Abstract
Background
Porphyromonas gingivalis (P. gingivalis) is a vital pathogen that can increase mortality. However, whether serum carotenoids, commonly known antioxidants, modify the association of serum P. gingivalis immunoglobulin G (IgG) antibody levels with mortality remains uncertain. We aimed to explore the modification of this association by serum carotenoids.
Methods
The associations of serum P. gingivalis IgG levels with mortality among 7,043 participants from the Third U.S. National Health and Nutrition Examination Survey were assessed. The modifications of serum carotenoids on the associations were also examined. We used the recommended sampling weights to account for the complex survey design. Surveycox proportional hazards regression models were used to calculate hazard ratios (HRs) and 95% confidence intervals (CIs) for all-cause and specific-cause mortality.
Results
We identified 2,646 deaths during a median follow-up period of 25.8 years. Compared with participants with low serum P. gingivalis IgG levels (< 69 EU), those with high levels (≥ 69 EU) were associated with an HR (95% CI) of 1.10 (1.04–1.16) for all-cause mortality. Similar results were found in mortality analyses for cardiovascular disease, cerebrovascular disease, and diabetes mellitus (all P < 0.05). Notably, significant interactions were detected between the serum α-carotene, β-carotene, or lutein/zeaxanthin concentrations and the serum P. gingivalis IgG levles on all-cause mortality (all Pint < 0.05). Compared with high P. gingivalis IgG participants with low carotenoids, low P. gingivalis IgG participants with high carotenoids had the lowest risk of all-cause mortality, with HR of 0.74 (0.68–0.80) for α-carotene, 0.78 (0.72–0.86) for β-carotene, 0.70 (0.64–0.76) for β-cryptoxanthin, 0.79 (0.73–0.85) for lutein/zeaxanthin and 0.64 (0.59–0.69) for lycopene, respectively.
Conclusion
Serum carotenoids mitigate the associations between serum P. gingivalis IgG levels and mortality.
Supplementary Information
The online version contains supplementary material available at https://doi.org/10.1186/s40795-026-01420-2.
Keywords: Serum carotenoids, Serum P. gingivalis IgG, Mortality, Cohort study
Introduction
Diverse and abundant microbial communities are present in the human oral cavity, and play a pivotal role in maintaining the delicate balance of the local oral environment. When the homeostasis of these microbial communities is disrupted, a wide array of pathological reactions can be triggered, ranging from periodontal diseases to cardiovascular and metabolic ailments, neurologic disorders, gastrointestinal issues, and even cancers [1]. Among the numerous microbiota, Porphyromonas gingivalis (P. gingivalis) stands out as a particularly troublesome and widely recognized gram-negative microbe [2]. This microorganism is one of the most severe pathogens associated with periodontal infections, posing a significant threat to oral health [3].
In most individuals afflicted with oral diseases, there is a marked elevation in serum immunoglobulin G (IgG) antibodies directed against P. gingivalis (P. gingivalis IgG) [4]. Serum P. gingivalis IgG titers persist at elevated levels for up to 15 years [5], which is indicative of both past and ongoing periodontal infections. Consequently, P. gingivalis IgG serves as a cumulative measure of periodontal infection over time [6]. P. gingivalis IgG has been linked to various health outcomes, including all-cause mortality [7], cardiovascular diseases (CVD) [8–10], rheumatoid arthritis [11], chronic kidney disease [12], cognitive impairment [13, 14], and various types of cancers [15, 16]. In previous research conducted by Damian et al., a nonlinear correlation was identified between serum P. gingivalis IgG and mortality via the data from the Third US National Health and Nutrition Examination Survey (NHANES III), which followed up to 2006 [7]. We further utilized NHANES III data followed up to 2019 to explore the associations between P. gingivalis and all-cause and specific-cause mortality.
Carotenoids are a widely recognized class of antioxidants and immunomodulatory compounds, with more than 700 unique varieties identified to date. Among them, α-carotene, β-carotene, β-cryptoxanthin, lycopene, lutein, and zeaxanthin account for more than 95% of the total carotenoids [17, 18]. Researches have shown that carotenoids can effectively prevent diseases such as periodontitis, indicating their potential to regulate the oral microbiota, including inflammation caused by P. gingivalis [19, 20]. Given the strong correlation between periodontitis and mortality, coupled with the antioxidant and anti-inflammatory functions of carotenoids, we hypothesize that serum carotenoids may modify the association between serum antibodies against P. gingivalis and all-cause and specific-cause mortality.
Therefore, in this study, we utilized data from NHANES III to further evaluate the associations between serum P. gingivalis IgG and mortality, with follow-up extending until 2019. Additionally, we examined the modified associations of carotenoids on these associations, providing valuable insights and evidence for the prevention of death linked to the oral microbiota.
Methods
Study population
The study participants were from NHANES III, which was described previously [21]. In short, NHANES III was conducted by the National Center for Health Statistics (NCHS) between 1988 and 1994 in two phases in which the civilian noninstitutionalized U.S. population was sampled to provide a multidimensional overview of physical health status and aspects of functional health status. After excluding participants with missing information on serum carotenoids, serum P. gingivalis IgG, mortality status, and those who were pregnant or already had cancer at the baseline, a total of 7,043 participants ≥ 20 years of age from NHANES III were included in this study (Figure S1). Patients or the public were not involved in the design, conduct, reporting, or dissemination plans of our research.
Covariate collection
Each participant’s baseline demographic characteristics, disease and medication history, and lifestyle were collected via face-to-face questionnaires, including sex (male or female), age groups (20–39,40–59, ≥ 60 years), race-ethnicity (non-Hispanic White, non-Hispanic Black, Mexican American, other), education (less than junior high school, junior high school, high school, college or above), annual total income of the family (< $20,000, ≥ $20,000), smoking status (current smokers, former smokers, never smokers), drinking status (≥ 12 drinks in the past 12 months, < 12 drinks in the past 12 months), exercise status (more active, less active, about the same), pregnancy status (yes or no), and self-reported cancer status (yes or no). Body mass index (BMI) was calculated by dividing weight in kilograms (kg) by height in meters squared (m2) and was divided into four groups (< 18.5, 18.5–24.9, 25–29.9, or > 30 kg/m2).
Assessment of the serum IgG antibody against P. gingivalis
Serum P. gingivalis IgG levels at baseline was chosen for this study to reflect evidence of systemic exposure to this bacterium, as it is a validated periodontal pathogen [22] and has well-described periodontal pathogenicity [23]. The levels of the serum P. gingivalis IgG were quantified via enzyme-linked immunosorbent assay, and the antibody measurements were reported in units of IgG (EU), which has been reported in detail in the documentation of periodontal pathogens of NHANES III [24]. Based on a previous study [15], we divided the P. gingivalis IgG levels into three categories: low (< 69 EU), intermediate (69–119.0 EU), and high (> 119.0 EU).
Assessment of serum carotenoids
In this study, we included six serum carotenoids represented by five serum measures including α-carotene, β-carotene, β-cryptoxanthin, lutein/zeaxanthin, and lycopene at baseline which are available from NHANES III. Serum concentrations of these carotenoids were measured via high-performance liquid chromatography (HPLC) in NHANES III. The laboratory procedures and quality control methods used for the serum carotenoids measurements were described in detail previously [25, 26]. Serum carotenoids concentrations were divided into low or high groups based on the median.
