Causal relationship between varicose veins and hydrocele: A bidirectional two-sample Mendelian randomization study

Abstract

Objective:

This study aimed to investigate the causal relationship between varicose veins and hydroceles.

Methods:

Genetic data for varicose veins and hydroceles were extracted from a Genome-Wide Association Study (GWAS). A total of 22,037 cases of varicose veins and 2634 cases of hydroceles were included for Mendelian randomization analysis. Inverse variance-weighted (IVW) analysis was employed as the primary method, with weighted median and MR-Egger analyses were used for supplementary validation. Several sensitivity analyses were performed to further assess the results.

Results:

The IVW analysis of varicose veins on hydroceles revealed an odds ratio (OR) of 1.117 (95% confidence interval (CI), 1.009–1.236; P = 0.014). Conversely, the IVW analysis of hydroceles on varicose veins showed an OR of 0.983 (95% CI, 0.944–1.025; P = 0.330). Heterogeneity and pleiotropy were negative in the bidirectional analysis.

Conclusions:

This study suggests that varicose veins may be a risk factor for hydroceles. However, there is no evidence to support hydroceles as a causal risk factor for varicose veins. Our findings provide genetic evidence for the potential role of varicose veins as a risk factor for hydroceles, offering new insights for clinical practice. By establishing the causal relationship, high-risk patients can be identified, allowing for early surveillance of hydrocele presence to facilitate timely intervention and optimized treatment strategies. A statistically significant causal relationship between varicose veins and hydroceles was found, whereas hydroceles did not exert a causal impact on varicose veins.

Introduction

Hydrocele is a common condition of the male reproductive system that primarily affects pediatric and adult populations. In adults, it is typically secondary, often resulting from inguinal surgery, trauma, or inflammation. ¹ During embryonic development, the testes descend through the inguinal canal, forming an open ring at the internal inguinal ring. If the closure of the processus vaginalis fails after the testes descend into the scrotum, it leads to a condition known as patent processus vaginalis. When this patent processus vaginalis has a restricted lumen, allowing only fluid passage, it results in hydrocele. ² Studies have shown that autopsy findings in adults reveal that 5%–37% of individuals have a persistently patent processus vaginalis. A patent processus vaginalis is a necessary predisposing factor for the development of indirect inguinal hernia, with 8%–12% of adults eventually developing this condition. ³ Studies also suggest that defects in lymphatic drainage may contribute to hydrocele development, although the exact mechanisms remain unclear.^4–6

Varicose veins are common venous disorders affecting individuals across various age groups, with a prevalence of approximately 30%–60% in adults; they can also occur in children. This condition is primarily caused by valvular incompetence, which impedes venous reflux and leads to vascular wall dilation, distortion, and a range of associated symptoms. Varicose veins can develop in various parts of the body, with the lower extremities, portal veins, esophagus, and spermatic cord vessels being the most commonly affected.

Studies have shown that prolonged venous hypertension caused by varicose veins can lead to symptoms such as edema, skin pigmentation, skin ulcers, and venous thrombosis. ⁷ Traditional studies indicate that 20%–25% of varicocele surgeries involve concomitant patent internal inguinal rings, which are an important cause of hydrocele formation. ⁸

Varicocele can cause venous reflux or impaired venous drainage, ⁹ consequently increasing intrascrotal pressure. Impaired venous return may also elevate testicular temperature, negatively impacting normal testicular function.^10–12 Previous studies have indicated that hydrocele similarly exerts adverse effects on testicular function, ¹³ suggesting a potential causal relationship between these two conditions. Furthermore, within the male reproductive system, particularly in the postoperative setting, a notable association exists between varicocele and hydrocele, with hydrocele being one of the most common complications following varicocelectomy. ¹⁴ Nevertheless, the precise causal relationship between primary varicocele and hydrocele remains unclear.

Mendelian randomization (MR) is considered a natural randomized clinical trial (RCT) that can effectively infer the causal relationship between exposure factors and disease outcomes. ¹⁵ Compared to traditional case–control and cross-sectional studies, MR better controls for confounding factors and reverse causality, providing more reliable causal inferences.^16,17 This study used bidirectional two-sample MR analysis to assess the potential causal relationship between varicose veins as a genetic phenotype and hydrocele.

