Association of lipid accumulation product and visceral adiposity index with the risk of hypertension among oil workers in Xinjiang, China

Study subjects

Using the whole-group random sampling method, a total of 2,312 employees aged 18–65 years old who participated in health checkups were randomly selected from four districts in Karamay City, Xinjiang, with operational units as clusters and six randomly selected oilfield enterprises and service units from April to June 2021. The examinees included oil recovery, gathering workers, thermal operators, painters, welders, drilling diesel engine workers, and chemical process testers. After excluding 126 of them with incomplete medical examination information, 2,186 subjects were finally included in the study. Inclusion criteria: (1) those with complete physical examination data related to this study; (2) those who were willing to cooperate with this investigation and signed the informed consent form. Exclusion criteria: (1) Patients with serious organic lesions, mental illnesses, and genetic diseases; (2) Those with incomplete physical examination data. The sample content estimation formula is: N = μ²_α ρ( 1 − ρ)/δ², the pre-survey showed that the prevalence of hypertension ρ = 0.16, α = 0.05 (bilateral), and the tolerance error δ = 0.02. N = 1,291 was calculated, taking into account a 20% loss, at least 1,549 cases need to be included in the study, and the number of people surveyed in this current study is 2,186. The sample size requirement was satisfied. All participants signed a written informed consent. The study was approved by the Ethics Review Committee of the First Affiliated Hospital of Xinjiang Medical University (No. 2015006).

Anthropometric and defintion

The general survey and physical examination information collected by the staff of the physical examination section of the central hospital in Karamay. The general conditions mainly include age, gender, alcohol consumption, smoking, and history of previous diseases. According to the relevant definition of WHO, Smoker: smoking ≥1 cigarette per day for 6 months or more; drinker: drinking ≥1 times a week with alcohol intake ≥50 g per drinking session regularly ≥1 year. The physical examination is done by the professional staff of the Physical Examination Center of Karamay Central Hospital, which mainly includes the measurement of height, weight, waist circumference, blood pressure, blood glucose and lipid, and the calculation of BMI, WHtR, LAP and VAI. Height and weight measurement by automatic recorder (SK-X80/TCS-160D-W/h) standard position without shoes. After the subject rested for 5 min, the blood pressure was measured in the sitting arm using an Omron (HEM-7211; Omron, Kyoto, Japan) electronic blood pressure monitor, and the average value was taken after three measurements. A fully automated biochemical analyzer (Beckman Coulter AU5800; Beckman Coulter, Brea, CA, USA) was used for biochemical analysis, detection of TG, TC, HDL-C, LDL-C, FPG and serum creatinine (Scr).

Blood pressure measurement and blood biochemical assays

Hypertension if they met one of the following standards (Joint Committee for Guideline Revision, 2019): (1) systolic blood pressure ≥ 140 mm Hg and/or diastolic blood pressure ≥ 90 mm Hg; (2) self-reported hypertension diagnosed by a physician and current antihypertensive treatment during the previous 2 weeks (China Hypertension Prevention and Control Guidelines Revision Committee, Hypertension Alliance China, Chinese Society of Cardiovascular Diseases Chinese Physicians Association Hypertension Professional Committee, 2019). ① BMI < 18.5 kg/m² is too low body weight (BW), BMI between 18.5 and 23.9 kg/m² is normal BW, BMI between 24.0 and 27.9 kg/m² is overweight, and BMI ≥ 28 kg/m² is obese (China Working Group on Obesity, 2004). ② WC ≥ 90 cm in men and 85 cm in women is considered centrally obese (Bao et al., 2008). ③ central obesity when WHtR ≥ 0.5.

Calculation of related indices

1. BMI was calculated as:

• $\mathrm{B}\mathrm{M}\mathrm{I}=\mathrm{B}\mathrm{W}(\mathrm{k}\mathrm{g})/\mathrm{b}\mathrm{o}\mathrm{d}\mathrm{y}\mathrm{h}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}(\mathrm{B}{\mathrm{H}}^2)(\mathrm{m})$ (China Working Group on Obesity, 2004).

2. WHtR was calculated as:

• $\mathrm{W}\mathrm{H}\mathrm{t}\mathrm{R}=\mathrm{W}\mathrm{C}(\mathrm{c}\mathrm{m})/\mathrm{B}\mathrm{H}(\mathrm{c}\mathrm{m})$ (Browning, Hsieh & Ashwell, 2010).

3. LAP was calculated as:

• $\mathrm{L}\mathrm{A}\mathrm{P}=(\mathrm{W}\mathrm{C}(\mathrm{c}\mathrm{m})-65)\times \mathrm{T}\mathrm{G}(\mathrm{m}\mathrm{m}\mathrm{o}\mathrm{l}/\mathrm{L}))$ in males, and

• $\mathrm{L}\mathrm{A}\mathrm{P}=(\mathrm{W}\mathrm{C}(\mathrm{c}\mathrm{m})-58)\times \mathrm{T}\mathrm{G}(\mathrm{m}\mathrm{m}\mathrm{o}\mathrm{l}/\mathrm{L})\mathrm{i}\mathrm{n}\mathrm{f}\mathrm{e}\mathrm{m}\mathrm{a}\mathrm{l}\mathrm{e}\mathrm{s}$ (Kahn, 2006).