Ascertainment of mortality
Data on all-cause and specific-cause of death were obtained from the public-use data files available from the National Center for Health Statistics (NCHS). Vital status was determined by the National Death Index (NDI), and the cause of death was determined by the basic cause of death listed on the death certificate. Vital status and the cause of death were linked to data from NHANES III for further analysis. The International Classification of Diseases (ICD), 9th Revision (ICD-9) was used for coding all causes of death occurring before 1999, while the ICD-10 was used for coding deaths after 1998. The NCHS created a UCOD_113 variable to recode all deaths into comparable underlying causes of death groups, to assist researchers in analyzing specific death periods throughout the entire survey. We selected all causes of death and several major causes of death, including CVD, cancer, cerebrovascular diseases (CeVD), and diabetes mellitus (DM) as outcomes. For each study participant, follow-up extended from baseline to the date of death or end of follow-up (31 December, 2019), whichever occurred first.
Statistical analysis
Baseline demographic characteristics were described according to serum P. gingivalis IgG levels (low [< 69 EU], intermediate [69–119.0 EU], and high [> 119.0 EU]). The demographic characteristics of the different P. gingivalis IgG groups were compared via the chi-square test (categorical variables) or one-way analysis of variance (continuous variables). Person-years were calculated from baseline to the date of death or the end of follow-up (31 December, 2019), whichever occurred first.
We used the recommended sampling weights to account for the NHANES' complex survey design. Specifically, we applied the MEC- and home-examined sample final weight (WTPFHX6), with variance estimated using the pseudo-stratum (SDPSTRA6) and pseudo-PSU (SDPPSU6) variables as recommended by NCHS. Two variable-adjusted Surveycox proportional hazards regression models were conducted to calculate hazard ratios (HRs) and 95% confidence intervals (CIs) for all-cause and specific-cause mortality in relation to serum P. gingivalis IgG levels. Model 1 was adjusted for sex and age groups. Model 2 was additionally adjusted for potential confounders: race-ethnicity, education, income, smoking status, drinking status, exercise, and BMI groups. The proportional hazards assumption was examined by creating a product term of follow-up time and exposure, and we found no significant deviation from the assumption.
We further conducted stratified analyses by groups of serum carotenoids (low or high) to investigate associations of the serum P. gingivalis IgG levels with all-cause and specific-cause mortality. To assess multiplicative interaction, we included in the Cox model the main terms for serum P. gingivalis IgG (dichotomized as low [< 69 EU] vs. high [≥ 69 EU]) and serum carotenoids (low vs. high), along with their product term. Both dichotomous variables were coded as 1 and 2, and the product term was calculated as the simple multiplication of the two. The Wald test for the coefficient of the product term was used to evaluate interaction on the multiplicative scale, and the corresponding hazard ratio (HR) with its 95% confidence interval (CI) is reported as the measure of interaction. To analyze the joint associations of P. gingivalis IgG concentration and carotenoids concentration on mortality, we divided the participants into four groups according to P. gingivalis IgG (< 69 EU vs. ≥ 69 EU) and the serum carotenoids (low vs. high) and estimated HRs (95% CIs) of mortality in different groups compared with those with serum P. gingivalis IgG ≥ 69 EU and low serum carotenes.
We also performed sensitivity analyses using tertiles of the serum P. gingivalis IgG distribution. In addition, the high-exposure group was redefined as the top two tertiles combined, and the stratified analyses, interaction analysis as well as the joint association models were repeated using the alternative cutoff.
Taking into account the complex survey sampling of NHANES, the survival probabilities for all-cause mortality and specific cause mortality were calculated and plotted via the R packages “survey” and “jskm”, respectively. Restricted cubic spline (RCS) curves for sampling weights were depicted to show the nonlinear relationships between serum P. gingivalis IgG levels and mortality. All analyses were performed via SAS 9.4 (SAS Institute Inc., Cary, NC, USA) and R 4.0.2 (R Core Team, Vienna, Austria), and a two-sided P < 0.05 was considered statistically significant.
Results
Descriptive statistics
A total of 7,043 (3,127 males; 44.4%) participants were included, and their characteristics according to the groups of serum P. gingivalis IgG (< 69 EU, 69–119 EU, and > 119 EU) are displayed in Table 1. Participants with higher serum P. gingivalis IgG levels tended to be older, non-Hispanic Black and Mexican American, non-current smokers, who had attended less than junior high school. These participants also tended to have a higher BMI, a lower income, were less likely to be physically active, and were less likely to drink alcohol (all P < 0.05).
Table 1.
Baseline characteristics of participants according to serum P. gingivalis IgG
| Characteristics | Total | Serum Porphyromonas gingivalis IgG (EU) | P valuea | ||
|---|---|---|---|---|---|
| < 69 | 69–119 | > 119 | |||
| N | 7043 | 1948 | 3384 | 1711 | |
| Follow-up years, median (IQR) | 25.83 (17.75–27.00) | 25.92 (19.92–27.08) | 25.92 (17.92–27.00) | 25.67 (15.92–26.92) | 0.001 |
| Gender (%) | 0.002 | ||||
| Male | 3127 (44.4) | 805 (41.3) | 1518 (44.9) | 804 (47.0) | |
| Female | 3916 (55.6) | 1143 (58.7) | 1866 (55.1) | 907 (53.0) | |
| Age (years), mean (SD) | 46.48 (18.56) | 45.20 (19.56) | 46.12 (18.43) | 48.64 (17.44) | < 0.001 |
| Age groups (years) (%) | < 0.001 | ||||
| 20–39 | 3099 (44.0) | 960 (49.3) | 1513 (44.7) | 626 (36.6) | |
| 40–59 | 1929 (27.4) | 455 (23.4) | 940 (27.8) | 534 (31.2) | |