Methods

Research design

In this study, varicose veins were considered as the exposure factor, with SNPs significantly associated with varicose veins used as instrumental variables. Hydrocele was selected as the outcome variable. A positive Mendelian Randomization (MR) analysis was conducted using a two-sample MR method. Cochran’s Q test was applied to assess heterogeneity, and a sensitivity analysis was performed to validate the robustness of the results. For reverse two-sample MR analysis, hydrocele was designated as the exposure factor, with SNPs significantly associated with hydrocele chosen as instrumental variables, and varicose veins set as the outcome variable. The research workflow is illustrated in Figure 1.

Figure 1.

Study flowchart.

Data sources

Genetic data regarding varicose veins and hydrocele were obtained from the website (https://gwas.mrcieu.ac.uk). ¹⁸ The MRC-IEU provides summary data for Genome-Wide Association Studies (GWAS) on European ancestry populations for varicose veins (dataset ID: ebi-a-GCST90018939), ¹⁹ which includes 22,037 cases and 437,665 controls. Summary data for hydroceles in individuals of European ancestry were also retrieved (dataset ID: finn-b-N14_HYDROCELE), comprising 2634 cases and 72,799 controls. All data used were in the public domain and did not require additional ethical approval.

To mitigate biases related to racial confounding factors, the genetic background of the study population was exclusively European. The basic information is presented in Table 1.

Table 1.

Basic information of the GWAS database in the two-sample MR study.

Name	ID	Sample size	Number of SNPs	ASce	Sex	Year
Varicose veins	ebi-a-GCST90018939	459,702	24,179,697	European	Male /Female	2021
Hydrocele	finn-b-N14_HYDROCELE	75,433	16,377,576	European	Male /Female	2021

GWAS: Genome-Wide Association Study.

Instrumental variable selection

Instrumental variables must fulfill three assumptions: (1) a strong association with varicose veins, (2) independence from confounding factors, and (3) no alteration of the outcome through pathways independent of varicose veins.

For the positive MR analysis, SNPs with significant associations (P < 5.0 × 10⁻⁸, F > 10) with hydrocele in varicose veins were selected. Given the limited number of SNPs, the threshold was slightly adjusted based on the genomewide significance level (5 × 10⁻⁸) to meet the requirements for further analysis. For the reverse MR analysis, SNPs with significant associations (P < 1.0 × 10⁻⁵, F > 10) with varicose veins in hydrocele were chosen. ²⁰ A threshold of r² = 0.001 and a kilobase pair (kb) of 10,000 were set to cluster SNPs. SNPs with r² > 0.001 and a distance < 10,000 kb were excluded to ensure no linkage disequilibrium (LD) correlation. The Human Genotype-Phenotype Association Database^21,22 was used to identify phenotypes related to the remaining SNPs, eliminating those associated with the outcome variable. Incompatible alleles and palindromic SNPs were removed, resulting in a set of effective SNPs significantly associated with both exposure and outcome variables.

Mendelian randomization analysis

After aligning the effect alleles for the GWAS of varicose veins and hydroceles, various MR methods, including Inverse Variance Weighted (IVW), ²³ MR-Egger regression, ²⁴ and Weighted Median (WME), ²⁵ were employed to determine the MR estimates for the association between varicose veins and hydroceles. The IVW method, which considers both heterogeneity and pleiotropy, was preferred when both analyses yielded negative results, as it is the most accurate in MR. ²⁶ This method uses the inverse variance of each gene pair to weight the data, enhancing sensitivity to data with smaller variances and providing a more accurate estimate of the regulatory effect of genes. The MR-Egger method introduces an intercept term into the Mendelian randomization model to detect potential horizontal bias. The weighted median and weighted mode methods offered consistent estimates of the causal effects, demonstrating good stability and reliability in the presence of outliers and extreme observations. The simple mode, which adopts the traditional mode estimation approach, offers a concise estimate of the regulatory effects by identifying the most frequently occurring expression patterns in all gene pairs.

