Parental smoking and risk of hypospadias: An updated meta-analysis of observational studies

Background Inconsistent relationships have been shown between cigarette smoking and hypospadias in offspring. The purpose of this study was to summarize epidemiological evidence to evaluate the relationship between parental smoking and the risk of hypospadias. Methods Up until October 2022, PubMed, EMBASE, Web of Science, and the Cochrane Library were systematically searched for qualified research. The summary RRs and 95% CIs were calculated using either a fixed-effects or a random-effects model. There were subgroup analyses undertaken to identify potential sources of heterogeneity. Results 44 studies with 16,637,830 participants were included in our meta-analysis. Overall, maternal active smoking [risk ratio (RR) = 0.94; 95% confidence interval (CI): 0.90–0.99; P < 0.01] was significantly associated with the risk of hypospadias. And neither paternal smoking (RR = 1.00; 95% CI: 0.86–1.15) nor maternal passive smoking (RR = 0.91; 95% CI: 0.60–1.23) was associated with the risk of hypospadias. Conclusion Our study discovered an association between maternal active smoking and a decreased risk of hypospadias, which may be due to the effect of smoking on androgen. However, as numerous studies have proved that cigarette smoking during pregnancy increases the risk of overall birth abnormalities in offspring, quitting cigarettes before pregnancy positively influences the health of offspring and should be advocated worldwide. Systematic review registration [www.crd.york.ac.uk/prospero], identifier [CRD42022319378].

Introduction musculoskeletal deformities, heart defects, and cryptorchidism (6). Even though smoking during pregnancy increases the risk of most congenital defects, many pregnant women continue to smoke (7).
A previous meta-analysis of 15 studies on this topic revealed that pregnant women who smoke have a decreased risk of hypospadias (6), while a recent meta-analysis of 12 studies indicates that pregnant women who smoke have an increased risk of hypospadias (8). However, the number of studies included in these two meta-analyses is quite limited. Moreover, despite the fact that paternal smoking and maternal passive smoking are more common, these two meta-analyses did not conduct a meta-analysis of the association between maternal passive smoking or paternal smoking and the risk of hypospadias. Numerous studies evaluating the risk of hypospadias associated with paternal active smoking and maternal passive smoking produced equivocal findings.
Notably, a number of studies, including a large number of cohort studies with adequate sample sizes, have been published, which will help provide an adequate number of studies to assess the risk of hypospadias with greater confidence and to investigate plausible causes for heterogeneity. Prior meta-analyses did not consider the duration of smoking exposure. Early in pregnancy, the important phase of fetal urethral development occurs. Examining the relationship between parental smoking in early pregnancy and hypospadias is crucial, as this may help explain causation. Therefore, with newly accumulating evidence, we aimed to conduct a meta-analysis with two aims: (i) to determine the risk of hypospadias associated with parental smoking, including maternal active smoking, paternal active smoking, and maternal passive smoking; and (ii) to identify potential moderators of heterogeneity using subgroup, including subgroups based on smoking exposure time, and other potential factors.

Search strategy
This research followed PRISMA standards and was pre-registered with PROSPERO: CRD42022319378, outlining our investigation's objectives and procedures.

Information sources
This meta-analysis included articles published before October 23, 2022, about maternal or paternal smoking and the risk of hypospadias. Using the Boolean approach, relevant papers were retrieved from PubMed, EMBASE, Web of Science, and the Cochrane Library. The following search terms were utilized in the systemic search: (smoking OR tobacco OR cigarette OR gestational cigarette exposure OR gestational smoking OR gestational tobacco exposure OR factor) AND (hypospadias OR congenital anomalies OR genitourinary anomalies). The reference lists of included papers were also searched to find any additional research that may have been overlooked.

Inclusion criteria
In accordance with the research question (PICO), the following inclusion criteria were established for the selection of articles: studies on women who gave birth to a live-born son (patients, P); studies that examined maternal active smoking/maternal passive smoking/ paternal smoking during pregnancy for risk of hypospadias (intervention, I); studies that used no cigarette exposure during pregnancy as a comparison (comparison, C); and studies that described the risk of hypospadias in their offspring (outcome, O).

Exclusion criteria
Studies were excluded if the following criteria were met: (1) statistical data for case and control groups or effect estimates were unavailable. (2) the absence of a non-smoking control group.

Data extraction
Two independent reviewers screened the abstracts and titles of the studies retrieved during the search procedure to identify those studies that potentially meet the inclusion criteria. Then, reviewers evaluated the entire text of the identified studies to determine which ones should be included in the study. Disputes were resolved through discussion or consultation with a third author.
The fundamental data from selected studies would be extracted, including the first author, the year of publication, the participants' country, the study period, the number of hypospadias cases and sample size of cohort studies, or the total sample size of case and control group for case-control studies, the exposure of interest, the assessment method of smoking status, the reported exposure time, the assessment method of hypospadias, whether the confounding factors were controlled, and quality score ( Table 1).

