportion of mast cells in periodontitis [[104]35, [105]36], indicating a close association between mast cells and the disease. Mast cells are capable of secreting various enzymes that degrade the extracellular matrix (ECM), thereby promoting the progression of periodontitis [[106]37]. In this study, scRNA-seq analysis revealed that mast cells exhibit abnormally active lipid metabolism in periodontitis, with high expression of inflammation-related genes such as IL7, IL15, and CTSG, which are associated with lipid accumulation [[107]38]. These genes exert their effects on fibroblasts and osteoblasts via corresponding receptor-ligand interactions, inhibiting the synthesis of collagen and other bone matrix components. Additionally, genes like CTSG activate extracellular matrix metalloproteinases at inflammatory sites through receptor-ligand interactions such as CTSG-PARD3, which was found up-regulated in periodontitis tissue in this study, leading to the degradation of ECM components [[108]39, [109]40], and stimulate the expression of Receptor Activator of Nuclear Factor Kappa-B Ligand (RANKL) to activate osteoclast precursors [[110]41], thereby promoting osteoclast differentiation and further collagen degradation and over-resorption of bone matrix. The bone matrix, primarily composed of collagen including COL1, is reduced in synthesis due to the downregulation of COL1A1/COL1A2 gene expression [[111]42]. The imbalance in bone remodeling, characterized by enhanced collagen degradation by osteoclasts and matrix metalloproteinases and reduced collagen synthesis in the inflammatory microenvironment, directly contributes to the loss of bone matrix. Furthermore, collagen itself exerts an inhibitory effect on osteoclasts [[112]43], and its degradation further activates osteoclasts, exacerbating bone matrix loss. Clinical studies have also shown a strong correlation between mast cell infiltration and the immaturity of collagen in periodontitis [[113]44]. Thereby, we propose that dyslipidemia, as a systemic factor, contributes to the inflammatory microenvironment by increasing systemic inflammatory burden and dysregulating lipid metabolism in immune cells, with mast cells playing a pivotal role. Dyslipidemia triggers the release of various inflammatory mediators and enzymes from immune cells, including mast cells, which regulate osteoclast and osteoblast activity through receptor-ligand interactions such as CTSG-PARD3. This regulatory mechanism promotes osteoclast differentiation via the upregulation of RANKL signaling while simultaneously suppressing bone matrix collagen synthesis through the downregulation of COL1A1 and COL1A2. Consequently, these processes disrupt periodontal bone remodeling, leading to bone loss and exacerbating the progression of periodontitis. Therefore, modulating lipid levels may represent a promising therapeutic strategy for the management of periodontitis. Although statin-induced improvements in lipid profiles have been shown to enhance gingival index (GI) scores, they do not significantly ameliorate periodontitis [[114]45]. Consequently, a dual-targeted therapeutic approach that simultaneously addresses lipid metabolism and inflammatory pathways may be essential to achieve superior clinical outcomes in periodontitis. Key genes identified in this study, including CTSG, COL1A1, and COL1A2, which potentially play pivotal roles in the interplay between dyslipidemia and periodontitis, could serve as novel therapeutic targets for intervention. One of the key strengths of this study is its inclusion of a substantial participant population, which enhances the reliability of the conclusion that periodontitis and dyslipidemia are associated. However, the NHANES database has inherent limitations in its data collection methods and variables, which do not fully align with the detailed parameters required by the 2017 World Workshop on the Classification of Periodontal and Peri-Implant Diseases and Conditions for accurate staging and classification of periodontitis. Consequently, the updated classification system could not be applied in this study. Additionally, the mechanisms discussed in this study have not yet been experimentally validated. To address these limitations, future studies should prospectively enroll participants, improve periodontal data collection by adopting the latest classification methods, and incorporate medication usage into the analysis to evaluate the impact of lipid level changes on periodontitis treatment outcomes. Moreover, scRNA-seq analysis of periodontal tissues from individuals with varying lipid profiles and periodontal conditions should be conducted to further elucidate the underlying mechanisms. Conclusions The accumulation of blood lipids is significantly associated with an increased risk of severe periodontitis. This association may be mechanistically explained by the infiltration and enhanced lipid metabolism of immune cells, particularly mast cells, as well as up-regulated interactions with fibroblasts through inflammatory signaling pathways in the periodontitis group, ultimately contributing to bone resorption. These findings provide potential therapeutic targets for managing periodontal disease, especially in patients with concurrent dyslipidemia. Furthermore, the results emphasize the importance of closely monitoring lipid levels in periodontitis patients with dyslipidemia, in conjunction with active periodontal therapy. Acknowledgements