Sensitivity analysis

Sensitivity analysis evaluates the uncertainties in the model to minimize their impact on the subsequent analysis results. The IVW random-effects model and Cochran’s Q test were used for the two-sample Mendelian Randomization analysis to assess heterogeneity between individual genetic variations. If Cochran’s Q test yielded a P-value < 0.05, ²⁷ the final MR results followed the IVW random effects model; otherwise, the fixed effects model was used. The intercept term in the MR-Egger model can be used to evaluate horizontal pleiotropy. The causal estimation results of the MR-Egger test were used when the analysis results indicated significant horizontal pleiotropy.

All analyses were conducted using the “TwoSampleMR” package in R version 4.1.2, and data visualization was accomplished using R software.

Results

Varicose veins and hydrocele: Causal relationship

Instrumental variable selection

From the GWAS data, 63 SNPs exhibiting significant associations were selected from the GWAS data. Each SNP had an F-value of > 10, indicating the absence of weak instrumental variables. Three incompatible and palindromic SNPs were excluded. Using phenoscanner, SNPs related to confounding factors were filtered, resulting in a final set of 60 SNPs that were deemed effective as instrumental variables(Supplemental Table 1).

Mendelian randomization analysis

In the MR analysis exploring the relationship between varicose veins and hydroceles, the IVW model, weighted median, and MR-Egger regression were applied, with results visualized in Figure 2. Using the IVW method as the gold standard, the results (OR = 1.117, 95% CI: 1.009–1.236, P = 0.014) (Supplemental Figure 1) indicated a causal relationship between varicose veins and hydroceles. The weighted median method (OR = 1.196, 95% CI: 1.042–1.374, P = 0.011) yielded similar findings.

Figure 2.

Forest plot of Mendelian randomization analysis for varicose veins and risk of hydrocele.

Sensitivity analysis

The MR-Egger intercept and Cochran’s Q test results (P = 0.054 and P = 0.054, respectively) suggested no evidence of horizontal pleiotropy or heterogeneity. The heterogeneity tests yielded an MR-Egger regression intercept of −0.012, P = 0.351, indicating no apparent horizontal pleiotropy. The leave-one-out analysis revealed no SNPs that significantly influenced the causal estimate values (Figure 3). The symmetric funnel plot confirmed the stability of the results (Figure 4).

Figure 3.

Sensitivity analysis using the leave-one-out approach.

Figure 4.

Funnel plot of the Mendelian randomization analysis for varicose veins and hydrocele.

Hydrocele and varicose veins: Causal relationship

Instrumental variable selection

Eighteen SNPs exhibiting significant associations were selected from the GWAS data. Each SNP had an F-value of > 10, indicating the absence of weak instrumental variables. Subsequently, incompatible and palindromic SNP were excluded. Using phenoscanner, SNPs related to confounding factors were filtered, resulting in a final set of 17 SNPs identified as instrumental variables(Supplemental Table 2).

Mendelian randomization analysis

In the MR analysis investigating the relationship between hydroceles and varicose veins, the IVW model, weighted median, and MR-Egger regression were applied, with results visualized in Figure 5. Considering the IVW method as the gold standard, the results (OR = 0.983, 95% CI: 0.944–1.025, P = 0.330) (Supplemental Figure 2) indicated no causal relationship between hydroceles and varicose veins. The weighted median method (OR = 0.968, 95% CI: 0.919–1.020, P = 0.225) and the MR-Egger method (OR = 0.971, 95% CI: 0.895–1.053, P = 0.495) produced similar conclusions.

Figure 5.

Forest plot of Mendelian randomization analysis for hydrocele and risk of varicose veins.

Sensitivity analysis

The MR-Egger intercept and Cochran’s Q test (P = 0.069 and P = 0.090, respectively) suggested no evidence of horizontal pleiotropy or heterogeneity. The heterogeneity tests yielded an MR-Egger regression intercept of 0.003 (P = 0.733), indicating no apparent horizontal pleiotropy. The leave-one-out analysis revealed no SNPs that significantly influenced the causal estimate values (Figure 6). The symmetric funnel plot suggests the stability of the results (Figure 7).