| ≥ 60 | 2015 (28.6) | 533 (27.4) | 931 (27.5) | 551 (32.2) | |
| Race-ethnicity (%) | < 0.001 | ||||
| Non-Hispanic white | 2536 (36.0) | 1159 (59.5) | 1089 (32.2) | 288 (16.8) | |
| Non-Hispanic black | 2177 (30.9) | 351 (18.0) | 1063 (31.4) | 763 (44.6) | |
| Mexican American | 1980 (28.1) | 386 (19.8) | 1026 (30.3) | 568 (33.2) | |
| Other | 350 (5.0) | 52 (2.7) | 206 (6.1) | 92 (5.4) | |
| Education (%) | < 0.001 | ||||
| Less than junior high school | 1654 (23.5) | 318 (16.3) | 839 (24.8) | 497 (29.0) | |
| Junior high school | 1100 (15.6) | 264 (13.6) | 571 (16.9) | 265 (15.5) | |
| High school | 2218 (31.5) | 629 (32.3) | 1062 (31.4) | 527 (30.8) | |
| College or above | 2027 (28.8) | 727 (37.3) | 895 (26.4) | 405 (23.7) | |
| Missing | 44 (0.6) | 10 (0.5) | 17 (0.5) | 17 (1.0) | |
| Income (%) | < 0.001 | ||||
| < $20,000 | 3343 (47.5) | 756 (38.8) | 1635 (48.3) | 952 (55.6) | |
| ≥ $20,000 | 3562 (50.6) | 1160 (59.5) | 1674 (49.5) | 728 (42.5) | |
| Missing | 138 (2.0) | 32 (1.6) | 75 (2.2) | 31 (1.8) | |
| Smoking status (%) | < 0.001 | ||||
| Current smokers | 1767 (25.1) | 535 (27.5) | 897 (26.5) | 335 (19.6) | |
| Former smokers | 1588 (22.5) | 426 (21.9) | 773 (22.8) | 389 (22.7) | |
| Never smokers | 3688 (52.4) | 987 (50.7) | 1714 (50.7) | 987 (57.7) | |
| Drinking status (%) | 0.001 | ||||
| ≥ 12 drinks in the past 12 months | 5548 (78.8) | 1581 (81.2) | 2670 (78.9) | 1297 (75.8) | |
| < 12 drinks in the past 12 months | 1318 (18.7) | 322 (16.5) | 639 (18.9) | 357 (20.9) | |
| Missing | 177 (2.5) | 45 (2.3) | 75 (2.2) | 57 (3.3) | |
| Exercise (%) | 0.15 | ||||
| More active | 2176 (30.9) | 617 (31.7) | 1060 (31.3) | 499 (29.2) | |
| Less active | 1556 (22.1) | 408 (20.9) | 734 (21.7) | 414 (24.2) | |
| About the same | 3217 (45.7) | 893 (45.8) | 1552 (45.9) | 772 (45.1) | |
| Missing | 94 (1.3) | 30 (1.5) | 38 (1.1) | 26 (1.5) | |
| BMI groups | < 0.001 | ||||
| < 18.5 | 136 (1.9) | 40 (2.1) | 70 (2.1) | 26 (1.5) | |
| 18.5 ~ < 25 | 2457 (34.9) | 792 (40.7) | 1182 (34.9) | 483 (28.2) | |
| 25 ~ < 30 | 2455 (34.9) | 653 (33.5) | 1186 (35.0) | 616 (36.0) | |
| ≥ 30 | 1988 (28.2) | 463 (23.8) | 942 (27.8) | 583 (34.1) | |
| Missing | 7 (0.1) | 0 (0.0) | 4 (0.1) | 3 (0.2) | |
| α-carotene (nmol/L), median (IQR) | 70 (40–110) | 70 (40–110) | 70 (40–110) | 60 (20–90) | 0.001 |
| β-carotene (nmol/L), median (IQR) | 260 (150–450) | 280 (170–470) | 260 (150–450) | 260 (150–430) | 0.07 |
| β-cryptoxanthin (nmol/L), median (IQR) | 160 (110–240) | 140 (110–220) | 160 (110–250) | 160 (110–250) | < 0.001 |
| Lutein/zeaxanthin (nmol/L), median (IQR) | 350 (260–470) | 325 (250–440) | 350 (260–490) | 370 (260–510) | < 0.001 |
| Lycopene (nmol/L), median (IQR) | 370 (260–540) | 410 (280–580) | 370 (260–520) | 350 (240–500) | < 0.001 |
Abbreviation: SD standard deviation, 95% CI 95% confidence interval, BMI Body mass index, IQR Interquartile Range
aP-values were based on Chi-square test (categorical variables) or one-way analysis of variance (normal continues variables) or Kruskal–Wallis Rank Sum Test (nonnormal continues variables)
Associations of serum P. gingivalis IgG levels with mortality
The survival probability of participants with all-cause mortality was highest among those with low serum P. gingivalis IgG levels (< 69 EU), followed by those with intermediate (69–119 EU) and high P. gingivalis IgG levels (> 119 EU) (P < 0.001; Fig. 1A). Similar results were found for CVD, cancer, and CeVD mortality (Figure S2). The dose–response association results revealed that serum P. gingivalis IgG levels were linearly associated with the risk of all-cause, CVD, cancer, CeVD, and DM mortality (all Pnonlinearity > 0.05; Fig. 1B and Figure S3).
Fig. 1.
The survival probability of all-cause mortality by serum P. gingivalis IgG groups and the dose–response associations between serum P. gingivalis IgG levels and all-cause mortality. A survival curve; B restricted cubic spline. Abbreviation: P. gingivalis, Porphyromonas gingivalis; HR, hazard ratio; CI, confidence interval; IgG, immunoglobulin G. The restricted cubic spline model was adjusted by sex (males or females), age (20–39, 40–59, ≥ 60 years), race-ethnicity (non-Hispanic white, non-Hispanic black, Mexican American and other), education (less than junior high school, junior high school, high school, college or above), income (< $20,000, ≥ $20,000), smoking status (current smokers, former smokers, never smokers), drinking status (≥ 12 drinks in the past 12 months, < 12 drinks in the past 12 months), exercise (more active, less active, about the same), body mass index (< 18.5, 18.5–24.9, 25–29.9, ≥ 30 kg/m.2)
The associations of classified P. gingivalis IgG with all-cause and specific mortality are shown in Table 2. A total of 2,646 all-cause deaths, 761 CVD deaths, 553 cancer deaths, 132 CeVD deaths, and 111 DM deaths were identified during a median follow-up period of 25.8 years. Compared with low P. gingivalis IgG levels (< 69 EU), those with intermediate (69–119 EU) and high P. gingivalis IgG levels (> 119 EU) were associated with multivariable-adjusted HRs (95% CIs) of 1.10 (1.04 to 1.16) and 1.11 (1.05 to 1.17) for all-cause mortality (Ptrend < 0.001); 1.23 (1.08 to 1.39) and 1.22 (1.03 to 1.44) for CVD mortality (Ptrend = 0.010); 1.06 (0.91 to 1.23) and 1.22 (1.09 to 1.37) for cancer mortality (Ptrend = 0.007); 1.31 (1.13 to 1.53) and 1.25 (1.04 to 1.49) for CeVD mortality (Ptrend = 0.012); and 2.81 (2.54 to 3.12) and 2.18 (1.85 to 2.58) for DM mortality (Ptrend < 0.001), respectively. Additionally, compared with low P. gingivalis IgG levels (< 69 EU), those with high P. gingivalis IgG levels (≥ 69 EU) have a higher risk of all-cause mortality (HR, 1.10; 95% CI, 1.05 to 1.16), CVD mortality (HR, 1.23; 95% CI, 1.09 to 1.38), CeVD mortality (HR, 1.30; 95% CI, 1.11 to 1.51), and DM mortality (HR, 2.66; 95% CI, 2.40 to 2.96), respectively. Sensitivity analyses using tertile-based cutoffs yielded generally similar results (Table S1).
Table 2.
