Association between lipid accumulation product and endometriosis: A cross-sectional study from NHANES 1999-2006

Endometriosis is a common, gynecologic condition defined as the formation of ectopic endometrial tissue in non-uterine regions [ 1 ]. According to reports, approximately 5% to 15% of reproductive-aged women suffer from this chronic systemic disease [ 2 – 4 ]. Endometriosis can also lead to female infertility in addition to causing severe pain and irregular menstrual cycles [ 2 ]. Women with endometriosis can have negative impacts on interpersonal relationships, quality of life, and work efficiency of individuals and families [ 5 ]. However, the main methods for treating endometriosis currently include drug therapy and surgical resection, and their therapeutic quality of life effects are limited [ 6 ]. Further research is urgently needed to identify more effective markers for evaluation, prevention, diagnosis, and management strategies. Although there are several factors, such as inflammation, hormones, metabolism, and immunology, that have been implicated in endometriosis, the specific pathological mechanism of endometriosis is still not fully understood [ 6 ]. Recent studies have shown that metabolic abnormalities are key to the occurrence and development of endometriosis [ 7 , 8 ]. Especially, women diagnosed with endometriosis have been recognized to have abnormal lipid and glucose metabolism [ 9 – 11 ]. Lipid accumulation product (LAP) is an index that combines fasting triglyceride (TG) and waist circumference (WC) [ 12 ]. Kahn studied LAP in 2005 and found it to be a reliable index of lipid accumulation [ 12 ]. Currently, studies have provided evidence that LAP can be used as a potential marker for metabolic syndrome (MetS), type 2 diabetes, insulin resistance, and cardiovascular diseases [ 13 – 16 ]. In addition, the LAP index is not affected by muscle mass and is suitable for the elderly and people with higher muscular density [ 10 ]. So far, there has been little exploration of the relationship between lipid accumulation products and endometriosis. Therefore, the study aimed to investigate the association between LAP and endometriosis using data from NHANES in the 1999–2006 cycles. The overall objective is to extend a new perspective to the precaution and therapy of endometriosis.

The baseline data of 2,216 participants in this study were summarized in Table 1 . Among these, 171 (7.72%) were diagnosed with endometriosis, whereas 2,045 cases were not. During the whole study process, the age range of the women participating in this study was 20–54 years. In contrast to 36.91 ± 10.02 years in those without endometriosis, patients indicated a higher incidence in elderly people( P  < 0.001). Family income to poverty ratio (PIR), LAP, race, educational level, marital status, and smoking status all showed a significant relationship with endometriosis. More precisely, those who were non-Hispanic White, completed at least high school, were in smoking condition, had higher PIR, and had a status of married or living with partners were more frequent in those with this condition. Mean ±SD for continuous variables; P value was calculated by the weighted linear regression model. % for categorical variables: P value was calculated by weighted chi-square test. PIR, family income to poverty ratio; BMI, Body mass index. Nevertheless, there was no statistically significant association between the chance of endometriosis prevalence and BMI, with a P -value above 0.05. The use of alcohol and the age of menarche represented similar trends. Furthermore, there is a significant difference in LAP between the two groups. Logistic regression models were investigated in Table 2 . The ln LAP and endometriosis have been demonstrated to have a strong positive association in the crude model (OR=1.23, 95% CI:1.03–1.47, P  = 0.023). After a full adjustment in Model 2, this positive association remained (OR=1.37, 95% CI:1.08–1.75, P  = 0.010), suggesting a 0.37-fold raise in the odds of endometriosis incidence for every unit elevation in the ln LAP. The crude model adjusts for: none. Model 1 adjusts for age and race. Model 2 adjusts for age, race, educational level, marital status, BMI, PIR, early menstruation before the age of 11, glycohemoglobin, smoking status, and alcohol use. To further conduct the association between ln LAP and endometriosis, ln LAP was divided into four tertiles. Compared to the Tertile-1- ln LAP in Model 2, the OR for the Tertile-4-ln LAP was 1.93(95% CI:1.08–3.46, P  = 0.027), suggesting a substantial 0.93-fold increase in the chance of endometriosis prevalence in Tertile 4. However, Tertile 1 and Tertile 2 did not differ statistically significantly. As shown in Fig 2 , the smoothing curve fitting results further revealed a positive association between the ln LAP and the chance of endometriosis incidence. Analyses showed that high ln LAP was associated with higher rates of infertility in women aged ＜35 years and those aged ≥35 years old ( Fig 3 ). This association existed in women with a BMI ＜ 25 kg/m 2 group ( Fig 4 ). The smoothing curve fitting results by race are shown in Fig 5 . Ln LAP was a natural log transformation of the LAP index. The solid red line represents the smooth curve fit between variables. Blue bands represent the 95% confidence interval from the fit. Ln LAP was a natural log transformation of the LAP index. The red line represents the population aged 20 to 35, while the blue line represents women over 35 years old. Ln LAP was a natural log transformation of the LAP index. The red, green, and blue lines are used to classify the population with BMI ＜ 25 kg/m 2 , 25-30 kg/m 2 and ≥30 kg/m 2 . Ln LAP was a natural log transformation of the LAP index.1: Non-Hispanic White; 2. Non-Hispanic Black; 3. Mexican American; 4: Other Race. Subgroup analysis and interaction tests were performed to evaluate the constancy of the relationship between ln LAP and endometriosis among factors. Stratification was based on age, race, level of education, age of menarche, glycohemoglobin, BMI, PIR, drinking, and smoking status ( Fig 6 ). Adjust for age, race, educational level, marital status, BMI, PIR, early menstruation before the age of 11, glycohemoglobin, smoking status, and alcohol use.

