Impact of Valve Morphology on the Prevalence of Coronary Artery Disease: A Systematic Review and Meta‐Analysis

Abstract

Background

Literature studies suggested a lower prevalence of coronary artery disease (CAD) in bicuspid aortic valve (BAV) than in tricuspid aortic valve (TAV) patients. However, this finding has been challenged. We performed a meta‐analysis to assess whether aortic valve morphology has a different association with CAD, concomitant coronary artery bypass grafting (CABG), and postoperative mortality.

Methods and Results

Detailed search was conducted according to the PRISMA (Preferred Reporting Items for Systematic reviews and Meta‐Analyses) guideline to identify all patients with BAV or TAV and presence of CAD, concomitant myocardial surgical revascularization, and the postoperative mortality. Thirty‐one studies on 3017 BAV and 4586 TAV patients undergoing aortic valve surgery were included. BAV patients showed a lower prevalence of CAD (odds ratio [OR]: 0.33; 95% CI: 0.17, 0.65), concomitant CABG (OR, 0.45; 95% CI: 0.35, 0.59), and postoperative mortality (OR, 0.62; 95% CI: 0.40, 0.97) than TAV. However, BAV subjects were significantly younger than TAV (mean difference: −7.29; 95% CI: −11.17, −3.41) were more frequently males (OR, 1.61; 95% CI: 1.33, 1.94) and exhibited a lower prevalence of hypertension (OR, 0.58; 95% CI: 0.39, 0.87) and diabetes (OR, 0.71; 95% CI: 0.54, 0.93). Interestingly, a metaregression analysis showed that younger age and lower prevalence of diabetes were associated with lower prevalence of CAD (Z value: −3.03; P=0.002 and Z value: −3.10; P=0.002, respectively) and CABG (Z value: −2.69; P=0.007 and Z value: −3.36; P=0.001, respectively) documented in BAV patients.

Conclusions

Analysis of raw data suggested an association of aortic valve morphology with prevalence of CAD, concomitant CABG, and postoperative mortality. Interestingly, the differences in age and diabetes have a profound impact on prevalence of CAD between BAV and TAV. In conclusion, our meta‐analysis suggests that the presence of CAD is independent of aortic valve morphology.

Introduction

Aortic valve stenosis (AVS) is considered the most prevalent form of valve disease.1, 2 The number of people affected by this progressive and debilitating pathology will increase because of the aging population, causing an ever‐increasing public health burden.3 Traditionally, it was thought that AVS was related to valve degeneration attributed to aging, caused by several years of mechanical stress and biological response to such injury. However, more recently, several risks factors have been linked to the development of AVS, including male sex, hypertension, hyperlipidemia, smoking, advanced age, and congenital bicuspid valve morphology.4, 5 Overall, AVS pathogenesis is a multifactorial process and seems to be related to coronary atherosclerosis. In detail, an important link between AVS and early stages of coronary atherosclerotic plaque has been hypothesized,6, 7, 8 and a multitude of studies have been published regarding the associations between AVS, coronary artery disease (CAD), and coronary artery bypass grafting (CABG).9, 10, 11, 12 In addition, a very recent study by Boudoulas et al.13 describes the association between aortic stenosis and CABG focusing on valve morphology and concluding that “the incidence of coronary artery disease is extremely high in patients with aortic stenosis and tricuspid aortic valve.” However, literature data are not consistent about this issue, and data about such an association have been challenged. In addition, the critical review of available studies highlights that, in most cases, only univariate unadjusted analyses were used, seldom taking into account for potential confounders. Thus, the aim of this study was to perform a meta‐analysis of literature studies enrolling patients undergoing aortic valve surgery to assess whether aortic valve morphology (tri‐ or bicuspid) impacts on prevalence of CAD, concomitant CABG, and postoperative mortality. In addition, to assess the presence of potential confounders, we evaluated the impact of distribution of major clinical and demographic variables between patients with bi‐ or tricuspid aortic valve.