Hazard ratios (95% CIs) of mortality in relation to serum P. gingivalis IgG
| Serum Porphyromonas gingivalis IgG | |||||
|---|---|---|---|---|---|
| < 69 EU | 69–119 EU | > 119 EU | P for trend | 69 + EU | |
| All-cause mortality | |||||
| Cases/person-years | 681/43,037 | 1238/73,783 | 727/36,267 | 1965/110,050 | |
| Age and gender-adjusted HR (95% CI) | 1.00 (ref) | 1.15 (1.10–1.20) | 1.15 (1.05–1.26) | 0.002 | 1.15 (1.09–1.21) |
| Multivariable-adjusted HR (95% CI)* | 1.00 (ref) | 1.10 (1.04–1.16) | 1.11 (1.05–1.17) | < 0.001 | 1.10 (1.05–1.16) |
| CVD mortality | |||||
| Mortality/person-years | 200/36,132 | 350/61,714 | 211/28,845 | 561/90,558 | |
| Age and gender-adjusted HR (95% CI) | 1.00 (ref) | 1.25 (1.10–1.43) | 1.15 (1.01–1.32) | 0.02 | 1.23 (1.08–1.39) |
| Multivariable-adjusted HR (95% CI)* | 1.00 (ref) | 1.23 (1.08–1.39) | 1.22 (1.03–1.44) | 0.010 | 1.23 (1.09–1.38) |
| Cancer mortality | |||||
| Mortality/person-years | 134/35,640 | 256/60,893 | 163/28,446 | 419/89,339 | |
| Age and gender-adjusted HR (95% CI) | 1.00 (ref) | 1.03 (0.96–1.10) | 1.15 (0.99–1.35) | 0.07 | 1.06 (0.99–1.14) |
| Multivariable-adjusted HR (95% CI)* | 1.00 (ref) | 1.06 (0.91–1.23) | 1.22 (1.09–1.37) | 0.007 | 1.10 (0.97–1.25) |
| CeVD mortality | |||||
| Mortality/person-years | 48/34,474 | 58/58,116 | 26/26,639 | 84/84,755 | |
| Age and gender-adjusted HR (95% CI) | 1.00 (ref) | 1.22 (1.07–1.40) | 1.20 (0.98–1.47) | 0.047 | 1.22 (1.05–1.42) |
| Multivariable-adjusted HR (95% CI)* | 1.00 (ref) | 1.31 (1.13–1.53) | 1.25 (1.04–1.49) | 0.01 | 1.30 (1.11–1.51) |
| DM mortality | |||||
| Mortality/person-years | 16/33,946 | 60/58,191 | 35/26,764 | 95/84,955 | |
| Age and gender-adjusted HR (95% CI) | 1.00 (ref) | 2.55 (2.38–2.74) | 2.30 (1.97–2.69) | < 0.001 | 2.49 (2.30–2.69) |
| Multivariable-adjusted HR (95% CI)* | 1.00 (ref) | 2.81 (2.54–3.12) | 2.18 (1.85–2.58) | < 0.001 | 2.66 (2.40–2.96) |
Abbreviations: HR hazard ratio, CI confidence interval, CVD cardiovascular disease, CeVD cerebrovascular disease, DM diabetes mellitus
*The cox proportional hazard models were adjusted by sex (males or females), age (20–39,40–59, ≥ 60 years), race-ethnicity (non-Hispanic white, non-Hispanic black, Mexican American and other), education (less than junior high school, junior high school, high school, college or above), income (< $20,000, ≥ $20,000), smoking status (current smokers, former smokers, never smokers), drinking status (≥ 12 drinks in the past 12 months, < 12 drinks in the past 12 months), exercise (more active, less active, about the same), body mass index (< 18.5, 18.5–24.9, 25–29.9, ≥ 30 kg/m2)
Interactions of serum carotenoids and P. gingivalis IgG levels on mortality
Significant interactions were found between serum α-carotene, β-carotene, or lutein/zeaxanthin and P. gingivalis IgG on all-cause mortality (all Pint < 0.05) (Fig. 2). Higher P. gingivalis IgG levels were associated with increased risks of all-cause mortality among individuals in various serum carotenoids subgroups, whereas the associations were weaker among those in the higher serum carotenoids subgroups (Fig. 2). Specifically, compared with low serum P. gingivalis IgG levels (< 69 EU), high P. gingivalis IgG levels (≥ 69 EU) were associated with multivariable-adjusted HRs (95% CIs) of 1.19 (1.12 to 1.26) in the lower α-carotene group and 1.06 (1.03 to 1.09) in the higher α-carotene group (Pint = 0.003), 1.24 (1.12 to 1.37) in the lower β-carotene group and 1.03 (0.97 to 1.10) in the higher β-carotene group (Pint = 0.04), and 1.21 (1.11 to 1.32) in the lower lutein/zeaxanthin group and 1.02 (0.95 to 1.09) in the higher lutein/zeaxanthin group (Pint = 0.03) (Fig. 2).
Fig. 2.
The interactions between serum carotenoids and serum P. gingivalis IgG levels on all-cause mortality. Abbreviation: P. gingivalis, Porphyromonas gingivalis; IgG, immunoglobulin G; CI, confidence interval. The cox proportional hazard models were adjusted by sex (males or females), age (20–39, 40–59, ≥ 60 years), race-ethnicity (non-Hispanic white, non-Hispanic black, Mexican American and other), education (less than junior high school, junior high school, high school, college or above), income (< $20,000, ≥ $20,000), smoking status (current smokers, former smokers, never smokers), drinking status (≥ 12 drinks in the past 12 months, < 12 drinks in the past 12 months), exercise (more active, less active, about the same), body mass index (< 18.5, 18.5–24.9, 25–29.9, ≥ 30 kg/m.2)
In addition, interaction analyses were also conducted for CVD, cancer, CeVD, and DM mortality, the results of which are shown in Figure S4-S7, respectively. Similar results were found for CeVD and DM mortality, except for β-cryptoxanthin for CeVD mortality (all Pint < 0.05) (Figure S6-S7). We found significant interactions of serum β-cryptoxanthin and serum P. gingivalis IgG levels on the risk of cancer and CeVD mortality (both Pint < 0.05), whereas the associations were weaker among those from the lower serum β-cryptoxanthin subgroups when stratified by serum β-cryptoxanthin (Figure S5-S6).
In sensitivity analyses using tertiles to categorize serum carotenoid levels (Table S2), the interaction patterns for all-cause, CVD, CeVD, and DM mortality remained generally similar. In the primary analysis for cancer mortality, no significant interaction was observed between serum levels of α-carotene, β-carotene and P. gingivalis IgG. However, in the sensitivity analysis, a significant interaction emerged (Table S2).
Joint analyses of serum carotenoid and P. gingivalis IgG levels with mortality
We observed significant joint associations of serum carotenoids and P. gingivalis IgG on all-cause mortality in an exposure–response manner (Fig. 3). Specifically, compared with participants with higher serum P. gingivalis IgG levels (≥ 69 EU) and lower serum carotenoids, participants with lower serum P. gingivalis IgG levels (< 69 EU) and higher serum carotenoids presented the lowest risk of all-cause mortality, with HRs (95% CIs) of 0.74 (0.68–0.80) for α-carotene, 0.78 (0.72–0.86) for β-carotene, 0.69 (0.64–0.76) for β-cryptoxanthin, 0.79 (0.73–0.85) for lutein/zeaxanthin and 0.64 (0.59–0.69) for lycopene, (Fig. 3). The results of the joint analyses of serum carotenoids and serum P. gingivalis IgG levels with CVD and cancer mortality were similar, but not for CeVD and DM (Figure S8). Sensitivity analyses using tertile-based cutoffs gave similar findings for joint analyses(Table S3).
Fig. 3.
The joint associations of serum carotenoids and serum P. gingivalis IgG levels on all-cause mortality. Abbreviation: P. gingivalis, Porphyromonas gingivalis; IgG, immunoglobulin G. The cox proportional hazard models were adjusted by sex (males or females), age (20–39, 40–59, ≥ 60 years), race-ethnicity (non-Hispanic white, non-Hispanic black, Mexican American and other), education (less than junior high school, junior high school, high school, college or above), income (< $20,000, ≥ $20,000), smoking status (current smokers, former smokers, never smokers), drinking status (≥ 12 drinks in the past 12 months, < 12 drinks in the past 12 months), exercise (more active, less active, about the same), body mass index (< 18.5, 18.5–24.9, 25–29.9, ≥ 30 kg/m.2)
Discussion
In the national population-based prospective cohort study conducted in NHANES III, we observed that serum P. gingivalis IgG levels were positively associated with all-cause mortality. We also found that the association was significantly modified by serum α-carotene, β-carotene, and lutein/zeaxanthin, and the associations between serum P. gingivalis IgG levels and all-cause mortality were weaker among those with higher carotenoid levels. The lowest risks of all-cause mortality were observed in those with low P. gingivalis IgG levels and high carotenoids levels. In addition, we observed that serum P. gingivalis IgG levels were positively associated with CVD, cancer, CeVD, and DM mortality. Significant interactions were found between serum carotenoid and serum P. gingivalis IgG levels with respect to CeVD and DM mortality.