Assessment of risk of bias
The Newcastle-Ottawa Scale (NOS), the Agency Guidelines for Healthcare Research and Quality (AHRQ), was used to assess study quality and the potential for bias. A star system (range, 0-9 stars) was developed to rate the quality of research in the three categories of participant selection, group comparability, and exposure and result of interest for investigations, respectively. A study with seven or more stars as being low risk of bias was defined. The details are listed in Table 2.

Statistical analysis
The association between parental smoking and hypospadias in offspring was evaluated using RRs. Due to the low incidence of hypospadias, ORs were regarded as RRs directly. First, we estimated the summary RRs and 95% CIs for the risk of hypospadias associated with parental smoking using the RRs and 95% CIs from each study. Then, using the chi-square test (with a p-value of <0.1 and ≥0.1indicating high and low heterogeneity, respectively) and I 2 index statistics (varied from 0% to 100%), we determined the heterogeneity of these studies, which was described as no (0%-25%), low (26%-50%), moderate (51%-74%), and high (75%-100%).   When there was evidence of heterogeneity (I 2 > 50%) across studies, random-effects models were used to generate the combined RRs and the corresponding 95% CIs, whereas fixed-effects models were employed otherwise. Subgroup analyses were undertaken based on the study's primary outcomes to investigate the potential causes of heterogeneity: study design, participants' region by continent, assessment methods of smoking, whether the confounding factors were controlled, smoking exposure time, and quality score. Furthermore, publication bias was to be examined visually using funnel plots if the number of studies in either cohort equalled or exceeded 10. Stata version 15.1 and the statistical software R 4.1.1 were used to conduct all analyses.

Results
Search results and study characteristics Figure 1 is a diagrammatic illustration of the study identification and inclusion process. 126,938 articles were identified using our systematic search approach in PubMed, the Cochrane Library, EMBASE, and Web of Science. We retained 107,865 articles after removing duplicate studies. Following a review of the titles and abstracts, it was determined that 107,761 papers did not meet the research objective. After comprehensively screening 204 full texts, 58 papers were found to meet our inclusion criteria. Additionally, we searched the reference lists of chosen papers for relevant studies. We found one additional article (9). After reviewing the entire texts of fifty-nine studies, we found that several studies used similar initial data, running the risk of the same patients being included in the meta-analysis twice [Kallen et al. (10)   Frontiers in Pediatrics to data overlap (23), while another study was a book, for which data were unavailable (35). Three of the studies included by Zhang et al.
The quality assessment scores of included studies are summarized in Table 2. Overall, 75.0% (33/44) of these studies received a score higher than 6. The mean value was 6.86 stars for 35 case-control studies and 7.78 stars for 9 cohort studies. The evaluation of the study's quality revealed little evidence of considerable bias that could have significantly impacted the study's outcomes.

Maternal active smoking and risk of hypospadias
The meta-analysis of the combined RR ( Figure 2) indicated that maternal active smoking was associated with a lower risk of offspring FIGURE 2 Forest plot for maternal active smoking and risk of hypospadias. Paternal smoking and risk of offspring hypospadias Figure 3 summarizes risk estimates between paternal smoking and hypospadias in offspring. There was no significant association between paternal smoking and the risk of hypospadias (RR = 1.00; 95% CI: 0.86-1.15). Moreover, low evidence of heterogeneity was found, and a fixed-effects model was employed (I 2 = 31.1%; P = 0.170).
Maternal passive smoking and risk of offspring hypospadias Figure 4 summarizes risk estimates between maternal passive smoking and hypospadias in offspring. Overall, there was no association between maternal passive smoking and the risk of hypospadias (RR = 0.91, 95% CI: 0.60-1.23). In addition, moderate heterogeneity was detected, and the random-effects model was utilized (I 2 = 69.3%; P = 0.039).

Subgroup analysis
We performed a subgroup analysis based on the risk of hypospadias and smoking exposure time to determine whether smoking exposure time was the source of heterogeneity ( Figure 6). However, no differences were identified (P = 0.804; Figure 6).
Other subgroup analyses of maternal active smoking failed to explain the overall analysis's heterogeneity and revealed consistent findings within subgroups ( Figure 6). We found no difference in RR between cohort and case-control studies, participants' region by continent, the assessment methods of smoking, quality scores, and whether the confounding factors were controlled.

Publication bias
The funnel plot ( Figure 5) indicated no evidence of publication bias in risk estimates for maternal active smoking and offspring hypospadias.
We did not analyze publication bias in risk estimates for paternal smoking or maternal passive smoking using a funnel plot because the number of included studies was less than 10. Forest plot for paternal smoking and risk of hypospadias.