Methods

A protocol for this review was prospectively developed, detailing the specific objectives, criteria for study selection, outcomes, and statistical methods.

Search Strategy

To identify all available studies pertaining to prevalence of CAD, defined by atherosclerosis of 1 or more arteries that supply blood to the heart causing oxygen deficiency in the myocardium, we included articles with anamnestic CAD also when it was not specified whether it was anatomical CAD (70% stenosis or greater in at least 1 major coronary artery) or clinical CAD (previous acute myocardial infarction or percutaneous coronary intervention). A detailed search was conducted according to PRISMA (Preferred Reporting Items for Systematic reviews and Meta‐Analyses) guidelines14 to identify all concomitant myocardial surgical revascularization (CABG) and postoperative mortality in patients with bicuspid aortic valve (BAV) and tricuspid aortic valve (TAV) undergoing aortic valve surgery. A systematic search was performed in the electronic databases (PubMed, Web of Science, Scopus, and EMBASE), using the following search terms in all possible combinations: tricuspid aortic valve; bicuspid aortic valve; coronary artery disease; coronary artery bypass; myocardial infarction; mortality; and death. The last search was performed in November 2015. The search strategy was developed without any language or publication year restriction. In addition, the reference lists of all retrieved articles were manually reviewed. In case of missing data, the authors were contacted by e‐mail to try to retrieve original data. Two independent authors (P.P. and M.N.D.D.M.) analyzed each article and performed the data extraction independently. In case of disagreement, a third investigator was consulted (L.C.). Discrepancies were resolved by consensus. Selection results have been reported according to the PRISMA flow chart (Figure 1).

Figure 1.

PRISMA flow diagram. BAV indicates bicuspid aortic valve; CT, computed tomography.

Data Extraction and Quality Assessment

According to the prespecified protocol, all studies evaluating prevalence of CAD or use of CABG or postoperative mortality in patients with BAV and in those with TAV undergoing aortic valve replacement were included in the analysis. Case reports, reviews, animal studies, and studies on natural history of patients with aortic valve disease (not undergoing surgery) were excluded. In each study, data regarding sample size, major clinical and demographic variables, number of patients with CAD, and number of those undergoing CABG and those dying postoperatively were extracted. Formal quality score adjudication was not used, because previous investigations failed to demonstrate its usefulness.15

Statistical Analysis and Risk of Bias Assessment

Statistical analysis was carried out using Comprehensive Meta‐analysis (version 2 [2005]; Biostat, Englewood, NJ). Differences among BAV and TAV subjects were expressed as mean difference (MD) with pertinent 95% CIs for continuous variables, and as odds ratio (OR) with pertinent 95% CI for dichotomous variables. Overall effect was tested using Z scores, and significance was set at P<0.05. Statistical heterogeneity among studies was assessed with the chi‐square Cochran's Q test and with the I² statistic, which measures the inconsistency across study results and describes the proportion of total variation in study estimates, that is, attributed to heterogeneity rather than sampling error. In detail, I² values of 0% indicate no heterogeneity, 25% low, 25% to 50% moderate, and 50% high heterogeneity.16

Publication bias was assessed by the Egger test and represented graphically by funnel plots of the standard difference in means versus SE. Visual inspection of funnel plot asymmetry was performed to address for possible small‐study effect, as well as the Egger test to address publication bias, over and above any subjective evaluation. P<0.10 was considered statistically significant.17 In order to be as conservative as possible, the random‐effect method was used to take into account the variability among included studies.

Metaregression Analyses

We hypothesized that differences in prevalence of CAD, use of CABG, or postoperative mortality between BAV and TAV patients may be affected by differences in clinical and demographic characteristics of patients included in different studies (mean age, sex, body mass index [BMI], hypertension, diabetes mellitus, hyperlipidemia, and smoking habit). To assess the possible effect of such variables in explaining the different results observed across studies, we planned to perform metaregression analyses after implementing a regression model with prevalence of CAD, use of CABG, or postoperative mortality between BAV and TAV patients as dependent variables (y) and the variables mentioned above as independent variables (x).