Our findings revealed a positive association between serum P. gingivalis IgG levels and all-cause mortality. Dose–response association analysis revealed a linear relationship between the serum P. gingivalis IgG concentration and the risk of all-cause mortality. However, a prospective cohort study conducted by Damian et al. in 2015 reported a nonlinear association of serum P. gingivalis IgG levels with mortality [7]. This difference in findings may suggest that the relationship between P. gingivalis IgG levels and mortality is more complex than previously understood. Although the precise mechanisms underlying the observed positive association between the serum P. gingivalis IgG and the all-cause mortality remain unclear, several potential mechanisms may explain this relationship. The pathogenicity of P. gingivalis is primarily attributed to its ability to interfere with the host immune system, through virulence factors and cellular immune responses, which are mediated mainly by T cells. These immune modulations may influence the etiology of various systemic diseases, such as atherosclerosis, hypertension, adverse pregnancy outcomes, inflammatory bowel disease, DM, nonalcoholic fatty liver disease, rheumatoid arthritis, and Alzheimer’s disease [27]. These pathological effects may ultimately contribute to increase risks in all-cause mortality.
In addition, our study revealed that serum P. gingivalis IgG levels were positively associated with CVD, cancer, CeVD, and DM mortality. A previous study revealed that the P. gingivalis antibody titer in the serum is significantly positively correlated with the aortic valve area and Doppler velocity index [10]. Additionally, P. gingivalis has been shown to regulate atherosclerosis through immune pathways [28], further indicating a substantial relationship between this oral bacterium and the pathogenesis of CVD. Possible oncogenic mechanisms of P. gingivalis include the production of virulence factors or active metabolites, the activation of oncogenic signaling pathways, the promotion of aberrant immune responses, and the inhibition of host cell apoptosis, all of which align with our findings related to cancer mortality [29]. This study also revealed a significant link between serum P. gingivalis IgG levels and DM mortality, possibly because P. gingivalis can trigger a systemic inflammatory response, disrupt insulin signaling, and induce β-cell hypofunction and decrease [30]. These findings suggest that P. gingivalis infection may play a significant role in the pathogenesis and prognosis of these life-threatening conditions.
For the first time, we found that serum carotenoids significantly modified the relationship between serum P. gingivalis IgG levels and all-cause mortality. This discovery suggests that serum carotenoids may reduce the risk of death associated with P. gingivalis. Serum carotenoids are well-established composite biomarkers of fruit and vegetable intake and systemic antioxidant status [31, 32]. Their modifying associations likely reflects a direct counteraction of P. gingivalis-induced oxidative stress and inflammation through multiple mechanisms. Carotenoids are potent lipophilic antioxidants that may scavenge reactive oxygen species (ROS) generated during P. gingivalis infection and other metabolic processes [33]. At the molecular level, carotenoids activate the transcription factor nuclear factor erythroid 2-related factor 2 (Nrf2), which in turn upregulates the expression of a battery of antioxidant and cytoprotective enzymes, such as heme oxygenase-1 (HO-1) [34]. Concurrently, carotenoids are known to inhibit the activation of nuclear factor-κB (NF-κB) [34], a key pro-inflammatory transcription factor, and suppress the production of major pro-inflammatory cytokines including tumor necrosis factor (TNF), interleukin (IL)−1β, and IL-6 [35]. This finding explains the specific positive impact that carotenoids exert on the body, highlighting their potential health benefits. Similarly, vitamin D has been reported to reduce P. gingivalis infection by promoting autophagy, enhancing innate immunity in gingival epithelial cells, and modulating periodontal inflammation [36–39]. These findings suggest that, like carotenoids, vitamin D may also exert a protective effect against P. gingivalis-related mortality, although this hypothesis warrants further investigation.
The interactions between serum carotenoid and serum P. gingivalis IgG levels on DM are obvious both in primary analyses and sensitivity analyses. Individuals with high carotenoid levels have a reduced risk of DM mortality associated with serum P. gingivalis IgG levels. These results indicated that carotenoids may effectively alleviate the association of P. gingivalis IgG levels on DM mortality. Under high glucose conditions, β-carotene significantly inhibited the production of TNF, IL-1 and monocyte chemotactic protein-1 (MCP-1) induced by P. gingivalis lipopolysaccharide (LPS) in human acute monocytic leukemia cells (THP)−1 monocytes [40]. This study suggested that β-carotene may be a potential anti-inflammatory nutrient for circulating P. gingivalis LPS-mediated cytokine production in diabetic patients with periodontitis [40]. β-carotene may help interfere with diabetic complications by targeting intracellular NF-kB. Diabetic patients may be able to prevent or delay the development of several complications mediated by periodontitis by using appropriate dietary treatment with β-carotene [20]. For cancer mortality, the modifying associations of serum carotenoids was less consistent. No significant interaction was observed in the primary analysis, but a significant interaction emerged in the sensitivity analysis. This discrepancy may reflect differences in cut-point selection and statistical power, as well as the complexity of the tumor microenvironment and cancer heterogeneity. Further studies are needed to validate these findings.
To the best of our knowledge, this is the first prospective study to assess the interaction between serum carotenoid and serum P. gingivalis IgG levels on mortality. The main purpose of this study was to explore the associations between serum P. gingivalis IgG levels and all-cause mortality and specific-cause mortality in a national prospective cohort survey population with a median follow-up period of 25.8 years and to evaluate the modifying associations of serum carotenoids on this association. The findings revealed that carotenoids may have a protective effect on mortality associated with serum P. gingivalis IgG levels. Our study also has several limitations. First, while P. gingivalis is a well-characterized oral bacterium, serum antibody assessment is an indirect measure of oral gastrointestinal bacterial exposure on the basis of individual variable vascular exposure and the immune response. Second, serum carotenoids and P. gingivalis IgG were measured only at baseline, and no repeated dietary data were collected during follow-up. This single-measurement approach cannot account for temporal changes in biomarkers or diet, potentially leading to regression dilution [41] and additional exposure misclassification. Third, although baseline age was adjusted for in the Cox modles, age-related mortality risk over the long-term follow-up may still influence the survival curves. The irregularities observed in later follow-up, such as rapid declines or jumps, likely reflect statistical instability due to the small number of participants still at risk. Fourth, we acknowledge that multiple comparisons were made across several carotenoids measures and mortality outcomes, which may increase the risk of type I error. Finally, as an observational cohort study, although we adjusted for multiple covariates, residual confounding from factors such as changes in dietary habits, smoking, oral hygiene practices, or socioeconomic variables during follow-up cannot be completely ruled out; therefore, causal inference in this study is limited.
Conclusion
Higher serum levels of selected carotenoids were associated with a weaker relationship between serum P. gingivalis IgG levels and all-cause, CeVD, and DM mortality. These findings suggest that carotenoids-related nutritional status may play a modifying role in the association between periodontal pathogen exposure and long-term mortality.
