Association between bicuspid aortic valve phenotype and patterns of valvular dysfunction: A meta‐analysis

Abstract Background Valvular dysfunction is a common complication in patients with bicuspid aortic valves (BAV). The aim of this study was to determine the relationship between BAV morphology patterns and valve dysfunction. Methods We searched the PubMed, The Cochrane Library, Web of Science, and CNKI until May 31, 2020, to identify all studies investigating the morphology of BAV and valvular dysfunction, and data were extracted according to the Preferred Reporting Items for Systematic reviews and Meta‐Analyses (PRISMA). Data were analyzed using Stata 15.1 software. The additional characteristics (gender, mean age) were collected to perform a meta‐regression analysis. Results Thirteen studies on BAV‐RL (n = 2002) versus BAV‐RN (n = 1254) and raphe (n = 4001) versus without raphe (n = 673) were included. The BAV‐RL patients showed a higher incidence of aortic regurgitation than BAV‐RN patients (OR = 1.46; 95% CI: 1.12 to 1.90, p = .005), while the BAV‐RL patients showed a lower incidence of aortic stenosis than BAV‐RN patients (OR = 0.66, 95% CI: 0.58 to 0.76, p = .000); BAV patients with raphe presents a higher incidence of aortic regurgitation than those without raphe (OR = 1.95, 95% CI: 1.12–3.39, p = .017). No differences were found between raphe and without raphe group in the incidence of aortic stenosis (OR = 0.97, 95% CI: 0.53 to 1.76, p = .907). Mean age and gender had no influence on observed differences. Conclusions Our results confirmed a relationship between different BAV phenotypes and aortic valve dysfunction. BAV‐RL and BAV with raphe are more likely to develop aortic regurgitation, while patients with BAV‐RN present a higher possibility to develop aortic stenosis.


| INTRODUCTION
The bicuspid aortic valve (BAV) is the most common congenital cardiac defect that observed in 1%-2% of general population, 1 with a male to female ratio of about 3:1. Patients with BAV are at a high risk of developing aortic valve dysfunction, either stenosis or regurgitation, or both. Studies have suggested that 33% of patients with BAV will suffer serious and life-threatening complications in their lifetime. Therefore, early detection and prevention of the complications caused by BAV are of paramount importance. 2 BAV appears to be inherited in an autosomal dominant fashion with incomplete penetrance. It has been postulated that the defective genes encoding the protein matrix structure could be responsible for developmental impairment of heart, and leading to valvular abnormalities. [3][4][5] presents several phenotypes, and an animal experiment demonstrated that different BAV phenotypes are caused by different developmental processes, suggesting that different BAV phenotypes should be considered as different etiological entities with different valvular lesions, aortic size, and elasticity. 6 Thus, more credit should be given to the association between BAV phenotypes with valvular dysfunction. 7 The most common BAV pattern is fusion of the right and left coronary cusps, and fusion of the right and noncoronary cusps. 8,9 Previous evidence suggests that various BAV types, distinguished by the morphology of the valve cusp fusion, may carry different relationships with valvular dysfunction; however, the published literature is incoherent in this regard. Several studies have reported an increased frequency of significant valvulopathy in pediatric patients with right and left coronary cusps fusion, 10 while another longitudinal follow-up study claimed that BAV phenotype failed to demonstrate a prognostic implication. 11 2 | AIM OF THE STUDY Therefore, the purpose of our study was to evaluate the impact of different BAV cusp fusion morphology on the incidence of valvular dysfunction, and provide clues and evidence for early clinical diagnosis and prevention of complications.

| Search strategy
A systematic search was performed in the electronic databases (PubMed, The Cochrane Library, Web of Science, and CNKI), using the following search terms in all possible combinations: bicuspid aortic valve, aortic regurgitation, aortic stenosis, valve dysfunction. Articles were rejected on initial screening if from the title or the abstract it was judged that the article does not report aortic valve dysfunction and BAV morphology. Subsequently, the full text of the remaining articles was retrieved. All the references were also scanned. The particular studies were examined to exclude duplicated and overlapped data. Finally, only studies evaluating aortic stenosis and aortic regurgitation were included. In case of missing data, the authors were F I G U R E 1 Flow diagram for study selection. This flow chart shows the initial search results and final review of 11studies after consideration of exclusions contacted by e-mail to try to retrieve the original data. Each article was analyzed by two independent individuals and data extraction was done independently. In case of disagreement, a third investigator was consulted. Discrepancies were resolved by consensus. Data extraction was conducted according to the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) (Figure 1). 12   The following data were also extracted from each study: first author, year of publication, used imaging modality, study population characteristics including mean age, male gender percentage, sample size (number of subjects in particular BAV subtypes), number of patients with AS, and number of patients with AR.

