Introduction
Bronchiectasis is a chronic respiratory disease with multiple etiologies, characterized by clinical and radiological abnormalities in the lungs [1]. When severe, it can cause pulmonary hypertension, a rare but serious cardiovascular complication [2]. Chronic airflow obstruction in bronchiectasis may lead to persistent hypoxemia in the vascular bed and higher pulmonary vascular resistance, in severe cases resulting in changes to the right ventricular function [3]. The development of pulmonary hypertension is significantly associated with increased mortality in patients with bronchiectasis [4,5]. Cardiopulmonary changes begin before symptoms appear [6]. Therefore in patient management, screening for suspected pulmonary hypertension and stratifying patients who require further testing are paramount [5].
The prevalence of pulmonary hypertension in bronchiectasis has not been well studied, because right heart catheterization, which is needed for a confirmatory diagnosis, is rarely performed [5]. According to the European Society of Cardiology (ESC)/European Respiratory Society (ERS) guidelines, right heart catheterization is recommended only for select patients with chronic lung disease, primarily due to its invasive nature and the limited evidence supporting approved treatments for group 3 pulmonary hypertension [7]. Physicians must integrate information from multiple modalities to identify patients who require referral to a specialized center to evaluate the need for a right heart catheterization [8].
Echocardiography has been suggested as a noninvasive modality to detect pulmonary hypertension in patients with chronic lung disease [9]. However, its accuracy in identifying pulmonary hypertension is limited, because peak tricuspid regurgitation velocity (TRVmax) can be inaccurate, leading to both overestimation and underestimation of systolic pulmonary artery pressure (sPAP) [7]. In addition, because echocardiography requires an experienced physician, screening all patients with bronchiectasis is impractical in clinical settings. However, since the gold standard for diagnosing bronchiectasis is high-resolution computed tomography (HRCT), screening patients for pulmonary hypertension using computed tomography (CT) parameters may be effective in real-world settings [5].
While the clinical impact of pulmonary hypertension on patients with bronchiectasis has been reported, so far as we know it only focuses on mortality outcomes. Understanding the relationship between pulmonary hypertension and hospitalization in patients with bronchiectasis is crucial, because hospitalization itself could be predictive of further exacerbations and mortality. Therefore in this paper, we attempt to find an association between pulmonary hypertension and hospitalization, and suggest reliable CT parameters for screening patients with bronchiectasis who are likely to develop pulmonary hypertension.
Materials and Methods
1. Study population and design
This was a single-center retrospective study of patients with bronchiectasis at the National Medical Center, Seoul, Korea. Patients diagnosed with bronchiectasis using HRCT between November 2012 and October 2022 were included in the study population. Patients without echocardiogram records during the study period were excluded from the study.
Patients were divided into two groups based on their mPA diameter. Those with mPA diameter greater than the 29 mm cutoff were considered to be at risk of pulmonary hypertension [6]. Hospital admission was defined as hospitalization due to bronchiectasis exacerbation in either an outpatient clinic or the emergency department. We adopted the 2017 consensus definition of bronchiectasis exacerbation, which requires deterioration in three or more key symptoms for at least 48 hours, and a clinician’s decision to change treatment [10].
The study protocol was approved by the Institutional Review Board of the National Medical Center (IRB No. NMC−2023−02−023) on February 23, 2023. The requirement for informed consent was waived.
2. Data collection
Baseline characteristics, such as age, sex, body mass index (BMI), smoking history, etiology of bronchiectasis, underlying pulmonary diseases, and comorbidities, were collected. The clinical characteristics of bronchiectasis were obtained from hospital records. HRCT images obtained during the study period were included in the analysis. If multiple scans were available for a patient, the image taken during symptom presentation was selected. Two clinicians independently interpreted the HRCT images, and the results were cross-checked for discrepancies. The main pulmonary artery (mPA) and ascending aorta diameters were determined by measuring the largest diameter at the bifurcation level (Supplementary Figure S1). Pulmonary artery to aorta (PA/A) ratio, with a cutoff value of 0.9, was obtained by dividing the diameter of the mPA by the diameter of the ascending aorta [8]. Echocardiographic parameters such as TRVmax, left ventricular ejection fraction, and estimated sPAP, were obtained from transthoracic echocardiography records. Pulmonary function test results were obtained from the most recent examination performed with the patient in a stable condition.
