A total of 200 participants (100 HTG cases and 100 controls) were recruited from the physical examination cohort at Guangzhou 11th People’s Hospital between January and December 2023. To control for population stratification, participants were proportionally matched between the observation and control groups. Only Han Chinese individuals were included, excluding other ethnic groups to avoid genetic heterogeneity confounding the results. All participants completed the TCM Dampness Syndrome Assessment Scale. Ages ranged from 18 to 75 years (mean: 47.15 ± 11.93), with 134 males (67%; mean age 47.42 ± 10.63) and 66 females (33%; mean age 46.61 ± 14.29). This study was approved by the Ethics Committee of Guangzhou Cadre Health Management Centre(Ethics Number: JGZX-2023-06), and written informed consent was obtained from all participants.

1.HTG diagnosis (Wang ZW, 2024): Fasting TG ≥ 1.7 mmol/L(150 mg/dL). Controls: Total cholesterol (TC) < 5.2 mmol/L(200 mg/dL), low-density lipoprotein cholesterol (LDL-C) < 3.4 mmol/L (130 mg/dL), HDL-C≥ 1.0 mmol/L(40 mg/dL) and TG < 1.70 mmol/L(150 mg/dL). 2.Dampness syndrome diagnosis (Lu Ty and Cai, 2021): Evaluated using the TCM Dampness Syndrome Diagnostic and Evaluation Scale (National Key Laboratory of TCM Dampness Syndrome, Ministry-Province Co-Constructed). This 30-item self-assessment scale (total score 120 points) defines: No dampness syndrome 0–19 points; Dampness syndrome ≥20 points.

The inclusion criteria were as follows: ① completed the TCM Dampness Syndrome Evaluation Scale and obtained a score greater than or equal to 0; ② was able to provide written informed consent, cooperate with the completion of the questionnaire and provide a blood sample; ③ aged ≥20 years old and ≤75 years old. The exclusion criteria were as follows: ① Inability to cooperate with the study; ② History of mental disorders; ③ Pregnant or lactating women; ④ Patients with diabetes, hypothyroidism, nephrotic syndrome, liver/kidney diseases, heavy alcohol consumption, or those taking lipid-altering medications (statins/fibrates/omega-3 fatty acids, retinoids, steroids, beta-blockers, antiretrovirals).

Demographic, clinical, and biochemical data were retrieved from hospital records. Genotyping was performed via PCR and Sanger sequencing. All investigators specialized in TCM or integrated Chinese-Western medicine and were trained in the study’s standard operating procedures. Participants were randomly selected from outpatient attendees, with trained investigators assisting in questionnaire completion to ensure data integrity and reduce bias. Questionnaire components: ① General demographics (age, gender, etc.); ② Medical history; ③ TCM Dampness Syndrome Assessment Scale. Physical examinations: Height, weight, body mass index (BMI), blood pressure, waist circumference (WC), etc. Laboratory assessments: ① Biochemical markers: TC, TG, LDL-C, HDL-C, apolipoprotein AI, apolipoprotein B; ② Genetic polymorphisms: APOA5 rs2075291 and CIDEB rs2144492 loci. Genomic DNA extraction, PCR amplification, and sequencing were conducted by Guangzhou Aiji Biotechnology Co., Ltd.

Primer sequences: For ApoA5 rs2075291: Forward primer: 5′-CAG​CAA​CTG​AAG​CCC​TAC​ACG-3′, Reverse primer: 5′-ATG​CCG​CTC​ACC​AGC​TCT​CG-3′, Product length: 227 bp.

For CIDEB rs2144492: Forward primer: 5′-CTT​ATG​GCT​TCT​CCA​GTA​GGT-3′, Reverse primer: 5′-GTA​TGT​GTG​TCT​TTG​GTG​ATG​A-3′, Product length: 194 bp.

PCR reaction conditions: Initial denaturation at 94 C for 5 min; Denaturation at 94 C for 30 s; Annealing at 56 C for 30 s; Extension at 72 C for 30 s; 35 cycles in total; Final extension at 72 C for 5 min after the last cycle; Storage at 4 C.

The amplification products were analyzed by 1.5% agarose gel electrophoresis. The genotyping success rate and repeat concordance rate for APOA5 rs2075291 and CIDEB rs2144492 SNPs both reached 100%, satisfying quality control criteria.

Data were analyzed using SPSS 26.0. Measurement data were expressed as mean ± standard deviation. Subgroup comparisons were performed via t-tests or analysis of variance. Pearson correlation analysis was applied for normally distributed data, while Spearman correlation was used for non-normally distributed data. Allele and genotype frequencies (calculated via genotype counting) were compared using chi-square tests or Fisher’s exact test. The false discovery rate (FDR) was corrected via the Benjamini–Hochberg method, with corrected significant results reported. The Hardy-Weinberg equilibrium (HWE) was assessed for polymorphic locus genotype distributions. The additive model served as the primary model, with dominant/recessive models as secondary. Genotypes were coded as 0/1/2 under the additive model. Models reported the Area Under the Curve (AUC), Hosmer-Lemeshow test (HL) p-value, and maximum variance inflation factor (VIF), including APOA5×dampness and CIDEB × dampness interaction terms. Post-hoc power analysis was conducted using the expected minor allele frequency (MAF) and observed OR values. Binary logistic regression was used to identify factors associated with HTG, with p < 0.05 denoting statistical significance.