Results

After excluding duplicate results, the search retrieved 282 articles. Of these studies, 232 were excluded because they were off the topic after scanning the title and/or the abstract and 19 because they were reviews/comments/case reports or they lacked data of interest.

Thus, 31 articles (3017 BAV patients and 4586 TAV patients) were included in the final analysis (Figure 1). In detail, 15 studies with data on prevalence of CAD (1163 BAV and 2234 TAV patients), 16 reporting on use of CABG (1782 BAV and 1886 TAV patients), and 16 on postoperative mortality rate (1067 BAV and 2399 TAV patients) were included.

Study Characteristics

Major characteristics of the 31 studies included in the meta‐analysis are shown in Table 1.

Table 1.

Demographic and Clinical Data of BAV and TAV

Author (Year)	Reported Outcomes	Observation Time		Subjects	Age, y	Males	Hypertension	Hyperlipidemia	Diabetes	BMI	Smoking
Abdulkareem 201318	CABG		BAV	192	58	71.3
			TAV	203	65	62.5
Ali 201019	CABG, mortality	7 years	BAV	90	63	78.9	31.1	12.2	5.6
			TAV	125	70	58.4	32.0	16.8	11.2
Asano 201220	CABG, mortality	5 years	BAV	86	46.3
			TAV	58	70
Badiu 201021	CAD, CABG, mortality	5 years	BAV	11	37	100.0	45.5	9.1
			TAV	91	48	63.7	69.2	36.3	4.4
Boudoulas 201513	CABG		BAV	95	62	71.6	70	61	31		53
			TAV	175	71	55.4	87	60	47		48
Branchetti 201422	CAD		BAV	74	55.5	64.9	31.1	27.0	5.4		43.2
			TAV	61	64.4	70.5	42.6	42.6	13.1		19.7
Costopoulos 201423	Mortality	1 year	BAV	21	76.7	57.1	66.7		28.6	26.6
			TAV	447	79.8	47.4	77.2		30.2	26.1
Davies 199624	CABG		BAV	296
			TAV	125
Delius 199825	Mortality	10 years	BAV	16
			TAV	31
Eleid 201326	CAD, CABG		BAV	47	58	76.6	68.1	48.9	8.5		23.4
			TAV	53	66	75.5	86.8	49.1	11.3		13.2
Etz 201527	CAD, mortality	Inhospital	BAV	32	46.7	71.9	46.9				15.6
			TAV	347	61.6	63.7	72.0		9.2		10.7
Girdauskas 201428	Mortality	10 years	BAV	153	54	73.2	48.4		11.1		35.9
			TAV	172	64	47.7	57.0		15.7		40.1
Holubec 201429	CAD, CABG, mortality	2 years	BAV	60	45a	81.7	45.0
			TAV	40	59a	67.5	75.0
Hwang 201130	CAD, mortality	10 years	BAV	45	59.6	60.0	44.4	2.2	6.7		26.7
			TAV	43	58.3	48.8	23.3	2.3	9.3		9.3
Jackson 20149	CAD		BAV	292	61.1	73.6	51.0		11.3
			TAV	355	717	69.9	74.6		14.9
Kochman 201431	CAD, mortality	1 year	BAV	28	77.6	46.4	60.7		39.3
			TAV	84	79.1	47.6	65.5		34.5
Kvitting 201332	CABG		BAV	63	43	79.4	36.5		4.8	26
			TAV	170	36	67.6	25.9		2.4	24
Liu 201533	CAD, mortality	30 days	BAV	15	75.4	60.0	33.3			23.6
			TAV	25	75.8	68.0	56.0		12.0	21.7
Mosalanezhad 201434	CABG, mortality	8 years	BAV	30	42	93.3
			TAV	20	59	75.0
Nakamura 201435	CAD		BAV	17	70	76.5	58.8	35.3	17.6		35.3
			TAV	59	77	52.5	79.7	33.9	20.3		40.7
Philip 201510	CAD		BAV	200	57		23.5	21.5	10.5
			TAV	200	78		76.5	78.5	31.5
Roberts 200311	CABG		BAV	232	64.7	72.0
			TAV	267	74	51.3
Roberts 2007a36	CABG, mortality	5 years	BAV	102
			TAV	18
Roberts 2007b37	CABG, mortality	4 years	BAV	187
			TAV	235
Roberts 2007c38	CABG		BAV	54
			TAV	142
Roberts 2007d39	CABG, mortality	13 years	BAV	180
			TAV	107
Rylski 201412	CAD, mortality	Inhospital	BAV	41	55a	63.4	56.1		9.8
			TAV	588	61a	64.1	81.1		9.2
Shim 201140	CAD		BAV	50	52	78.0	40.0	20.0	12.0	24.7	32.0
			TAV	50	52	78.0	50.0	28.0	8.0	25.2	42.0
Stephan 199741	CABG		BAV	57	67	57.9
			TAV	57	73	54.4
Warner 201342	CAD		BAV	10	46.5	60.0	50.0	40.0	10.0	30.1	10.0
			TAV	13	46.3	76.9	38.5	46.2			23.1
Yuan 201043	CAD		BAV	241	56.1	77.2	30.7	24.1	10.8
			TAV	225	62.8	64.0	41.8	24.9