Supplementary Information
Additional file 1: Figure S1. The flowchart of this study. Figure S2. The survival probability of specific cause mortality by serum P. gingivalis IgG groups. (A) CVD mortality; (B) Cancer mortality; (C) CeVD mortality; (D) DM mortality Abbreviation:P. gingivalis, Porphyromonas gingivalis; IgG, immunoglobulin G; CI, confidence interval; CVD, cardiovascular disease; CeVD, cerebrovascular disease; DM, diabetes mellitus. Figure S3. The dose-response associations between serum P. gingivalis IgG levels and specific cause mortality. (A) CVD mortality; (B) Cancer mortality; (C) CeVD mortality; (D) DM mortalit. Abbreviation: P. gingivalis, Porphyromonas gingivalis; HR, hazard ratio; CI, confidence interval; IgG, immunoglobulin G; CVD, cardiovascular disease; CeVD, cerebrovascular disease; DM, diabetes mellitus. The restricted cubic spline models were adjusted by sex (males or females), age (20-39, 40-59,≥60 years), race-ethnicity (non-Hispanic white, non-Hispanic black, Mexican American and other), education (less than junior high school, junior high school, high school, college or above), income (<$20,000, ≥$20,000), smoking status (current smokers, former smokers, never smokers), drinking status (≥12 drinks in the past 12 months,<12 drinks in the past 12 months), exercise (more active, less active, about the same), body mass index (<18.5, 18.5-24.9, 25-29.9, ≥30kg/m2). Figure S4. The interactions between serum carotenoids and serum P. gingivalis IgG levels on cardiovascular disease mortality. Abbreviation: P. gingivalis, Porphyromonas gingivalis; IgG, immunoglobulin G; CI, confidence interval; CVD, cardiovascular disease. The cox proportional hazard models were adjusted by sex (males or females), age (20-39, 40-59, ≥60 years), race-ethnicity (non-Hispanic white, non-Hispanic black, Mexican American and other), education (less than junior high school, junior high school, high school, college or above), income (<$20,000, ≥$20,000), smoking status (current smokers, former smokers, never smokers), drinking status (≥12 drinks in the past 12 months,<12 drinks in the past 12 months), exercise (more active, less active, about the same), body mass index (<18.5, 18.5-24.9, 25-29.9, ≥30kg/m2). Figure S5. The interactions between serum carotenoids and serum P. gingivalis IgG levels on cancer mortality. Abbreviation: P. gingivalis, Porphyromonas gingivalis; IgG, immunoglobulin G; CI, confidence interval. The cox proportional hazard models were adjusted by sex (males or females), age (20-39, 40-59, ≥60 years), race-ethnicity (non-Hispanic white, non-Hispanic black, Mexican American and other), education (less than junior high school, junior high school, high school, college or above), income (<$20,000, ≥$20,000), smoking status (current smokers, former smokers, never smokers), drinking status (≥12 drinks in the past 12 months,<12 drinks in the past 12 months), exercise (more active, less active, about the same), body mass index (<18.5, 18.5-24.9, 25-29.9, ≥30kg/m2). Figure S6. The interactions between serum carotenoids and serum P. gingivalis IgG levels on cerebrovascular disease mortality. Abbreviation:P. gingivalis, Porphyromonas gingivalis; IgG, immunoglobulin G; CI, confidence interval; CeVD, cerebrovascular disease. The cox proportional hazard models were adjusted by sex (males or females), age (20-39, 40-59, ≥60 years), race-ethnicity (non-Hispanic white, non-Hispanic black, Mexican American and other), education (less than junior high school, junior high school, high school, college or above), income (<$20,000, ≥$20,000), smoking status (current smokers, former smokers, never smokers), drinking status (≥12 drinks in the past 12 months, <12 drinks in the past 12 months), exercise (more active, less active, about the same), body mass index (<18.5, 18.5-24.9, 25-29.9, ≥30kg/m2). Figure S7. The interactions between serum carotenoids and serum P. gingivalis IgG levels on diabetes mellitus mortality. Abbreviation:P. gingivalis, Porphyromonas gingivalis; IgG, immunoglobulin G; CI, confidence interval; DM, diabetes mellitus. The cox proportional hazard models were adjusted by sex (males or females), age (20-39, 40-59, ≥60 years), race-ethnicity (non-Hispanic white, non-Hispanic black, Mexican American and other), education (less than junior high school, junior high school, high school, college or above), income (<$20,000, ≥$20,000), smoking status (current smokers, former smokers, never smokers), drinking status (≥12 drinks in the past 12 months,<12 drinks in the past 12 months), exercise (more active, less active, about the same), body mass index (<18.5, 18.5-24.9, 25-29.9, ≥30kg/m2). Figure S8. The joint associations of serum carotenoids and serum P. gingivalis IgG levels on specific-cause mortality. Abbreviation: P. gingivalis, Porphyromonas gingivalis; IgG, immunoglobulin G; CI, confidence interval. The cox proportional hazard models were adjusted by sex (males or females), age (20-39, 40-59, ≥60 years), race-ethnicity (non-Hispanic white, non-Hispanic black, Mexican American and other), education (less than junior high school, junior high school, high school, college or above), income (<$20,000, ≥$20,000), smoking status (current smokers, former smokers, never smokers), drinking status (≥12 drinks in the past 12 months,<12 drinks in the past 12 months), exercise (more active, less active, about the same), body mass index (<18.5, 18.5-24.9, 25-29.9, ≥30kg/m2).
Acknowledgements
We thank all the participants and staff of the NHANES study for their valuable contributions and thank all participants and staff for their cooperation and assistance in the study.
Authors’ contributions
Weihong Qiu, Conceptualization, Methodology, Data curation, Formal analysis, Visualization, Writing—Original Draft, Writing—Review & Editing; Lijun Zheng, Conceptualization, Methodology, Investigation, Resources, Data curation, Writing-Original Draft, Writing-Review & Editing; Yingying Yu, Conceptualization, Methodology, Writing—Original Draft, Writing—Review & Editing; Zhiquan Zhang, Investigation, Writing—Review & Editing; Zhenzhen Li, Investigation, Writing—Review & Editing; Zihan Zhang, Investigation, Writing—Review & Editing; Heng He, Conceptualization, Methodology, Formal analysis, Resources, Data curation, Writing—Original Draft, Writing—Review & Editing, Visualization, Supervision, Project administration, Funding acquisition.
Funding
This study was supported by the National Natural Science Foundation of China (Grant number: 82304214), the Fujian Provincial Natural Science Foundation of China (Grant number: 2023J01631), the Joint Funds for the Innovation of Science and Technology, the Fujian Province (Grant number: 2025Y9265), and the Zonglian Plan Talents Start-Up Project of Fujian Medical University (Grant number: ZLJH2025002). The funders did not play any role in study design; in the collection, analysis, and interpretation of data; in the writing of the report; nor the preparation, review, or approval of the manuscript.
Data availability
The datasets supporting the conclusions of this article are available in the NHANES repository, https://wwwn.cdc.gov/nchs/nhanes/Default.aspx.
Declarations
Ethics approval and consent to participate
This study involves human participants and the National Center for Health Statistics and Ethics Review Board approved the protocol for NHANES, and all participants provided written informed consent. This study was conducted in accordance with the Declaration of Helsinki.
Consent for publication
Not applicable.
Competing interest
The authors declare no competing interests.
Footnotes
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Weihong Qiu and Lijun Zheng contributed equally to this work.
