The association between non-high-density lipoprotein cholesterol to high-density lipoprotein cholesterol ratio (NHHR) and prevalence of periodontitis among US adults: a cross-sectional NHANES study

The non-high-density lipoprotein cholesterol to high-density lipoprotein cholesterol ratio (NHHR) is a recently developed lipid parameter, but there is currently a lack of research exploring its relationship with periodontitis. This study aims to identify the potential association between NHHR and periodontitis. The association between NHHR and periodontitis were examined through univariate and multivariate weighted logistic regression utilizing the National Health and Nutrition Examination Survey data from 2009 to 2014. The participants were grouped based on the type of periodontitis. This study included a total of 9023 participants, with 1947 individuals having no periodontitis, and an additional 7076 individuals suffering from periodontitis. Patients in periodontitis group demonstrated a statistically significant elevation in NHHR values 2.82 (2.05–3.80) compared to those in no periodontitis group (p < 0.001). Logistic regression analysis of variables demonstrated a positive association between NHHR and periodontitis [1.07 (1.02, 1.12) p = 0.0067]. The study revealed a positive association between NHHR and an elevated prevalence of periodontitis development. For each unit increase in NHHR, there is a 7% increase in the prevalence of periodontitis. Further investigations into NHHR may enhance our understanding of preventing and treating periodontitis. However, additional studies are required to validate these findings.


Study population
The data utilized in this study are sourced from the NHANES 2009-2014, a research program aimed at evaluating the wellbeing and nutritional condition of adults and children in the United States.The survey comprises demographic, socio-economic, dietary, and health-related inquiries.The screening segment encompasses medical, dental, and physiological measurements alongside laboratory tests conducted by trained medical personnel.The survey results will establish the occurrence rates of significant diseases and disease-causing prevalences.Further details are available online (https:// www.cdc.gov/ nchs/ nhanes/ index.htm).
A total of 30,468 individuals participated in NHANES between 2009 and 2014.However, after applying exclusion criteria, the sample size for this study was reduced to 9,023 participants.The exclusion criteria included incomplete demographics data, missing data on covariates (such as smoking status, BMI, alcohol use, the use of dental floss, the number of tooth loss, diet level, history of hypertension, and hyperglycemia), and lack of periodontal examination or NHHR-related data (total cholesterol and high-density lipoprotein data).Figure 1 shows the screening process flowchart.

The calculation of NHHR
The data source for the NHHR calculations is derived from laboratory data in NHANES called 'HDL.Doc' which provides HDL data and 'TCHOL.Doc' , which provides total cholesterol data.The NHHR data is obtained using the formula for total cholesterol minus HDL, then divided by HDL.

Assessment of periodontitis
The NHANES examination data includes a file named "OHXPER.Doc" that encompasses data on periodontal examinations involving clinical attachment loss (CAL) and probing depth (PD).This data classified participants as periodontitis or not based on criteria provided by the Centers for Disease Control and Prevention and the American Academy of Periodontology (CDC/AAP) 30 .
Severe periodontitis was determined by the presence of at least two interproximal regions exhibiting a clinical attachment level (CAL) of no less than 6 mm and not located on the same tooth, alongside at least one interproximal site exhibiting a probing depth (PD) of no less than 5 mm.Moderate periodontitis was defined as the identification of two or more interproximal regions with probing pocket depths that are greater than or equal to 5 mm and not situated on the same tooth, or two or more interproximal areas with clinical attachment levels that are greater than or equal to 4 mm and not found on the same tooth.Participants diagnosed with mild periodontitis must exhibit at least two interproximal sites with a clinical attachment loss of at least 3 mm, and at least two interproximal sites with a probing depth of at least 4 mm (not on the same tooth) or one site with a probing depth of at least 5 mm.In this study, we grouped mild, moderate, and severe periodontitis into one category (having periodontitis), and no periodontitis in another 31,32 .

Statistical analyze
All statistical analyses were conducted using R (version 4.2), SPSS (version 26.0) and Empowerstats (version 5.0), and the NHANES guidelines were used to weight all data.Natural cubic spline was used to identify the relationship of NHHR with periodontitis.Then, we performed logistic regression analyses to examine the correlation between NHHR and periodontitis 34 , followed by subgroup analyses to investigate potential differences in the correlations regarding gender, age, race, income, education, and marital status etc.

Characteristics of participants
The characteristics of participants are shown in Table 1.The study includes 9023 individuals with an average age of 52.00 years.Among them, 78.42% were diagnosed with periodontitis.The NHHR of this group was 2.82 (2.05-3.80)which was statistically significant when compared to the NHHR of none periodontitis patients [2.67 (1.96-3.61),(P < 0.001)].Significant differences in characteristics were observed between participants with none periodontitis and periodontitis with regards to age, gender, race, education level, marital status, ratio of family income to poverty, HEI-2015, the number of tooth loss and total cholesterol.

Association between NHHR and periodontitis
The association between NHHR and periodontitis were displayed in Table 2. Following multiple adjustments, NHHR was found to be positively correlated with periodontitis [1.07 (1.02, 1.12), p = 0.0067] in the model 3, while the smoothed curve fitting confirmed this positive correlation (Fig. 2).The goodness of fit results in Table 2 showed that Model 3 (R 2 = 17.5%) has a better predictive ability than the other models.

Subgroup analyses by potential effect modifiers
The findings of the subgroup analyses can be found in Fig. 3.The logistic regression analysis results for the subgroups showed statistically significant findings for participants with 9-11 grade education level or non-drinkers.Figure 3 displays that the OR of participants with 9-11 grade education level was 1.14 (1.00, 1.30), indicating that the prevalence of periodontitis increased by 14% for each unit increase in NHHR in this subgroup.However, the difference between the subgroups was not significant at 0.7284.