^aData reported as median value. BAV indicates bicuspid aortic valve; BMI, body mass index; CABG, coronary artery bypass grafting; CAD, coronary artery disease; TAV, tricuspid aortic valve.

The number of patients varied from 23 to 647, mean age from 36 to 79.8 years, and prevalence of male sex from 47.4% to 100%. Presence of hypertension was reported by 23.3% to 87.0% of patients, smoking habit by 9.3% to 53.0%, diabetes mellitus by 2.4% to 47.0%, and hyperlipidemia by 2.2% to 78.5%. Mean BMI varied from 21.7 to 30.1 kg/m². Length of follow‐up for mortality assessment ranged from in‐hospital stay period to 13 years with a median of 5 years.

Coronary Artery Disease

Fifteen studies,¹ for a total of 1163 BAV and 2234 TAV patients, showed that CAD was reported by 13.5% of BAV and 30.3% of TAV patients (OR, 0.33; 95% CI: 0.17, 0.65; P=0.001; Figure 2). Heterogeneity among studies was significant (I²=86%; P<0.001), and it was not reduced by the exclusion of one study at a time. In addition, after excluding 2 studies by Kochman et al.31 and Liu et al.,33 including patients undergoing transcatheter aortic valve implantation (TAVI), the results were entirely confirmed (OR, 0.31; 95% CI: 0.14, 0.66; P=0.002; I²=88%; P<0.001).

Figure 2.

Prevalence of coronary artery disease in patients with bicuspid (BAV) and tricuspid aortic valve (TAV).

Concomitant CABG

Sixteen studies,² reporting on 1782 BAV and 1886 TAV patients, showed that the number of patients undergoing concomitant CABG was significantly lower between BAV than TAV patients (31.0% vs 45.7%; OR, 0.45; 95% CI: 0.35, 0.59; P<0.001; Figure 3). Heterogeneity among studies was significant (I²=56%; P=0.003). However, after excluding the study by Asano et al.,20 all results were confirmed without heterogeneity (OR, 0.49; 95% CI: 0.39, 0.61; P<0.001; I²=37%; P=0.07).

Figure 3.

Prevalence of coronary artery bypass grafting (CABG) in patients with bicuspid (BAV) and tricuspid aortic valve (TAV).

Postoperative Mortality

The 16 studies³ evaluating postoperative mortality showed a slightly significant difference in rate of mortality between BAV and TAV patients (16.2% vs 18.8%; OR, 0.62; 95% CI: 0.40, 0.97; P=0.04; I²=65%; P<0.001; Figure 4). Interestingly, a metaregression analysis showed that the mortality rate was independent by the length of the observation (Z value: −1.17; P=0.240). In addition, we conducted a subanalysis removing the 3 studies23, 31, 33 that analyzed patients undergoing TAVI; the results did not differ from the previous ones (OR, 0.51; 95% CI: 0.32, 0.84; P=0.006; I²=63%; P=0.001).

