Abstract
Objective
Pulmonary-to-aortic diameter ratio (P/A) correlates with pulmonary hypertension (PH), which is linked to adverse outcomes. Our aim was to evaluate the impact of the P/A ratio, measured using computed tomography (CT), on adverse postoperative outcomes in patients undergoing transcatheter aortic valve implantation (TAVI).
Methods
This study included 387 patients with aortic valve disease who underwent TAVI between January 2018 and June 2024. The clinical endpoint was the composite outcome of all-cause mortality and postoperative hospitalization for heart failure. The cut-off value for the P/A ratio was obtained by analyzing the receiver operating characteristic (ROC) curve. Cox proportional hazards models were used to examine the association between the P/A ratio and the composite outcome, with subsequent subgroup and sensitivity analyses.
Results
The median follow-up time was 19 months [interquartile range (IQR): 1.00–59.0], and a composite outcome occurred in 14% of patients in this study. A P/A ratio of 0.84 was identified as the optimal cut-off value. Compared with the low P/A group, the high P/A group had a higher proportion of females and patients with diabetes, higher triglyceride levels, and higher systolic pulmonary artery pressure (sPAP) levels. Multivariable Cox analysis confirmed that a high P/A ratio [hazard ratio (HR): 1.96, 95% confidence interval (CI): 1.03–3.72; p = 0.04] was associated with the composite outcome. Subgroup and sensitivity analyses yielded consistent results.
Conclusions
The P/A ratio measured using CT may serve as a novel prognostic factor in patients undergoing TAVI. A high P/A ratio (≥ 0.84) is an independent risk factor for the composite outcome.
Keywords
Introduction
Transcatheter aortic valve implantation (TAVI), a cardiovascular interventional technique, has enabled remarkable clinical results for the treatment of aortic stenosis in recent years, 1 bringing new hope to many elderly patients, those at high risk, or contraindicated for surgical operations.2,3 With the continuous promotion and application of TAVI technology, the indications for TAVI have rapidly expanded to include individuals at intermediate and low risk as well as patients with aortic regurgitation.4–7 Multiple studies have revealed that the pulmonary artery-to-aortic (P/A) ratio, an imaging assessment indicator, is closely associated with pathological conditions such as right ventricular dysfunction and pulmonary hypertension (PH)8,9; these conditions are common in patients with aortic valve disease.10,11 However, the specific clinical value of the computed tomography (CT)-derived P/A ratio in patients undergoing TAVI, particularly its correlation with clinical outcomes, remains unclear. Therefore, this study aimed to explore the impact of the CT-derived P/A ratio on clinical outcomes in patients undergoing TAVI.
Materials & methods
Study population
This retrospective cohort study included patients with aortic valve disease who underwent TAVI at the First Affiliated Hospital of University of Science and Technology of China (USTC) between January 2018 and June 2024. This study was conducted in accordance with the Helsinki Declaration of 1975 as revised in 2024 and was approved by the Ethics Committee Review Board of the First Affiliated Hospital of USTC (No.22025-RE-252). Informed consent was waived as the study used fully anonymized, retrospectively collected medical records. The reporting of this study conforms to STROBE guidelines. 12 All patients were evaluated by a cardiac team and considered suitable for TAVI. A total of 36 patients who lacked preoperative coronary computed tomography angiography (CCTA) or were lost to follow-up were excluded. Finally, 387 patients were included in the analysis.
P/A ratio measurement
The most commonly used method for measuring the P/A ratio is axial CT imaging, 13 which was obtained by a multidetector CT scanner (GE Healthcare, Gemstone Spectral Imaging, DiscoveryHD750). Two experienced doctors, who were blinded to the patients’ clinical characteristics, simultaneously measured the main pulmonary artery and ascending aorta diameters on the enhanced images at the level of the pulmonary artery bifurcation. The P/A ratio was calculated by dividing the diameter of the main pulmonary artery by that of the ascending aorta.
Clinical variables and outcome definition
Demographic and clinical characteristics, as well as postoperative outcomes, were retrospectively collected from patient medical records. Hypertension was defined based on any of the following criteria: mean systolic blood pressure ≥140 mmHg, mean diastolic blood pressure ≥ 90 mmHg, and current use of antihypertensive drugs. Patients with diabetes mellitus were identified by any of the following conditions: random blood glucose ≥ 11.1 mmol/L, HbA1c ≥ 6.5%, and treatment with insulin or oral hypoglycemic drugs. Diagnoses of chronic obstructive pulmonary disease (COPD), cerebrovascular disease, peripheral vascular disease, and preoperative arrhythmia were obtained based on the preoperative examination results or the patient's medical history. Systolic pulmonary artery pressure (sPAP) was estimated from tricuspid regurgitation velocity on echocardiography. A history of cardiac surgery was defined as any open or interventional cardiac surgery that the patient had received.