Discussion
According to this study, NHHR serves as a satisfactory diagnostic biomarker for periodontitis.Our findings show a positive association between NHHR and an increased prevalence of periodontitis in an adjusted model.
For each unit increase in NHHR, there is a 7% increase in the prevalence of periodontitis.Metabolic abnormalities in lipid metabolism are common in patients with periodontitis and are associated with the occurrence and development of periodontitis 35,36 .NHHR is an emerging comprehensive indicator of atherosclerotic lipids, including HDL and NHDL, and is associated with dyslipidemia related diseases 28 .HDL are small lipoproteins that circulate throughout the body and are present near most cells 37 .In the mid-1950s, John Gofman and his colleagues discovered a negative correlation between levels of HDL cholesterol (HDL-C) and the likelihood of coronary heart disease 38 .Although it has been long acknowledged that high levels of HDL are inversely and linearly associated with the risk of systemic diseases like cardiovascular disease [39][40][41] , recent epidemiological studies have proposed that there is a "U-shaped" curve between HDL-C levels and cardiovascular disease risk, wherein patients with cardiovascular disease and excessively low or high levels of HDL-C have higher mortality rates compared to those with intermediate HDL-C levels [42][43][44] .Patients with low and high levels of HDL cholesterol have a higher mortality rate than those with intermediate levels, marking a significant deviation from earlier studies.
Over the past few years, there has been a growing interest in the connection between HDL and periodontitis.While certain studies have not detected disparities in serum HDL levels between healthy individuals and patients with periodontal disease [45][46][47] , two recent systematic reviews and meta-analyses have demonstrated a correlation between periodontitis and reduced serum HDL levels 20,21 .No clear "U-shaped" curve was found between NHHR and periodontitis in this study, potentially due to non-HDL-C's influence on the relationship or the complex relationship between HDL levels and periodontitis, which may be mediated by multiple factors.Additional adjustment of relevant covariates is still necessary.
Further studies found that periodontal treatment significantly increased serum HDL-C levels in patients with periodontitis 16 .A bidirectional relationship between HDL and periodontitis has also been proposed: upregulation of proinflammatory factors caused by periodontitis can have adverse effects on serum lipid metabolism, as proposed by Fentoğlu et al. 48A significant correlation was identified between the pro-inflammatory cytokines TNF-α and IL-1b and the TC/HDL ratio in the gingival sulcus fluids and serum of hyperlipidemic patients with periodontitis.Furthermore, Anniina's study validated these findings and proposed that HDL may play a role in the association between the number of teeth with deep periodontal pockets and C-reactive protein levels 49 .
Apart from HDL-C, NHDL is another crucial component of NHHR, capable of measuring low-density lipoprotein cholesterol (LDL-C), very-low-density lipoprotein cholesterol (VLDL-C), apolipoprotein A (apo A), and intermediate-density lipoprotein (IDL).Macri et al. 50, based on their study involving experimental periodontitis in animals, found that a cholesterol-rich diet increases serum NHDL levels and exacerbates alveolar bone loss caused by periodontitis.Furthermore, various studies have reported higher levels of NHDL, including LDL-C, VLDL-C, and IDL, in patients with periodontitis compared to healthy individuals [51][52][53][54] .In a Mendelian randomization study published in 2023, it was also discovered through genetic predictions that apo A1 is associated with the risk of periodontitis 19 .The association between NHHR and periodontitis may be based on a joint analysis of the potential impact of two cholesterol categories on periodontitis, namely HDL and NHDL.However, this association and the exploration of mechanisms between them needs to be further explored with more welldesigned basic and clinical studies based on large samples.
There were several limitations to this study: (1) Cohort studies are more reliable for validating the results of this study.The study was based on cross-sectional data, and the results may be influenced by selection bias; (2) Lipid profiles were assessed and recorded only once in this study.The lack of repeated measurements of lipid profiles may be subject to acute stress and incidental effects; (3) Some of the covariates have potential confounding power, are affected by the 2009-2014 NHANES database, there are problems with missing data, and there may be some bias in the results.Meanwhile, detailed medication information for patients, including types of medications, frequency, and duration, was not investigated in this study, and the impact of medication on periodontitis could not be ruled out.

Conclusion
In summary, current research suggests a positive correlation between NHHR and the increased prevalence of periodontitis, potentially serving as a novel predictive factor for periodontal events.This also provides valuable evidence for primary prevention in individuals at high prevalence of periodontitis.Clinicians can use NHHR to assist in identifying high-risk populations for periodontitis, thereby improving screening efficiency.

Figure 3 .
Figure 3. Subgroup analysis of the association between NHHR and periodontitis. )

Table 1 .
Characteristics of participants.NHHR non-high-density lipoprotein cholesterol to high-density lipoprotein cholesterol ratio, HEI-2015 The Healthy Eating Index 2015, BMI body mass index, HDL-C high density lipoprotein cholesterol.

Table 2 .
Association between NHHR and periodontitis among participants in NHANES 2009-2014.Model 1 did not adjust for any potential confounders; Model 2 adjusted for: gender, age, race; Model 3 adjusted for: gender, age, race, education level, pir, smoking status, hypertension, hyperglycemia, HDL-c, total cholesterol, BMI, marry status, Alcohol use, HEI-2015 score, dental floss use status, tooth loss number.*Goodness-of-fit results.The goodness of fit results showed that Model 3 (R 2 = 17.5%) has a better predictive ability than the other models.