Figure 4.

Post‐operative mortality in patients with bicuspid (BAV) and tricuspid aortic valve (TAV).

Clinical and Demographic Variables

As showed in Figure S1, when major clinical and demographic characteristics have been compared between BAV and TAV patients, we found that BAV subjects were significantly younger than TAV (MD, −7.29; 95% CI: −11.17, −3.41; P<0.001). In addition, BAV subjects were more frequently males (72.7% vs 60.1%; OR, 1.61; 95% CI: 1.33, 1.94; P<0.001) and exhibited a lower prevalence of hypertension (43.0% vs 65.4%; OR, 0.58; 95% CI: 0.39, 0.87; P<0.001) and diabetes (11.6% vs 16.3%; OR, 0.71; 95% CI: 0.54, 0.93; P=0.01) than TAV. In contrast, no significant differences were found in prevalence of hyperlipidemia and smoking habit as well as in mean BMI between BAV and TAV patients.

Interestingly, a metaregression analysis showed that younger age and lower prevalence of diabetes were associated with lower prevalence of CAD (Z value: −3.03; P=0.002 and Z value: −3.10; P=0.002, respectively; Figure 5) and CABG (Z value: −2.69; P=0.007 and Z value: −3.36; P=0.001, respectively; Figure 6) documented in BAV patients as compared to TAV patients.

Figure 5.

Meta‐regression analysis. Effect of the difference in mean age (A) and in prevalence of diabetes (B) on prevalence of coronary artery disease in patients with bicuspid (BAV) and tricuspid aortic valve (TAV).

Figure 6.

Meta‐regression analysis. Effect of the difference in mean age (A) and in prevalence of diabetes (B) on prevalence of concomitant coronary artery bypass grafting in patients with bicuspid (BAV) and tricuspid aortic valve (TAV).

Sensitivity Analysis

Given the statistically significant difference in distribution of age, sex, hypertension, and diabetes between BAV and TAV patients, all the analyses have been repeated after including only studies enrolling patients comparable for all these variables.30, 33, 40, 41, 42 Interestingly, the difference in prevalence of CAD (OR, 0.65; 95% CI, 0.33, 1.26; P=0.20) and mortality rate (OR, 1.77; 95% CI, 0.81, 3.87; P=0.15) were no longer significant. None of the 5 studies enrolling patients comparable for age, sex, hypertension, and diabetes provided information about concomitant CABG in BAV and TAV subjects. Thus, the sensitivity analysis was not performed for this outcome. When stratifying results according to the study design (retrospective or prospective), we found that the difference in CAD, concomitant CABG, and postoperative mortality between BAV and TAV patients were consistently confirmed only by retrospective studies. Interestingly, all the differences between the 2 groups were no longer significant in prospective studies (Table 2).

Table 2.

Stratification of the Studies: Retrospective and Prospective

	No. of Studies	OR (95% CI)	Heterogeneity
CAD
Retrospective studies	8	0.37 (0.18, 0.76; P=0.007)	I2: 84%; P<0.001
Prospective studies	7	0.30 (0.07, 1.32; P=0.11)	I2: 87%; P<0.001
CABG
Retrospective studies	15	0.45 (0.35, 0.59; P<0.001)	I2: 59%; P=0.002
Prospective studies	1	0.43 (0.02, 7.90; P=0.57)	Not evaluable
Mortality
Retrospective studies	13	0.57 (0.34, 0.94; P=0.03)	I2: 72%; P<0.001
Prospective studies	3	1.08 (0.41, 2.86; P=0.88)	I2: 0%; P=0.95

CABG indicates coronary artery bypass grafting; CAD, coronary artery disease; OR, odds ratio.