As shown in Table 1, the HTG group had significantly higher waist circumference, BMI, TG, TC, LDL-C, apolipoprotein B, and dampness syndrome scores compared to controls (p < 0.05). Conversely, HDL-C and apolipoprotein AI were lower in the HTG group (p < 0.05). There was no significant age difference between groups (p = 0.152). Collinearity assessment for BMI and waist circumference showed a VIF of 1.

The polymorphic genotypes of the APOA5 rs2075291 locus in both groups were in Hardy-Weinberg equilibrium (p > 0.05). An interaction was observed between APOA5 rs2075291 and dampness syndrome in the overall population (F = 5.796, p = 0.004). As presented in Table 2, after FDR correction (p ≤ 0.033), the genotype of APOA5 rs2075291 differed significantly between the HTG and control groups (OR = 2.916, 95% CI:1.160–7.334, χ² p = 0.019), with significant intergroup differences in allele frequencies (OR = 2.727, 95% CI:1.113–6.682, χ² p = 0.023). In the dampness syndrome subgroup, neither genotype (OR = 2.241, 95% CI:0.667–7.527, χ² p = 0.183) nor allele frequencies (p = 0.200) showed significant differences. In the non-dampness syndrome subgroup, the HTG group showed no significant difference from the control group in APOA5 rs2075291 genotype frequency (OR = 4.083, 95% CI:0.977–17.063, χ² p = 0.041) or allele frequencies (p = 0.05).

The genotype distribution of the CIDEB rs2144492 locus polymorphism in both groups was in Hardy-Weinberg equilibrium (P > 0.05). An interaction between CIDEB rs2144492 and dampness syndrome was observed in the overall population (F = 5.796, p = 0.004). As shown in Table 3, after FDR correction (p ≤ 0.042), Fisher’s exact test showed significant differences in CIDE-B rs2144492 genotype (OR = 1.688, 95% CI:0.886–3.141, χ² p = 0.042) and allele frequency (OR = 1.837, 95% CI:1.040–3.244, χ² p = 0.034) between HTG and control groups. In the dampness syndrome subgroup, the HTG group exhibited a higher frequency of CIDEB rs2144492 TG/TT genotypes, though the difference did not reach statistical significance (OR = 2.065, 95% CI:0.816–5.226, χ² p = 0.146). Similarly, no significant difference was observed in allele frequencies between the two groups (p = 0.043). In the non-dampness subgroup, neither CIDEB rs2144492 genotype (OR = 1.604, 95% CI:0.639–4.023, χ² p = 0.281) nor allele frequency (p = 0.250) differed significantly between groups.

Using log-transformed mean TG as the dependent variable, interactions of APOA5×dampness syndrome and CIDEB × dampness syndrome were tested. Interaction p-values were: APOA5×dampness syndrome (p = 0.322) and CIDEB × dampness syndrome (p = 0.966). As shown in Table 4, mean triglyceride levels did not differ significantly between dampness and non-dampness groups for APOA5 rs2075291 GG/GT genotypes or CIDEB rs2144492 GG/TG/TT genotypes (p > 0.05). An ANCOVA model for log-transformed TG was fitted, adjusting for age, gender, BMI, dampness syndrome, and genotype. Adjusted mean differences are reported in Table 4.

As shown in Table 5, in the overall population, HTG showed significant correlations with age, gender, body mass index, APOA5 rs2075291 genotype, and dampness syndrome (p < 0.05). In the dampness syndrome subgroup, HTG was significantly associated with gender and body mass index, whereas in the non-dampness syndrome subgroup, it was significantly correlated with gender, body mass index, and APOA5 rs2075291 genotype (p < 0.05 for all).

Variables were selected based on correlation analysis results and primary research objectives, with genotypes encoded additively. Binary multivariate logistic regression was performed, using HTG status (presence/absence) as the dependent variable and including dampness syndrome, APOA5 rs2075291 genotype (GG = 0, GT = 1), and CIDE-B rs2144492 genotype (GG = 0, TG = 1, TT = 2) as independent predictors. Collinearity assessment showed all tolerance values > 0.1 and max VIF = 1.310, indicating no multicollinearity. AUC values for variables (age = 0.586, gender = 0.670, APOA5 = 0.555, CIDEB = 0.556, BMI = 0.721, dampness syndrome = 0.585) all exceeded 0.5. The Hosmer-Lemeshow test (χ² = 11.884, p = 0.156) indicated good model fit. Logistic regression identified male gender, higher BMI, dampness syndrome, and APOA5 rs2075291 genotype as independent HTG risk factors (p < 0.05), while CIDEB rs2144492 genotype trended toward significance (p = 0.05), as shown in Table 6.

This study recruited participants from outpatient clinics, which may introduce selection bias. For example, outpatients differ from the general population in disease severity, treatment compliance, and help-seeking behaviors, potentially biasing the observed associations between APOA5/CIDEB polymorphisms and HTG. Additionally, the single-center outpatient sample limits external validity to populations with similar healthcare-seeking patterns and clinical profiles, hindering generalizability to non-visited or geographically distinct groups.

Post-hoc power analysis using expected MAF and observed ORs showed that with n = 200, the power was only ∼51% for APOA5 rs2075291 (MAF = 0.1, OR = 2.241) and ∼47.2% for CIDEB rs2144492 (MAF = 0.19, OR = 1.8)-both far below the 80% statistical benchmark. This indicates insufficient precision in effect estimation. For low-frequency variant association studies, small sample sizes may miss true effects or yield non-reproducible results due to random error. Study designs should pre-calculate sample size based on MAF and target OR to avoid unreliable conclusions. Future research could integrate community epidemiological surveys to comprehensively assess genotype-TCM dampness syndrome interactions in general populations.