The clinical end point of this study was the composite outcome of all-cause mortality and hospitalization for heart failure after TAVI. The time to reach the clinical endpoint was calculated as the time of death or the first hospitalization for heart failure after TAVI.
Processing of missing values
In the dataset, the following variables had missing values: height (n = 3), diabetes mellitus (n = 10), hypertension (n = 10), COPD (n = 10), peripheral vascular disease (n = 10), cerebrovascular disease (n = 10), arrhythmia (n = 10), history of cardiac surgery (n = 10), white blood cell (WBC; n = 14), hemoglobin (n = 14), platelet (n = 14), alanine aminotransferase (n = 9), aspartate aminotransferase (n = 9), total bilirubin (n = 9), direct bilirubin (n = 10), albumin (n = 9), blood urea nitrogen (BUN; n = 9), serum creatinine (SCr; n = 9), uric acid (n = 10), total cholesterol (n = 44), triglyceride (n = 44), and fasting blood glucose (n = 10). Missing values were imputed using the random forest method in the R software package, RandomForest. 14
Statistical analysis
Continuous variables were presented as mean ± SD for data with normal distributions and median (interquartile range, IQR) for data with non-normal distributions, whereas categorical data were presented as percentages. Group differences were analyzed using the unpaired student's t-test, Mann–Whitney U, or χ2 test, as appropriate.
Receiver operating characteristic (ROC) curve analysis was performed to identify the optimal cut-off value for the P/A ratio. The cut-off value obtained at the maximum Youden index was used to categorize the P/A ratio as a binary variable for subsequent analyses.
Univariable and multivariable Cox proportional hazards models were utilized to explore the association between the P/A ratio and the composite outcome, and hazard ratios (HRs) were expressed with 95% confidence intervals (95% CIs). In the unadjusted model, no covariates were adjusted; gender and age were adjusted for in model one; gender, age, body mass index (BMI), left ventricular ejection fraction (LVEF), diabetes mellitus, COPD and History of cardiac surgery were adjusted for in model two. Kaplan–Meier survival curves were plotted and compared between the groups using the log-rank test.
Subgroup analyses and sensitivity analysis
Subgroup and interaction analyses were conducted based on gender, age (< 70 years old or ≥ 70 years old), BMI (< 23 or ≥ 23 kg/m2), and the presence of hypertension using the adjusted Cox proportional hazards model.
Sensitivity analysis was performed to verify the robustness of the results. Using the dataset before imputation and excluding data with covariates missing values, a Cox proportional hazards model was used to explore the association between the P/A ratio and composite outcome.
In all analyses, p < 0.05 (two-sided) was considered statistically significant. All statistical analyses were performed using R software (version 4.3.1).
Results
Study population and baseline characteristics
Among the 387 patients included in this study (Table 1), the median follow-up time was 19 (IQR: 1.00–59.0) months. The composite outcome of death and heart failure was observed in 54 patients (14%). The mean age of the patients was 73.2 ± 6.88 years old, among which 61.5% were male. The mean P/A ratio was 0.702, and the average length of the postoperative hospital stay was 8.42 days. Compared to patients without a composite outcome, patients with a composite outcome had a larger pulmonary artery diameter on preoperative CT (2.97 ± 0.506 vs. 3.14 ± 0.509 cm; p = 0.029). However, no significant difference in sPAP was observed between the two groups.
Baseline characteristics of patients with or without composite outcomes.
Abbreviations: BMI, body mass index; COPD, chronic obstructive pulmonary disease; BUN, blood urea nitrogen; SCr, serum creatine; eGFR, estimated glomerular filtration rate; WBC, white blood cell count; LVEF, left ventricular ejection fraction; sPAP, systolic pulmonary artery pressure; NTproBNP, N-terminal pro-B-type natriuretic peptide.
ROC curve analysis showed that the P/A ratio at the maximum Youden index was 0.84. The patients were divided into low and high P/A groups with a cut-off value of 0.84. Table 2 details the baseline characteristics between the two groups. Compared with the low P/A group, the high P/A group had a higher proportion of females, a higher proportion of patients with diabetes mellitus, higher triglyceride levels, and higher sPAP levels. There were no significant differences in other baseline data between the two groups.
Baseline characteristics of patients in low and high P/A ratio groups.
Abbreviations: BMI, body mass index; COPD, chronic obstructive pulmonary disease; BUN, blood urea nitrogen; SCr, serum creatine; eGFR, estimated glomerular filtration rate; WBC, white blood cell count; LVEF, left ventricular ejection fraction; sPAP, systolic pulmonary artery pressure; NTproBNP, N-terminal pro-B-type natriuretic peptide.