Publication Bias

Because it is recognized that publication bias can affect results of meta‐analyses, we attempted to assess this potential bias using funnel plot visual analysis (Figure S2).

Visual inspection of funnel plots of effect size versus SE for studies evaluating CAD, CABG, and mortality in BAV and in TAV patients suggested a symmetric distribution of studies around the effect size, and the Egger test confirmed the lack of publication bias for all these outcomes (CAD, P=0.868; CABG, P=0.914; mortality, P=0.403).

Discussion

This meta‐analysis, which includes more than 7500 patients undergoing aortic valve surgery, shows, in agreement with previously published data, a lower prevalence of CAD (13.5% vs 30.3%) and concomitant CABG (31% vs 45.7%) in BAV than in TAV patients, accompanied by a marginally lower rate of postoperative mortality (16.2% vs 18.8%). We also show that BAV patients enrolled in the included studies were ≈7 years younger, more often males (72.7% vs 60.1%), and exhibited lower prevalence of hypertension (43.0% vs 65.4%) and diabetes (11.6% vs 16.3%) compared to TAV patients (Table S1).

Altogether, these data suggest that the lower cardiovascular risk reported in BAV patients may be partly explained by younger age and lower prevalence of some cardiovascular risk factors in this clinical setting. Interestingly, a meta‐regression analysis confirmed and extended this hypothesis, showing that age and diabetes have a profound impact on the difference in prevalence of CAD and CABG between BAV and TAV patients. Moreover, when the analyses have been repeated after including only those studies enrolling patients matched for age, sex, hypertension, and diabetes, the difference in prevalence of CAD and postoperative mortality were no longer significant between the 2 groups of patients. Thus, patients with BAV typically develop aortic stenosis at a younger age, usually before 65 years, compared to aortic stenosis in patients with TAV, which more often develops after age 70.44

Several studies analyzed prevalence of CAD in BAV and TAV patients, concluding that this atherosclerotic disease is uncommon in BAV, but is associated with TAV disease.9, 24 However, recent developments suggest that “incidence of CAD is high in patients with aortic valve degeneration, both in those with tricuspid and bicuspid aortic valve.”13

Our analysis shows that although the incidence of all the considered variables is lower in BAV compared to TAV patients, a higher mean age and a higher prevalence of hypertension and diabetes is found in the TAV group. The potential impact of these confounding covariates should be taken into account. Indeed, our sensitivity analysis shows that once hypertension, sex, diabetes, and age are taken into account and only studies with patients comparable for these variables are included, the differences in prevalence of CAD between BAV and TAV patients is no longer significant.

The finding that, in the TAV group, higher mean age and increased prevalence of diabetes and hypertension are paralleled by a higher prevalence of CAD, concomitant CABG, and postoperative mortality somehow supports the hypothesis that the etiopathogenetic mechanism underling aortic valve degeneration is similar or complementary to coronary atherosclerosis. Indeed, recent evidence suggested that risk factors responsible for pathogenesis of atherosclerosis are also related to development of aortic calcification and stenosis.13, 45, 46 However, the reasons why many patients with CAD do not develop aortic stenosis are still under debate. The major explanation may reside in the anatomic variation in size and diameter in normal tricuspid aortic valve47 and the genetic predisposition to aortic calcification, such as lipoprotein(a) expression levels encoded by the lipoprotein(a) gene.48

Another finding of our analysis is the marginally lower postoperative mortality in BAV than in TAV patients. Evaluating the risk of post operative mortality in BAV and TAV patients, it is interesting to highlight that the only study19 providing an adjusted multivariate analysis showed that valve morphology did not impact on mortality rate. Moreover, similar findings have been confirmed by Roberts et al.37, 39 by means of an unadjusted analysis. On the other hand, it is noteworthy to stress that 4 studies20, 27, 28, 32 consistently highlighted that an increasing age was the main predictor of postoperative mortality. Overall, these evidences confirm and extend our data suggesting that the difference in mean age between BAV and TAV might, at least in part, explain the difference in postoperative mortality documented in these 2 groups. Indeed, when we have analyzed only studies on patients comparable for age, sex, hypertension, and diabetes, the differences in postoperative mortality rate between BAV and TAV patients were no longer significant. In addition, after stratifying results according to study design, we found that differences between BAV and TAV patients were confirmed only by retrospective studies. Interestingly, all the differences between the 2 groups were no longer significant in prospective studies.

