jpad journal

AND option

OR option



C. Zheng1, R. Zeng1


1. Shantou University Medical College, Shantou 515000, Guangdong, China

Corresponding Author: Dr. Ruijie Zeng, Shantou University Medical College, Shantou, Guangdong, China, Email:

J Prev Alz Dis 2022;
Published online September 16, 2022,



Dear Editor,

With great interest, we read the recent study by Chun-Yu Cheng demonstrating that patients with varicose veins had an increased incidence of Alzheimer’s disease (AD), especially those with complicated cases (1). We thank the author for his significant work and meaningful results, and we would like to share our findings that there is a predictably genetically pathogenic effect of AD on varicose veins. Our results might indicate that the increased AD incidence in varicose vein patients is due to reverse causality, which is usually found in evaluating the association between two diseases in observational studies.
Due to various confounding factors and bias in observational studies, it is usually difficult to identify the causal relationship between two diseases simply by epidemiologic studies and analyses. Although the results of epidemiological analysis may show that increased incident AD is associated with individuals with varicose veins, such observational results are likely to result from reverse causality and need further investigation. Bi-directional two-sample Mendelian randomization (MR) (2) is therefore applied to effectively minimize the influences of confounding factors by treating genetic variants as instrumental variables (IVs) for exposures, and to assess the potential reverse causation between AD and varicose veins.
The publicly available genome-wide association studies (GWAS) summary statistics were accessed to investigate the causal effect of AD (Ncases = 21,982; Ncontrols = 41,944) (3) on varicose veins (Ncases = 17,027; Ncontrols = 190,028) (FinnGen) of European descent. Single-nucleotide polymorphisms (SNPs) of AD with P values less than 5E–07 were selected as IVs and were clumped based on the linkage disequilibrium (LD) structure (R2 < 0.001 within 10,000 kb). Using two-sample MR analysis, horizontal pleiotropic effects, heterogeneity, and leave-one-out sensitivity were detected for the validation of the causal association of AD on varicose veins. Weighted median, inverse variance weighted (IVW), weighted mode methods were applied to calculate the overall odd ratio (OR). Meanwhile, the causal association of varicose veins on AD was also detected by the same methods. All statistical analyses were performed using the TwoSampleMR package (4) with default parameters in R.
Generally, our results demonstrated that genetic susceptibility to AD were associated with an increased risk of varicose veins (IVW: OR = 1.074, 95% confidence interval [CI] = 1.026 to 1.124; P = 0.002). Weighted median (OR = 1.076, 95% CI = 1.023 to 1.132; P = 0.005), and weighted mode (OR = 1.075, 95% CI = 1.020 to 1.132; P = 0.015) methods all indicated a causal association for AD on varicose veins. These MR analysis results showed that AD was correlated with increased susceptibility of varicose veins and was statistically significant (Fig. 1A). No obvious horizontal pleiotropic effects and heterogeneity were identified in Mendelian randomization pleiotropic (IVW Q = 24.752, P = 0.074; MR-Egger Q = 25.055, P = 0.094) and outlier MR-PRESSO global (MR-PRESSO RSSobs = 27.025; P = 0.148) and MR-Egger test (MR–Egger intercept: 0.003, P = 0.664). Leave-one-out sensitivity analysis also showed no potential outliers resulting in inconsistent causal estimation (Fig. 1B). No evidence of causal effects of varicose veins on AD was identified (IVW: OR = 1.074, 95% CI = 1.026 to 1.124, P = 0.914; Weighted median: OR = 1.076, 95% CI = 1.023 to 1.132, P = 0.710; Weighted mode: OR = 1.075, 95% CI = 1.020 to 1.132, P = 0.895). Our findings demonstrated a predictably genetically pathogenic effect of AD on varicose veins rather than varicose veins on AD, which is opposite to the results reported in Cheng’s study.

Figure 1. (A) Forest plots illustrating the causal effect of AD on varicose veins. (B) Leave-one-out sensitivity analysis for AD on varicose veins

Taken together, our data from bi-directional two-sample MR analysis remind clinicians that AD might be a trigger for the development of varicose veins. The positive association between varicose veins and AD indicated in Cheng’s study might be due to reverse causality, and the observed relationship between varicose veins and AD might be interpreted directionally. More concern should be given and more preventive measures should be taken to prevent varicose veins development in AD patients.


Competing interests: The authors declared no conflict of interest.

Ethical standard: No ethical approval was required for the present study. All analyses were performed based on publicly available summary statistics.



1. Cheng CY. Increased Incident Alzheimer’s Disease among Individuals with Varicose Veins: A Population-Based Cohort Study. J Prev Alzheimers Dis. 2022;9(3):441-8. doi: 10.14283/jpad.2022.23.
2. McKinnon SJ, Lehman DM, Kerrigan-Baumrind LA, Merges CA, Pease ME, Kerrigan DF, et al. Caspase activation and amyloid precursor protein cleavage in rat ocular hypertension. Invest Ophthalmol Vis Sci. 2002;43(4):1077-87.
3. Kunkle BW, Grenier-Boley B, Sims R, Bis JC, Damotte V, Naj AC, et al. Genetic meta-analysis of diagnosed Alzheimer’s disease identifies new risk loci and implicates Aβ, tau, immunity and lipid processing. Nat Genet. 2019;51(3):414-30. doi: 10.1038/s41588-019-0358-2.
4. Hemani G, Zheng J, Elsworth B, Wade KH, Haberland V, Baird D, et al. The MR-Base platform supports systematic causal inference across the human phenome. Elife. 2018;7. doi: 10.7554/eLife.34408.