Figure 6.

Sensitivity analysis was carried out using the leave-one-out approach.

Figure 7.

Funnel plot of the Mendelian randomization analysis for hydrocele and varicose veins.

Discussion

This study examined the bidirectional association between varicose veins and hydroceles using MR. The results provide evidence supporting a causal relationship between varicose veins and hydrocele; however, no causal link was found between hydrocele and varicose veins.

Our analysis employed varicose veins as a genetic phenotype exposure, revealing genetic evidence for a causal relationship between varicose veins and hydroceles. Specifically, the risk of developing hydrocele was found to increase with the incidence of varicose veins, whereas the risk of developing varicose veins did not rise with the occurrence of hydrocele.

This randomized study utilized a large sample size from a GWAS database and provided clinical research-based evidence for the causal relationship between varicose veins and hydrocele. MR utilizes genetic variations extracted from GWAS as instrumental variables to eliminate confounding effects potentially caused by age or other environmental factors, thereby focusing on the causal relationship between exposure factors and outcome variables. Therefore, this study provides evidence of a potential causal relationship between varicose veins and hydroceles through genetic susceptibility.

Varicose veins, particularly varicoceles, are a known cause of male infertility, while hydroceles may impact testicular function. Previous investigations have identified increased venous pressure and capillary stasis as consequences of varicocele, leading to interstitial edema in the scrotum. ²⁸ Notably, numerous surgeons have documented postvaricose vein surgery hydrocele as a common complication, primarily attributed to the intraoperative disruption of lymphatic reflux. Since our Mendelian randomization study demonstrated a causal relationship between varicose veins and hydroceles, early surveillance using ultrasound or other imaging modalities can be performed in patients with varicose veins, a known high-risk factor. If a persistently patent processus vaginalis is detected, early intervention may be warranted to prevent the progression of hydrocele or its potential transformation into an indirect inguinal hernia. Similarly, early intervention may also be beneficial in cases of secondary hydrocele.

It is well-established that hydrocele and inguinal hernia share several common characteristics in their occurrence. ²⁹ A prospective cohort study has shown a positive correlation between varicocele and the development of inguinal hernia. ³⁰ However, there is currently no direct biological evidence establishing varicose veins as a risk factor for hydrocele. Our MR analysis supported the causal impact of varicose veins on hydrocele formation (OR = 1.117, 95% CI: 1.009–1.236, P = 0.014).

Additionally, a clinical study indicated that inguinal hernia is frequently concomitant with varicocele during surgery, suggesting a mechanistic similarity in the onset of inguinal hernia and varicocele. The mechanisms contributing to varicocele-induced inguinal hernia involve not only venous compression but also biochemical changes due to enzymatic abnormalities. ³¹ Clarifying the causal relationship between varicose veins and hydrocele can help optimize the treatment of male infertility. In addition to addressing varicocele, simultaneous management of hydrocele could be considered to improve treatment outcomes. Particularly in patients with infertility, these findings may encourage clinicians to adopt a dual-treatment approach, addressing both varicose veins and hydroceles, thereby enhancing fertility rates.

Our study has some limitations. First, our dataset comprises only European populations, which limits the generalizability of the results to non-European populations. Second, there was heterogeneity in the findings. Although two of the primary analysis methods in our forward MR analysis were statistically significant and our results were relatively stable, heterogeneity may have impacted the reliability of our findings. Third, as a linear MR study, our analysis could only demonstrate causal relationships, making the quantification of indicators challenging. Therefore, substantial large-scale prospective studies are required to qualitatively and quantitatively explore this association.

Conclusion

Our MR analysis provides genetic evidence supporting an association between varicose veins and hydroceles. A causal relationships was observed between varicose veins and hydroceles, while no such connection was found between hydroceles and varicose veins. Clinicians should, therefore, consider the potential risk of hydrocele in patients with varicose veins.