We recognize that our study has several potential limitations. The studies included in our meta‐analysis have different inclusion and exclusion criteria with no clear definition of CAD. In addition, most of the patients included in the analysis had concomitant cardiovascular risk factors. Given that meta‐analysis is performed on aggregate data and some missing information is present in each study, the multivariate approach allowed for the adjustment for some (but not all) potential confounders. Thus, although results of metaregression analyses were able to refine analyses by assessing the influence of most clinical and demographic variables on the observed results, caution is necessary in overall results interpretation. Moreover, heterogeneity among the studies was generally significant. Although it was not possible to conclusively ascertain sources of heterogeneity, publication bias did not affect results of our meta‐analysis.

In conclusion, patients undergoing aortic valve replacement might represent a potential bias. Patients undergoing surgery are a subset of patients with BAV or TAV, and this could influence the results. This might limit the reproducibility of reported results. However, a large study49 reporting on natural history of BAV and TAV patients not undergoing surgery confirmed the lack of difference in cardiovascular mortality between the 2 groups. In addition, the studies included analyzed not only aortic valve stenosis pathology, but also aortic valve regurgitation. However, in the studies analyzing both aortic valve pathologies,18, 22, 26, 27, 35 there are no significative differences between BAV and TAV groups.

A further potential confounder is represented by inclusion of studies analyzing patients’ candidate for TAVI, who, besides the older age, have a greater prevalence of comorbidities. However, a subanalysis excluding these 3 studies23, 31, 33 confirmed all our findings.

In conclusion, the results of the meta‐analysis here presented shows that analysis of raw data clearly suggests an association of aortic valve morphology with prevalence of CAD, concomitant CABG, and postoperative mortality. Interestingly, differences in age and diabetes have a profound impact on prevalence of CAD and concomitant CABG between BAV and TAV patients. Thus, based on our analysis, BAV patients do not exhibit a lower risk of CAD compared to TAV patients. However, further studies or patient‐level meta‐analysis are needed to adjust results for potential confounders and definitely address this issue.

Author Contributions

Paolo Poggio and Matteo Nicola Dario Di Minno conceived and designed the study, performed statistical analysis, interpreted results, and drafted the manuscript; Laura Cavallotti, Paola Songia, Alessandro Di Minno, Pasquale Ambrosino, and Liborio Mammana acquired clinical data and drafted the manuscript; Alessandro Parolari and Elena Tremoli interpreted results and performed critical revisions. All authors read and approved the final version of the manuscript. Paolo Poggio and Matteo Nicola Dario Di Minno had full access to all data in the study and take responsibility for the integrity of the data and the accuracy of data analysis.

Sources of Funding

This work was supported by the Fondazione Gigi e Pupa Ferrari ONLUS, the Italian Ministry of Health (RC2014 BIO61 and RC2015 BIO30), and the Fondazione Umberto Veronesi.

Disclosures

None.

Supporting information

Figure S1. Distribution of major clinical and demographic characteristics in patients with bicuspid (BAV) and tricuspid aortic valve (TAV). A, Sex, (B) age, (C) hypertension, (D) diabetes, (E) hyperlipedimia, (F) BMI, and (G) smoking. BMI indicates body mass index.

Figure S2. Funnel plots of effect size vs standard error for studies evaluating the prevalence of coronary artery disease (A), coronary artery bypass grafting (B), and postoperative mortality (C) in patients with bicuspid (BAV) and tricuspid aortic valve (TAV).

Table S1. Summary Table of Meta‐Analysis Results

(J Am Heart Assoc. 2016;5:e003200 doi: 10.1161/JAHA.116.003200)