References
- 1.Baker JL, Mark Welch JL, Kauffman KM, McLean JS, He X. The oral microbiome: diversity, biogeography and human health. Nat Rev Microbiol. 2024;22(2):89–104. 10.1038/s41579-023-00963-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Reyes L. Porphyromonas gingivalis. Trends Microbiol. 2021;29(4):376–7. 10.1016/j.tim.2021.01.010. [DOI] [PubMed] [Google Scholar]
- 3.Hajishengallis G, Darveau RP, Curtis MA. The keystone-pathogen hypothesis. Nat Rev Microbiol. 2012;10(10):717–25. 10.1038/nrmicro2873. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Vlachojannis C, Dye BA, Herrera-Abreu M, et al. Determinants of serum IgG responses to periodontal bacteria in a nationally representative sample of US adults. J Clin Periodontol. 2010;37(8):685–96. 10.1111/j.1600-051X.2010.01592.x. [DOI] [PubMed] [Google Scholar]
- 5.Lakio L, Antinheimo J, Paju S, Buhlin K, Pussinen PJ, Alfthan G. Tracking of plasma antibodies against Aggregatibacter actinomycetemcomitans and Porphyromonas gingivalis during 15 years. J Oral Microbiol. 2009. 10.3402/jom.v1i0.1979. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Merchant AT, Shrestha D, Chaisson C, Choi YH, Hazlett LJ, Zhang J. Association between serum antibodies to oral microorganisms and hyperglycemia in adults. J Dent Res. 2014;93(8):752–9. 10.1177/0022034514538451. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Sanchez-Torres D, Gutierrez-Bejarano D, Hurtado-Roca Y, Guallar-Castillon P, Muntner P, Laclaustra M. Non-linear association of periodontal pathogen antibodies with mortality. Int J Cardiol. 2015;187:628–36. 10.1016/j.ijcard.2015.03.315. [DOI] [PubMed] [Google Scholar]
- 8.Joshi C, Bapat R, Anderson W, Dawson D, Cherukara G, Hijazi K. Serum antibody response against periodontal bacteria and coronary heart disease: systematic review and meta-analysis. J Clin Periodontol. 2021;48(12):1570–86. 10.1111/jcpe.13550. [DOI] [PubMed] [Google Scholar]
- 9.Bohnstedt S, Cullinan MP, Ford PJ, et al. High antibody levels to P. gingivalis in cardiovascular disease. J Dent Res. 2010;89(9):938–42. 10.1177/0022034510370817 [DOI] [PubMed]
- 10.Kataoka A, Katagiri S, Kawashima H, et al. Association between periodontal bacteria and degenerative aortic stenosis: a pilot study. J Periodontal Implant Sci. 2021;51(4):226–38. 10.5051/jpis.2006040302. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Svard A, Kastbom A, Ljungberg KR, et al. Antibodies against Porphyromonas gingivalis in serum and saliva and their association with rheumatoid arthritis and periodontitis. Data from two rheumatoid arthritis cohorts in Sweden. Front Immunol. 2023;14:1183194. 10.3389/fimmu.2023.1183194. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Iwasaki M, Taylor GW, Manz MC, et al. Serum antibody to Porphyromonas gingivalis in chronic kidney disease. J Dent Res. 2012;91(9):828–33. 10.1177/0022034512455063. [DOI] [PubMed] [Google Scholar]
- 13.Liu S, Butler CA, Ayton S, Reynolds EC, Dashper SG. Porphyromonas gingivalis and the pathogenesis of Alzheimer's disease. Crit Rev Microbiol. 2023:1–11. 10.1080/1040841X.2022.2163613 [DOI] [PubMed]
- 14.Huang Y, Wei Y, Xu J, Wei X. A comprehensive review on the prevention and regulation of Alzheimer's disease by tea and its active ingredients. Critical reviews in food science and nutrition. 2022:1–25. 10.1080/10408398.2022.2081128 [DOI] [PubMed]
- 15.Ahn J, Segers S, Hayes RB. Periodontal disease, Porphyromonas gingivalis serum antibody levels and orodigestive cancer mortality. Carcinogenesis. 2012;33(5):1055–8. 10.1093/carcin/bgs112. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Zhou B, Lu J, Beck JD, et al. Periodontal and other oral bacteria and risk of lung cancer in the Atherosclerosis Risk in Communities (ARIC) Study. Cancer Epidemiol Biomarkers Prev. 2023;32(4):505–15. 10.1158/1055-9965.EPI-22-0601. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Rodriguez-Concepcion M, Avalos J, Bonet ML, et al. A global perspective on carotenoids: metabolism, biotechnology, and benefits for nutrition and health. Prog Lipid Res. 2018;70:62–93. 10.1016/j.plipres.2018.04.004. [DOI] [PubMed] [Google Scholar]
- 18.Maiani G, Caston MJ, Catasta G, et al. Carotenoids: actual knowledge on food sources, intakes, stability and bioavailability and their protective role in humans. Mol Nutr Food Res. 2009;53(Suppl 2):S194-218. 10.1002/mnfr.200800053. [DOI] [PubMed] [Google Scholar]
- 19.Kaur G, Kathariya R, Bansal S, Singh A, Shahakar D. Dietary antioxidants and their indispensable role in periodontal health. J Food Drug Anal. 2016;24(2):239–46. 10.1016/j.jfda.2015.11.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Naruishi K. Carotenoids and periodontal infection. Nutrients. 2020. 10.3390/nu12010269. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Burt VL, Harris T. The third National Health and Nutrition Examination Survey: contributing data on aging and health. Gerontologist. 1994;34(4):486–90. 10.1093/geront/34.4.486. [DOI] [PubMed] [Google Scholar]
- 22.Dye BA, Herrera-Abreu M, Lerche-Sehm J, et al. Serum antibodies to periodontal bacteria as diagnostic markers of periodontitis. J Periodontol. 2009;80(4):634–47. 10.1902/jop.2009.080474. [DOI] [PubMed] [Google Scholar]
- 23.Offenbacher S, Barros SP, Singer RE, Moss K, Williams RC, Beck JD. Periodontal disease at the biofilm-gingival interface. J Periodontol. 2007;78(10):1911–25. 10.1902/jop.2007.060465. [DOI] [PubMed] [Google Scholar]
- 24.National Health and Nutrition Examination Survey Codebook for Data Production (1991–1994) (NHANES III) Periodontal Pathogens: Serum Actinobacillus actinomycetemcomitans Antibody and Porphyromonas gingivalis Antibodies http://www.cdc.gov/nchs/data/nhanes/nhanes3/depp.pdf.
- 25.Gunter EW, Lewis BG, Koncikowski SM. Laboratory procedures used for the third National Health and Nutrition Examination Survey (NHANES III), 1988–1994.1996.Accessed 16 August 2023. Available from https://stacks.cdc.gov/view/cdc/45776. 10.1186/s12874-021-01484-7
- 26.Centers for Disease Control and Prevention. National Center for Health Statistics (NCHS). National Health and Nutrition Examination Survey Questionnaire (or Examination Protocol,orLaboratory Protocol).2006. Accessed 16 August 2023. Available from https://www.cdcgov/nchs/nhanes.htm.