In summary, the results of this study revealed a positive association between the LAP index and endometriosis. The findings suggested the LAP index could impact preventive measures in the future. However, it is necessary to conduct extensive and prospective investigations for subsequent studies.

NHANES is a project that began in the early 1960s to provide data for this study. NHANES is a research program that combines physical examinations and interviews to estimate the nutrition and overall health of the US population. Every year, it applies a multi-stage sampling design to select a sample nationwide of approximately 5,000 individuals [ 17 ]. The NHANES survey plan was developed by the National Research Ethics Review Committee, and study individuals gave informed permission [ 18 ]. Relevant data came from the NHANES cycle from 1999 to 2006. 41,474 people participated in this cross-sectional study. Through this extensive examination and analysis of NHANES, we excluded male individuals (n = 20,264) and female participants aged 20 and younger and 55 and older (n = 6,508). Then we excluded individuals without complete data on endometriosis. Additionally, participants with missing information on educational levels (n = 5), marital status (n = 152), the ratio of family income to poverty (n = 340), BMI (n = 52), age when first menstrual period occurred (n = 79), smoking status (n = 1) and alcohol use (n = 2) were excluded. Subsequently, participants with incomplete waist circumstance (n = 36) and triglyceride levels (n = 2,674) data were also excluded from the study. Ultimately, 2,216 participants were enrolled ( Fig 1 ). The main outcome was self-reported endometriosis from the Reproductive Health Questionnaire (variable name in the questionnaire: RHQ360). Participants were asked, “Has a doctor or other health professional ever told you that you have endometriosis?” Those who answered “Yes” were considered to have endometriosis. Serum triglyceride levels (TG, mmol/L) and waist circumference (WC, cm) are variables in the gender-specific LAP calculation. The formula is [WC (cm)-58] × [TG (mmol/l)] for women [ 19 – 20 ]. Ln LAP was a natural log transformation of the LAP index. All participants’ LAP index and ln LAP were subsequently calculated using these values, and then they were divided into quartiles (Q1-Q4) for analysis. Based on previous literature [ 21 , 22 ] and clinical experience, potential covariates included age (years), race, educational level, marital status, body mass index (BMI), early menstruation before the age of 11, glycohemoglobin, poverty income ratio (PIR), smoking, and alcohol use. The level of education was reported as below high school, high school, and above high school [ 23 ]. BMI was categorized as BMI ＜ 25 kg/m 2 , 25 kg/m 2  ≤ BMI ＜ 30 kg/m 2 , and BMI ≥ 30 kg/m 2 . Age at menarche was collected by using ‘RHQ10’ questionnaires. Participants were asked, “Age when first menstrual period occurred” [ 24 ]. PIR was assessed as PIR ＜ 1.35, 1.35 ≤ PIR ＜ 3.0, and PIR ≥ 3.0 by the degree of poverty [ 25 ]. Glycohemoglobin was collected in laboratory data(＜6% and ≥6%). Other covariates relevant to health, such as alcohol consumption and smoking status, were analyzed in the MEC. The smoking situation was divided into yes or no. Participants with a minimum of 100 cigarettes throughout their lifespan are defined as smokers. Alcohol use was classified as never, former, or current. Participants with no drinking alcohol were defined as not having consumed 12 alcohol drinks/lifetime. This study’s data analysis utilized package R version 3.4.3 and EmpowerStats software. The individuals were firstly divided into two groups: one for endometriosis and the other for control; these groups were referred to as the “Endometriosis” and “Control” groups, respectively. There was a 0.05 significance limit. To more fully indicate the significance level of the results, use asterisks (*) to mark them. Significant levels were 0.05 (*), 0.01 (**), and 0.001 (***). The grouping variable was shown as percentages and continuous variable data was shown as Mean±SD (standard deviations). A natural log transformation, identified as ‘ln LAP’ was utilized to convert the non-normal distribution of LAP into a normal distribution. Two approaches were applied in the study to evaluate group differences: weighted linear regression was performed for continuous variables, and the weighted chi-square test was used for categorical data. Multivariable logistic regression models were used to assess the association between ln LAP and endometriosis. Model 1 (the crude model) had no adjustment. Model 2 (the minimally adjusted model) was adjusted for age and race. Model 3 (the fully adjusted model) was adjusted for the variables in Model 2 plus level of education, marital status, BMI, PIR, early menstruation before the age of 11, glycohemoglobin, alcohol use, and smoking status. Then ln LAP was regarded as a categorical variable by quartile. To investigate the potential association between LAP and odds of endometriosis incidence, a smoothed curve fitting was employed. Subgroup stratified by age and race, educational level, BMI, marital status, PIR, early menstruation before the age of 11, glycohemoglobin, drinking, and smoking status were also utilized. The logistic regression and subgroup analysis are expressed as odds ratios (OR) and 95% confidence intervals (CI).