Association between the P/A ratio and composite outcome
Three Cox proportional hazards models were used to investigate the association between the P/A ratio and composite outcome. In the unadjusted model, the HR for patients in the high P/A group was 2.01 (95% CI: 1.10–3.70; p = 0.02) for the composite outcome, as compared with the low P/A group (Table 3). In model one, the multivariable-adjusted HR was 2.10 (95% CI: 1.13–3.89; p = 0.01). In the fully-adjusted model, the multivariable-adjusted HR was 1.96 (95% CI: 1.03–3.72; p = 0.04). Additionally, analysis of the association between the P/A ratio and all-cause mortality and postoperative heart failure rehospitalization revealed that the P/A ratio was significantly associated with heart failure rehospitalization (fully-adjusted HR 2.15, 95% CI: 1.06–4.37; p = 0.03), but not with all-cause mortality (fully-adjusted HR 1.35, 95% CI: 0.27–6.61; p = 0.71).
Association between the P/A ratio and composite outcome.
Model 1: adjusted for gender, age; Model 2: further adjusted for BMI, LVEF, diabetes mellitus, COPD, history of cardiac surgery based on Model 1.
Abbreviations: BMI, body mass index; LVEF, Left ventricular ejection fraction.
Survival curves are shown in Figure 1. The patients in the high P/A group had significantly lower composite-outcome survival rates than those in the low P/A group (log-rank, p = 0.022).
Kaplan–Meier curves for the composite outcome stratified by P/A ratio.
Subgroup analyses and sensitivity analysis
The relationship between the P/A ratio and composite outcome was further explored in different subgroups. No significant interactions were found in any subgroup, including age, gender, BMI, and presence of hypertension (Table 4).
Subgroup analyses stratified by gender, age, BMI and hypertension.
Adjusted for gender, age, BMI, LVEF, diabetes mellitus, COPD, history of cardiac surgery except the stratification factor itself.
After excluding covariates missing values, 374 patients were included in the sensitivity analysis (Supplementary Table 1). In the unadjusted model, the HR of the high P/A group was 2.15 (95% CI: 1.16–3.97; p = 0.01). The adjusted HRs of the high P/A ratio group in model one and model two were 2.23 (95% CI: 1.19–4.17; p = 0.01) and 2.06 (95% CI: 1.08–3.95; p = 0.03), respectively. These findings are overall consistent with findings from the main analysis.
Discussion
In this single-center retrospective study, we explored the relationship between the P/A ratio and the composite outcome of postoperative all-cause mortality and hospitalization for heart failure in patients who underwent TAVI. After adjusting for multiple covariates, we found that a high P/A ratio was associated with a worse prognosis in patients who underwent TAVI. This association showed no interaction in any subgroup, including gender, age, BMI and presence of hypertension, consistent with results of the sensitivity analysis. To the best of our knowledge, this study is the first to explore the association between the P/A ratio and the composite outcome of postoperative all-cause mortality and hospitalization for heart failure after TAVI.
The P/A ratio is primarily used in research on PH and pulmonary diseases. The guidelines of the European Society of Cardiology and European Respiratory Society indicate that the P/A ratio may provide important information for patients with suspected or confirmed PH. 15 Through the study of patients with idiopathic pulmonary fibrosis, Mitsuaki et al. found that the P/A ratio is a useful and convenient method for predicting elevated mean pulmonary artery pressure, and a high P/A ratio (> 0.9) was associated with a poor prognosis in this patient population. 16 In a study of 118 patients with acute exacerbations of COPD, Cheng et al. found that patients with a P/A ratio ≥ 1 had a worse prognosis during hospitalization. 17 An analysis of the Rotterdam Study revealed that a larger P/A ratio was not associated with mortality in the general elderly population but was an independent determinant of mortality in moderate to severe COPD. 18 Moll et al. found that the P/A ratio was the primary imaging predictor of all-cause mortality in COPD. 19 Additionally, the P/A ratio is a useful parameter for assessing the prognosis of acute pulmonary embolism. 20
Compared with PH and pulmonary diseases, there are few studies on the correlation between the P/A ratio and heart disease. In a retrospective observational study, Ferrufino et al. found that the P/A ratio could predict right ventricular failure and 1-year mortality after LVAD implantation. 21 Ieki et al. analyzed 447 patients with acute decompensated heart failure and found that the P/A ratio is a reliable alternative indicator for estimating PH in patients with heart failure, as well as a reliable biomarker for prognosis prediction in these patients. 22 Studies have shown that PH is associated with the risk of progressive death in patients with aortic insufficiency/stenosis and poor prognosis after TAVI surgery.23,24 In a study of 579 patients who continuously received TAVI treatment and underwent preoperative right heart catheterization, the presence of PH could identify patients with a higher risk of postoperative death and rehospitalization. 25 A meta-analysis of 22 studies showed that PH was associated with an increase in long-term cardiac and all-cause mortality. 26 However, the association between the P/A ratio and the outcomes of all-cause mortality as well as hospitalization for heart failure following TAVI was previously unexplored. Our study addressed this critical gap in the literature and revealed that a higher P/A ratio was associated with the composite outcome of all-cause mortality and heart failure hospitalization in patients who underwent TAVI. After adjusting for other covariates, a higher PA/A ratio remained an independent risk factor of the composite outcome in patients who underwent TAVI.