| Statistical analyses and risk of bias assessment
The presented meta-analysis was performed using Statistica 15.1. The frequency variable is expressed as n (%). Differences among AS and AR between the two types of BAV patients were expressed as odds ratio (OR) with pertinent 95% CI for dichotomous variables. Overall effect was tested using Z scores, and significance was set at p < .05.
Statistical heterogeneity among studies was assessed with the chisquare Cochran's Q test and with the I 2 statistic, which measures the inconsistency across study results and describes the proportion of total variation in study estimates. To evaluate the individual impact of each study on the overall effect size, sensitivity analysis was conducted using the leave-one-out approach, by estimating the weighted mean difference in the absence of each single study.  1.04-2.67, p = .032). Heterogeneity among studies was significant (I 2 = 80.3%, p = .001), the combined effect quantity OR was determined using Random effect model. Forest plot summarizing the meta-analysis of studies comparing aortic regurgitation between raphe and without raphe BAV groups is shown in Figure 3A.  Heterogeneity among studies was significant (I 2 = 90.3%; p = 0.000), the combined effect quantity OR was determined using Random effect model. Forest plot summarizing the meta-analysis of studies comparing aortic stenosis between raphe and without raphe BAV groups is shown in Figure 3B.

| Sensitivity analysis
Presented pooled results were found to be robust in the performed leave-one-out sensitivity analysis, removing 1 study at a time.
Obtained stability of the presented results confirms a significant difference in the frequency of aortic stenosis and aortic regurgitation between the BAV-RL and BAV-RN groups. For the analysis of the association between BAV phenotype and aortic stenosis, I 2 ranged from 8.5% to 42.7%, showing increased heterogeneity (Table S1). For the analysis of the association between BAV phenotype and aortic regurgitation, I 2 ranged from 60.0% to 69.5%, the results did not differ from the previous ones. (Table S2).

| Publication bias analysis
Because it is recognized that publication bias can affect results of metaanalyses, we attempted to assess this potential bias using funnel plot suggested also no evidence of publications bias when comparing the incidence of aortic stenosis between BAV-RL and BAV-RN patients.

| DISCUSSION
Our meta-analysis shows that BAV patients with right and left cusp fusion are incline to develop aortic regurgitation, while patients with right and noncoronary cusp fusion are more likely to develop aortic stenosis. Moreover, bicuspid aortic valves with raphe showed a higher incidence of aortic regurgitation. However, with or without raphe does not affect the incidence of aortic stenosis. This meta-analysis is the first to assess the effect of BAV phenotype on valvular dysfunction differences.
BAV has diverse morphologic variants, and might result in different pathogenesis and clinical manifestations, the BAV phenotype has been an interesting topic for many investigators. There are multiple classifications of the pathological types of bicuspid aortic valve malformations, and the most common classification is based on the presence or absence of fused spine formation, leaflet fusion type, and leaflet spatial location. This study classifies the presence or absence of fused spine formation and type of leaflet fusion.
The most common complication of the BAV in adults is AV dysfunction necessitating surgical aortic repair or AV replacement (AVR) (population-based 25-year risk of AVR is up to 53%), 27 and is most commonly driven by presence of severe AS followed by severe aortic regurgitation (AR) and mixed AV disease. As a cause of AVR, AS has been reported between 61% and 88% in population-based studies and studies from tertiary-referral centers, conversely, AR is only responsible for 15 and 29% AV surgery. 28  Because a raphe is commonly seen in patients with BAV, the clinical significance of raphe is of interest. A global registry showed that the presence of a visible raphe is associated with significant (moderate or greater) AS and AR and higher future incidence of AVR. 25 Furthermore, BAV patients with raphe had higher rates of AVR compared with patients without raphe. 25 It is reported that the raphe of a BAV and a higher tendency for calcium deposition are important causes of significant valve dysfunction. 29 Therefore, patients with BAV with raphe tend to develop significant valvular dysfunction at a younger age. Our meta-analysis found that aortic regurgitation was more frequently observed among patients with raphe.
In addition, evaluation of inter-ethnic differences in valve morphology and function in patients with BAV is important for the worldwide spread of transcatheter aortic valve replacement (TAVR). Kong et al. 30 reported that there is significant heterogeneity in BAV across European and Asian population, type 0 (without raphe) is more frequently observed in Europeans and fusion raphe between the right and the noncoronary cusps is more frequently observed in Asians.
The European group had higher incidence of significant aortic regurgitation than the Asian group (44.2% vs. 26.8%, respectively; p < .001).
There was no difference in the grades of aortic stenosis between these two populations.
BAV patients show obvious heterogeneity in many different clinical aspects, including the BAV phenotype and the severity of valve The present meta-analysis has some limitations. First, due to the limited number of studies included, the heterogeneity of the studies about the relationship between the BAV and aortic regurgitation is greater, and the source of heterogeneity is not further analyzed. However, we have at least partially reduced the effect of observed heterogeneity on the overall effect size by selecting a random effects model analysis. Second, there is inevitably a risk shift in the study of any population, especially the confounder in retrospective studies. These differences may be responsible for the heterogeneity between studies. Third, a systematic approach to the detailed classification of BAV should be routinely applied in clinical practice to provide new insights into this common disease entity in the future.
Our results confirmed a relationship between different BAV phenotypes and aortic valve dysfunction. BAV-RL and BAV with raphe are more likely to develop aortic regurgitation, while patients with BAV-RN present a higher possibility to develop aortic stenosis.