3. Statistical analysis
For baseline characteristics, Student’s t-test was performed for continuous variables, while the chi-squared or Fisher’s exact test was used for categorical variables. Logistic regression analysis was used in univariate and multivariate models to determine the prognostic factors for hospital admission.
Three models were constructed for multivariable analysis. HRCT and echocardiographic parameters that were significant in the univariate model were included. Multicollinearity among the variables was assessed, and if present, excluded. Model 1 included demographic and HRCT parameters; Model 2 included demographic and echocardiographic parameters; and Model 3 included demographic, HRCT, and echocardiographic parameters. The results are presented as odds ratios (OR) with 95% confidence intervals (CI). The accuracy of each model was evaluated by calculating the area under the curve (AUC) of the receiver operating characteristic (ROC) curve. A two-sided p<0.05 was considered statistically significant. All statistical analyses were performed using SPSS version 18.0 (SPSS Inc., Chicago, IL, USA).
Results
1. Baseline characteristics
Among the 415 patients diagnosed with bronchiectasis, 241 who did not undergo echocardiography and 44 who did not have measurable HRCT images during the study period were excluded, resulting in a final analysis of 130 patients. Of these 130 patients, 69 (53.1%) were admitted to the hospital due to exacerbations of bronchiectasis. Among those who required hospitalization, the mean mPA diameter was 30.3 mm (Supplementary Figure S2). Sixty-nine (53.1%) of the 130 patients were male, and the mean age was 66.6±11.2 years.
Table 1 summarizes the baseline characteristics of patients with mPA diameter ≤29 mm (n=69), and those with mPA diameter >29 mm (n=61). There were no significant differences between the two groups in terms of age, sex, BMI, etiology of bronchiectasis, other underlying pulmonary diseases, or comorbidities.
HRCT findings revealed a more severe form of bronchiectasis in patients with mPA diameter >29 mm, compared to those with mPA diameter ≤29 mm (Table 2). Patients with mPA diameter >29 mm tended to have a greater number of involved lobes, while a higher proportion of them had cavitary lesions in the lungs.
Echocardiographic parameters indicative of pulmonary hypertension, such as sPAP and TRVmax, were significantly higher in patients with mPA diameter >29 mm. The mean TRVmax value in this group was 2.7±0.7 m/sec.
Among patients with mPA diameter >29 mm, 29 (56.9%) had airflow limitation, defined as forced expiratory volume in 1 second (FEV1)/forced vital capacity (FVC) ratio <0.7. Both the mean FEV1/FVC ratio and mean FEV1 were significantly lower in this group, compared to those with mPA diameter ≤29 mm. Bronchodilators were more commonly used in patients with pulmonary artery enlargement. Although the diffusing capacity of the lungs for carbon monoxide (DLCO) tended to be lower in patients with mPA diameter >29 mm, the difference was not statistically significant.
Figure 1 shows the correlations between the mPA diameter observed on HRCT and echocardiographic parameters. The associations of mPA with sPAP (=0.71×mPA diameter+13.17; p=0.008) and TRVmax (=0.03×mPA diameter+1.79; p=0.019) were statistically significant. Supplementary Figure S3 shows the correlations between mPA diameter and PA/A ratio, as well as between sPAP and TRVmax, while Supplementary Figure S4 shows the correlations with DLCO.
2. Prognostic factors for hospital admission in univariate and multivariate analyses
In the univariate analysis, several factors were significantly associated with hospital admission in patients with bronchiectasis: previous exacerbation history, baseline modified Medical Research Council (mMRC) grade ≥2, larger mPA diameter, PA/A ratio >0.9, involvement of more than two lung lobes, the presence of cavities and nodules, FEV1/FVC <0.7, FEV1% <80, increased TRVmax, and higher sPAP (Table 3). Supplementary Figure S5 presents the associated ROC curves.