Discussion
There is considerable debate over the association between paternal smoking and offspring hypospadias. Our study demonstrated an association between maternal active smoking and a decreased risk of hypospadias in offspring. However, no association was found between maternal passive smoking or paternal smoking and the risk of offspring hypospadias.
Our subgroup analysis showed that the risk of offspring hypospadias was consistent regardless of whether the mother smoked in early pregnancy or during the entire pregnancy, which accorded with the results of a prior Danish study. Eighty-eight percent of pregnant women who smoked at the 16th week continued to do so throughout their pregnancy (68). Consequently, even if some studies collected data on smoking during early pregnancy, it could be a good approximation of exposure status throughout pregnancy.
There are two meta-analyses examining the association between maternal smoking and the risk of hypospadias in offspring (6,8). However, no meta-analysis has evaluated the risk of hypospadias related to maternal passive smoking or paternal smoking. The 15study meta-analysis by Hackshaw et al. reported an odds ratio (OR) of 0.90 (95% CI: 0.85; 0.95) for having offspring with hypospadias (6), whereas 12-study meta-analysis by Zhang et al. reported an odds ratio (OR) of 1.16 (95% CI: 1.01; 1.33) for having offspring with hypospadias. Our study has some significant advantages over the previous meta-analysis; for example, we included more existing articles with a total of 16,637,830 participants to more precisely estimate the association between parental smoking and the risk of hypospadias in offspring. We also performed the first meta-analysis on the association between maternal passive smoking and paternal smoking and the risk of offspring hypospadias. Additionally, we performed subgroup analyses to explore potential sources of heterogeneity.
The specific mechanisms underlying the association between maternal smoking and hypospadias in offspring remain unknown FIGURE 4 Forest plot for maternal passive smoking and risk of hypospadias. Funnel plot of the risk of hypospadias associated with maternal active smoking.
Ye et al. 10.3389/fped.2023.1003037 Frontiers in Pediatrics and require further study, but various hypotheses based on epidemiological and animal data have been offered. One hypothesis is that a higher androgenic steroid level in the fetuses of pregnant smokers promotes normal urethral closure. According to a prior study, estrogens may inhibit the activity of fetal androgens (69).
Smoking may have anti-estrogenic effects by increasing estradiol metabolism to metabolites with little estrogenic activity via increased 2-hydroxylation (70), which is consistent with the finding that smoking women have a higher androgenic steroid level (71,72). In addition, smoking reduces the activity of 21-or 11-hydroxylase in the adrenal cortex, increasing adrenal androgen secretion (73,74). A higher androgenic steroid level could compensate for intrinsic fetal endocrine testicular insufficiency, fetal androgen receptor deficiency, or decreased fetal 5-reductase activity (65), which promotes normal urethral closure, in fetuses of pregnant smokers. Our study should also be viewed with certain limitations. First, only one study indicated the exclusion of passive smokers from the comparison of active smokers and no-smokers, as well as the exclusion of active smokers from the comparison of passive smokers and no-smokers (14). The imprecision of the categorization of included studies may introduce bias into the study's findings. Second, there were a limited number of studies on maternal passive smoking, which may have led to biased results. The relationship between maternal passive smoking and the risk of hypospadias requires additional research. Third, cigarette smoking is associated with an increased risk of the majority of congenital defects (6). Our study did not assess the association between smoking and other congenital defects, which may have affected the study's credibility.
In conclusion, the purpose of our study, which included a significant number of participants to ensure statistical power, was to examine the association between parental smoking and hypospadias. Our study demonstrated an association between maternal active smoking and a decreased risk of hypospadias. No association was identified between parental smoking or passive smoking and the risk of hypospadias in offspring. However, quitting cigarettes before pregnancy indeed has a good effect on the health of the offspring and should be advocated worldwide.

Data availability statement
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.

Author contributions
The meta-analysis was conceived and designed by Z-hY and XL. Z-hY and H-sC independently extracted the data from PubMed, the Cochrane Library, EMBASE, and Web of Science. Z-hY led the data analysis and interpretation, drafted the manuscript, and revised the Summary forest plot of subgroup analyses. Summary forest plot of subgroup analyses, including study design, participants' region by continent, assessment methods of smoking, whether the confounding factors were controlled, smoking exposure time, and quality score.
Ye et al. 10.3389/fped.2023.1003037 Frontiers in Pediatrics content in response to feedback. Z-CZ played the role of the second reviewer. XW assisted with database retrieval and data collecting. H-sC assisted with data interpretation and gave crucial draft revisions. G-hW contributed to the conception and design, data interpretation, and revision of drafts. XL acted as the corresponding author, provided funding support, assisted with data interpretation, provided critical revision of drafts, and served as the third (mediating) reviewer. All authors contributed to the article and approved the submitted version.