- 27.Li C, Yu R, Ding Y. Association between Porphyromonas gingivalis and systemic diseases: focus on T cells-mediated adaptive immunity. Front Cell Infect Microbiol. 2022;12:1026457. 10.3389/fcimb.2022.1026457. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Ruan Q, Guan P, Qi W, et al. Porphyromonas gingivalis regulates atherosclerosis through an immune pathway. Front Immunol. 2023;14:1103592. 10.3389/fimmu.2023.1103592. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Kong J, Liu Y, Qian M, Xing L, Gao S. The relationship between Porphyromonas gingivalis and oesophageal squamous cell carcinoma: a literature review. Epidemiol Infect. 2023;151:e69. 10.1017/S0950268823000298. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Jia S, Li X, Du Q. Host insulin resistance caused by Porphyromonas gingivalis-review of recent progresses. Front Cell Infect Microbiol. 2023;13:1209381. 10.3389/fcimb.2023.1209381. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Bohm V, Lietz G, Olmedilla-Alonso B, et al. From carotenoid intake to carotenoid blood and tissue concentrations - implications for dietary intake recommendations. Nutr Rev. 2021;79(5):544–73. 10.1093/nutrit/nuaa008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Terao J. Revisiting carotenoids as dietary antioxidants for human health and disease prevention. Food Funct. 2023;14(17):7799–824. 10.1039/d3fo02330c. [DOI] [PubMed] [Google Scholar]
- 33.Tapiero H, Townsend DM, Tew KD. The role of carotenoids in the prevention of human pathologies. Biomed Pharmacother. 2004;58(2):100–10. 10.1016/j.biopha.2003.12.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Ba W, Xu W, Deng Z, Zhang B, Zheng L, Li H. The Antioxidant and Anti-Inflammatory Effects of the Main Carotenoids from Tomatoes via Nrf2 and NF-kappaB Signaling Pathways. Nutrients. 2023;15(21) 10.3390/nu15214652 [DOI] [PMC free article] [PubMed]
- 35.Kose O, Arabaci T, Yemenoglu H, et al. Influences of fucoxanthin on alveolar bone resorption in induced periodontitis in rat molars. Mar Drugs. 2016. 10.3390/md14040070. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Figgins EL, Arora P, Gao D, et al. Enhancement of innate immunity in gingival epithelial cells by vitamin D and HDAC inhibitors. Front Oral Health. 2024;5:1378566. 10.3389/froh.2024.1378566. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Moszura J, Gawlak-Socka S, Peksa J, Bielecka-Kowalska N, Klosek S. The Role of Vitamin D(3) in Periodontal Health: Implications for Bone Metabolism, Immune Modulation and Inflammation Control. Nutrients. 2026;18(4) 10.3390/nu18040577 [DOI] [PMC free article] [PubMed]
- 38.Meghil MM, Cutler CW. Influence of vitamin D on periodontal inflammation: a review. Pathogens. 2023. 10.3390/pathogens12091180. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Niu L, Chen S, Yang X, et al. Vitamin D decreases Porphyromonas gingivalis internalized into macrophages by promoting autophagy. Oral Dis. 2021;27(7):1775–88. 10.1111/odi.13696. [DOI] [PubMed] [Google Scholar]
- 40.Kajiura Y, Nishikawa Y, Lew JH, Kido JI, Nagata T, Naruishi K. beta-carotene suppresses Porphyromonas gingivalis lipopolysaccharide-mediated cytokine production in THP-1 monocytes cultured with high glucose condition. Cell Biol Int. 2018;42(1):105–11. 10.1002/cbin.10873. [DOI] [PubMed] [Google Scholar]
- 41.Clarke R, Shipley M, Lewington S, et al. Underestimation of risk associations due to regression dilution in long-term follow-up of prospective studies. Am J Epidemiol. 1999;150(4):341–53. 10.1093/oxfordjournals.aje.a010013. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Additional file 1: Figure S1. The flowchart of this study. Figure S2. The survival probability of specific cause mortality by serum P. gingivalis IgG groups. (A) CVD mortality; (B) Cancer mortality; (C) CeVD mortality; (D) DM mortality Abbreviation:P. gingivalis, Porphyromonas gingivalis; IgG, immunoglobulin G; CI, confidence interval; CVD, cardiovascular disease; CeVD, cerebrovascular disease; DM, diabetes mellitus. Figure S3. The dose-response associations between serum P. gingivalis IgG levels and specific cause mortality. (A) CVD mortality; (B) Cancer mortality; (C) CeVD mortality; (D) DM mortalit. Abbreviation: P. gingivalis, Porphyromonas gingivalis; HR, hazard ratio; CI, confidence interval; IgG, immunoglobulin G; CVD, cardiovascular disease; CeVD, cerebrovascular disease; DM, diabetes mellitus. The restricted cubic spline models were adjusted by sex (males or females), age (20-39, 40-59,≥60 years), race-ethnicity (non-Hispanic white, non-Hispanic black, Mexican American and other), education (less than junior high school, junior high school, high school, college or above), income (<$20,000, ≥$20,000), smoking status (current smokers, former smokers, never smokers), drinking status (≥12 drinks in the past 12 months,<12 drinks in the past 12 months), exercise (more active, less active, about the same), body mass index (<18.5, 18.5-24.9, 25-29.9, ≥30kg/m2). Figure S4. The interactions between serum carotenoids and serum P. gingivalis IgG levels on cardiovascular disease mortality. Abbreviation: P. gingivalis, Porphyromonas gingivalis; IgG, immunoglobulin G; CI, confidence interval; CVD, cardiovascular disease. The cox proportional hazard models were adjusted by sex (males or females), age (20-39, 40-59, ≥60 years), race-ethnicity (non-Hispanic white, non-Hispanic black, Mexican American and other), education (less than junior high school, junior high school, high school, college or above), income (<$20,000, ≥$20,000), smoking status (current smokers, former smokers, never smokers), drinking status (≥12 drinks in the past 12 months,<12 drinks in the past 12 months), exercise (more active, less active, about the same), body mass index (<18.5, 18.5-24.9, 25-29.9, ≥30kg/m2). Figure S5. The interactions between serum carotenoids and serum P. gingivalis IgG levels on cancer mortality. Abbreviation: P. gingivalis, Porphyromonas gingivalis; IgG, immunoglobulin G; CI, confidence interval. The cox proportional hazard models were adjusted by sex (males or females), age (20-39, 40-59, ≥60 years), race-ethnicity (non-Hispanic white, non-Hispanic black, Mexican American and other), education (less than junior high school, junior high school, high school, college or above), income (<$20,000, ≥$20,000), smoking status (current smokers, former smokers, never smokers), drinking status (≥12 drinks in the past 12 months,<12 drinks in the past 12 months), exercise (more active, less active, about the same), body mass index (<18.5, 18.5-24.9, 25-29.9, ≥30kg/m2). Figure S6. The interactions between serum carotenoids and serum P. gingivalis IgG levels on cerebrovascular disease mortality. Abbreviation:P. gingivalis, Porphyromonas gingivalis; IgG, immunoglobulin G; CI, confidence interval; CeVD, cerebrovascular disease. The cox proportional hazard models were adjusted by sex (males or females), age (20-39, 40-59, ≥60 years), race-ethnicity (non-Hispanic white, non-Hispanic black, Mexican American and other), education (less than junior high school, junior high school, high school, college or above), income (<$20,000, ≥$20,000), smoking status (current smokers, former smokers, never smokers), drinking status (≥12 drinks in the past 12 months, <12 drinks in the past 12 months), exercise (more active, less active, about the same), body mass index (<18.5, 18.5-24.9, 25-29.9, ≥30kg/m2). Figure S7. The interactions between serum carotenoids and serum P. gingivalis IgG levels on diabetes mellitus mortality. Abbreviation:P. gingivalis, Porphyromonas gingivalis; IgG, immunoglobulin G; CI, confidence interval; DM, diabetes mellitus. The cox proportional hazard models were adjusted by sex (males or females), age (20-39, 40-59, ≥60 years), race-ethnicity (non-Hispanic white, non-Hispanic black, Mexican American and other), education (less than junior high school, junior high school, high school, college or above), income (<$20,000, ≥$20,000), smoking status (current smokers, former smokers, never smokers), drinking status (≥12 drinks in the past 12 months,<12 drinks in the past 12 months), exercise (more active, less active, about the same), body mass index (<18.5, 18.5-24.9, 25-29.9, ≥30kg/m2). Figure S8. The joint associations of serum carotenoids and serum P. gingivalis IgG levels on specific-cause mortality. Abbreviation: P. gingivalis, Porphyromonas gingivalis; IgG, immunoglobulin G; CI, confidence interval. The cox proportional hazard models were adjusted by sex (males or females), age (20-39, 40-59, ≥60 years), race-ethnicity (non-Hispanic white, non-Hispanic black, Mexican American and other), education (less than junior high school, junior high school, high school, college or above), income (<$20,000, ≥$20,000), smoking status (current smokers, former smokers, never smokers), drinking status (≥12 drinks in the past 12 months,<12 drinks in the past 12 months), exercise (more active, less active, about the same), body mass index (<18.5, 18.5-24.9, 25-29.9, ≥30kg/m2).
Data Availability Statement
The datasets supporting the conclusions of this article are available in the NHANES repository, https://wwwn.cdc.gov/nchs/nhanes/Default.aspx.