In the multivariate analysis of Model 1, mPA diameter >29 mm (adjusted OR [aOR], 2.47; 95% CI, 1.14 to 5.32), involvement of more than two lobes (aOR, 2.57; 95% CI, 1.14 to 5.77), and the presence of cavitary or nodular lesion (aOR, 5.15; 95% CI, 1.28 to 20.74) were significant prognostic factors for hospital admission. In Model 2, significant factors included TRVmax >2.8 m/sec (aOR, 2.49; 95% CI, 0.97 to 6.40), involvement of more than two lobes (aOR, 2.52; 95% CI, 1.10 to 5.78), and the presence of a cavitary or nodular lesion (aOR, 13.24; 95% CI, 1.61 to 109.02). In Model 3, the significant predictors were mPA diameter >29 mm (aOR, 1.98; 95% CI, 0.88 to 4.49), involvement of more than two lobes (aOR, 2.41; 95% CI, 1.04 to 5.58), the presence of cavitary or nodular lesion (OR, 15.16; 95% CI, 1.81 to 127.26), and TRVmax >2.8 (aOR, 2.12; 95% CI, 0.81 to 5.58). Supplementary Figure S6 summarizes the corresponding ROC curves, with AUC values of 0.718, 0.717, and 0.735, respectively.
Discussion
This study suggests that in patients with bronchiectasis, increased mPA diameter and the involvement of two or more lobes may be associated with a higher risk of hospitalization. Patients with mPA diameter greater than 29 mm, suggestive of pulmonary hypertension, tended to have a more severe form of the disease. These patients also showed more extensive bronchiectasis on CT, greater impairment in lung function, and significantly higher baseline mMRC grades. The mean TRVmax value in the group with pulmonary artery enlargement was comparable to the echocardiographic threshold suggested in the 2022 ESC/ERS guidelines [7].
Clinical characteristics of patients with pulmonary artery enlargement were consistent with parameters of the bronchiectasis severity index, such as higher Medical Research Council (MRC) score, greater radiographic extent, and decreased FEV1 [11]. More patients with pulmonary artery enlargement required supplementary oxygen, consistent with a previous study [12]. Notably, our model also found that the involvement of more than two lobes served as a valid cutoff. Other prognostic factors suggested in other studies to predict hospitalization were observed in the hospitalization group, including greater involvement of the lobes [13], low FEV1 [14], and concurrent chronic obstructive pulmonary disease or asthma (Supplementary Table S1) [15].
Severe exacerbation of bronchiectasis is a clinical sign of high disease activity, requiring greater attention and earlier intervention [16]. Hospitalization due to bronchiectasis exacerbation is associated with increased mortality [13]. Most of our study population had clinically significant disease, characterized by worsening bronchiectasis-related symptoms and the need for in-depth clinical evaluation. In this population, factors associated with exacerbations—such as a history of previous exacerbations and coexisting airway disease [17]—were more common in patients with pulmonary artery enlargement. Bronchodilator use, suggested as a treatment option for bronchiectasis patients with airflow limitation [18], was observed in more than half of the patients with mPA diameter >29 mm.
To the best of our knowledge, this is the first study to suggest prognostic CT parameters related to pulmonary hypertension for the hospitalization of patients with bronchiectasis. Given that HRCT is the gold standard for diagnosing bronchiectasis, all patients underwent CT at diagnosis [19]. Our findings suggest that CT scans may help detect subclinical changes in the cardiopulmonary system, stratify patients at high risk of future hospitalization, and identify those who require further evaluation. Compared to echocardiography, CT parameters such as mPA diameter and the number of involved lobes are relatively easy to assess, with results that are reproducible, and less dependent on the examiner’s experience [20].