Although the P/A ratio measured by CT or MRI is positively correlated with pulmonary artery pressure,27,28 the cut-off value of the P/A ratio used to estimate PH is not exactly the same in existing studies. We performed ROC curve analysis with a cutoff value (0.84) based on the maximum Youden index and divided the patients into two groups based on this value. This cut-off value is consistent with the cut-off value of the P/A ratio used by Chen et al. to predict PH. 29 The presence of PH was estimated using the P/A ratio. On the one hand, it is not disturbed by the surface area. 30 On the other hand, as the P/A ratio is easy to measure on CT images, it may ultimately have more prognostic value worthy of concern. In our study, the pulmonary artery diameter of patients with a composite outcome was significantly larger than that of patients without a composite outcome, with no significant differences in other baseline characteristics. The proportion of females in the high P/A ratio group was higher than that of males, and more patients had diabetes in this group. Several studies have shown that the prevalence of PH is higher in women than in men.31,32 This difference may be related to the changes in hormone levels in female patients.33–35 There is a direct relationship between diabetes and pulmonary pressure, as well as right ventricular function. 36 A study involving the Jackson Heart Cohort and another study on American veterans found that individuals with diabetes had an increased risk of PH.37,38 Diabetes mellitus and adverse pulmonary vascular remodeling share many common pathogenic mechanisms, such as chronic inflammation, endothelial cell dysfunction, impaired lipid processing, and mitochondrial metabolic disorders,39,40 which may be plausible explanations for their association. Further subgroup analysis showed that a high P/A ratio was more strongly associated with the composite outcome of postoperative death and heart failure in TAVI patients who were female, older than 70 years of age, had a low BMI, and did not have hypertension. However, no interaction was found between these four factors and the P/A ratio.
This study has some limitations. First, the data were collected from a single center. The generalizability of these results to other regions and populations remains unclear. Multicenter studies involving people of different races are required to further verify these findings. Second, this retrospective study could not determine the causal relationship between the P/A ratio and the composite outcome of postoperative all-cause mortality and hospitalization for heart failure in patients undergoing TAVI. Prospective studies are needed to better control for confounding factors. Third, although many studies have found a positive correlation between the P/A ratio and pulmonary artery pressure, we did not collect hemodynamic data from the right heart catheterization procedure to evaluate the direct correlation between pulmonary artery pressure and the P/A ratio. Fourth, the P/A ratio is a radiological measurement derived from CT and serves as a prognostic marker rather than a direct assessment of PH; it should not replace echocardiographic or invasive hemodynamic evaluation for diagnosing PH. Furthermore, in this study, a high P/A ratio was defined as ≥ 0.84, which has been used in a past study; however, whether the different definitions in previous studies (such as > 0.9 or ≥ 1) are applicable to the conclusions of this study remains to be further investigated.
Conclusion
Our research indicates that the P/A ratio measured using CT may serve as a novel prognostic factor for patients undergoing TAVI. A high P/A ratio (≥ 0.84) is an important risk factor for the composite outcome of all-cause mortality and hospitalization for heart failure. However, these results need to be verified in further prospective studies.
Supplemental Material
sj-docx-1-sci-10.1177_00368504261431852 - Supplemental material for Impact of pulmonary-artery-to-aorta ratio on clinical outcomes in patients undergoing transcatheter aortic valve implantation: A retrospective study
Supplemental material, sj-docx-1-sci-10.1177_00368504261431852 for Impact of pulmonary-artery-to-aorta ratio on clinical outcomes in patients undergoing transcatheter aortic valve implantation: A retrospective study by Yang Wang, Ruhua Shen and Yuyong Liu in Science Progress
Footnotes
Ethical considerations
This study was approved by the Ethics Committee Review Board of the First Affiliated Hospital of USTC (No.22025-RE-252). Informed consent was waived as the study used fully anonymized, retrospectively collected medical records.
Author contributions
Yang Wang: Writing – original draft, Methodology, Data curation, Investigation, Software. Ruhua Shen: Writing – review & editing, Writing – original draft, Methodology. Yuyong Liu: Writing – review & editing, Validation, Supervision, Conceptualization.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data availability
The datasets examined in this study can be obtained from the corresponding author upon a reasonable request.
Supplemental material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