4. VAI score was calculated as described previously (Amato et al., 2010) using the following sex-specific equations (with TG levels in mmol/L and HDL-cholesterol levels in mmol/L):

• $\text{VAI = }\left(\begin{array}{ccc}\frac{\text{WC}\left(\text{cm}\right)}{39.68+\left(1.88\times \text{BMI}\right)}\text{\hspace{0.17em}}\times & \frac{\text{TG(mmol/I)}}{1.03}\text{\hspace{0.17em}}\times & \frac{\text{1.31}}{\text{HDL(mmol/I)}}\end{array}\right)$ in males,

• $\text{VAI = }\left(\begin{array}{ccc}\frac{\text{WC}\left(\text{cm}\right)}{36.58+\left(1.89\times \text{BMI}\right)}\text{\hspace{0.17em}}\times & \frac{\text{TG(mmol/I)}}{0.81}\text{\hspace{0.17em}}\times & \frac{\text{1.52}}{\text{HDL(mmol/I)}}\end{array}\right)$ in females.

Statistical analysis

Analyses were performed using SPSS Version 26.0 (IBM, Chicago, IL, USA) and R version 4.0.5 data were expressed as mean ± standard deviation for normally distributed continuous variables and median (P₂₅, P₇₅) for not normally distributed continuous variables, and the t-test or non-parametric test was used to compare the two groups. Count data were expressed as rates or composition ratios and data were compared using chi-square tests. Spearman Rank correlation was used for correlation analysis. LAP and VAI indices were grouped using the quartile method (LAP: “Q1” is ≤12.8, “Q2” is 12.8–27.51, “Q3” is 27.51–50.71, “Q4” is >50.71; VAI: “Q1” is ≤0.86, “Q2” is 0.86–14.6, “Q3” is 14.6–2.39, “Q4” is >2.39). The dose-response relationship between LAP and VAI index and hypertension was analyzed by using restricted cubic spline and logistic regression model. The test level was α = 0.05.

Baseline characteristics of oil workers

A total of 2,186 persons were included, 1,372 of whom were men, aged (41.58 ± 10.03) years. There were 814 females with an age of (39.59 ± 9.00) years. Statistically significant differences were found in the comparison of age, history of alcohol consumption, history of smoking, hypertension, BMI, WC, WHtR, SBP, DBP, TC, TG, HDL-C, LDL-C, FPG, Scr, LAP and VAI in different gender groups (P < 0.01) (Table 1).

Prevalence of hypertension in oil workers with different individual characteristics

The prevalence of hypertension among oil workers was 10.1%, with 13.9% among men and 3.6% among women. The prevalence of hypertension among oil workers with different individual characteristics was statistically significant (P < 0.05) (Table 2).

Correlation analysis of hypertension prevalence and each body fat index in oil workers

In this study, Spearman rank correlation was used to analyze the correlation between the prevalence of hypertension and various body fat indicators such as WC, BMI, WHtR, LAP, and VAI in the study subjects (Table 3). The results suggested that the prevalence of hypertension in oil workers was positively associated with WC, BMI, WHtR, LAP, and VAI (P < 0.001).

Multifactorial logistic regression analysis of hypertension and LAP and VAI in oil workers

LAP and VAI quartiles were grouped (Q1, Q2, Q3, Q4) to establish a logistic regression model for hypertension in oil workers (Table 4). The results suggest that higher LAP and VAI among oil workers are independent risk factors for the development of hypertension (P < 0.001). After adjustment by model 2, the OR (95% CI) values of LAP in the third and fourth quartile groups were 3.71 (1.71–7.75) and 5.69 (2.72–11.89), respectively, using the lowest quartile group as a reference. The OR (95% CI) values were 2.63 (1.49 to 4.67) and 3.56 (2.03 to 6.23) in the third and fourth quartile groups compared to the lowest quartile group of VAI. The risk of hypertension increases gradually with the increase of LAP and VAI levels.

Predictive value of LAP, VAI, and LAP combined with VAI for hypertension in oil workers

According to the receiver operating characteristic curve (ROC), the area under curve (AUC) of LAP, VAI, and LAP combined with VAI were 0.658 (95% CI [0.619–0.696], P < 0.001), 0.614 (95% CI [0.574–0.654], P < 0.001), 0.661 (0.620–0.703, P < 0.001) in the male group; 0.787 (95% CI [0.710–0.865], P < 0.001) in the female group, 0.732 (95% CI [0.640–0.825], P < 0.001), 0.792 (0.719–0.864, P < 0.001). This suggests that LAP, VAI and LAP combined with VAI have some diagnostic value for hypertension prevalence. The cut-off values for LAP, VAI, and LAP combined with VAI were 43.245, 1.575, and 0.129 for the male group and 35.730, 1.759, and 0.025 for the female group, respectively (Table 5, Figs. 1 and 2).

Dose-response relationship between LAP, VAI index and hypertension prevalence

After adjusting for relevant confounders, the restricted cubic spline results illustrate a nonlinear dose-response relationship between LAP, VAI, and risk of hypertension (P < 0.001 for overall trend and P < 0.001 for nonlinearity). The risk of hypertension increased rapidly before a VAI index of four and then began to decrease gradually; as the LAP index increased, the risk of hypertension increased, and as the LAP continued to increase above 80, the risk of hypertension began to decrease, but the decrease was small (Figs. 3 and 4).

Limitations

This study has the following limitations. Firstly, the present study is a cross-sectional study and the causal relationship between LAP, VAI and hypertension could not be analyzed. Secondly, the sample of this study was a single-center sample from the occupational population in Xinjiang, and there are some limitations of extrapolation. Future longitudinal epidemiological studies with large samples from different regions and populations are still needed.