In the multivariate analysis, increased mPA diameter appeared to be a significant risk factor for hospitalization. While the AUC values for Models 1−3 were not very high, AUC values ranging 0.7 to 0.8 are generally considered acceptable [21]. Model 1 (using CT data alone) was comparable to Model 2 (using TRVmax instead of mPA diameter) and Model 3 (which combined TRVmax with CT data). Model 1 did not show a significant decrease in predictive power, suggesting that CT data alone may be sufficient to evaluate clinically significant pulmonary hypertension in these patients.
Certain CT parameters have been studied to predict pulmonary hypertension and clinical outcomes. Studies by the Medical College of Wisconsin Lung Transplant group and the Framingham Heart Study reported the mPA diameter for prediction, with cutoff values of 29 mm overall, and 29 mm for men and 27 mm for women, respectively [6]. Devaraj et al. [5] suggested use of the average diameters of the right and left main pulmonary arteries [5], while Gao et al. [22] presented PA/A ratio to predict mortality. In our study, we selected the reference values from previous research utilizing mPA diameters to allow easily reproducible evaluation. The mean mPA diameter in the hospitalized group was 30.4 mm, which exceeded the cutoff value in earlier studies, and was found to be significant in both univariate and multivariate models.
The number of lung lobes involved was also associated with pulmonary hypertension and proved to be an effective predictor of clinical outcomes in patients with bronchiectasis. In a Turkish study, patients with bronchiectasis and pulmonary hypertension had more involved lung lobes on HRCT than those without pulmonary hypertension [4]. In a study of patients with bilateral bronchiectasis, mPAP >35 mm Hg was observed in 75% of the patients, and mPAP was inversely correlated with arterial oxygen saturation values [23]. Although the mechanisms of pulmonary hypertension in bronchiectasis have not been fully elucidated, our study suggests that airflow obstruction and parenchymal destruction caused by a greater extent of bronchiectasis contribute to hypoxemic changes and increased pressure in the pulmonary vessels.
It is notable that the presence of cavitary and nodular lesions in CT scans were significant factors to predict hospitalization in bronchiectasis patients. Both pulmonary tuberculosis and non-tuberculous mycobacterial (NTM) infection can cause bronchiectasis, and present cavities and nodular patterns [24]. In the case of NTM infection, a new onset disease can develop in lungs with bronchiectasis [25], and is well-established as one of the treatable causes of bronchiectasis [26,27]. Among the 39 patients with cavitary lesion, 23 (59.0%) had a history of tuberculosis. This could be explained by an intermediate tuberculosis burden in the Korean population. In our study population, five (3.8%) of 130 patients had NTM pulmonary disease, despite none of the patients receiving treatment.
This study has several limitations. First, echocardiography was performed only in patients presenting with moderate to severe bronchiectasis-related symptoms. As a result, the exclusion of 241 patients who did not undergo echocardiography may have introduced selection bias. Second, the timing of assessments was inconsistent between groups, and a time-to-event analysis was not feasible. A prospective study design incorporating baseline CT and serial CT follow-ups would be more suitable to address this limitation. Third, right heart catheterization was not performed to confirm the presence of pulmonary hypertension. Although none of the patients required catheterization to guide pulmonary hypertension-specific therapy, future studies may help determine the true prevalence of pulmonary hypertension in this population. Fourth, DLCO—an important parameter in the progression of pulmonary hypertension—was not routinely collected in the study population, and was therefore excluded from the multivariate analysis.
In conclusion, increased mPA diameter and extensive bronchiectasis on HRCT were identified as potential risk factors for hospitalization in bronchiectasis patients. CT parameters, such as mPA diameter, may be useful in screening for pulmonary hypertension and estimating the risk of hospital admission. These findings underscore the potential value of HRCT in identifying patients at higher risk of severe outcomes, and support its role in guiding patient management through targeted screening.
PDF Links
PubReader
ePub Link
XML Download
Data Sharing Statement
Full text via DOI
Supplement1
Print
Download Citation
