Baseline characteristics of E-cigarette users and Non-users
The study included 872 participants, with 188 (21.6%) classified as e-cigarette users (≥ 1 day of use) and 684 (78.4%) as non-users. E-cigarette users were significantly younger than non-users (29.7 ± 10.2 vs. 42.3 ± 16.8 years, P < 0.001) and had a higher proportion of Non-Hispanic White individuals (56.4% vs. 45.6%, P = 0.003). Diabetes prevalence was lower in users (7.4% vs. 14.7%, P = 0.012), and users exhibited lower BMI (27.1 ± 5.8 vs. 28.9 ± 6.5 kg/m², P = 0.004). Serum cotinine levels were markedly higher in users (223.6 ± 498.7 vs. 85.2 ± 312.4 ng/mL, P < 0.001), consistent with greater nicotine exposure. Systolic blood pressure was slightly lower in users (118.9 ± 15.3 vs. 122.4 ± 17.6mmHg, P = 0.021). No significant differences were observed in gender distribution (P = 0.078)(Table 1).
Multivariable logistic regression
In weighted multivariable logistic regression analyses, e-cigarette use demonstrated a progressively stronger association with CKD risk across sequentially adjusted models. After adjustment for age, gender, and race (Model 1), e-cigarette users showed 2.10-fold higher odds of CKD (95% CI = 1.40–3.15, P = 0.003). Further adjustment for BMI (Model 2) strengthened this association (OR = 2.30, 95% CI = 1.60–3.30, P = 0.001). In the fully adjusted model including diabetes, hypertension, and current cigarette smoking (Model 3), e-cigarette use exhibited the strongest association with CKD (OR = 2.50, 95% CI = 1.80–3.48, P < 0.001), indicating a dose-response relationship with covariate adjustment and confirming independence from key demographic, metabolic, and smoking-related confounders (Table 2).
Dose-Response relationship between E-cigarette use frequency and CKD
A clear dose-response relationship was observed between e-cigarette use frequency and CKD risk (P for trend = 0.002)(Table 3). Compared to non-users, participants reporting 1–2 days/week of e-cigarette use had 1.80-fold higher odds of CKD (95% CI = 1.20–2.70), while those using e-cigarettes ≥ 3 days/week exhibited a 2.60-fold increased risk (95% CI = 1.70–4.00). A Wald test confirmed significant differences between frequency categories (P = 0.041), supporting a graded association. Restricted cubic spline analysis confirmed a linear trend (P-nonlinearity = 0.35), with no evidence of threshold effects. The ≥ 3 days/week threshold represented the upper tertile of use frequency (75th percentile = 4 days), optimizing between exposure intensity and sample size.These associations persisted after adjusting for age, gender, race, BMI, diabetes, and hypertension, supporting a graded relationship between vaping frequency and CKD (Fig. 2A).
Association Between E-cigarette Use and CKD. A Multivariable Logistic Regression. B Restricted Cubic Spline Analysis of E-cigarette Use Frequency and CKD Risk
Subgroup analysis by diabetes status, obesity, and current smoking status
Subgroup analyses revealed significant heterogeneity in the association between e-cigarette use and CKD by diabetes status (P-interaction = 0.032) and current smoking status (P-interaction = 0.021). Among non-diabetic individuals, e-cigarette use was associated with a 2.40-fold increased odds of CKD (95% CI = 1.65–3.50), whereas no significant association was observed in diabetic participants (OR = 1.30, 95% CI = 0.80–2.10). Stratification by current smoking status showed a stronger effect in smokers (OR = 2.35, 95% CI = 1.55–3.56) compared to non-smokers (OR = 1.78, 95% CI = 1.05–3.02). In contrast, the association did not differ significantly by obesity status (P-interaction = 0.215), with comparable effect sizes in both BMI < 30 (OR = 2.20, 95% CI = 1.45–3.35) and BMI ≥ 30 subgroups (OR = 1.85, 95% CI = 1.10–3.10)(Table 4). These findings suggest that the detrimental effects of e-cigarettes on kidney health are modified by metabolic and exposure factors, with amplified risk in non-diabetic individuals and current smokers, potentially reflecting synergistic toxicity pathways and dose-dependent biological effects (Fig. 2B).
Sensitivity analysis after propensity score matching (PSM)
Propensity score matching (PSM) was performed to further address potential confounding(Table 5). After matching 188 e-cigarette users with 188 non-users based on age, gender, race, BMI, diabetes, systolic BP, and current smoking, OR = 1.89 (95% CI = 1.18–3.01, P = 0.009), with SMD < 0.10 for all covariates. Using cotinine ≥ 10 ng/mL (n = 203, excluding dual users of traditional cigarettes), e-cigarette exposure was associated with CKD (OR = 2.25, 95% CI = 1.43–3.54, P = 0.001), validating self-reported use. After excluding all current cigarette smokers, e-cigarette use retained a significant association with CKD (OR = 2.05, 95% CI = 1.32–3.18, P = 0.002), confirming the relationship among exclusive e-cigarette users. Excluding dual users (n = 156), e-cigarette use remained associated with CKD (OR = 1.98, 95% CI = 1.28–3.06, P = 0.003), indicating isolation of e-cigarette-specific effects. Covariate balance was achieved with standardized mean differences (SMD) < 0.10 for all matched variables, confirming adequate overlap between groups. The attenuated but persistent effect size in the matched cohort (vs. unmatched OR = 2.10) suggests residual confounding in observational analyses, yet the robustness of the association underscores its clinical relevance(Fig. 3).
Standardized Mean Differences Before and After PSM
link
About the Author
